Intensive Care Units (ICUs): Clinical Outcomes, Costs, and ... · Recommended Citation: Berenson, R. A., Intensive Care Units (ICUs): Clinical Outcomes, Costs, and Decisionmaking

Intensive Care Units (ICUs): ClinicalOutcomes, Costs, and Decisionmaking

November 1984

NTIS order #PB85-145928

HEALTH TECHNOLOGY CASE STUDY 28

Intensive Care Units (ICUs)Clinical Outcomes, Costs,

and Decisionmaking

NOVEMBER 1984

This case study was performed as a part of OTA’S Assessment of

Medical Technology and Costs of the Medicare Program

Prepared under contract to OTA by:

Robert A. Berenson, M.D.

1

OTA Case Studies are documents containing information on a specific medical tech-nology or area of application that supplements formal OTA assessments. The materialis not normally of as immediate policy interest as that in an OTA Report, nor doesit present options for Congress to consider.

CONGRESS OF THE UNITED STATES

Otlke of Technology AssessmentWashington, D C 20510

Recommended Citation:Berenson, R. A., Intensive Care Units (ICUs): Clinical Outcomes, Costs, and Decisionmaking(Health Technology Case Study 28), prepared for the Office of Technology Assessment,U.S. Congress, OTA-HCS-28, Washington, DC, November 1984.

Library of Congress Catalog Card Number 84-601138

For sale by the Superintendent of DocumentsU.S. Government Printing Office, Washington, D.C. 20402

PrefaceIntensive Care Units (ICUs): Clinical Outcomes,

Costs, and Decisionmaking, is Case Study 28 inOTA’S Health Technology Case Study Series. Thiscase study has been prepared in connection withOTA’S project on Medical Technology and Costsof the Medicare Program, requested by the HouseCommittee on Energy and Commerce and its Sub-committee on Health and the Environment andthe Senate Committee on Finance, Subcommit-tee on Health. A listing of other case studies inthe series is included at the end of this preface.

OTA case studies are designed to fulfill twofunctions. The primary purpose is to provideOTA with specific information that can be usedin forming general conclusions regarding broaderpolicy issues. The first 19 cases in the Health Tech-nology Case Study Series, for example, were con-ducted in conjunction with OTA’S overall projecton The Implications of Cost-Effectiveness Anal-ysis of Medical Technology. By examining the 19cases as a group and looking for common prob-lems or strengths in the techniques of cost-effec-tiveness or cost-benefit analysis, OTA was ableto better analyze the potential contribution thatthose techniques might make to the managementof medical technology and health care costs andquality.

The second function of the case studies is toprovide useful information on the specific tech-nologies covered. The design and the funding lev-els of most of the case studies are such that theyshould be read primarily in the context of the as-sociated overall OTA projects. Nevertheless, inmany instances, the case studies do represent ex-tensive reviews of the literature on the efficacy,safety, and costs of the specific technologies andas such can stand on their own as a useful contri-bution to the field.

Case studies are prepared in some instances be-cause they have been specifically requested bycongressional committees and in others becausethey have been selected through an extensive re-view process involving OTA staff and consulta-tions with the congressional staffs, advisory panelto the associated overall project, the Health Pro-gram Advisory Committee, and other experts invarious fields. Selection criteria were developedto ensure that case studies provide the following:

● examples of types of technologies by func-

●

●

●

●

●

●

●

tion (preventive, diagnostic, therapeutic, andrehabilitative);examples of types of technologies by physicalnature (drugs, devices, and procedures);examples of technologies in different stagesof development and diffusion (new, emerg-ing, and established);examples from different areas of medicine(e.g., general medical practice, pediatrics,radiology, and surgery);examples addressing medical problems thatare important because of their high frequen-cy or significant impacts (e.g., cost);examples of technologies with associated highcosts either because of high volume (for low-cost technologies) or high individual costs;examples that could provide information ma-terial relating to the broader policy and meth-odological issues being examined in theparticular overall project; andexamples with sufficient scientific literature.

Case studies are either prepared by OTA staff,commissioned by OTA and performed under con-tract by experts (generally in academia), or writ-ten by OTA staff on the basis of contractors’papers.

OTA subjects each case study to an extensivereview process. Initial drafts of cases are reviewedby OTA staff and by members of the advisorypanel to the associated project. For commissionedcases, comments are provided to authors, alongwith OTA’S suggestions for revisions. Subsequentdrafts are sent by OTA to numerous experts forreview and comment. Each case is seen by at least30 reviewers, and sometimes by 80 or more out-side reviewers. These individuals may be fromrelevant Government agencies, professional so-cieties, consumer and public interest groups, med-ical practice, and academic medicine. Academi-cians such as economists, sociologists, decisionanalysts, biologists, and so forth, as appropriate,also review the cases.

Although cases are not statements of officialOTA position, the review process is designed tosatisfy OTA’S concern with each case study’sscientific quality and objectivity. During the vari-ous stages of the review and revision process,therefore, OTA encourages, and to the extentpossible requires, authors to present balanced in-formation and recognize divergent points of view.

,..111

Health Technology Case Study Seriesa

Case Study Case StudySeries Case study title; author(s); Series Case study title; author(s);number OTA publication numberb number OTA publication numberb

1

2

3

4

5

6

7

8

9

10

11

12

13

Formal Analysis, Policy Formulation, and End-StageRenal Disease;

Richard A. Rettig (OTA-BP-H-9(1))C

The Feasibility of Economic Evaluation ofDiagnostic Procedures: The Case of CT Scanning;

Judith L. Wagner (OTA-BP-H-9(2))Screening for Colon Cancer: A TechnologyAssessment;

David M. Eddy (OTA-BP-H-9(3))Cost Effectiveness of Automated MultichannelChemistry Analyzers;

Milton C. Weinstein and Laurie A. Pearlman(OTA-BP-H-9(4))

Periodontal Disease: Assessing the Effectiveness andCosts of the Keyes Technique;

Richard M Scheffler and Sheldon Rovin(OTA-BP-H-9(5))

The Cost Effectiveness of Bone Marrow TransplantTherapy and Its Policy Implications;

Stuart O. Schweitzer and C. C. Scalzi(OTA-Bp-H-9(6))

Allocating Costs and Benefits in Disease PreventionPrograms: An Application to Cervical CancerScreening;

Bryan R. Luce (Office of Technology Assessment)OTA-BP-H-9(7))

The Cost Effectiveness of Upper GastrointestinalEndoscopy;

Jonathan A. Showstack and Steven A. Schroeder(OTA-Bp-H-9(8))

The Artificial Heart: Cost, Risks, and Benefits;Deborah P. Lubeck and John P. Bunker(OTA-BP-H-9(9))

The Costs and Effectiveness of Neonatal IntensiveCare;

Peter Budetti, Peggy McManus, Nancy Barrand,and Lu Ann Heinen (OTA-BP-H-9(1O))

Benefit and Cost Analysis of Medical Interventions:The Case of Cimetidine and Peptic Ulcer Disease;

Harvey V. Fineberg and Laurie A. Pearlman(OTA-BP-H-9(11))

Assessing Selected Respiratory Therapy Modalities:Trends and Relative Costs in the Washington, D.C.Area;

Richard M. Scheffler and Morgan Delaney(OTA-BP-H-9(12))

Cardiac Radionuclide Imaging and CostEffectiveness;

William B. Stason and Eric Fortess(OTA-BP-H-9(13))

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

Cost Benefit/Cost Effectiveness of MedicalTechnologies: A Case Study of Orthopedic JointImplants;

Judith D. Bentkover and Philip G. Drew(OTA-BP-H-9(14))

Elective Hysterectomy: Costs, Risks, and Benefits;Carol Korenbrot, Ann B. Flood, Michael Higgins,Noralou Roos, and John P. Bunker(OTA-BP-H-9(15))

The Costs and Effectiveness of Nurse Practitioners;Lauren LeRoy and Sharon Solkowitz(OTA-BP-H-9(16))

Surgery for Breast Cancer;Karen Schachter Weingrod and Duncan Neuhauser(OTA-BP-H-9(17))

The Efficacy and Cost Effectiveness ofPsychotherapy;

Leonard Saxe (Office of Technology Assessment)(OTA-BP-H-9(18))d

Assessment of Four Common X-Ray Procedures;Judith L. Wagner (OTA-BP-H-9(19))e

Mandatory Passive Restraint Systems inAutomobiles: Issues and Evidence;

Kenneth E. Warner (OTA-BP-H-15(20))f

Selected Telecommunications Devices for Hearing-Impaired Persons;

Virginia W. Stern and Martha Ross Redden(OTA-BP-H-16(21))g

The Effectiveness and Costs of AlcoholismTreatment;

Leonard Saxe, Denise Dougherty, Katharine Esty,and Michelle Fine (OTA-HCS-22)

The Safety, Efficacy, and Cost Effectiveness ofTherapeutic Apheresis;

John C. Langenbrunner (Office of TechnologyAssessment) (OTA-HCS-23)

Variation in Length of Hospital Stay: TheirRelationship to Health Outcomes;

Mark R, Chassin (OTA-HCS-24)Technology and Learning Disabilities;

Candis Cousins and Leonard Duhl (OTA-HCS-25)Assistive Devices for Severe Speech Impairments;

Judith Randal (Office of Technology Assessment)(OTA-HCS-26)

Nuclear Magnetic Resonance Imaging Technology:A Clinical, Industrial, and Policy Analysis;

Earl P. Steinberg and Alan Cohen (OTA-HCS-27)Intensive Care Units (ICUs): Clinical Outcomes,Costs, and Decisionmaking;

Robert A. Berenson (OTA-HCS-28)available for sale by the Superintendent of Documents, U.S. Government dBackground pavr #3 to The Implications of Cost-Effectiveness Analysis of

Printing Office, Washington, D. C., 20402, and by the National Technical Medical Technology.Information Service, 5285 Port Royal Road, Springfield, Va., 22161. Call ‘Background Paper #5 to The implications of Cost-Effectiveness Analysis ofOTA’S Publishing Office (224-8996) for availability and ordering infor- Medical Technology.

IBackground paper #1 to OTA’S May 1982 report Technology and lfandi-mation.boriginal publication numbers appear in Parenth=s. capped People.‘The first 17 cases in the series were 17 separately issued cases in Background gBackground paper #2 to Technology and Handicapped People.Paper ,#2: Case Studies of Medical Technologies, prepared in conjunctionwith OTA’S August 1980 report The Implications of Cost-Effectiveness Anal-ysis of Medical Technology.

iv

OTA Project Staff for Case Study #28

H. David Bantal and Roger C. Herdman,2 Assistant Director, 0TAHealth and Life Sciences Division

Clyde J. Behney, Health Program Manager

Anne Kesselman Burns, Project Director

Pamela Simerly, Research Assistant

Virginia Cwalina, Administrative Assistant

Beckie I. Erickson,3 Secretary/Word Processing Specialist

Brenda Miller, PC Specialist

Jennifer Nelson,3 Secretary

Mary Walls,’ Secretary

1Until August 1983.‘Since December 1983.3Since January 1984.‘Until January 1984.

Medical Technology

— —

and Costs of the Medicare Program Advisory Panel

Stuart Altman, Panel ChairDean, Florence Heller School, Brandeis University

Frank BakerVice PresidentWashington State Hospital Association

Robert BlendonSenior Vice PresidentThe Robert Wood Johnson Foundation

Jerry CromwellPresidentHealth Economics ResearchChestnut Hill, MA

Karen DavisChair, Department of Health Policy and

ManagementSchool of Hygiene and Public HealthJohns Hopkins University

Robert DerzonVice PresidentLewin & AssociatesSan Francisco, CA

Howard FrazierDirectorCenter for the Analysis of Health PracticesHarvard School of Public Health

Clifton GausPresident, Foundation for Health Services

ResearchWashington, DC

Jack HadleyDirectorCenter for Health Policy StudiesGeorgetown University

Kate IrelandChair, Board of GovernorsFrontier Nursing ServiceWendover, KY

Judith LaveProfessorDepartment of Health EconomicsUniversity of Pittsburgh

Mary MarshallDelegateVirginia House of Delegates

Walter McNerneyProfessor of Health PolicyJ. L. Kellogg Graduate School of ManagementNorthwestern University

Morton MillerImmediate Past PresidentNational Health CouncilNew York, NY

James MorganExecutive DirectorTruman Medical CenterKansas City, MO

Seymour PerryDeputy DirectorInstitute for Health Policy AnalysisGeorgetown University Medical Center

Robert SigmondDirector, Community Programs for

Affordable Health CareAdvisor on Hospital Affairs Blue Cross/Blue

Shield Association

Anne SomersProfessorDepartment of Environmental and

Community and Family MedicineUMDNJ—Rutgers Medical School

Paul TorrensSchool of Public HealthUniversity of California, Los Angeles

Keith WeikelGroup Vice PresidentAMIMcLean, VA

vi

Contents

ChapterCHAPTER 1: INTRODUCTION ANDIntroduction . . . . . . . . . . . . . . . . . . . . . . .Executive Summary . . . . . . . . . . . . . . . . .

Utilization oflCUs . . . . . . . . . . . . . . .Outcomes of lntensive Care . . . . . . .Payment for ICU Services . . . . . . . . .Decision making in the ICU . . . . . . . .Foregoing Life-Sustaining TreatmentPossible Future Steps, . . . . . . . . . . . . .

EXECUTIVE. . . . . . .. . . . . . .. . . . . . .. . . . . . .. . . . . . .. . . . . ,.. . . . . . .. . . . . ,.

CHAPTER 2: EVOLUTION, DISTRIBUTION,OF INTENSIVE CARE UNITS . . . . . . . . . . . .

The Development of the ICU . . . . . . . . . . . . . . .Advantages and Disadvantages of ICU Care..Definitions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Requirements of an ICU . . . . . . . . . . . . . . . . . . . .Specialty. Multispecialty ICUs. . . . . . . . . . . . .Distribution of ICU Beds . . . . . . . . . . . . . . . . . . .Expansion of ICU Beds . . . . . . . . . . . . . . . . . . . . .Regulation of ICUs . . . . . . . . . . . . . . . . . . . . . . . .

CHAPTER 3: COST OF ICU CARE . . . . . . . . .Components of ICU Costs . . . . . . . . . . . . . . . . . .Costs of an ICU Day . . . . . . . . . . . . . . . . . . . . . .Total National Costs of Intensive Care.. . . . . .

CHAPTER 4: UTILIZATION OF ICES. . . . . . .Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Utilization by Type of ICU . . . . . . . . . . . . . . . . .ICU Admission Rates . . . . . . . . . . . . . . . . . . . . . .Sex and Age Distribution of ICU Use . . . . . . . .ICU Case Mix. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Diagnoses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Other Case Mix Parameters . . . . . . . . . . . . . . .Readmission . . . . . . . . . . . . . . . . . . . . . . . . . . .Length of Stay . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

SUMMARY. . . . . . . . . . .. . . . . . . . . . .. . . . . . . . . . .. . . . . . . . . . .. . . . . . . . . . .. . . . . . . . . . .. . . . . . . . . . .. . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .

AND REGULATION. . . . . . . . . . . . . . .. . . . . . . . . . . . . . .. . . . . . . . . . . . . . .. . . . . . . . . . . . . . .. . . . . . . . . . . . . . .. . . . . . . . . . . . . . .. . . . . . . . . . . . . . .. . . . . . . . . . . . . . .. . . . . . . . . . . . . . .

. . . . . . . . . . . ...$

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .. . . . . . . . . . . . . . .

CHAPTER 5: OUTCOMES OF INTENSIVE CARE: MEDICALAND COST EFFECTIVENESS . . . . . . . . . . . . . . . . . . . . . . . . . . .

Difficulties in Assessing Effectiveness. . . . . . . . . . . . . . . . . . . . . . . .Clinical Outcomes of ICU Care . . . . . . . . . . . . . . . . . . . . . . . . . . . .Functional Outcome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Characteristics of ICU Nonsurvivors . . . . . . . . . . . . . . . . . . . . . . . .

Age . . . . . . . . . . . . . . . . . . . . . . . . .Severity of Illness . . . . . . . . . . . .Resource Use . . . . . . . . . . . . . . . .

Distribution of ICU Costs AmongMonitored Patients . . . . . . . . . . . . .Adverse Outcomes of ICU Care..

Iatrogenic Illness . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . .Patients . . . . . . . . . . . . . . . . . ....)., .!$.... . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .

BENEFITS. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .

334556677

111112131414151617

21212122

25252525272828282929

333334353535363637383939

Contents—continued

Nosocomial Infections. . . . . . . . . . . . . . . . .Psychological Reactions . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Cost-Effectiveness Analysis of Adult Intensive Care. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

CHAPTER 6: PAYMENT FOR ICU SERVICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Traditional Hospital Reimbursement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Patient Copayments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Utilization Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Prospective Payment Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Medicare’s Current Inpatient Hospital Payment System . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Medicare Utilization of ICUs by DRGs.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Applicability of DRGs to Ices. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Physician Payment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

CHAPTER 7: THE ICU TREATMENT IMPERATIVE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .The Highly Technological Nature of ICU Care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .The Nature of ICU Illnesses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Traditional Moral Distinctions in Medicine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .The Diffusion of Decision making Responsibility ........,.. . . . . . . . . . . . . . . . . . . . . . . . .Problems of Informed Consent in the ICU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Legal Pressures: Defensive Medicine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Payment and the Treatment Imperative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .The Absence of Clinical Predictors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

CHAPTER 8: FOREGOING LIFE-SUSTAINING TREATMENT . . . . . . . . . . . . . . . . . . . . .Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .The Natural Process of Death . . . . . . . . . . . . . . . . . . . . . . . . .Fundamental Ethical, Moral, and Legal Considerations. . .Procedures for Review of Decisionmaking . . . . . . . . . . . . . .Rationing ICU Care.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Explicit or Implicit Rationing of ICU Care? . . . . . . . . . . . . .

Explicit Rationing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Implicit Rationing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

CHAPTER 9: CONCLUSIONS AND POSSIBLE FUTURE STEPS . . . . . . . . . . . . . . . . . . .

APPENDIX A: ACKNOWLEDGMENTS AND HEALTH PROGRAMADVISORY COMMITTEE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

APPENDIX B: COST ESTIMATES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Page

404041

45454545464646484851

55555556575859606263

676767687071727273

77

8184

89

Tables

Table No. Page

1. Distribution of ICU Beds in Short-Term, Non-Federal Hospitals,by Size of Hospital 1982 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

2. ICU/CCU Beds as Percent of Total Beds by Hospital Size for Short-TermNon-Federal Hospitals, 1982 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

3. Distribution of ICU and CCU Beds, by Region, 1981 . . . . . . . . . . . . . . . . . . . . . . . . . . . 164. Percentage of ICU/CCU Beds in Short-Term Hospitals,

by Hospital Sponsorship, 1976 and 1982 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165. Summary of Selected ICU Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

Contents—continued

6.

7.

8.

9.10,

Page

Use and Percentage of Hospital Charges Incurred in ICUs and CCUs for MedicareBeneficiaries Discharged From Short-Stay Hospitals, 1979 . . . . . . . . . . . . . . . . . . . . . . . 26Use and Percentage of Hospital Charges Incurred in ICUs and CCUs for MedicareBeneficiaries Discharged From Short-Stay Hospitals, by Geographic Region, 1979.. 27Use and Percentage of Hospital Charges Incurred in ICUs and CCUs for MedicareBeneficiaries Discharged From Short-Stay Hospitals, by Age, 1980 . . . . . . . . . . . . . . . 27Retrospective Outcome Studies of ICU Care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35Estimated Number of Special Care Days by Primary Diagnosis Basedon HCFA 20-Percent Sample of Medicare Discharges, 1980. . . . . . . . . . . . . . . . . . . . . . 49

——

Glossary of Acronyms

AHA –APACHE –

CBA –CCC –CEA –CON –D H H S –

DRG –HCFA –

HMO –ICU –

American Hospital AssociationAcute Physiology and Chronic HealthEvaluation Scalecost-benefit analysiscoronary care unitcost-effectiveness analysiscertificate-of-needU.S. Department of Health andHuman ServicesDiagnosis Related GroupHealth Care Financing Administration(U.S. Department of Health andHuman Services)health maintenance organizationintensive care unit

IvLOSNIH

ORPSRO

SCUTISS

UCR

UR

— intravenous– length of stay– National Institutes of Health (U.S.

Department of Health and HumanServices)

— operating room— Professional Standards Review

Organization— special care unit— Therapeutic Intervention Scoring

System— usual, customary and reasonable

physician charges for paymentpurposes

— utilization reviewIRB – institutional review board

OTA Note

These case studies are authored works commissioned by OTA. Each authoris responsible for the conclusions of specific case studies. These cases are not state-ments of official OTA position. OTA does not make recommendations or endorseparticular technologies. During the various stages of review and revision, therefore,OTA encouraged the authors to present balanced information and to recognizedivergent points of view.

1 .

Introduction andExecutive Summary

“The patient’s recovery will be watched not only by nurses but by electric eyestoo. Sensing devices will constantly monitor his heart rate, his temperature, his respi-ration rate, his electrocardiogram, and the blood pressure both in his veins and in hisarteries. The nurses will not rouse the patient early in the morning to poke a glassthermometer between his gums and then spend much of the day checking up on hisand the other patients’ conditions. They will simply push a button at the console oftheir station to get as many readouts as they want. The patient will not have to hopethat if he enters a crisis somebody may spot it.

If any single bodily function or combination of functions deviates beyond the fixedlimits the patient’s Physician has programmed into a computer, lights will flash anda buzzer will sound the-alarmplete array of equipment will

“Physicians tend to be

Within seconds, nurses, technicians, doctors, and-be in action at his bedside.”

–Life Magazine, December 2, 1966unun

unimpressed with the published descriptions of units andtheir working. It often seems to them that the assessment of the results is naive, sur-vival being taken as equivalent of a life saved. They suspect that, however expert thehandling of the apparatus, there is often a shallow understanding of the disease andan over-readiness to employ the most dramatic treatment;. . . One is tempted to saythat treatment is often more intense than careful . . .

I believe, therefore, with many of my colleagues, that the attempt to segregateall medical emergencies on a basis of apparatus need will prove to have been anaberration.”

—Professor A. C. Dornhorst, April 1, 1966

1

Introduction and Executive Summary

INTRODUCTION

Intensive care units (ICUs) exemplify the bestthat American medicine has to offer—teams ofdedicated professionals using the latest technol-ogy to save lives that in the past would havealmost surely been lost. Formally developed onlyin the late 1950s, ICUs are present in almost 80percent of hospitals in the United States. They areestimated to consume between 15 and 20 percentof the Nation’s hospital budget, or almost 1 per-cent of the gross national product. Yet, despitesuch large expenditures of public and privateresources, there has been remarkably little criticalevaluation of the effectiveness of ICU care byeither the public or the medical profession.

In recent years, however, there has been grow-ing public and professional awareness of the emo-tional torment suffered by the patients and theirfamilies related to the use of “lifesaving” medicalcare which does not really benefit the patient.Correspondingly, there has been increasing sup-port for the notion that patients have the rightto reject measures that will prolong their liveswithout improving their condition.

Along with the increasing public recognitionthat there are times when extraordinary medicalcare should not be employed, three key develop-ments have made this an opportune time to ana-lyze the costs and benefits of ICU care. First, thePresident’s Commission for the Study of EthicalProblems in Medicine and Biomedical and Behav-ioral Research issued a comprehensive report inMarch 1983 on the medical, ethical, and legalissues underlying decisions on whether to foregolife-sustaining treatment for seriously ill patients(191), The recommendations of the expert com-mission have direct bearing on decisionmaking formany ICU patients.

Also in March 1983, a Consensus DevelopmentConference sponsored by the National Institutesof Health (NIH) formally evaluated the efficacy

and appropriatenessthe first time (176).

of critical care medicine]

The Conference Reportforex-

amines the evidence for efficacy of critical care.medicine for various clinical problems and pro-vides recommendations for organization andadministration of ICUs.

Finally, in April 1983, Congress enacted a pro-spective payment system for Medicare in theSocial Security Act Amendments of 1983 (PublicLaw 98-21). This new payment system, whichbegan to be phased in over a 3-year period begin-ning in October 1983, will dramatically alter pay-ment for services provided in ICUs by placing alimit on the amount of reimbursement availablefor different categories of illnesses. These limitsmay have a significant impact on the servicesavailable for critically ill patients.

This case study has two purposes. The first isto present what is currently known about ICUsin terms of the distribution of ICU beds, the costsof maintaining ICUs, the utilization of ICUs, thecharacteristics of ICU patients, and the outcomeof ICU care. There are still important gaps in thedata, but a substantial body of knowledge existsabout the technical aspects of ICU care. The ICUis examined as a discrete medical technology.

The second purpose of the study is to establisha framework for considering some of the clinical,moral, and legal issues that arise with respect toICU care. The study explores, for example, thefactors unique to the ICU that sometimes leadphysicians to continue life-support for patientswho have minimal hope of improving. It discussesways in which patients can make known theirwishes about foregoing or discontinuing life-support if their condition deteriorates and howphysicians and family members can decide wheth-er to terminate life-support when the patient isnot capable of making such a decision. It also con-

IThis case study defines both “’intensive care” and “critical care”as care provided in separate hospital units generally known as “in-tensive care units. ” See ch. 2 for a discussion of definitions.

3

4 . Health Case Study 28: intensive Care Units: Costs, Outcome, and Decisionmaking

siders how ICU treatment might be rationed inthe future if it becomes necessary to do so.

As is shown in the review of data on costs andbenefits of ICU care, the ICU is often an effec-tive, lifesaving technology. However, it is effec-tive at a high cost. Indeed, partially because ofits success in many clinical situations, it will notbe easy to simply find and eliminate the “waste”in ICUs. Changing the economic incentives forprovision of ICU care, as under Medicare’s newhospital payment system, has not made it anyeasier for patients, families, and ICU staffs whofrequently face difficult decisions about how ag-gressively to treat individual patients. Indeed, asthe case study explores, the new prospective pay-ment system may make ICU decisionmaking evenmore difficult and contentious than in the past.

EXECUTIVE SUMMARY

The ICU has been called the hallmark of themodern hospital but has come into existence onlyover the last 25 years. Initially, the ICU was anexpansion of the surgical recovery room and wassubsequently an outgrowth of the respiratory careunits made possible by the development of the me-chanical ventilator.

Today, almost 80 percent of short-term generalhospitals have at least one ICU. Overall, 5.9 per-cent of total hospital beds in non-Federal, short-term community hospitals in 1982 were beds inICU and coronary care units (CCUS). Beds inother types of special care units, includingpediatric, neonatal, and burn units, add another1 percent to the total complement of special carebeds.

ICU beds are reasonably evenly distributedamong all sizes of hospitals, regions of the coun-try, and types of hospital sponsorship. Over thelast 6 years, the number of ICU beds has risenabouts percent a year, compared to a rise of gen-eral hospital beds of only 1 percent a year. A ma-jor rise of ICU beds occurred between 1979 and1981, particularly in hospitals of greater than 500beds. Federal and State policy, particularly cer-tificate-of-need laws and Medicare reimbursement

The case study focuses on adult ICUS and notneonatal, burn, or cardiac units. While some ofthe issues raised here are applicable to these otherspecialized care units, these other units generallypresent different clinical, ethical, and public pol-icy issues. Certainly, all units treat seriously illpatients. However, the moral, ethical, and legalproblems raised by withholding care for seriouslyhandicapped newborns, for example, differ fromthe problems raised by withholding care for anelderly person with a terminal condition. Theissues related to treatment of such infants, whichhas been the center of the recent “Baby Doe” con-troversy, deserve separate attention. Likewise, asthe study emphasizes, coronary care patients areclinically different from general intensive carepatients.

policy until 1982, probably contributed to thecontinued expansion of ICU beds and ICU utili-zation.

For a number of technical and conceptualreasons, an accurate estimate of the cost of ICUcare is difficult to make. For example, there isdisagreement on whether consideration of ICUcosts should include the room and board costs ofICU care only, the room and board and ancillarycare costs of patients while in the ICU, or the in-cremental costs of ICUs above that which the hos-pital would have to bear in any case for seriouslyill patients. The national average per diem chargein 1982 of an ICU bed was $408 compared to aregular bed per diem charge of $167, a ratio ofabout 2.5:1. However, it is likely that the true costratio is closer to 3-3.s:1. In addition, ICU patientsconsume a greater proportion of ancillary serv-ices, particularly laboratory and pharmacy serv-ices, than regular floor patients.

Based on these and other considerations, it isestimated that the costs of adult ICU and CCUcare—the cost to the hospital patients while theyare in the special care unit—represents about 14to 17 percent of total inpatient, community hos-pital costs, or $13 billion to $15 billion in 1982.

Ch. l–Introduction and Executive Summary ● 5

Inclusion of the other types of specialized and Fed-eral hospital ICUs would bring the percentage upto about 20 percent.

Utilization of ICUs

According to 1979 Medicare data, 18 percentof Medicare discharges included a stay in inten-sive care (including coronary care) in that year.Unfortunately, similar data are not available forthe entire population. From reports from individ-ual hospitals, however, certain general utilizationpatterns do emerge (these reports are weightedtowards large and teaching hospitals). The rep-resentation of the elderly in ICUs seems to be thesame or slightly more than in the hospital as awhole. Poor chronic health status, rather thanage, appears to be a predominant factor limitinguse of ICUs in individual cases in the UnitedStates. In comparison to the United States, ICUpatients in other countries have a significantlylower mean age.

There is no accepted classification scheme thatdescribes the clinical characteristics of ICU pa-tients, largely because ICU patients are a hetero-geneous population who have multiple underlyingmedical problems and who exhibit varying phys-iologic disturbances. ICU patients range fromthose who are in the ICU primarily for monitor-ing for potential disturbances to those who arecritically ill and receive life-supporting treatmentand continuous intensive nursing and physiciancare.

Outcomes of Intensive Care

Unfortunately, it is difficult to separate the in-tensity of care from the setting in which it is pro-vided, and therefore, to know whether intensivecare would be as effective if provided on the gen-eral hospital floor as in the physically and ad-ministratively separate ICU. Many believe thatrandomized clinical trials of ICUs, at least forunstable patients, are currently unethical, becauseICU care has become the accepted and standardmode of treatment in the United States for mostseverely ill and injured patients.

A recent NIH-sponsored consensus panel foundthat it is impossible to generalize about whether

ICU care improves outcome for the varied ICUpatient population. The panel felt that ICU in-tervention is unequivocally lifesaving for someconditions, particularly where there is an acute,reversible disease such as drug overdose or ma-jor trauma. There is less certainty about the ef-fectiveness of ICU care in other conditions, par-ticularly in the presence of a severe, debilitatingchronic illness, such as cancer or cirrhosis of theliver. Investigators believe that underlying diseaseis probably the most significant predictor of theoutcome of ICU care, although patient age andseverity of illness are also important.

Recent data have emphasized the inverse rela-tionship between the cost of ICU care and sur-vival. At this time, however, there are no acceptedmethods for determining ahead of time which pa-tients will benefit from additional ICU care. Froma number of studies, it is clear that the sickest ICUpatients, many of whom do not survive, consumea highly disproportionate share of ICU charges.Two recent studies, for example, found that 17and 18 percent, respectively, of the ICU patientpopulation generated half of the ICU charges.Moreover, charges do not account for the substan-tial cross-subsidization of costs between ICU pa-tients. It is likely, then, that the true proportionof costs consumed by the sickest ICU patients aresubstantially greater than even the charge datasuggest.

At the other end of the ICU patient spectrumare patients in the ICU primarily for monitoringof the development of a life-threatening complica-tion. Some of these patients may be able to becared for safely and more cost effectively outsideof the ICU, either in intermediate care units oron regular medical floors. On the other hand,there may be a population of ICU patients whoare discharged prematurely from ICUs. Researchhas only recently begun to better define which pa-tients should be routinely monitored in an ICUand which would do as well or even better if caredfor on other floors in the hospital.

Another consideration in deciding whether apatient should be cared for in the ICU is the realityof adverse effects of ICU care, so-called iatrogenicillness. A list of major iatrogenic complicationsof prolonged ICU care has been identified. Noso-

—. —

6 ● Health Case Study 28: Intensive Care Units: Costs, outcome, and Decisionmaking

comial infections—i. e., infections that were notpresent or incubating at the time of hospital ad-mission—and various serious psychological re-actions are particular complications of ICU care.

Payment for ICU Services

To the extent that insurers distinguish ICU carefrom other hospital care for purposes of payment,the result has been to reward ICU care relativeto care in intermediate level special care units oron general floors of the hospital. For example, in1980, Medicare tightened the existing paymentlimits on routine bed costs but not on ICU bedcosts—the so-called “section 223 limits. ” Further-more, utilization review efforts generally have notconsidered the appropriate level of care within thehospital.

Medicare’s inpatient hospital payment policies,however, have now changed dramatically as a re-sult of the passage of the Social Security ActAmendments of 1983 (Public Law 98-21). Underthe relatively new system, hospitals receive a fixedpayment per discharge based on the patient’s prin-cipal diagnosis. The classification system, whichidentifies 467 different clinical conditions calleddiagnosis-related groups (DRGs) appears ill-suitedfor describing certain types of patients cared forin ICUs. DRG payments are based largely on asingle diagnosis. Yet, ICU patients often havemultiple serious underlying illnesses. For these pa-tients, designation of a single, principal diagno-sis is likely to be arbitrary, and the resources useddue to the presence of additional diagnoses wouldnot be accounted for.

In addition, the DRG scheme does not take se-verity of illness into account. For some diagno-ses, particularly noncardiac medical conditions,the DRG category does not reflect the use of ICUsfor the more severely ill patients with that prin-cipal diagnosis. For example, only 3.5 percent ofthe average total hospital stay for Medicare pa-tients with cirrhosis (DRG 202) represent ICUdays. Yet, the sickest patients with cirrhosis areamong the highest cost ICU patients.

Furthermore, the outlier policy that the HealthCare Financing Administration has implementedpays hospitals less than the marginal costs of car-

ing for the sickest ICU patients. In short, it ap-pears that under Medicare’s DRG payment sys-tem, the sicker ICU patients will be substantialfinancial “losers” to the hospital.

Decisionmaking in the ICU

The new incentives of the DRG payment systemmay conflict with an ICU decisionmaking envi-ronment in many hospitals in which the cost ofcare has been of minor concern in the past. In-deed, a number of factors, some of which aresomewhat unique to the ICU, have led to a deci-sionmaking process that often has led physiciansto provide life-support care in the ICU after theinitial rationale for doing so no longer exists. Fac-tors that have created an ICU treatment im-perative include:

●

●

●

●

●

●

●

The highly technological nature of ICU care,which often results in focus on the technicaldetails of treatment rather than the rationalefor continued treatment.The nature of ICU illnesses, which often re-quire “technologically oriented” treatmenteven when the primary intent is to providecomfort rather than cure to a desperately illpatient.Traditional moral distinctions in medicinethat in some cases result in more care thanthe patient would choose if able to do so.Diffusion of decisionmaking responsibility,especially in relation to decisions to foregoor terminate life-support.Problems of informed consent in the ICUwhere many patients are temporarily or per-manently incompetent.The practice of defensive medicine by physi-cians, which involves taking or not takingcertain actions more as a defense against po-tential legal actions than for the patient’s ben-efit. Defensive medicine may be a particularproblem in the ICU, because of the life-and-death nature of ICU care, the relative visi-bility of ICU decisions, and great uncertaintyabout likely court decisions on these kindsof cases.A payment environment which, until 1982,provided financial rewards to hospitals andphysicians for provision of ICU care. Physi-

Ch. I—Introduction and Executive Summary ● 7

cian payment methods continue to pay gen-erously for the procedure-oriented ICU care.

● The absence of a data base for the commonICU conditions on which to make reliableclinical predictions of individual ICU pa-tients’ chances of immediate and long-termsurvival.

Foregoing Life=Sustaining Treatment

The Critical Care Consensus DevelopmentConference sponsored by NIH has concluded thatit is not appropriate to devote limited ICUresources to patients without reasonable prospectof significant recovery or to simply prolong thenatural process of death.

In general, a terminally ill patient’s right toforego or discontinue life-sustaining treatment hasbeen established and is usually protected by theconstitutional right to privacy. Practical dif-ficulties arise when the patient is not competentto decide, and when other decisionmakers, in-cluding physicians, families; and patient sur-rogates, do not agree on what medical treatmentto pursue. State courts have differed on the deci-sionmaking procedures to use when a patient isnot able to choose for himself.

Recent court decisions differ even over whena patient is considered “terminal” and over whatconstitutes “medical” treatment. Likewise, manycourts have continued to invoke a distinction be-tween ordinary and extraordinary care, whilesome have explicitly rejected the distinction.

Possible Future Steps

Because of the increasing burden of medical carecosts on individuals and on society as a whole,it is likely that the funds available for intensivecare will be much more strictly limited than inthe past. Because Medicare’s DRG payment sys-tem in general makes many ICU Medicare patientsfinancial losers for the hospital it may, therefore,alter the prevailing provider attitudes about theappropriateness and extent of ICU care in indi-vidual situations.

In recent years, the number of ICU beds hasexpanded to meet increased demand for beds, ex-cept in public hospitals in financial distress or at

times when there was a shortage of ICU nursesto staff available beds. In the future, there willneed to be greater attention paid to how to ra-tion ICU beds. The DRG system used by Medicareis a form of “implicit” rationing, because the pay-ment limitations place greater pressures on physi-cians and hospitals to make resource allocationchoices without setting “explicit” limitations onservices or eligible patients. Under this form ofrationing, there will be a need to consider expand-ing the procedural safeguards used on behalf ofpatients who become major financial losers forthe hospital. ICU decisionmaking will becomeeven more difficult than it has been in the pastdue to potential financial conflict between pa-tients, physicians, and hospitals.

A number of steps might improve the environ-ment for intensive care decisionmaking:

Research on developing accurate predictorsof survival for patients with acute andchronic illnesses could be expanded in orderto permit better informed decisions based onthe likelihood of short- and long-term sur-vival. In the absence of valid and reliabledata, hospitals could consider formalizing aninstitutional prognosis committee whosefunction would be to advise physicians, fam-ilies, and patients on the likelihood of sur-vival with ICU care.The suitability of the current DRG methodof payment for ICUs should be tested andmodified if necessary to take sufficient ac-count of severity of illness.The legal system may need to recognize thepossible conflict between malpractice stand-ards which assume quality of care that meetsnational expert criteria, and a decisionmak-ing environment in which resources may beseverely limited.Health professionals who are involved in de-cisionmaking on critically ill patients mightbenefit from more education in medical ethicsand relevant legal procedures and obli-gations.The actual decisionmaking process for criti-cally ill patients may need greater attention.For example, hospitals might explore formal-izing decisionmaking committees to lessen the

8 . Health Case Study 28: ]ntensive Care Units: Costs, Outcome, and Decisionmaking

burden on individuals faced with difficult through formal hospital committees, throughchoices about terminating life-support. More government-imposed procedures which cangenerally, society will need to decide how it follow fixed rules and regulations, or other,wishes conflicts over decisions on terminating perhaps more decentralized, mechanisms.life-support to be resolved—i.e., in courts,

2.Evolution, Distribution, and

Regulation of Intensive Care Units

2 ■

Evolution, Distribution, andRegulation of Intensive Care Units

THE DEVELOPMENT OF THE ICU

The intensive care unit (ICU) has been calledthe hallmark of the modern hospital (205), yet itis a recent development, having come into ex-istence only in the last 25 years. The developmentof ICUs was preceded by the rapid growth of post-operative recovery rooms (115) following WorldWar II. As early as 1863, however, FlorenceNightingale had foreseen the utility of a separatearea for observing patients recovering from theimmediate effects of surgery (172).

To a large extent, the initial stimulus for a sep-arate recovery area for specialized care was amanagerial response to overwhelming medicaldemands. The Massachusetts General Hospital,for example, when suddenly faced with treating39 survivors of the Boston Coconut Grove Firein 1942, set up a makeshift “burn unit” which itmaintained for 15 days, until the majority of pa-tients had been sent home (115). In the NorthAfrican and Italian campaigns of World War II,shock wards were established to resuscitate bat-tlefield casualties and to care for injured soldiersbefore and after surgery (115). After the war, anacute shortage of nurses provided much of the im-petus for the spread of recovery rooms in theUnited States.

Although recovery rooms were established ini-tially as a means of managing large numbers ofpatients more efficiently, the medical benefits ofbetter postoperative nursing care soon becameapparent, and recovery rooms flourished. In 1951,only 21 percent of community hospitals had re-covery rooms; a decade later, virtually all hospi-tals had them (205).

During the 1950s, using the recovery room asa model, a few ICUs began appearing on bothsides of the Atlantic. An early version of whathas become known as a respiratory ICU, for ex-ample, was set up in Denmark during the 1952polio epidemic in Scandinavia. After 27 of 31 pa-tients suffering from respiratory or pharyngeal

paralysis at Copenhagen’s Blegdam Hospital died,the hospital’s senior anesthetist performed a tra-cheotomy on a 12-year-old girl and inserted acuffed endotracheal tube. The patient underwentprolonged manual ventilation and survived.

With this new lifesaving, if laborious, technol-ogy in hand, a separate area to care for polio vic-tims was established in the hospital. “At an earlystage the following measures were adopted: 1) pa-tients who were likely to develop respiratory com-plications were transferred to special wards forobservation and recording vital signs, etc.; 2)tracheotomies were done under general anesthe-sia and cuffed tubes were used; 3) manual, inter-mittent positive-pressure ventilation was used in-stead of or to supplement respirators; and 4)secondary shock was treated” (121).

In addition, the hospital developed an elaboratepersonnel system, involving anesthetists, epidemi-ologists, nurses, medical students, and hospitalworkers, to provide continuous care for patientsand to maintain the machinery being used. As aresult of these measures, the mortality rate forpolio victims was reduced from 87 to 40 percent.

With the exception of Danish experience, ICUs,like recovery rooms, were established initiallymore for managerial than for medical reasons. Amajor factor in their early development was theneed to relieve nurses who were so busy caringfor a few critically sick patients that they wereneglecting the remaining patients on the wards(30). In addition, ICUs were even seen as a meansof reducing the cost of medical care (115).

By the late 1950s, the rapid development of themechanical ventilator provided the medical ra-tionale for establishing ICUs. This life-supportingtechnology needed to be monitored too closelyto be dispersed throughout the hospital (136,200).In a number of hospitals, the general ICU was adirect outgrowth of a respiratory ICU set up to

11

——.

12 ● Health Case Study 28 : ]ntensive Care Units: Costs, Outcome, and Decisionmaking

care for patients suffering respiratory paralysis ing an ICU. By the last half of the 1960s, mostcaused by polio (36) or tetanus (155). U.S. hospitals had established at least one ICU

In 1958, only about 25 percent of community (205).

hospitals with more than 300 beds reported hav-

ADVANTAGES AND DISADVANTAGES

Early advocates of ICUs identified a number ofadvantages for establishing a separate intensivecare unit (frequently called an “intensive therapyunit” in England and Europe) (25,47,178,208,231):

●

●

●

●

●

●

●

maintenance of high standards of care forseriously ill patients by using specially trainedphysicians and nurses;provision of more continuous observationand frequent measurements of relevant in-dicators of clinical condition;concentration of technologies in one locationto avoid duplication of equipment and per-sonnel;direct access to patients for major proceduresand therapies, including resuscitation;avoidance of upsetting the regular ward rou-tine and disturbing less ill ward patients;fostering high staff morale and team work;andopportunities for concentrated education and.research.

From the outset, there was disagreement onwhich patients would benefit from ICU care. Earlyunits attempted to exclude “terminal care cases,chronic cases, and disturbed or disturbing pa-tients” (23). Some emphasized that intensive ther-apy should be provided to support vital functionsuntil the underlying disease process could be cor-rected or run its course (200). Other early com-mentators saw the ICU simply as the place for the“critically ill” (187), or advocated the use of theICU as a last resort for a “final desperate attempt”to save a life (36). Lack of agreement persists onwhich patients should have priority access to ICUcare.

While the advantages of the ICU were recog-nized early, so were the potential disadvantages(25,64,178):

OF ICU CARE

a noisy, intrusive environment for seriouslyill patients;interrupted continuity of medical responsi-bility;mental and physical strain on the ICU staff;overenergetic treatment—for both hopelessand less serious cases;decreased nursing skills on the general wardsas the sickest patients are removed;potential for high cost with unfair claims onthe hospital budget; andincreased cross-infections among seriously illpatients in the same area.

Stated another way, in some situations, applica-tion of intensive care maybe unnecessary becausethe condition is not serious enough; unsuccessfulbecause the condition is too far advanced; unsafebecause the risk of complications is too great; un-sound because it serves no useful purpose for thepatient; or unwise because it utilizes too manyresources (125).

Despite recognized patient care problems and,more recently, cost concerns, ICU beds have con-tinued to proliferate. There is substantial evidencethat, at least for some types of patients, care pro-vided in ICUs is extremely effective. For manymedical problems, care of patients outside an ICUwould be unthinkable to the modem clinician. Atthe same time, it is remarkable that such an all-pervasive and cost-generating innovation has de-veloped primarily because of “a priori” considera-tions, with few critical evaluations of its effective-ness (198). The growth of ICUs has been fosteredby a highly favorable reimbursement system (6o),by the development of professional medical andnursing critical care societies which constitute astrong constituency for continued expansion ofICUs (166), and by Federal policies which either

Ch. 2—Evolution, Distribution, and Regulation of Intensive Care Units .• 13

have directly stimulated ICU development (e.g., preferentially to exempt ICUs from expansionthe Regional Medical Program) or have tended restraints (205).

DEFINITIONS

In the broadest sense of the term, “critical caremedicine” has been used to include managementof critical illness or injury at the scene of onset,during transportation to a medical facility, in theemergency department, during surgical interven-tion in the operating room, and finally in thehospital-based ICU (207). Some consider criticalcare to be the highly technical treatment that isprovided to the most severely ill or injured subsetof the population receiving concentrated care ina specialized unit (128,208). Thus, critical caremay be considered a higher level of managementthan intensive care. This case study, however, willfollow the lead of the 1983 NIH Consensus De-velopment Conference on Critical Care Medicineand not distinguish the two terms (262); it willconsider both intensive care and critical care tobe the care provided in separate units generallyknown as “intensive care units.”

From the original recovery rooms and ICUs,other types of units providing specialized carehave evolved. In fact, the Joint Commission forthe Accreditation of Hospitals provides standardsfor “special care units,” which encompass a broad-er spectrum of functions than ICUs (126). Sincethe early 1960s, when the ability to identify andtreat potentially life-threatening arrhythmias wasfirst developed, most cardiac patients have beentreated in coronary care units (CCUs) (59). CCUsgenerally developed independently of ICUs toutilize the new technology of rhythm monitoringto preserve the health of relatively stable patients,rather than to relieve nurses faced with caring forward patients, which was the primary impetus forthe development of ICUs (205). Today, CCUs

treat patients with a relatively narrow range ofdiagnoses, primarily patients with suspected oractual heart attacks and related problems. CCUpatients are not as ill, have fewer physiologic sys-tems involved, require fewer therapeutic services(67), have better outcomes (31,249), have a greaterneed for a quiet, stress-free environment (28), andpose different evaluation and policy issues thando patients in ICUs. In short, CCUs serve a dif-ferent primary function from ICUs (238), andmost hospitals with more than 100 beds have sep-arate CCUs and ICUs (4). Because they cannotafford to operate separate units, smaller hospi-tals frequently combine the separate functions ofcoronary and intensive care. As a result, some ofthe data sources cited in this study, includingMedicare cost reports, have necessarily combinedICUs and CCUs as critical care or special careunits.

In recent years, special care units have diver-sified in other ways (166). First, they have evolvedalong specialty or subspecialty lines. Thus, burn,cardiovascular surgery, pediatric, neonatal, andrespiratory as well as medical and/or surgical in-tensive care units are now common. Neonatal, pe-diatric, and burn units raise distinct issues and willnot be considered in this case study. Second, unitshave differentiated into increasingly distinct levelsof intensity of care, e.g., step-down and inter-mediate care units. These newer types of units,usually adjacent to the coronary or intensive careunit, generally provide more concentrated nurs-ing levels than those on the general medical orsurgical floors, but they do not provide intensivetherapy.

.

14 • Health Case Study 28 : Intensive Care Units: Costs, Outcome, and Decisionmaking

REQUIREMENTS OF AN ICU

A detailed consideration of the design, orga-nization, staffing levels, skills, personnel policies,and other components of an ICU is beyond thescope of this study. Yet in general, all intensivecare units meet these requirements:

care for severely ill or potentially severely illpatients;employ specially trained registered nurses ona one-nurse to one- to three-patient basis;identify a physician as the director of patientcare and administrator of the unit;have 24-hour acute care laboratory support;andprovide a wide range of technological serv-ices, with the help- of expert medical sub-specialists and ancillary personnel (51,166).

While the availability of physicians in ICUs varieswith the size and type of hospital, all ICUs com-bine intensive nursing care and constant patientmonitoring (116). In community hospitals, theICU medical director is frequently not full-timeand shares patient care responsibilities with otherstaff physicians who also have major non-ICUresponsibilities. In these units, day-to-day man-

agement and administrative decisions are madeby the head nurse of the ICU (283). Large hospi-tal ICUs tend to have full-time medical directors.

The NIH Consensus Panel has identified theminimal technological capabilities that an ICUshould provide, regardless of the type of facilityin which it is located (176):

A. cardiopulmonary resuscitation;B. airway management, including endotracheal

incubation and assisted ventilation;C. oxygen delivery systems and qualified res-

piratory therapists or registered nurses todeliver oxygen therapy;

D. continual electrocardiographic monitoring;E. emergency temporary cardiac pacing;F. access to rapid and comprehensive, speci-

fied laboratory services;G. nutritional support services;H. titrated therapeutic interventions with in-

fusion pumps;I. additional specialized technological capa-

bility based on the particular ICU patientcomposition; and

J. portable life-support equipment for use inpatient transport.

SPECIALTY V. MULTISPECIALTY ICUs

Since their development two decades ago, hos-pitals have differed on whether to establish oneor more multispecialty ICUs to treat the range ofseriously ill medical and surgical patients or toset up separate ICUs for patients with similarproblems (208). For reasons of efficiency andeconomy, smaller hospitals generally have a com-bined medical and surgical ICU. The smallest hos-pitals also combine coronary care with intensivecare in a single unit (4)0

Larger hospitals, particularly teaching hospi-tals, often have separate general medical and sur-gical units as well as separate subspecialty unitsfor specific types of medical problems, e.g., car-diac surgery and respiratory care. The Massachu-setts General Hospital, for example, has nine sep-

arate subspecialty ICUs (248). However, evenhospitals of similar size and type have adopteddifferent approaches to the issue of multispecialtyv. separate specialty ICUs (136).

The major rationale for multispecialty ICUS isa medical one, namely, that regardless of theunderlying disease, many life-threatening physi-ological disturbances are quite similar in seriouslyill patients (43,208,265). Thus, a basic purpose ofICU care is to support general physiologic re-sponses to stress in order to provide time for aspecific therapy for the underlying illness to takeeffect (89,116,199,222). At times, ICUs primar-ily treat physiologic disturbances, not diseases;they save lives primarily by supporting oxygena-tion, often with respirators (209), and by prevent-

Ch. 2–Evolution, Distribution, and Regulation of intensive Care Units ● 15

ing circulatory collapse and shock (222). Sincephysiologic complications are similar regardlessof precipitating factors, there is a strong medicalrationale for multispecialty intensive care pro-vided by comprehensively trained generalists (8).

Increasingly, concerns about efficiency and ris-ing costs have supported maintaining multispe-cialty units rather than separate subspecialty units.With multispecialty units, there maybe less dupli-cation of expensive equipment, although ICUsgenerally do not utilize “big ticket” technologies(6). More importantly, because of highly variableclinical demands for ICU care, ICU occupancy canvary dramatically, and combining medical andsurgical specialty and subspecialty units permitsgreater efficiency in the use of personnel, particu-larly nurses, which is a major cost factor in ICUs(212).

Traditionally, however, demand for ICUs hasdeveloped along subspecialty lines, usually in re-sponse to the availability of new medical technol-ogy. For example, the mechanical respirator ledto the respiratory ICU, and the advent of cor-onary artery bypass surgery led to the postcar-diac surgery ICU. In addition, specialists often feelthat physicians trained in other fields do not havesufficient understanding and skill to care for pa-tients with particular “subspecialty” problems. In-deed, some have advocated a separate surgical

DISTRIBUTION OF ICU BEDS

It is difficult to estimate precisely the numberof ICUs and ICU beds in this country because ofthe ways in which hospitals report their bed ca-pacity. This is particularly a problem with smallerhospitals, which may designate their ICUs asCCUs or mixed ICU/CCUs in the annual Ameri-can Hospital Association (AHA) survey. In ad-dition, the annual AHA survey includes multipleICUs reported from single hospitals. From 1981AHA survey tapes, it can be estimated that 78 per-cent of short-term general hospitals have at leastone ICU or CCU, and that 93 percent of hospi-tals larger than 200 beds have a separate ICU(106). Overall, in 1982, 5.9 percent of the total

ICU for each surgical specialty in a large hospi-tal (81). Others feel that nursing personnel skilledin one subspecialty, such as cardiology, may beunsuited by temperament, motivation, and train-ing for work in other subspecialties (147).

In short, the debate over the desirability ofgeneralists v. specialists which exists in medicinegenerally is also being waged in the intensive careworld. The trend, which is supported by theSociety for Critical Care Medicine, is to crosstraditional departmental and specialty lines andto create a “multidisciplinary specialty” equallyskilled at caring for medical and surgical prob-lems (95,274). An attempt to define the bound-aries of critical care medicine by examination andprescribed training has recently been developedby the American Board of Medical Specialties (8).In 1980, the Boards of Internal Medicine, Pedi-atrics, Anesthesiology, and Surgery joined to-gether to offer a certificate of special competencein critical care medicine (95). This examinationhas yet to be given. In 1982, some 50 fellowshipprograms in critical care medicine in the UnitedStates were training approximately 150 physiciansto become critical care generalists (91,92). Another36 programs were training fellows in pediatriccritical care medicine. Despite the new cadre ofcritical care generalists, however, many hospitalscontinue to maintain separate specialty and sub-specialty ICUs along departmental lines.

hospital beds in non-Federal, short-term commu-nity hospitals were ICU and CCU beds. This fig-ure does not include pediatric ICU beds, neonatalbeds, or burn care beds, which add another 0.2percent, 0.7 percent and 0.1 percent, respectively,to the total number ICU beds (4).

Table 1 shows the distribution of reported ICUbeds by size of hospital. In general, ICU beds arefairly evenly distributed across all sizes of hospi-tals. In 1982, for example, hospitals larger than500 beds, which account for 22.6 percent of totalshort-term general hospital beds (4), have 24.8percent of reported ICU beds. Table 2 shows the

-—— — ——

16 ● Health Case Study 28: intensive Care Units: Costs, Outcome, and Decisionmaking

Table 1 .- Distribution of ICU Beds in Short-Term, Non-Federal Hospitals, by Size of Hospital, 1982

Hospital bed size Total hospital beds Percent of total Total ICU/CCU beds Percent of total<100 . . . . . . . . . . . . . . . . . . . . . 146,706 14.5 5,889 9.9100-199 . . . . . . . . . . . . . . . . . . . 195,425 19.3 10,677 17.9200-299 . . . . . . . . . . . . . . . . . . . 179,312 17.7 11,302 18.9300-399 . . . . . . . . . . . . . . . . . . . 144,012 14.2 9,312 15.6400-499 . . . . . . . . . . . . . . . . . . . 120,682 11.9 7,692 12.9>500 . . . . . . . . . . . . . . . . . . . . . 229,043 22.6 14,826 24.8

Total . . . . . . . . . . . . . . . . . . . 1,015,180 99.3 59,698 100.0SOURCE: American Hospital Association, Hospital Statistics, 1983 edition.

Table 2.-lCU/CCU Beds as Percent of Total Bedsby Hospital Size for Short-Term Nonfederal

Hospitals, 1982

Hospital bed size Percent ICU/CCU beds<1oo . . . . . . . . . . . . . . . . . . . . . . . . 4.0100-199 . . . . . . . . . . . . . . . . . . . . . . 5.5200-299 . . . . . . . . . . . . . . . . . . . . . . 6.3300-399 . . . . . . . . . . . . . . . . . . . . . . 6.5400-499 . . . . . . . . . . . . . . . . . . . . . .>500 . . . . . . . . . . . . . . . . . . . . . . . . : : :

Total . . . . . . . . . . . . . . . . . . . . . . 5.9SOURCE: American Hospital Association, Hospital Statistis, 1983 edition,

percent of ICU/CCU beds as a percentage of totalbeds by hospital size in 1982. For hospitalsof200beds or more, the ICU/CCU bed percentage isvery consistent.

Table 3 indicates the distribution of combined,non-Federal intensive and coronary care bedsbyregion as of 1981. (Coronary care beds makeupabout 25 percent of the total.) There are somevariations in the number of these beds as a per-cent of total beds, with the Pacific, East NorthCentral and Mountain States having the highestpercentages. However, as Russell pointed out, thedistribution of ICU/CCU beds is much more uni-form when considered in relation to population,rather than to hospital beds (205).

Finally, as shown in table 4, the distributionof ICU beds varies somewhat according to hos-pital sponsorship.

EXPANSION OF ICU BEDS

While the number of community hospital bedsincreased only about 6 percent between 1976 and1982, reported ICU and CCU beds in community

Table 3.–Distribution of ICU and CCU Beds,by Region, 1981

Per 10,000 Per 100Region population hospital bedsNew England . . . . . . . . . . . . . . 5.8Middle Atlantic . . . . . . . . . . . . ; : :South Atlantic . . . . . . . . . . . . . 2.9 : : :East North Central . . . . . . . . . 3.3 6.7East South Central . . . . . . . . . 5.3West North Central . . . . . . . . ; : ; 5.0West South Central . . . . . . . . 2.6 5.2Mountain . . . . . . . . . . . . . . . . . 2.6 6.2Pacific . . . . . . . . . . . . . . . . . . . 2.7 7.0

Total . . . . . . . . . . . . . . . . . . . 2.9 5.9a

aHospital data in this table includes Federal hospitals and specialty service short-term hospitals.

SOURCE: American Hospital Association, Hospital Statistics, 1982 edition; andU.S. Department of Commerce, Bureau of the Census, State andMetropolita Area Data Book, 1982,

Table 4.—Percentage of ICU/CCU Beds in Short-TermHospitals, by Hospital Sponsorship, 1976 and 1982

Percent of hospitalbeds that are ICU

or CCU beds

hospitals increased by 29 percent, or an averageof almost 5 percent a year. Moreover, over halfof that reported increase occurred between 1979

Ch. 2–Evolution, Distribution, and Regulation of Intensive Care Units ● 17

to 1981. In this 2-year span, reported ICU bedsincreased 14.3 percent and reported CCU bedsgrew 15.4 percent (4), despite the absence of anydramatic medical breakthroughs that would ex-plain such a sharp rise. While the number of cor-onary artery bypass graft surgery procedures per-formed in the country was increasing by perhaps20 percent a year during these years (257), the in-crease in the number of such operations could ex-plain only a very small increase in ICU beds.

One can speculate, therefore, that the Medicarepolicy implemented in 1980 (73) that tightenedlimits on routine bed charges—commonly knownas the “section 223 limits” —but not on special care

REGULATION OF ICUs

Along with the medical and organizational rea-sons for their expansion, ICUs and CCUs wereencouraged by the Federal Government in the1960s initially in the Regional Medical Programs(205).

In the 1970s, State certificate-of-need (CON)statutes were passed in most States. CON statutesrequire a prior determination by a governmentalagency that certain major capital expenditures orchanges in health care facilities are needed (19).Early evaluations showed that CON programshelped forestall the addition of general hospitaland long-term care beds (19). However, ICU bedshave generally been approved by CON agencies.

In addition, Salkever and Bice (211) found thatwhile CON programs controlled expansion in bedsupply to some extent, they stimulated other typesof hospital investment. Specifically, they foundthat assets per hospital bed, for equipment andother nonlabor products, actually increased as aresult of CON. A subsequent, more definitivestudy confirmed the findings that the CON re-quirement generally has been successful in limitingthe number of beds, but not the intensity of re-source use or costs (188). Ironically, the threat ofCON review may have encouraged hospitals to

bed charges or ancillary services, created a strongstimulus for hospitals to add more ICU beds (60)or, perhaps, to reassign beds to special care wherepossible. The most dramatic rise in ICU/CCUbeds between 1979 and 1981 occurred in hospi-tals with more than 500 beds, which accountedfor almost 55 percent of the total increase in ICU/CCU beds in these two years (4). In 1982, thenumber of ICU/CCU beds increased 4 percent,while total community hospital beds increasedonly 1 percent. Thus, while ICU bed expansionhas continued at a much faster rate than hospitalbeds generally, the pace of growth found in 1980and 1981 has slowed. ,

convert low-asset routine care beds into compara-tively high-asset ICU beds (166).

Equipment used in ICUs rarely requires CONapproval. The national threshold for requiringCON approval in the National Health Planningand Resources Development Act of 1974 (PublicLaw 93-641) was $150,000, and most ICU equip-ment is well below that level. The cost per bedof typical ICU cardiac monitoring equipment in1978, for example, ranged from $6,000 to $8,500(6). A new ICU respirator costs between $10,000to $15,000 (87).

The construction costs of each patient unit inthe ICU was estimated to cost between $44,oooand $75,000 in 1978 dollars (6), Renovation costswere much less. Thus, hospitals with sufficientcapital can escape CON review altogether bygradually expanding and upgrading already ex-isting ICUs (119,166). As was noted earlier, hos-pitals reported about a 15-percent increase in ICUbeds between 1979 and 1981, a time when CONprograms were functioning in virtually everyState. The current trend toward raising CONthresholds practically assures that CON regula-tion of ICUs will remain a minor issue.

3 ■

Cost of ICU Care

3.Cost of ICU Care

COMPONENTS OF ICU COSTS

The cost of intensive care units (ICUs) can bedivided into the direct costs of operating the ICUand the indirect costs for central services that areallocated to the ICU (6). Sanders estimates (212)that for Massachusetts General Hospital in Bos-ton about 65 percent of ICU costs (for labor,equipment, etc. ) are direct, and that about 35 per-cent of costs (for hospital overhead, housekeep-ing, etc. ) are indirect.

Direct costs include fixed costs and variablecosts. Fixed costs exist no matter how many pa-tients are treated in the ICU and include deprecia-tion for the cost of construction, renovation, and

COSTS OF AN ICU DAY

It has become increasingly clear that hospitalcharges do not represent the true costs of provid-ing hospital services (80). Generally, charges aregreater than operating costs, in order to pay forbad debts, to support nonreimbursable educa-tional and preventive health programs, and to payfor costs disallowed by cost-based insurers, in-cluding many Blue Cross plans, Medicaid, andMedicare (80). For example, by analyzing cost andbilling data, the Health Care Financing Adminis-tration has calculated the national ratio of al-lowable Medicare inpatient operating costs toMedicare inpatient charges at 0.72 (74).

ICUs are different from most hospital services(including generaI room and board’), however,in that charges for ICU room and board are oftenset below cost (6,212,240). In one detailed econo-metric analysis, ICU charges for room and boardin one hospital were found to be only slightlymore than half of calculated costs (109). ICU datafrom U.S. hospitals consist mostly of room andboard charge data, unadjusted for actual cost. The

1Overall, room and board charges make up slightly less than halfof total hospital inpatient charges; the rest is made up of variouscategories of ancillary services.

equipment, as well as equipment maintenance (6).Variable costs are dependent on the volume ofservices provided. Some variable costs, such aspersonnel costs, are fixed over a specific range inpatient volume, but change when the patient vol-ume exceeds the range. Other variable costs, suchas nondurable equipment and oxygen, are de-pendent directly on total patient days (6). Datafrom both foreign and domestic ICUs indicate that50 to 80 percent of direct costs are variable per-sonnel costs, primarily for nursing (42,101,155,212). On average, ICUs use almost three times asmany nursing hours per patient day as do gen-eral floors (205).

charges or costs for the ancillary services used byICU patients are not matched to their ICU stays,because hospitals report their charges for thevarious ancillary services by department, not bysite of patient location. If one considers only ICUroom and board charges in estimating ICU costs,one may significantly underestimate the relativecostliness of ICU care, then, because ICU chargesunderestimate ICU costs and because the costs ofancillary services that are performed when pa-tients are in the ICU are not included.

With the exception of certain administrativecosts that support ICU physician staff, the costsof physician services to ICU patients generally arenot included in hospital cost reports or in hospi-tal charges. As will be discussed further in chapter6, there is reason to believe that ICU patients re-ceive a greater intensity of billable physician serv-ices than non-ICU patients.

Cost data from other countries provide an op-portunity to determine relative costliness of ICUv. non-ICU care, particularly in countries wherehospitals receive operating budgets. In those fixedrevenue systems, hospitals do not need to chargemore than costs in some departments to make up

21

—.

22 ● Health Case Stud y 28: Intensive Care Units: Costs, Outcome, and Decisionmaking

for losses in other departments. Estimates of costsfor a day of ICU care compared to a day of wardcare have ranged from a 2.5:1 ratio in France(182), to 3:1 in Canada and Australia (29,89), andto 4:1 in Great Britain (174). An attempt in theearly 1970s to estimate actual costs (includingancillary services) in the United States yielded anestimate of 3.5:1 in a large, teaching hospital (97).But anecdotal reports now suggest that relativecosts of ICU to non-ICU care in some institutionsare as much as 5:1 (93).

Numerous U.S. studies of the per diem chargeratio for room and board in the ICU comparedto non-ICU floors have shown a range of 2:1 to2.5:1 in small community hospitals (43,140) toabout 3:1 in large community and teaching hos-pitals (140).

The Equitable Life Assurance Hospital DailyService Charge Survey of 2,519 hospitals in 1982(71) showed an average charge of $408.50 for an

intensive care bed and $167.50 for a private bed,a ratio of about 2.5:1.

Patients in ICUs have a relatively greater per-centage of their charges attributed to ancillaryservices than to accommodations compared togeneral floor patients. In a recent study of alarge-sized community hospital, for example, 45.7percent of the total charges for ICU patients werefor room and board, while 57.1 percent of thetotal charges for non-ICU patients were for roomand board (175). Generally speaking, the moreacutely ill the patient, the greater the percentageof the bill attributable to ancillary services(49,67,162,271).

In short, ICU patients consume more directresources, mostly for nursing, than regular floorpatients, as well as a greater proportion of ancil-lary services, particularly laboratory and phar-macy services (49,101) than regular floor patients.

TOTAL NATIONAL COSTS OF INTENSIVE CARE

There is a notable lack of precision in estimatesof the portion of hospital care costs that can beattributed to intensive care. In a major review ofICUs in Technology in Hospitals (205), LouiseRussell provided a method for indirectly estimat-ing the national cost of ICU care. Recent reviewsusing Russell’s method (described in app. B) esti-mate that 15 to 20 percent of total costs of hospi-tal care can be attributed to intensive care (40,136,206).

Before refining and updating this estimate, itis important to present the alternative ways ofanalyzing the costs of intensive care, includingcalculations of: 1) the direct and indirect costs ofoperating an ICU; 2) the total hospital costs, in-cluding the costs of ancillary services as well asICU costs, incurred by patients when they are inthe ICU; 3) the total hospital costs attributableto patients who spend any time in ICUs; and 4)the incremental cost generated by ICUs above thecost that a hospital would have to absorb fortreating very sick patients who would remain inthe hospital even if ICUs did not exist. The lastdefinition is particularly relevant to this case

study, since it is consistent with the concept thatthe ICU is a separate technology, independent ofthe patients treated in it.

Estimates of the total hospital cost of patientswhen in an ICU (Definition 2) and of the incre-mental costs of operating an ICU (Definition 4)are probably the most relevant in terms of publicpolicy considerations, but are not easily madefrom available hospital accounting sources (267).The direct and indirect costs of an ICU (Defini-tion 1) and the total costs of intensive care pa-tients (Definition 3) are more easily estimatedfrom hospital accounting data, but have muchmore limited policy relevance.

Based on these considerations, estimates of thepercentage of total national inpatient hospitalcosts attributable to intensive care according tothe

●

●

different definitions can be made:

Definition 1: The direct and indirect costs ofrunning the ICU, as reflected in charges forICU room and board—8 to 10 percent.Definition 2: The total hospital costs of pa-tients when in the ICU—14 to 17 percent.

Ch. 3–Cost of ICU Care ● 23

●

●

Definition 3: The total hospital costs for pa- costs associated with most physician services,tients who spend any time in the ICU dur- neonatal, pediatric, or burn units, or the provi-ing a hospitalization—28 to 34 percent. sion of intensive care in Federal hospitals, oper-Definition 4: The incremental cost generated ated mainly by the Veterans Administration andby ICUs above the cost that a hospital wouldhave to absorb for treating ICU-type patientsif the ICU did not exist—cannot be estimated.

The assumptions underlying the estimates and thecalculations are available in appendix B.

Given these percentages, one can estimate thenational cost of adult intensive care. It should beemphasized that these estimates necessarily in-clude the costs of coronary care, but not those

the Department of Defense. In 1982, total nationalexpenditures for hospital care were $136 billion,of which 84 percent were for acute care in com-munity hospitals —or $114 billion (87a). Since anestimated 87 percent of community hospital costsare inpatient costs (4), $13 billion to $15 billionwere spent in 1982 for costs associated with pa-tients in adult ICUs and coronary care units,according to Definition 2 above.

25-338 0 - 84 - 3

4Utilization of ICUs

4Utilization of ICUs

INTRODUCTION

there is little system-characteristics of in-

For a number of reasons,atic information about thetensive care unit (ICU) patients, i.e., their age,sex, length of stay, and case mix. Hospitals andphysicians vary considerably, for example, in theway they treat patients with the same disease. Fur-thermore, as was noted earlier, there is no singlemodel of ICU organization—some hospitals havean ICU combined with a coronary care unit (CCU),while others have separate units; some combinemedical and surgical ICUs, and others do not; stillothers have multiple subspecialty ICUs. Commu-nity hospitals, which usually do not have full-timesalaried physicians, may put less sick patients inICUs primarily to provide them with concentratednursing care (67).

There is no national data base which describesICU utilization in any detail. The American Hos-pital Association (AHA) survey data provides in-formation only on ICU and CCU beds and daysby hospital size and type (see ch. z). A moredetailed profile of ICU patients is based on pub-lished studies from individual hospitals. A com-

UTILIZATION BY TYPE OF ICU

Surgical ICU patients tend to be younger (49,155,175,227), to have more limited or reversiblediseases with reasonably well-defined therapeu-tic endpoints (50,56,129,175,178), and to be morehomogeneous than medical ICU patients (49).Even so, there are substantial differences amongsurgical ICU patients. The patient profile of sur-gical trauma patients, for example, differs signif-icantly from that of postcardiac surgery patients.Trauma patients on average are younger and have

ICU ADMISSION RATES

It is not known what percentage of the popula-tion, or even how many hospitalized patients areplaced in an ICU for any defined period of time.Relman suggests that 15 to 20 percent of all pa-

pilation of many, but not all, such studies is pre-sented in table 5. It should be emphasized thatthese studies are from teaching hospitals and largecommunity hospitals and may not be represent-ative of the ICU care provided in small commu-nity hospitals.

Recently, the Health Care Financing Adminis-tration (HCFA) has developed a profile of Medi-care hospital utilization, including ICU/CCU uti-lization, based on its short-stay hospital inpatientstay record file for 1979 and 1980 (111,112). Thisfile, called the MEDPAR file¹, is generated by link-ing information from three HCFA master programfiles for a 20-percent sample of Medicare benefi-ciaries. The MEDPAR file is the only data basewhich provides population-based rather thanhospital-based ICU utilization data, and, of course,it only profiles the Medicare population.

‘The MEDPAR file also contains billed charge data and clinicalcharacteristics, such as principal diagnosis and principal procedure,in addition to utilization data.

longer ICU stays than postcardiac surgery pa-tients.

Medical ICU patients tend to be older, havemore progressive, chronic diseases (29,174,248,265) and have more concurrent illnesses (265).These differences must be kept in mind when eval-uating reports of utilization and outcome fromparticular ICUs.

tients are cared for in an ICU or CCU at somepoint during their hospital stay (195).

According to the 1979 MEDPAR sample, 18percent of Medicare patients who were discharged

25

———

26 ● Health Case Study 28: lntensive Care Units: Costs, Outcome, and Decisionmaking

Table 5.—Summary of Selected ICU Studies

Dates ofType of data Number Mean ICU Percent ICU Percent hospital

Study author* Country ICU collection in study age LOS mortality mortality for ICU patients

Safar . . . . . . . . . . . U.S.Bates . . . . . . . . . . . CanadaBoyd . . . . . . . . . . . U.S.Crockett . . . . . . . . G.B.Callahan . . . . . . . . U.S.BMA a . . . . . . . . . . . G.B.Rogers . . . . . . . . . U.S.Carroll . . . . . . . . . . U.S.Skidmore . . . . . . . G.B.Safar and

Grenvik(1971). . . . . . . . . Us.

Pessi . . . . . . . . . . . FinlandSpagnolo . . . . . . . U.S.Bell . . . . . . . . . . . . G.B.Petty (1974) . . . . . U.S.Nun. . . . . . . . . . . G.B.Turnbull . . . . . . . . U.S.Tagge (1975) . . . . . U.S.Tomlin . . . . . . . . . . G.B.McLeave . . . . . . . . AustraliaVanholder . . . . . . . BelgiumChassis . . . . . . . . . U.S.Byrick . . . . . . . . . . CanadaFedulo . . . . . . . . . . U.S.P o r n o . . . . . . . . , . . U . S .Thibault . . . . . . . . . U.S.Legal . . . . . . . . . . FranceMurata. . . . . . . . . . U.S.Hauser . . . . . . . . . U.S.Franklin . . . . . . . . . U.S.Knaus, et al.

( C C M , 1 9 8 2 )b. . U.S.

M-SR

M-SM-SM-CM-S

RM

M-S

M-S

iM-SM-SM-S

M-S(ca)M-SM-SM-S

M,R

M-SM-CM-SMMM

M-S

1959-19611958-1962

19631963-19651964-19661966-19671965-1968

19681965-1969

1965-19701965-19711970-19711966-19721964-19731970-19741971-19741972-19741973-19761975-1976

19761977197819781978

1977-19791978-1979

19791978-19801979-1980

1980-1981

561

336608

1,0005,521

20095

1,162

4,9181,001

2312,8961,598

4221,0352,8781,718

843380489

58182558

2,693228149724512

1,408

—48.0

—44.3

———

54.0—

—50.056.045.2

——

63.0

53.053.054.059.165.054.760.050.062.7

——

54.0

——

5.0—

3.94.0—

—

—6.24.84.4——

5.2—

3.03.4—

5.18.0

3.63.4—

3.9——

4.1

30.343.021.018.010.714.718.0

29.8

18,520.128.016.625.316.422.38.213.514,432.6

21.011.76.0

16.719.326.0

——

26.0——

—28.947.0

—

38.6—

19.7

42.614.0

—29.017.310.034.026.8

——

16.9“Full citations found in References section.aweighted average from 14 ICUs.bWeighted average from 6 ICUs.KEY: M-S Medical-Surgical ICU; M Medical ICU; M-C Medical-Cardiac ICU; R Respiratory ICU; S Surgical ICU.

SOURCE: Office of Technology Assessment.

from the hospital used intensive or coronary care.Fifteen percent used both general ward and ICU/CCU beds, while 3 percent used only ICU/CCUbeds. As table 6 indicates, use of ICU/CCU bedsby Medicare patients does not vary significantlyby hospital size, except for hospitals under 100beds. Table 6 also shows that there is little varia-tion in ICU use by Medicare patients by size ofhospital when ICU/CCU use is considered as apercentage of the patients’ total charges. In-terestingly, there was also little variation in ICU/CCU charges as a percent of total charges by typeof hospital sponsorship (not shown); 7 percent ofall charges for Medicare patients in voluntary,

Table 6.—Use and Percentage of Hospital Chargesincurred in ICUs and CCUs for Medicare Beneficiaries

Discharged From Short-Stay Hospitals, 1979

Percent Percent totalusing charges incurred

Hospital bed size lCU/CCU in ICU/CCU

1-99 beds. . . . . . . . . . . . . . . . 12 5100-199 beds. . . . . . . . . . . . . 18 7200-299 beds. . . . . . . . . . . . . 20 8300-499 beds. . . . . . . . . . . . . 19 8>500 beds . . . . . . . . . . . . . . 20 7

All hospitals . . . . . . . . . . . m 7SOURCE: C. Helbing, “Medicare: Use of and Charges for Accommodation and

Ancillary Services in Short-Stay Hospitals, 1979,” Office of Research,Health Care Financing Administration, U.S. Department of Health andHuman Services, undated.

Ch. 4–Utilization of ICUs ● 27

proprietary, and public, non-Federal hospitalswere room and board charges for ICU/CCUs.

Given the significant regional variations in theconcentration of ICU/CCU beds (see ch. 2), it isnot surprising that utilization of ICU/CCU bedsby Medicare patients also varied somewhat ac-cording to region (see table 7). Perhaps part ofthe explanation for the higher per diem costs andshorter lengths of stay in ICUs on the west coastis a result of the greater use of relatively costlyICU/CCUs in that region (255).

There are also variations by State in the use ofICU/CCUs by Medicare patients; with a rangefrom 12 percent of Medicare hospital dischargesin Louisiana, Kansas, and South Dakota, to 27percent in Connecticut.

Table 7.—Use and Percentage of Hospital ChargesIncurred in ICUs and CCUs for Medicare Beneficiaries

Discharged From Short-Stay Hospitals,by Geographic Region, 1979

Region

New England . . . . . . . . .Middle Atlantic . . . . . ., .South Atlantic. . . . . . . . .East North Central ., ., .East South Central . . . .West North Central . . . .West South Central . . . .Mountain . . . . . . . . . . . . .Pacific ... , . . . . . . . . . . .

Percent usingI c w c c u

Percent totalcharges incurred

in ICU/CCU

SOURCE: C. Helbing, “Medicare: Use of and Charges for Accommodation andAncillary Services in Short-Stay Hospitals, 1979,” Office of Research,Health Care Financing Administration, U.S. Department of Health andHuman Services, undated.

Studies of ICU patients demonstrate a remark-ably consistent male to female ratio of about 3:2(16,47,56,67,146,175,178,248). Only Chassin re-ports a slight female predominance (40). In gen-eral, the ratio represents the prevalence of seriouscardiovascular diseases among males and femalesunder the age of 70. Above that age, female rep-resentation in ICUs increases (248).

A major issue with respect to Medicare is therepresentation of elderly people in ICUs. Withaging comes an increase in the incidence of criticalillness. Thus, elderly people might be expected torequire more intensive care than their proportionof the general population (34) and, possibly, morethan their proportion of the hospitalized popula-tion (76,175). On the other hand, to the extentthat ICU beds are in short supply (248,265) or thatpoor patient prognosis is considered (34,54,56,76),elderly patients might receive less intensive carethan younger patients.

In the United States, the representation of elderlypatients in ICUs seems to be the same or onlyslightly more than as it is in the hospital as a whole(76,139,175). Data from ICUs do not address theeffect of screening on the basis of age that maytake place prior to ICU entry. Speculation on theextent of such screening differs (33,76,137). Therecent HCFA MEDPAR data is somewhat helpful

201918171515151823

77776767

10

SEX AND AGE DISTRIBUTION OF ICU USE

on this issue. As table 8 shows, use of ICU/CCUsby elderly people does not vary from that of thegeneral population until age 85. Even for people85 and older, however, the decrease in ICU/CCUuse is slight.

Once in the ICU, elderly patients generally re-ceive more interventions than younger patients(34). However, when an attempt is made to con-trol for acute severity of illness, the age of ICUpatients does not appear to be a factor in theamount of resources expended in the ICU (137,140). Rather, health status, independent of age,

Table 8.-Use and Percentage of Hospital ChargesIncurred in ICUs and CCUs for Medicare BeneficiariesDischarged From Short-Stay Hospitals, by Age, 1980

Percent totalBeneficiary age Percent using charges incurredgroup lCU/CCU in ICU/CCU

<65 . . . . . . . . . . . . . 18 765-69 . . . . . . . . . . . . 18 870-74 . . . . . . . . . . . . 18 775-79 . . . . . . . . . . . . 18 780-84 . . . . . . . . . . . . 17 7>85 . . . . . . . . . . . . . 15 6

Total all ages . . . 18 7SOURCE: C Helbing, Supervisory Statistician, Office of Research, Division of

Beneficiary Studies, Health Care Financing Administration, U.S. Depart-ment of Health and Human Services, personal communication, June6, 1983. Data derived from the MEDPAR file.

—

28 ● Health Case Study 28: ]ntensive care Units: Costs, Outcome, and Decisionmaking

seems to be the key factor influencing the useof ICU resources once the patient is in the ICU(33,137).

Age does appear to be an important determi-nant of ICU admission in other countries. Whilethe populations are not strictly comparable, table5 clearly demonstrates a younger mean age of ICUpatients in foreign countries. Knaus compared theICUs in five U.S. teaching hospitals and sevenFrench teaching hospitals and found that 45.5 per-cent of U.S. emergency ICU admissions were 60years or older compared to only 31 percent of theFrench patients (142). Vanholder in Belgium ac-knowledged that when there is a lack of space inthe ICU, older patients are less apt to be admitted(265). With many fewer ICU beds per capita avail-able in Britain, age appears to be a primary fac-

ICU CASE

Diagnoses

One characteristic of the ICU, particularly incomparison to other special care units (i.e., cor-onary, burn, and neonatal units), is the wide va-riety of underlying diseases that are present. AsChassin emphasized, medical ICUs treat a widespectrum of illnesses; any specific disease repre-sents a very small proportion of the total numberof diseases that are present (40,238,265). Similarfindings have been described for mixed ICUs andnonsubspecialty surgical ICUs (49,54,129, 139).Even respiratory ICUs treat a variety of primarydiseases (10,29).

In surgical ICUs in major regional centers,trauma patients may represent 40 to 50 percentof the ICU population (129,178). In other surgicalICUs and mixed ICUs, trauma victims representa much smaller percent of the overall ICU popula-tion (139), but are still a large proportion of themost critically ill patients (54). Trauma patientsare much younger than the overall ICU profile(54,129).

There is no accepted classification scheme thatdescribes the clinical characteristics of ICU pa-tients. Perhaps the major problem with identify-ing ICU case mix is the fact that many critically

tor for limiting access to the scarce ICU beds (1).

When they were first developed, use of renaldialysis machines were rationed partly on the basisof age, and it has been suggested that age was sim-ilarly a factor in the United States in rationingscarce beds in the early days of ICUs (248). In fact,as can be seen in table 5, in the last 15 years orso, there has been no dramatic trend toward olderICU patients even though the mean age of thepopulation has increased. Unfortunately, data onthe age of ICU patients in the late 1950s and early1960s, when ICUs were first opened, are not avail-able. In addition, there appears to be no consist-ent age difference in ICU use based on size or typeof hospital. Finally, it should be pointed out thatmean ages reported in ICU studies are a few yearslower than the median ages (248).

ill patients have multiple underlying medical prob-lems which interact to produce severe physiologiccomplications. Vanholder found, for example,that, excluding coronary care patients, each pa-tient in his ICU had an average of 4.39 signifi-cant, distinct diagnoses (265). Questionable diag-noses, disorders not likely to have vital conse-quences, and previous diseases that had beencured at the time of admission to the ICU werenot included in his calculation. The sicker the pa-tient, the more likely it is that the ICU is treatingfailure of major organ systems, in addition to theunderlying disease or the disease that precipitatedthe failure.

Other Case Mix Parameters

Recognizing that the complexity and severityof illness of ICU patients are generally not re-flected by the primary diagnosis, other descrip-tions of ICU case mix have been used. Patientscan be grouped according to those referred di-rectly from emergency rooms, those transferredfrom the regular hospital floors, and thosetransferred from other hospitals (31). InterhospitalICU transfer of patients is relatively infrequent

Ch. 4–Utilization of ICUs ● 29

in the United States, but common in some othercountries (81,142,146).

ICU admissions can be characterized as emer-gency or elective, the latter usually referring topostoperative admissions. Medical ICU admis-sions are usually emergencies, whereas the ma-jority of surgical admissions are elective (49,52,227), unless the hospital is a major trauma cen-ter. Elective, postoperative patients may, never-theless, be critically ill, or at least need closemonitoring and observation (54).

ICU patients can be characterized as those re-quiring close observation and monitoring andthose requiring intensive therapy. As was pointedout earlier, there is no general agreement on howto classify patients into these groups. Some haveemployed subjective medical assessments of sever-ity of illness and treatment needs (42,163,179).Others have employed objective measures of ther-apeutic resource use developed by Cullen andcolleagues at Massachusetts General Hospital inBoston to separate patients into discrete groupsrequiring different personnel and treatment re-quirements (51,129,144). Recent work has at-tempted to ascribe a severity-of-illness score toeach patient and has found a good correlation be-tween scores of severity and treatment require-ments (144,270).

Because authors use varying approaches to de-scribe the intensity of ICU therapy, it is difficultto summarize the data. Nevertheless, it would ap-pear from the literature —most of which is fromteaching or major community hospitals—that pa-tients receiving the most concentrated intensivetreatment, involving fairly continuously directphysician involvement and various forms of lifesupport, represent less than half and sometimesas little as 10 to 20 percent of the ICU patientpopulation (54,129,144). At the other end of thespectrum, patients who receive technical monitor-ing and nursing care but only routine physiciancare probably represent about 20 to 30 percentof patients in general ICUs (136,137,178,246,269).The remaining 30 to 70 percent of ICU patientsare those that receive actual therapeutic interven-tion to maintain and stabilize one or more phys-iologic functions, but do not require constantphysician involvement in their care or nurse-to-

patient ratios of greater than 1:1. The percentageof “monitor patients” is much higher in ICUs thatalso serve a CCU function (31,249).

Because most research has come from teachinghospitals, the pattern of case mix in communityhospitals may be different, although anecdotalreports do not indicate a consistent difference be-tween teaching and community hospitals (67,163,175).

Readmission

Recently, attention has focused on the fact thathigh-cost users of hospital care are often patientswith chronic illnesses who have repeated hospi-tal admissions (161,218). This pattern is being in-creasingly recognized for intensive care as well(231,248). As might be expected, readmission tothe same unit are less frequent for surgical ICUpatients (178). In a 5-year period, almost 19 per-cent of all patients seen in a major teaching hos-pital medical/cardiac ICU were readmissions, and6 percent were patients readmitted to the ICU dur-ing the same hospital stay (so-called “bouncebacks”) (248).

Length of Stay

The mean length of stay (LOS) in an ICU forall Medicare ICU/CCU patients in 1980 was 4.2days (112). The LOS in ICUs is about 0.5 dayslonger than in CCUs (49). The LOS is reportedlylonger in non-U.S. ICUs (29,88,142,178), prob-ably because there are fewer monitor patients inthese ICUs. The average LOS in U.S. hospitalshas been notably stable over the past 15 years (seetable 5).

As expected, mean LOS is significantly longerthan median LOS (42). The mean does not reflectthe great variation in LOS of ICU patients. In astudy of 1,001 consecutive patients in a surgica]ICU, Pessi (178) found that 27 percent stayed lessthan 2 days, while 15 percent stayed longer than10 days. In one medical ICU, Chassin (40) foundthat 10 percent stayed longer than 10 days. ICUstays of more than a month are not uncommon(49).

While the mean hospital LOS before the recentchanges in Medicare reimbursement in U.S. hos-

— ————

30 ● Health Case Study 28: : intensive Care Units: Costs, Outcome, and Decisionmaking

pitals was 7.6 days (4) and 10.4 days for Medicare sumably because of case mix differences (40,49,patients (113), ICU patients have significantly 175). Part of the variation in published studieslonger total hospital stays. From the few reports may also represent the general pattern of shorterthat present both ICU and total hospital LOS, hospital lengths of stay on the west coast (256).there is significant variation in hospital LOS, pre-

5 .

Outcomes of Intensive Care:Medical Benefits and Cost

Effectiveness

5 mOutcomes of Intensive Care:

Medical Benefits and Cost Effectiveness

DIFFICULTIES IN ASSESSING EFFECTIVENESS

Evaluating the effectiveness of the care providedin the general adult intensive care unit (ICU)presents a number of problems. Unfortunately,it is difficult to separate the intensity of the carefrom the setting in which it is provided (97,98),and therefore, to know whether the same carewould have been equally effective whether it wasprovided in an ICU or in a general hospital floor.

Theoretically, at least, intensive therapy couldbe provided on regular medical floors (120). Infact, there are institutional differences about whois treated in ICUs and for how long (142). More-over, the level and style of intensive care for simi-lar health problems differ significantly amongICUs (67). These differences have developed be-cause of the particular circumstances of individ-ual hospitals, rather than because established cri-teria were available (247).

For some complex medical problems, manyphysicians feel that the necessary care can onlybe provided in an ICU (65). In the late 1960s and1970s, admission to an ICU became routine fora number of medical problems, despite the lackof evidence that ICU care improved outcome.There have been no prospective clinical trials inwhich patients with similar problems were ran-domly allocated to two groups, one of which wastreated in an ICU while the other received inten-sive care outside the ICU (98,222). There is gen-eral agreement that such randomized studies wouldbe unethical (262,279), and it is felt that for manyproblems, treatment in an ICU is necessary if apatient is to have a chance of survival (50).

Since, as noted, randomized clinical trials ofICUs are considered by many to be unethical,most ICU outcome studies have been historicalcontrols and pre-ICU/post-ICU designs (166).These types of studies, however, have been seri-ously flawed by the absence of acceptable criteria

for stratifying ICU patients by diagnosis andseverity of illness to assure comparability of pa-tient populations between different ICUs and inthe same ICU over time (226,248,281).

In the coronary care unit (CCU), for example,it is felt that patients suffering myocardial infarc-tion should be stratified into clinically coherentsubpopulations based on the type of myocardialinfarction suffered in order to assess outcomeproperly (28). The problem of stratification isespecially complicated in the ICU, because pa-tients often have multiple diagnoses, which makecategorization difficult (16,265), and because theirseverity of illness varies (136).

There are other practical problems in conduct-ing research on ICU outcome, including:

1.

2 .

3 .

4 .

In

the fact that any individual institution willhave a relatively small number of patientsin any clinical subset;the lack of a standard format for collectingdata;the difficult yin obtaining informed consentfrom ICU patients in need of immediate, life-saving intervention (176); andthe difficulties in conducting studies that fol-low patients after their discharge from thehospital.

short, because of the absence of an acceptedclassification scheme for stratifying ICU patientsinto accepted subpopulations and because pro-spective clinical trials have not been performed,very little is known about the effectiveness of theICU as a distinct, discrete technology. Investi-gators who report on changes in ICU mortalityrates or lengths of stay can only speculate onwhether their patient populations have changedover time (227,248).

33

34 . Health Case Study 28 : ]ntensive Care Units: Costs, Outcome, and Decisionmaking

Finally, while the primary measure for assess- presence of ICUS may adversely affect the qualitying the effectiveness of ICUs is patient outcome, of nursing care on the regular medical and surgi-it should be recognized that the ICU as a discrete cal floors (25,136). As difficult as it is to measureunit within the hospital may be a focus for edu- the effectiveness of ICU treatment for patients incation and research activities which have positive the ICU, it is nearly impossible to assess objec-“trickle down” effects on care for non-ICU pa- tively the benefits or drawbacks of the ICU fortients (55,86,97). At the same time, however, the the hospital as a whole.

CLINICAL OUTCOMES OF ICU CARE

Because of the varied case-mix in ICUs, it is im-possible to generalize about whether ICU care im-proves outcome. The NIH consensus panel, whichwas asked to assess this issue, concluded thatevidence of the benefit of ICU care was unequiv-ocal for a portion of the heterogeneous ICU pa-tient population (176). The NIH panel identifieddifferent outcomes for three categories of patients(176):

First is the patient with acute reversible diseasefor whom the probability of survival without ICUintervention is low, but the survival probabilitywith such interventions is high. Common clini-cal examples include the patient with acute revers-ible respiratory failure due to drug overdose, orwith cardiac conduction disturbances resulting incardiovascular collapse but amenable to pace-maker therapy. Because survival for many ofthese patients without such life-support interven-tions is uncommon, the observed high survivalrates constitute unequivocal evidence of reducedmortality for this category of ICU patients. Thesepatients clearly benefit from ICU care.

Another group consists of patients with a lowprobability of survival without intensive carewhose probability of survival with intensive caremay be higher—but the potential benefit is notas clear. Clinical examples include patients withseptic or cardiogenic shock. The weight of clini-cal opinion is that ICUs reduce mortality for manyof these patients, though this conviction is sup-ported only by uncontrolled or poorly controlledstudies. Often these studies do not allow one todistinguish between ICU effectiveness and/or dif-ferences in cointerventions that do not require theICU.

A third category is patients admitted to theICU, not because they are critically ill, but be-cause they are at risk of becoming critically ill.The purposes of intensive care in these instancesare to prevent a serious complication or to allow

a prompt response to any complication that mayoccur. It is presumed that the prompt responseto a potentially fatal complication made possibleby continuous monitoring plus the concentrationof specialized personnel in the ICU increases theprobability of a favorable outcome. The risk ofcomplication may be high (as in the patient withan acute myocardial infarction and complex ven-tricular ectopy) or low (as in the patient withmyocardial infarction suspected because of chestpain in the absence of electrocardiographic abnor-malities). Also, the differences in probability ofa favorable outcome following a complication in-side rather than outside the ICU may be large (asin the patient with postcraniotomy intracranialbleeding) or small (as in the patient with gastro-intestinal bleeding). The strength of evidence sup-porting the effectiveness of the ICU varies withthe probability of a complication and with the dif-ference in expected outcome inside and outside theICU. When the risk of complication is high andthe potential gain large, a decrease in mortality

is likely. Similarly, when the risk is low and thepotential gain small, an observable decrease inmortality is unlikely. These patients are not likelyto benefit from ICU care.

The differences in outcomes of ICU care bydiagnosis has been demonstrated in all studies thathave looked at the issue, from the earliest studies(17) to the most recent (248). Table 9 gives ex-amples of specific retrospective outcome studieson the effect of ICU care for certain illnesses.(Note that contradictory findings are sometimesfound for the same condition. ) In general, condi-tions which respond well to ICU care are reversi-ble illnesses without significant underlying chronicillness (e.g., respiratory arrests as a result of drugoverdoses, major trauma, reversible neuromus-cular diseases such as Guillain-Barre Syndrome,and diabetic ketoacidosis) (198,214). Conditionswhich generally do not respond well are exacer-

Ch. .5-Outcomes of intensive Care: Medical Benefits and Cost Effectiveness • 35

Table 9.—Retrospective Outcome Studies of ICU Care

Study ConditionA. Studies showing definite reduction inmortality for condltlon:Petty (1975). . Respiratory failure treated with ventilatorsRogers. . . . . . Respiratory failure treated with ventilatorsBates . . . . . . . Status asthmatics and emphysemaDrake . . . . . . . Non-hemorrhagic strokesSkidmore . . . Postoperative trauma patientsFeller. . . . . . . Severe burns

B. Studies showing no reduction in mortality for condition:Pitner. . . . . . . StrokesPiper . . . . . . . Drug overdoseJennet . . . . . . Head injuries with comaCasali . . . . . . Postoperative acute renal failureGriner . . . . . . Pulmonary edemaHook . . . . . . . Pneumococcal bacteremiaNOTE Studies are cited in the Reference section

SOURCE Off Ice of Technology Assessment

bations of chronic conditions for which there hasbeen no definitive treatment (e.g., cirrhosis withgastrointestinal hemorrhaging, and advancedcancer).

FUNCTIONAL OUTCOME

Different investigators have used varying meas-ures of functional status to gauge outcomes otherthan mortality. These measurements have beensubjective and depend to a large extent on the pa-tient’s prehospital functional status. For patientswith a chronic disability, posthospital functionalstatus is almost never better than their prehospitalfunctional status (34,40,146), although improve-ment has occasionally been found (29).

Surgical patients suffering an acute injury orillness have a reasonable chance of returning to

Most studies have looked at mortality in theICU or in the hospital as a measure of the efficacyof ICU care. However, for some physiologic con-ditions, such as cardiac arrest, ICU care may belifesaving in the short term but may not affect theultimate course of the underlying illness (174,214).Indeed, in some instances, patients with severeunderlying illnesses, such as terminal cancer andcystic fibrosis, have not been offered ICU care be-cause of the dismal prognosis associated with theunderlying illness (58,110,252,253).

Investigators have only recently begun to lookat posthospital survival. As might be expected,the ability to follow patients for 6 months orlonger after their ICU stay depends to a great ex-tent on the population being studied. In general,chronically ill and medical patients are more likelythan acutely ill and surgical patients to die shortlyafter discharge from the hospital (29,34,50,129,146,174,175,178,248) .

a normal functional status (54,178). In a followupstudy, Cullen reported that the l-year mortalityrate was similar to the rate in a previous studyof similarly critically ill patients, but that the pa-tients’ quality of life as measured by the numberof patients who were fully recovered or returnedto full productivity was significantly improved(54,56). This finding suggests that Outcome COme meas-ures other than survival should also be examinedwhen determining effectiveness of ICU care.

CHARACTERISTICS OF ICU NONSURVIVORS

As noted above, certain diseases and conditions Ageare associated with particularly high ICU mortal-ity rates. Underlying disease is probably the most A number of investigators have looked at thesignificant single predictor of outcome of ICU care association of age and mortality in ICUS. Most(54,139). Other factors, including age and sever- have found a direct relationship between increas-ity of illness, are important as well. ing age above 65 and hospital mortality (54,107,

—- —————

36 ● Health Case Study 28: Intensive Care Units: Costs, Outcome, and Decisionmaking

116,178,214,248). In addition, for medical patientsin particular, some have found that patients 70and over who leave the hospital have very highposthospital mortality rates (29,174,248,249).Others, however, have found either a small or noassociation between age and survival (40,50,76,165,265).

When an attempt is made to control for chronichealth status in a multivariate logistic regressionanalysis, age has been found to remain a reliableindependent predictor of mortality (268). Thisfinding suggests that age is not simply a surrogatefor chronic health status. Fedullo (76), on the otherhand, suggests that with the passage of time, eld-erly patients have already gone through a proc-ess of selection, and therefore “healthy” elderlypatients are as able as younger patients to sur-vive an acute major illness.

Severity of Illness

Vanholder (265) found that ICU survivors hadan average of 3.13 major diagnoses whereas non-survivors had 6.09 diagnoses. LeGall (146) founda strong positive correlation between the numberof organ system failures and the likelihood of notsurviving a stay in an ICU. In a number of set-tings, the George Washington University ICU Re-search group in Washington, DC (143) has testedan acute severity-of-illness measure based primar-ily on the deviation from normal of certain clini-cal and laboratory measurements. Using theirscoring system, they found a direct relationshipbetween acute severity of illness and ICU mor-tality and concluded that acute physiologic de-rangement (i.e., acute severity of illness) is sec-ond only to the underlying disease as a risk factorof hospital mortality (139). Less sophisticatedseverity-of-illness classification systems have con-sistently demonstrated a positive relationship be-tween increasing severity of illness and likelihoodof mortality (51,178).

Resource Use

In comparing resource use of ICU nonsurvivorsto survivors, it is necessary to look at the patient’sentire hospitalization, not just the stay in the ICU.In a number of studies from different types of hos-

pitals, 25 to 40 percent of ICU patients who diedin the hospital did so after they were transferredfrom the ICU to the regular medical floor (seetable 5 in ch. 4). Presumably, many of these non-ICU deaths were anticipated and represented thetransfer of “hopeless” patients out of the ICU.

It is now recognized that a significant numberof deaths in the ICU occur after “no resuscitation”orders have been written. In two large medicalcenters, as many as 40 to 70 percent of ICU deathsoccurred under these circumstances (9,96). In alarge community hospital, 19 percent of ICU non-survivors had no hope of recovery and were inthe ICU solely for terminal care (165). In short,a substantial portion of ICU care for nonsur-vivors occurs after hope of recovery has beenabandoned.

Some nonsurvivors have very short and somehave very long ICU stays. Pessi (178) found thatone-third of surgical ICU nonsurvivors died within2 days and 80 percent died within 10 days of ICUadmission. More recently, Cromwell (49) foundthat while 20 percent of ICU nonsurvivors diedwithin 3 days of ICU admission, 10 percent diedafter 2 months in the ICU. On average, nonsur-vivors stay in the ICU about 1.5 to 2 times longerthan survivors (42,48,76,248)

In 1973, Civetta (42) first described the inverserelationship between ICU charges and survival.Since then, whenever it has been examined, thesame relationship has been found—ICU nonsur-vivors accumulate up to two times more hospi-tal charges than survivors (40,49,61). Byrick (29)found the same correlation in Canada when heconsidered actual ICU costs rather than charges.Furthermore, nonsurvivors have incurred propor-tionately higher charges for ancillary services(e.g., laboratory tests, X-rays, and blood) thansurvivors (61,76). Only Parno (175), in a studyinvolving a large community hospital, found nosubstantial difference in ICU charges between sur-vivors and nonsurvivors.

The inverse relationship between charges andsurvival is not as simple as it might first appear,however. Detsky (61) looked at the relationshipbetween charges and patients assigned to varioussubjective prognostic categories. He found the

Ch. 5—Outcomes of Intensive Care: Medical Benefits and Cost Effectiveness ● 37

highest per capita charges in two groups: sur-vivors who initially had been thought to have apoor chance of survival, and nonsurvivors whohad initially been felt to have the best chance ofsurvival. Predicted nonsurvivors who died andpredicted survivors who lived consumed fewerresources. The two groups with highest chargeswould logically be the ones who might benefit themost from intensive medical care.

In another study utilizing a severity-of-illnessmeasure, Scheffler (214) found a nonlinear, U-shaped relationship between the use of resourcesavailable in the ICU and the probability of sur-vival. The first segment —45 percent of patientsand 19 percent of therapeutic interventions-ex-hibited an overall decrease in the probability ofdeath as therapy increased. The second segment,found at the bend of the curve, showed little cor-

relation between probability of death and resourceuse. However, in the third segment, the rising por-tion of the U-shaped curve, there was an overallincrease in the probability of death as resourceuse increased. This last segment represented only9 percent of the ICU population, but those pa-tients consumed as much as 30 to 40 percent ofthe ICU resources. Thus, many patients, even themost seriously ill, may benefit from additionalICU resources applied to their care. While, in ret-rospect, some resources may prove to have been“wasted” in the sense that individuals did not sur-vive despite consuming these ICU resources, it isclear that many patients do benefit from increaseduse of ICU resources. The patients who will ben-efit from additional ICU resources cannot cur-rently be identified ahead of time with any cer-tainty.

DISTRIBUTION OF ICU COSTS AMONG PATIENTS

The data demonstrate that a small percentageof the ICU patient population consumes a substan-tial proportion of total ICU resources. Cromwell’sgroup (49) found that 1 percent of all ICU patientsincurred 10 percent of hospital charges, and 5 per-cent of ICU patients incurred 25 percent of thecharges. In Chassin’s ICU study (40), 7.4 percentof the patients incurred 31 percent of the charges,and 17 percent of patients incurred so percent ofthe charges. The 7.4 percent subgroup averaged$63,000 in charges in 1977 dollars. In general, thehigh cost subgroup was broadly representative ofthe total ICU patient population in terms of age,diagnosis, and other patient characteristics. Sim-ilarly, Parno (175) found that 18 percent of theICU population in his hospital generated half ofthe ICU charges.

In addition, it is likely that within the ICU,there is substantial cross-subsidization of charges.As noted in chapter 4, ICU populations includepatients who are there primarily to be observedand monitored for the development of complica-tions as well as patients who are receiving com-plex life-sustaining therapy. The nurse-to-patientratio can vary from 1:4 or 1:5 for patients withcardiac arrhythmias to 1:1 or greater for thesickest patients (176). While a portion of fixed di-

rect costs and allocated indirect costs should bedistributed evenly among all patients, the ICUcharge structure does not reflect the substantialdifferences in variable labor costs between pa-tients.

The Therapeutic Intervention Scoring System(TISS) (53,130) is a relative value scale which re-duces most of the tasks commonly performedwithin an ICU to 75 items which are assignedvarying weights. It has been used as a direct meas-ure of the use of labor in the ICU. Wagner (270)found that patients recuperating from coronary by-pass surgery utilized 2.5 times more TISS points perday than ICU patients recovering from brainsurgery.

The difference in labor resource use appears tobe even greater for other types of patients (51,54).The distribution of TISS points suggests that all ICUpatients receive a minimum amount of treatment be-yond that provided on the regular wards (67). Thedata also suggest, however, that even if indirect andfixed ICU costs are distributed evenly among all pa-tients, perhaps 50 percent of actual ICU resourcecosts—particularly labor costs—vary dramaticallyamong patients.

25-338 0 - 84 - 4


As noted earlier, the sickest ICU patients incursubstantially more total hospital charges than thosewho are relatively less sick. Yet, the actual cost dif-ferences between these two groups is even greater.Under the new Medicare payment system, which

MONITORED PATIENTS

Increased attention has been paid recently toICU patients who do not receive active intensivetherapy but rather are monitored and observedfor the development of potentially fatal complica-tions which must be responded to promptly (176).Progress has been made in identifying the char-acteristics of coronary patients who do not rou-tinely require coronary intensive care (85,90,141,189,190), and in recognizing CCU patients whocan be discharged to the general floor after 24hours rather than the usual 3 days (163). Simi-larly, national and regional data on intensive carefor patients with burns suggest that a substantialnumber of patients suffering relatively minorburns do not benefit from treatment in an inten-sive bum unit but receive it nevertheless (78,151).

Researchers at George Washington University(269) found that 513 of 1,148 admissions (45 per-cent) to a mixed medical-surgical ICU in a teach-ing hospital could be considered “monitoringonly” patients. Using a multivariate logistic regres-sion analysis of several variables, including aseverity-of-illness measure, they found that 154patients (13 percent of the total ICU patient pop-ulation) had less than a 5-percent predicted riskof requiring active intensive therapy. For thosepatients, the authors felt that the risks of iatro-genic illness’ might outweigh the benefits of ICUmonitoring. In fact, only of the 154 low-risk pa-tients actually received intensive therapy, and inno case did those patients require therapy for animmediately life-endangering condition. After up-dating their data base and looking at preliminarydata from other university hospitals, the authorsconcluded that all ICUs have significant propor-tions of predictably low-risk, monitor admissions(141). The conclusions were supported in a recent

‘An iatrogenic illness is an illness that results from clinical ther-apy rather than from the patient’s disease.

pays a fixed price per diagnosis regardless of actualcost of the treatment provided, hospitals ‘maybecome more aware of the highly disproportionateshare of ICU resources consumed by the mostseverely ill, long-term ICU patients (see ch. 6).

study by Fineberg that looked at patients with arisk of myocardial infarction that is low, but notlow enough for home care to be desirable (about5 percent). He calculated that admission to anintermediate care unit, rather than a CCU, washighly cost effective (79).

Others who have studied monitored patientsare not as sanguine about the ability to predictlow risk. In a coronary care-oriented ICU, Thibault(248) found that 1 of 10 patients admitted forcareful monitoring subsequently required a ma-jor ICU intervention. Using primarily subjectivecriteria, he could not predict which of the moni-tored patients would do well.

Teplick, et al. (246), studied patients routinelyadmitted to a surgical ICU after uneventful, ma-jor surgery of various types. Using a fairly con-servative definition of benefit, the authors foundthat overall, 33 percent of the patients benefitedmedically from an overnight stay in the ICU.There was a broad range in the percentage of pa-tients who benefited from ICU care across typesof surgery, from 44 percent of patients who hadvascular surgery to no patients who had anteriorcervical Iaminectomies. A number of the unan-ticipated complications were immediately life-threatening. Furthermore, using both a preopera-tive risk assessment and an evaluation of intra-operative problems, the authors were unable toidentify the patients within each surgical categorywho were more likely than others to develop seri-ous postoperative problems.

Another study of the same ICU, however, foundthat less than 1 percent of patients routinely ad-mitted overnight to the ICU for certain other sur-gical conditions suffered significant adverse post-operative effects (220). These contrasting findingsdemonstrate the importance of stratifying eventhe monitored ICU patients in order to determine

Ch. 5—Outcomes of lntensive Care: Medical Benefits and Cost Effectiveness Ž 39

which subgroups of monitor patients do well with-out routine admission to the ICU.

Attention has also been focused recently on pa-tients who may be discharged from the ICU pre-maturely. Schwartz (220) found that 15 percentof patients electively discharged from the ICU,and 23 percent of patients transferred out of theICU because of lack of space, suffered a signifi-cant adverse effect on the surgical floor. Adverseeffects included death, return to the ICU, orresidence in hospital 1 month after completion ofthe study. The researchers also found that approx-imately one-third of patients undergoing ab-dominal vascular surgery developed serious res-piratory and/or circulatory conditions after dis-charge from the ICU. They did not speculate onwhether outcomes for these patients would havebeen different had the complications occurred inthe ICU.

In a retrospective chart review, Franklin (82)noted that 62 percent of readmission to a mixedICU might have benefited if they had not beendischarged from the ICU initially. The authors didnot indicate whether the patients readmitted tothe ICU differed in any predictable manner from

patients who did not need to be readmitted. Nordid the study address how many lives were lostbecause of early discharge. Mulley (163), who rec-ommended identification of low-risk patients forearly transfer from the ICU, acknowledged that2 percent of the low-risk group had major com-plications during their stay in the ICU that wouldhave occurred after transfer if an early transferpolicy had been in effect.

By stratifying ICU-monitored patients, it maybe possible to reduce or eliminate ICU stays forsome patients with a low risk of resulting adverseeffects. This risk may, in fact, be lower than therisk of iatrogenic ICU illness for some patients.At the same time, other moderately sick ICU pa-tients are probably discharged too soon or not ad-mitted to the ICU at all because of lack of bedspace or recognition that the patients are at riskfor serious complications. As a result, they suf-fer avoidable adverse health effects.

Work is only now beginning on attempts to pre-dict which ICU discharge patients are most likelyto suffer adverse effects on the regular medicalor surgical floor.

ADVERSE OUTCOMES OF ICU CARE

Iatrogenic Illness

The possibility that the adverse effects of ICUcare may outweigh the potential benefits for somepatients is being increasingly recognized (176,275).However, the rates of iatrogenic illness and otheruntoward physical and psychological reactions toICU care are not known with any precision (176).

As with the problems of measuring the positiveeffects of ICU care, it is difficult to distinguish be-tween the negative effects that occur among crit-ically ill patients regardless of location and thosethat are specific to the ICU.

An iatrogenic illness is any illness or otherharmful occurrence that results from a diagnos-tic procedure or therapy that is not a natural con-sequence of the patient’s diseases (239). The ma-jor iatrogenic complications that result from pro-longed ICU care include nosocomial infections

(defined below), stress-induced gastrointestinalbleeding, alterations of consciousness associatedwith metabolic disorders, coagulation disordersassociated with multiple transfusions and infec-tion, drug interactions, complications of intra-vascular catheterization, complications of pro-longed endotracheal and nasogastric incubation,and sleep disorders and psychoses (41,275). Someof these complications, such as drug interactionsand bleeding, would likely occur in seriously illpatients regardless of location. Nosocomial infec-tions and various psychological reactions are oftena result of the ICU itself.

Recently, Steel found that 36 percent of patientson the medical service of a university teachinghospital had an iatrogenic illness (239). In 9 per-cent of the cases, the incident was life-threatening

. —

40 ● Health Case Study 28:: Intensive Care Units: Costs, outcome, and Decisionmaking

or produced considerable disability. In 2 percentof the cases, the iatrogenic illness was believedto have contributed to the death of the patient.The authors did not specify which problems spe-cifically occurred within the CCU or ICU sectionof the medical service. Nevertheless, a number ofthe complications came from drugs, such as lido-caine, and procedures, such as Swan-Ganz cath-eterization, that are, for the most part, only usedin ICUs.

In a different teaching hospital, Abramson (3)identified 145 reports of significant adverse oc-currences in 4,720 ICU admissions during a 4-yearperiod. Ninety-two of these incidents were felt tobe the result of human error, and 53 were equip-ment malfunctions. However, 43 of the 92 in-cidents linked to human error involved equip-ment, mostly mechanical ventilators. Thus, abouttwo-thirds of the adverse events involved thetechnically complex equipment used in ICUs. Theincidence of equipment-related adverse occur-rences would probably be much higher if theequipment and the staff operating it were dis-persed throughout the hospital (208). On the otherhand, ICU technology may sometimes be used un-necessarily for less sick patients, producing someincidence of avoidable iatrogenic illness (198). Asnoted in chapter 7, the ICU milieu provides a biasto the use of technology, which at times may beof only marginal benefit and can produce adversereactions (242).

Finally, it is clear that the sophisticated careprovided in the ICU requires skilled nurses andother technicians. Adverse effects in ICUs havebeen particularly noted during periods of nurs-ing shortages (3,136). The ICU environment pro-duces “technology-oriented” treatment protocols(100), and physicians are less apt to tailor thera-py based on the specific skills of the nurse andtechnicians on duty or on the particular nurse-to-patient ratios during a particular shift. In otherwords, certain ICU monitoring and therapy pro-tocols may work well under ideal circumstancesbut may be particularly subject to human and me-chanical error under less favorable circumstances.

Nosocomial Infections

Nosocomial infections are infections occurringduring hospitalization that were not present, andnot incubating, at the time of hospital admission(117). All patients in an ICU are at increased riskof developing nosocomial infections (117). Therate of significant nosocomial infection in an ICUis about 20 percent, or three to four times thatof a patient on a general ward (63,173). This in-creased rate stems in part from unalterable fac-tors, including the severity of the underlying ill-ness; the greater use of invasive procedures; andthe greater use of prior antibiotic therapy, whichmay predispose a patient to a superimposed in-fection (63,117,192). However, at least part of theincreased rate of ICU infection is due to cross-infection between very sick patients in the con-fined area of the ICU (63,204). Nosocomial in-fection “outbreaks” in ICUs are not uncommon(63). Bacterial infections may be spread directlyfrom one person to another, often via personnel,or may require an intermediate reservoir, such asrespirator nebulizers or tubing (117). While dif-ficult to estimate precisely, the costs of nosocomialinfections in terms of increased morbidity, mor-tality, and hospital charges are undoubtedly sub-stantial (108).

Psychological Reactions

There is a substantial body of literature on thepsychological reactions of patients in ICUs. It ap-pears that the frequency of psychiatric syndromesis considerably less in a CCU, where patients arerelatively stable, than in an ICU, where seriouslyill patients suffer organic impairments of cerebral,renal, and pulmonary function (104,131,156).

The so-called “intensive care syndrome” (156)described a “madness,” or acute delirium, that hadoriginally been seen in the postoperative recoveryroom (168). However, many psychiatric syndromeshave been noted, from acute anxiety, fear, andsustained tension to agitated depression and acutedelirium (132).

Ch. 5–Outcomes of Intensive Care: Medical Benefits and Cost Effectiveness ● 41

The unique environment of the ICU has beengraphically implicated as a cause of the varied andoften dramatic psychological reactions:

Immobilized, weak, inhibited from moving bya network of wires and tubes which connect everyorifice in his body with bottles and machines, helies watching the light pattern move from left toright on the monitor, disappear, then start again.He listens to the suction of the draining appara-tus, the on and off of the pulmonary respirator,the hissing sound of the steam from the vaporizedoxygen; steam which sometimes clouds his visionin the tent. He adds his own fantasies to thisbewildering environment. Fear and tension mount. . . . In the ICU, the lights are on constantly, andthere is little or no change in the level or type ofsensory input. The activity, in spite of its decreasetoward early morning, remains high. Hours anddays merge and blend. Privacy is almost impos-sible. The patient is exposed; his most private actsbecome public. . . . Strangers control the ma-chines. Their authority is absolute. In this seem-ingly irrational environment, he is deprived ofany volitional control. He becomes an object

rather than a participant in the struggle for life(62).

Sleep deprivation, sensory deprivation, sensoryoverload, medications, and various emotional fac-tors related to coping with serious illness havebeen cited as causes for ICU psychiatric syndromes(38,104,131,145).

Given the dramatic behavioral responses to ICUcare, it is remarkable that most patients remembervery little about the “terror in the ICU” (216). Insurveys taken both shortly after transfer out ofthe ICU and many months later, ICU patients gen-erally remember few details of their stay (24,29,115,127,162,216). Whether due to the seriousnature of the underlying illnesses (104,127), thelack of sleep, which produces general fogginess(24,127), or a powerful psychological defensemechanism of denial called “psychoplegia” (104,216,217), survivors of ICU care generally do notcarry unique psychological scars of their ICU ex-perience.

COST-EFFECTIVENESS ANALYSIS OF ADULT INTENSIVE CARE

Cost-effectiveness analysis (CEA) is intendedprimarily to measure and compare the costs ofdifferent ways of arriving at similar outcomes(256). This type of analysis has not been done forICUs, because it is considered unethical to denyICU care for most ICU patients (see ch. 6). Thefew “before ICU/after ICU” studies focused onrelatively small ICU subpopulations and are clear-ly dated (99,183).

For the low-risk monitored patient, it may beethically permissible to compare ICU observationwith non-ICU observation to determine the costeffectiveness of ICU care. Both Mulley (163) andWagner (269) have projected cost savings thatwould be generated by more selective admissionand earlier discharge policies. Using conservativeeconomic assumptions, Mulley found that a moreselective policy would result in a 6-percent reduc-tion in ICU charges. Similarly, Wagner estimateda 4-percent reduction in total ICU days with ear-lier discharge of low-risk patients. Neither authoraccounted for the possibility that earlier transferfrom the unit might either increase or, conceiva-

bly, decrease the rate of major complications,which, in turn, would affect costs (163). Finebergestimated that for patients with about a 5-percentprobability of having sustained a myocardial in-farction, admission to a CCU would cost $2.04million per life saved and $139, ooo per year oflife saved, as compared to care in an intermediatecare unit (79). Teplick (246) concluded that rou-tine overnight ICU admission for postoperativepatients at an additional cost of $3OO would re-duce overall patient costs if only 13 of the 88routinely admitted patients in their study whobenefited from the ICU were prevented from be-coming critically ill.

Another factor in considering the overall costeffectiveness of earlier discharges of low-risk ICUpatients is the fact that the costs of caring for thesepatients on the regular floors would increase,mostly because of the need for additional nurs-ing, probably from private duty nurses (97,220).There might also be a need for additional monitor-ing equipment on the regular floors. Finally, pro-jecting savings based on charges probably over-


estimates the savings from early discharge oflow-risk patients because of the cross-subsidiza-tion that is reflected in the ICU charges (see ch. 6).

Attempts have been made to assess averagecharges necessary to achieve one survivor forvarious subpopulations of ICU patients. For ex-ample, Parno (175) found that hospital chargesin 1978 dollars for a survivor alive 2 years afterdischarge averaged $15,000, with a range of $1,650for drug overdose patients to $46,000 for renalmedical patients. In a population of the most crit-ically ill surgical ICU patients, Cullen (50) foundthat in 1977-78 dollars, it required $71,000 in hos-pital charges to achieve a survivor alive 1 yearafter hospital discharge. Neither additional post-hospitalization expenses nor physician chargeswere included in this estimate. For the categoryof illness that includes gastrointestinal bleeding,cirrhosis, and portal hypertension, Cullen foundthat it cost $260,000 to achieve one survivor.

An interesting variation on this approach is tolook at “life-years” saved (134). The method isnot a true cost-benefit analysis (CBA), however,since CBA requires that benefits be assigned amonetary value in order to provide a direct com-parison of the costs and benefits of a particulartechnology (256). Assigning monetary values tothe varied and controversial outcomes of the ICUhas not been done. Theoretically, the life-yearssaved method could be extended into CBA. Rec-ognizing that longevity is generally considered abenefit, Bendixen used the life-year saved modelto view the cost of ICU care in relation to pre-dicted remaining lifespan. He used the followingequation:

cost = (cost per day) X (duration of stay)(survival fraction) X (predicted remaining lifespan)

This approach assumes not only that survivalis a benefit, but also that survival value is a multi-ple of survival time, i.e., that 2 years of survivalhas twice the value of 1 year of survival. The ap-proach theoretically permits one to weigh the fac-tors of a patient’s age and the prognosis associ-ated with chronic disease. The formula, however,does not discount the future value of costs andbenefits into present dollars; in essence, it over-states the importance of predicted remaining life-span (256).

The unavailability of disease-adjusted actuarialdata for diagnostic subgroups makes predictionof life expectancy for chronic diseases inexact(215). ICU survival fraction and predicted remain-ing lifespan are the major determinants of cost ef-fectiveness according to this formula. Using thisapproach in 1977, Bendixen estimated a cost-per-year saved of $84 for barbiturate overdose and$180,000 for hepatorenal failure.

When better estimates of life expectancy for pa-tients with chronic illnesses become available, thiscost-effectiveness approach may be more useful.Nevertheless, application of this approach docu-ments the importance of the underlying diseaseprocess and the patient’s age in determining thecost effectiveness of ICU care (215). The formulacurrently does not permit quantitative consider-ation of quality of life, which is obviously impor-tant for patients with debilitating chronic illnesses(18). Methods for adjusting life-years saved forquality of life have been attempted (213), but havebeen criticized as representing “bad science” andfor ignoring considerations of justice and equity(7).

6Payment for ICU Service;

6.Payment for ICU Services

TRADITIONAL HOSPITAL REIMBURSEMENT

Derzon (60) emphasized that several featuresof the American health financing and paymentsystem operate to reinforce use of expensive tech-nology, such as intensive care units (ICUs). Thesefactors include payment certainty, consumer in-surability, government assumption of risk, andbenefits based on “medical” necessity. These fac-tors have provided the major impetus to expan-sion of ICUs in the 1960s, 1970s, and early 1980s.

Most patients are covered for all or part of theirhospital costs through private or government in-surance. With the exception of small indemnityinsurance companies, which pay a fixed dollaramount per day or a fixed coinsurance rate basedon hospital charges, most private insurance com-panies pay full hospital charges, sometimes afteran initial deductible. Other major payers, in-cluding many Blue Cross plans, Medicare, andMedicaid, have traditionally reimbursed hospi-tals on the basis of the actual cost of providingthe service to their beneficiaries.

To the extent that many insurers distinguishICU care from other hospital care for purposesof reimbursement, the result has been both toreward ICU care and to penalize intermediate levelspecial care units. For example, until 1982, Medi-care paid hospitals different per diem rates foronly two levels of hospital care—routine care and“special care” (ICU and coronary care unit (CCU)care. ) Levels of care below special care were reim-bursed at routine care levels. The other cost-basedpayers have tended to follow Medicare’s defini-tional guidelines (166). Also, in 1979 and 1980,as was noted in chapter 2, Medicare tightened ex-isting payment limits on routine bed costs but noton ICU bed costs—the so-called “section 223limits” (73).

Furthermore, two reimbursement mechanismsdesigned, in part, to curb unnecessary utilizationof care (including ICU care) have no impact on

ICU utilization. These mechanisms—patient co-payments and utilization review—are discussedbelow.

Patient Copayments

Little is known directly about the effect of di-rect patient payments on the utilization and costof ICU care. Cullen (56) found that only 100 of189 seriously ill patients in a Boston surgical ICUwere billed directly for any amount. The aver-age bill for patients who did get billed was $1,856,which was equal to 9 percent of their total hospi-tal bills. Cromwell (49) found that 80 percent ofICU patients in a different Boston teaching hos-pital had direct bills of less than $100, and mostof the patients, 42 percent of whom had Medi-care coverage, were well covered for the costs ofICU care. Only 2.5 percent of the sample had out-of-pocket bills above $3,000, and they were re-sponsible for 67 percent of all uncovered hospi-tal charges for ICU care. This pattern may differin other parts of the country where private insur-ance coverage is not as extensive.

Finally, Cromwell found little correlation be-tween coinsurance rates and the utilization of ICUbeds and ancillary services after the completelyuninsured patients were discounted. He did find,however, that patients with no insurance cover-age had hospital and ICU stays about half as longas those patients with more extensive insurancecoverage. Uninsured patients may exhibit a dif-ferent case mix that explains at least part of thedifference in utilization.

Utilization Review

Theoretically, hospital utilization review (UR)programs have the potential for limiting hospitalreimbursement for ICU care by denying paymentsto patients “inappropriately” in the ICU. In re-

45

— -.


ality, hospital UR programs, administered for the than on appropriate level of care within the hos-last decade in accordance with the Federal Pro- pital (234). It would be most unusual for a URfessional Standards Review Organization pro- committee or insurance company to deny pay-gram, have focused almost exclusively on whether ments for a patient in the ICU or recommendthe admission is appropriate and whether the transfer to a lower cost unit in the hospital.length of the hospitalization is necessary, rather

PROSPECTIVE PAYMENT PROGRAMS

Evidence is accumulating that State-based hos-pital prospective payment programs have beensomewhat successful in reducing hospital cost in-flation (20,45). There is almost no published data,however, on how ICUs have fared relative toother hospital services in States with prospectivepayment systems.

OTA’S analysis shows that between 1976 and1981, in the eight States that had established hos-pital prospective payment (rate-setting) demon-stration programs (Connecticut, Maryland, Mas-sachusetts, New Jersey, New York, Rhode Island,Washington, and Wisconsin), increases in ICU/CCU beds were below the national average (4).However, some of these States, including Con-necticut and Washington, had relatively highlevels of ICU and CCU beds to begin with, so adecreased rate of increase may not necessarilybe attributable to the State regulatory paymentsystem.

Indeed, as demonstrated by Cromwell andKanak (48), it is difficult to separate the effect of

the presence of a State’s hospital prospective pay-ment program from many other factors that mayinfluence hospital costs and utilization of specifictechnologies. These factors include the mix of hos-pitals, the effectiveness of a complementary cer-tificate-of-need program, the length of time theprospective payment program has been in effect,the type of rate review, and the baseline level ofcosts and services. When they reviewed the periodbetween 1969 and 1978, Cromwell and Kanak (48)did find that in some States, the presence of a pro-spective payment system appeared to retard thediffusion of ICUs. It should be noted, however,that their analysis looked at the diffusion of in-tensive and coronary units, not at ICU/CCU beds.By 1969, the majority of hospitals already hadestablished ICUs (205).

There is no systematic information available onwhether ICU length of stay (LOS) is reduced inStates with prospective payment programs.

MEDICARE’S CURRENT INPATIENT HOSPITAL PAYMENT SYSTEM

Description

Title VI of the Social Security Act Amendmentsof 1983 (Public Law 98-21) provided a dramat-ically new payment system for Medicare inpatienthospital services. A full discussion of the impli-cations of Medicare’s prospective payment sys-tem for ICU care is beyond the scope of this casestudy.1 Nevertheless, a few preliminary observa-

tions on likely effects of the system on the provi-sion of ICU care can be offered.

In brief, the current payment system is basedon the concept of diagnosis-related groups (DRGs),Under this DRG system, which began to be phasedin over a 3-year period on October 1, 1983, hos-pitals receive a fixed payment per discharge basedon the patient’s diagnosis. Hospitals that treat pa-

‘See the Office of Technology Assessment’s technical memoran-dum, entitled Diagnosis Related Groups (DRGs) and the Medicare

Program: Implications for Medical Technology, which describes thepotential impact of the new payment system on medical technol-ogy (254).

—

Ch. 6—Payment for ICU Services ● 47

tients for less than Medicare’s payments are al-lowed to keep the difference. Those hospitals thatspend more have to absorb the loss.

More specifically, under the DRG payment sys-tem, rates are set for each of 470 different DRGs.2

More complex DRGs, such as kidney transplants(DRG 302), receive much higher payments thansimpler cases, such as hernia repairs (DRG 161).Certain types of cases with complications or a sec-ondary diagnosis receive a higher payment thancases without complications. For example, heartattacks with complications (DRG 121) receive asomewhat higher payment than uncomplicatedheart attacks (DRG 122).

The DRG classification system, however, doesnot directly take into account severity-of-illnessvariations of patients who have the same primarydiagnosis. For example, in one teaching hospitala group of only four patients in DRG 206 (dis-orders of the liver, excluding malignancy, cir-rhosis, alcoholism, and hepatitis, age less than 70without complications or comorbidities) had arange of charges from $1,171 to $114,515 (118).

The U.S. Department of Health and HumanServices (DHHS), which proposed the DRG-basedpayment system, has recognized that within someDRGs, some patients may be more severely ill(264). DHHS argues that in DRGs where sever-ity of illness is strongly associated with treatmentcost, most hospitals will have patients who ex-hibit a range of severity levels, thus producing onbalance only minor financial advantages or dis-advantages to most general hospitals. In addition,as enacted, the DRG payment system provides foradditional payments in “outlier” cases—atypicalcases which have particularly long lengths of stayor which are unusually expensive. For those cases,the additional costs, which must range between5 and 6 percent of the total national payments fordischarges in a year, are based on the marginalcost of care beyond established LOS or cost cut-off points.

Regulations implementing the new law werepublished on January 3, 1984 (75). Under the reg-ulations, a discharge could become either a “day”

‘Although there are 467 DRGs for clinical conditions, there are3 additional categories for payment purposes. Two of these cate-gories involve reassigning the original classification and have norates assigned.

outlier or a “cost” outlier. A day outlier is a dis-charge that exceeds the mean LOS for dischargeswithin that DRG by the lesser of 20 days or 1.94standard deviations. The mean LOS for each DRGare included in the regulations. If the dischargeis considered a day outlier, the hospital will bepaid 60 percent of the average per diem Federalrate for the excess days considered medically nec-essary. The 60-percent factor is intended to ap-proximate the marginal cost of care for the ex-cess days. However, a hospital will not be paid60 percent of the actual costs of outlier days, butrather 60 percent of the average DRG per diemrate based on the DRG price.

Additional payments will be made for costoutliers if a hospital requests such payment andif the cost of a discharge exceeds the greater of1.5 times the wage-adjusted Federal DRG paymentor $12,000. Additional payment will equal 60 per-cent of the difference between the hospital’s ad-justed cost for the discharge and the cutoff amount.The adjusted cost will be determined by multiply-ing the billed charges for the covered services by72 percent, the charge-to-cost adjustment factor.Importantly, a discharge will not be considereda cost outlier if it qualifies as a day outlier.

DHHS estimates that initially 5.1 percent of alldischarges will qualify as day outliers and only0.9 percent as cost outliers. Indeed, DHHS inten-tionally established criteria that would result insubstantially more day outliers than cost outliersfor two reasons: the information necessary to de-termine day outliers is automatically and routinelyavailable in the bill processing system; and pay-ments to hospitals that may simply be high-cost,inefficient providers of care will be minimized.

Another payment decision in the DRG paymentregulations could have specific relevance to ICUcare. Hospitals transferring a patient to anotherinstitution are paid a per diem rate based on theaverage LOS for the DRG treated. Full paymentfor the DRG treated is made to the hospital fromwhich the patient is finally discharged. For exam-ple, if hospital A treats a patient in a DRG withan average LOS of 10 days for an initial 4 daysand then transfers the patient to hospital B, hos-pital A will receive 40 percent of the DRG pay-ment and hospital B will receive a full DRG pay-ment, regardless of the actual LOS in hospital B.

48 ● Health • Study 28: Intensive Care Units: Costs, Outcome, and Decisionmaking

Finally, Medicare previously reimbursed hos-pitals for the reasonable costs of capital, whichinclude depreciation, interest, and rent. Under thecurrent law, capital expenses are specifically ex-cluded from the prospective payment system andcontinue to be reimbursed on a reasonable costbasis until October 1, 1986. At that time, Con-gress will decide whether to continue to pay rea-sonable costs or to incorporate payment for cap-ital into the DRG system.

Medicare Utilization of ICUs by DRGs

Because ICU patients often have multiple diag-noses and suffer serious physiologic abnormalitiesthat frequently do not correspond to diseaseentities, the DRG classification scheme may bepoorly suited to describing ICU patients. Never-theless, a preliminary analysis has been performedof the DRG case mix of Medicare ICU/CCU3 pa-tients based on available Health Care FinancingAdministration (HCFA) data for 1979 and 1980(259,260). For the purpose of this analysis, multi-ple DRGs for the same primary diagnosis werecombined. For example, DRGs 121 to 123—myo-cardial infarctions with differing clinical charac-teristics—were considered together.

Of the 15 DRG-based primary diagnoses withthe longest average LOS in special care in 1979,14 involved operating room procedures. The ex-ceptions were the DRGs for myocardial infarc-tions. Another way to view DRG case mix is toconsider special care as a percentage of total hos-pital stay. Of the 16 primary diagnoses in whichspecial care represented at least 10 percent of thetotal hospital stay, 9 were medical diagnoses.However, these medical diagnoses were mainlyfor the cardiovascular system—mostly related tocoronary artery disease. One can conclude thatfor DRGs involving certain operating room pro-cedures and coronary artery disease, stays inspecial care units are standard and, therefore, cap-tured by the DRG category. For example, 92 per-cent of cases for cardiac valve procedure withpump support (DRG 105) included special care.For many common surgical procedures and car-diac diagnoses, special care was utilized in more

than !70 percent of cases. For the remaining,predominately medical diagnoses, the DRG cat-egory does not reflect the use of special care unitsfor the more severely ill patients with that prin-cipal diagnosis. For example, a number of ICUstudies indicate that gastrointestinal bleeding inpatients with cirrhosis is one of the ICU problemsassociated with both long ICU lengths of stay andhigh cost (40,50). Yet, the DRG for this condi-tion, “cirrhosis and/or alcoholic hepatitis” (DRG202), has a mean special care length of 0.6 days,or only 4.5 percent of the average total hospitalLOS for discharges with this DRG.

A somewhat different picture of ICU use emergeswhen frequency of diagnosis is taken into account.By multiplying the number of discharges in the20-percent MEDPAR sample4 by the average LOSin special care, the number of special care daysby diagnosis can be estimated. Table 10 showsthe 15 diagnoses which use the most special caredays. Again, cardiovascular disease predomi-nates. However, diseases involving operatingroom procedures become less important as ma-jor special care diagnoses.

Applicability of DRGs to ICUs

As noted above, the current DRG classificationsystem may not be suitable for describing certaintypes of patients cared for in ICUs. DRGs arebased on a principal diagnosis, with some addi-tional categories available for patients with asingle substantial secondary diagnosis, called a“comorbidity,” or a significant “complication. ”Yet ICU patients often have multiple, seriousunderlying illnesses. In one study (265), ICU pa-tients had on average over four major diagnoses,and the high-cost nonsurvivors had over six diag-noses. For these patients, designation of a prin-cipal diagnosis is likely to be arbitrary andunreliable at times. Furthermore, the additionaldiagnoses would not be accounted for.

As discussed earlier, many cardiac diseases,particularly those involving coronary diseases,and many of those surgical diagnoses involvingoperating room procedures, include stays in theICU and CCU as a matter of routine. For exam-

3Available HCFA data combines ICU and CCU patients as specialcare patients. 4For a description of HCFA’s MEDPAR data base, see ch. 4.

Ch. 6—Payment ICU Services ● 4 9

Table 10.-Estimated Number of Special Care (lCU/CCU) Days by Primary DiagnosisBased on HCFA 20-Percent Sample of Medicare Discharges, 1980a

Special care Routine care

Percent of totalDiagnosis DRG Total days hospital days Total days

1. Myocardial infarctions . . . . . . . . . . . . . . . . . . . . . ..121-123 176,963 33 ”/0 362,0132. Atherosclerosis . . . . . . . . . . . . . . . . . . . . . ........132-133 103,781 14 625,4503. Heart failure and shock . . . . . . . . . . . . . . . . . . . . . . 127 87,347 11 693,4394. Pneumonia and pleurisy . . . . . . . . . . . . . . . . . . . . . 89-91 78,211 13 555,1155. Unrelated OR procedure . . . . . . . . . . . . . . . . . . . . . 468 66,451 9 734,6846. Arrhythmia . . . . . . . . . . . . . . . . . . . . . . . . . . .......138-139 54,464 21 200,9237. Angina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140 53,926 22 194,6538. Ungroupable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 470 51,100 6 734,6849. Cerebrovascular accident . . . . . . . . . . . . . . . . . . . . 14 42,120 5 715,668

10. Chronic obstructive pulmonary disease. . . . . . . . 88 41,203 8 467,82511. Pacemaker implant. . . . . . . . . . . . . . . . . . . . . . ....115-118 37,109 30 83,58612. Coronary artery bypass surgery . ..............106-107 30,169 32 64,96813. Pulmonary edema and respiratory failure . . . . . . . 87 28,371 25 83,27614. Major bowel OR procedure . ..................148-149 27,191 10 242,18815. Major reconstructive vascular procedure . ......110-111 20,543 18 94,077aMultiple DRGs for the same primary diagnosis were combined for this analysis.

SOURCE: Office of Technology Assessment.

pie, in the United States it is standard to treat allpatients with acute myocardial infarctions (heartattacks) in CCUs or ICUs. The average DRG priceper discharge will reflect the portion of the hos-pital costs consumed in the higher cost special careunit.

However, the DRG categories for many medi-cal diagnoses are so broad that ICU days repre-sent only a small proportion of total hospital days.For example, in 1980, hospital stays for chronicobstructive pulmonary disease (DRG 88) and forcirrhosis of the liver (DRG 202) averaged only0.82 and 0.60 days of intensive care, respectively(260). Yet, the sick patients within these DRGsmay spend many days in the ICU and use moretotal hospital resources than patients within DRGsthat include a much longer average special carestay. In other words, it appears that variationsin severity of illness are particularly great for non-coronary, medical diagnoses that represent themedical patients in medical or mixed ICUs. Like-wise, the DRG classification system does notsatisfactorily account for patients with a primarysurgical diagnosis who suffer major medical com-plications. For example, in a series of criticallyill surgical patients, Cullen (54) found that renalfailure (a costly medical complication) was apowerful predictor of ultimate survival. Many cli-nicians might agree that renal failure had becomea patient’s major clinical problem, but the DRG

system requires that the operating room proce-dure take precedence in DRG assignment. Thepresence of renal failure, then, would not signifi-cantly affect DRG payment.

Unfortunately, there is no data base availableto test whether there are systematic differences byhospital type in severity of illness in ICU popula-tions. DHHS’S initial evaluation found that teach-ing hospitals do have higher costs per case, sug-gesting, at least in part, that they treat moreseriously ill patients (75). Survey tapes of theAmerican Hospital Association document thatmajor teaching hospitals do have 50 percent moreICU days as a percentage of total hospital daysthan nonteaching hospitals (106). These additionalICU days probably explain some of the highercosts per case in teaching hospitals.

However, without an accurate severity-of-illness measure, one does not know whether theadditional ICU use in teaching hospitals representsthe presence of a sicker population or a differentthreshold for transferring and maintaining pa-tients in the ICU. Likewise, differences in resourceuse between ICUs may represent differences inseverity of illness or differences in intensity andstyle of care. Preliminary results from 15 tertiarycare hospitals recently surveyed by Knaus’ groupat George Washington University in Washington,DC, suggest that severity of illness, in fact, ac-

50 ● Health Case Study 28:: ]ntensive Care Units: Costs, outcome, and Decisionmaking

counts for a substantial portion of the differencesin ICU resource use for patients with the same pri-mary diagnosis (268).

Under Medicare’s DRG payment system, manycostly ICU cases will likely become outliers forwhom only marginal costs above a day or costthreshold are paid. As was described earlier, bydesign, day outliers will predominate over costoutliers. Utilizing HCFA’S 1980 MEDPAR data,OTA has estimated that 12 percent of cases in-volving special care would be classified as dayoutliers, in comparison to 9 percent of total cases.By definition, the marginal costs for day outliersare calculated based on the DRG price, not theactual cost for that patient. Yet, as was noted inchapter 3, the cost per day in the ICU is over threetimes greater than the cost for a general hospitalday. Thus, a hospital may receive far less thanthe actual marginal costs for caring for a long-term ICU patient. In short, the outlier paymentrules generally favor less severely ill, non-ICU,long-stay patients, such as those with strokes orcertain types of cancer, over more severely ill,long-stay ICU patients.

It would appear, then, that severely ill Medi-care patients, especially if they are in the ICU,will be “revenue losers” to the hospital, even withan outlier policy in effect. This fact, combinedwith the lack of a financial penalty for transfer-ring patients to a second hospital, may result inmore interhospital transfers of the sickest ICU pa-tients to tertiary care hospital ICUs. A region-alized system of ICU care that is common in someparts of Europe might thereby be stimulated inthe United States, perhaps desirably. It should benoted, however, that unless either a severity-of-illness measure or a different outlier policy isadopted, the tertiary care hospital receivingseverely ill transferred patients will be likely tolose financially. These hospitals would then facethe dilemma of either not accepting these patientsin transfer or of accepting these patients into theirhigh-quality ICUs at a financial loss. At the ex-treme, tertiary care hospitals could, in effect,become large ICUs (212). Public hospitals and

some teaching hospitals, however, may simply beunable to sustain the costs of ICU care and beforced to ration care even more strictly than theydo now (212).

The 3-year capital cost exclusion in the DRGlaw is not likely to affect ICUs, at least in the shortrun. ICU care is relatively costly largely becauseit is so labor-intensive. Common ICU technol-ogies, such as cardiac monitors, respirators, pul-monary artery (Swan-Ganz) catheters, centralfeeding lines, etc., are labor-generating rather thanlabor-reducing technologies, because they requirefairly constant attention.

As was noted in chapter 5, the monitor-onlyand other less severely ill ICU patients have beensubsidizing the care of the most critically ill ICUpatients. Under the DRG system, there may bea new incentive to treat monitor patients on reg-ular floors or perhaps in intermediate care units.In addition, hospitals will attempt to pass on tocharge payers the unreimbursed cost of ICU careto Medicare patients. The additional “pass-on,”combined with nonadmission and earlier dischargeof some of the less sick ICU patients, should re-sult in substantially increased charges for an ICUday. The current 2.5:1 ratio of ICU bed chargesto routine bed charges (71) will correspondinglyrise.

In short, ICU care to Medicare patients will notbe financially rewarding to hospitals under DRGpayment. Almost all ICU cases are likely to be“losers” to the hospital—ICU days are about 3 to3.5 times more costly than non-ICU days and ICUpatients have longer hospital stays than non-ICUpatients. The new incentives of the DRG paymentsystem will be imposed on an ICU decisionmak-ing environment in many hospitals in which thecosts of care had previously been a relatively mi-nor concern. The implications of the collision be-tween the hospital’s new interest in reducing thecost of ICU care and a decisionmaking environ-ment that results in expanding ICU care will bediscussed in chapters 7 and 8.

Ch. 6–Payment for ICU Services ● 51

PHYSICIAN PAYMENT

In a fee-for-service system that pays on the basisof “usual, customary, and reasonable” standards,“technological and procedural” medicine has beenrewarded (202). The ICU is a focal point for tech-nological and procedural medicine within the hos-pital. Incubation, use of respirators, and arterialline placement are among the many ICU proce-dures that generally require ICU admission andnumerous followup ICU visits by the patient’s pri-mary and consulting physicians. Payment for ICUprocedures and visits is generally high and israrely questioned by insurers (166).

Patients in ICUs have multiple diagnoses andoften multiple organ system failures. It is not sur-prising, therefore, that ICU patients have manyphysicians. Murata and Ellrodt (164), found in alarge community hospital in which the ICU hadfull-time housestaff that at least one physicianconsultation was requested in 65 percent of theprivate admissions. In this study, private ICU pa-tients had an average of nearly 2.5 physicians car-ing for them, in addition to round-the-clock house-staff coverage.

The situation is somewhat different in teachinghospitals and other large nonteaching hospitals.In these hospitals, there is usually one or morefull-time staff physicians who help administer theICU, provide staff education and, to varyingdegrees, participate in direct patient care. Al-though the specific payment method adopted bya particular hospital may be unique, compensa-tion arrangements can generally be classified intoone of four categories: fee-for-service, percentage

of income arrangements, salary only, and com-binations of the first three (77). Straight salaryarrangements represent the only compensationmethod that does not include a financial incen-tive component (77). In terms of ICU care, ICUstaff physicians who are not paid on strict salarybasis have a financial incentive to keep the unitfilled and to perform procedures and provide tech-nical services (166). Surveys on the prevalence ofvarious compensation methods in U.S. hospitalshave not specifically included ICU physicians (77).Similarly, the extent of ICU physician double bill-ing (submitting fees for reimbursement for pro-fessional services while receiving salaries foradministrative and educational activities, whichare reimbursed as a hospital cost) is unknown.

Under DRG payment, ICU staff physicians mayface conflicting payment incentives unless they arepaid on a strictly salary basis. Given the high costsof ICU care, it may be in a hospital’s interest toincrease the cost control function of ICU staff phy-sicians and to pay them salaries as their primaryform of remuneration. Hospitals could even pro-vide incentive bonuses for reduced costs or de-creased lengths of stays. In addition, hospitalswhich do not currently have ICU directors mayfind it in their economic interests to hire one tomonitor the costs of care provided by private phy-sicians who admit patients to the ICU. Thus, itis possible that a hospital’s attempt to reduce hos-pital ICU costs, paid under Part A of Medicare,might also indirectly result in a reduction in PartB physician payments.

7The ICU Treatment Imperative

7 ■

The ICU Treatment Imperative

INTRODUCTION

Medical decisionmaking involving seriously illpatients is often difficult and uncertain. In manycases, physicians do not know ahead of timewhether the treatment they prescribe will benefittheir patient. Physicians in the intensive care unit(ICU) frequently face the similar dilemma of notknowing whether to employ available life-supportfor critically ill patients and whether or when towithdraw such support when it seems clear thatcontinued treatment will merely prolong his lifewith no improvement in his grave condition. Onereason ICU decisionmaking is so difficult is thatit is so successful; most ICU patients survive. Yet,it is also clear that in some cases ICU care isprovided—at a very high cost—to patients whoare beyond help. In other cases, ICU care maybe immediately lifesaving but results in returningthe patient either to a condition that still has avery short life expectancy or to a condition witha severely limited functional status.

At the present time, there is no reliable way topredict outcome for most critically ill patients, and

therefore, it is usually reasonable and appropri-ate to initiate intensive care treatment for severelyill patients. However, because of certain factorssomewhat unique to the ICU, care is sometimescontinued beyond the point of benefit to thepatient.

This chapter explores those factors—includingthe underlying chronic illnesses suffered by manyICU patients, the diffused nature of decisionmak-ing that often prevails in the ICU, the frequentinability of patients themselves to make informedchoices about continuing therapy in the face ofa hopeless situation, the concern over the possi-bility of malpractice lawsuits or even criminalprosecution and the inability to predict out-come—that often lead physicians to provide life-support after the initial rationale for doing so nolonger exists. Together, these factors create anICU treatment imperative.

THE HIGHLY TECHNOLOGICAL NATURE OF ICU CARE

The “technological imperative,” which has beendefined by Fuchs as the desire of physicians to doeverything that they have been trained to do,regardless of the benefit-cost ratio (84), flourishesin the ICU. ICU technology can dramatically andconsistently sustain life for long periods of time.The ICU is a prototype of what Thomas has calleda “halfway technology, ” one that attempts tocompensate for the incapacitating effects of cer-tain diseases whose courses one is unable to af-fect. It is a technology designed to make up fordisease or to postpone death (250). Many of theindividual technologies used in an ICU, includingrespirators, defibrillators, and balloon pumps,

sustain vital functions but do little to correctunderlying disease processes.

In a well-functioning ICU, patients rarely dieimmediately of respiratory failure or circulatorycollapse, because the available technology can de-lay these complications (50). Some patients, par-ticularly those with the common ICU problemsof cardiovascular, respiratory, and necrologicfailure (139) have their vital functions sustainedby technology so as to forestall death, but theirbasic disease or diseases do not improve. For somedisease processes, then, ICU care does not changethe ultimate outcome, but rather results in a pro-

.55

——

56 ● Health Case Study 28: ]ntensive Care Units: Costs, Outcome, and Decisionmaking

longed, yet inexorable course, with death occur-ring sometimes from complications of ICU care(135) or after a decision is finally made to ter-minate the special life-support.

Measurements and monitoring are often pur-sued as ends in themselves in the ICU (198). Pa-tient care may become depersonalized. As onecritical care specialist noted, the paradox is thatICU staff treasure life highly and go to any lengthto salvage lives, yet often ignore, or actuallydebase, the very qualities that render patientsuniquely human (35). The technological impera-tive, which frequently results in more effectivemethods of managing very sick patients, can leadto the uncritical adoption of harmful therapies onthe assumption that the most critically ill have lit-tle to lose from new approaches (198). In addi-tion, new ICU therapies that are demonstrably ef-ficacious in expert hands for specific problemsmay become widely adopted and routinely usedin situations and under conditions where demon-stration of their effectiveness is absent (243).

Physicians who become intensive care special-ists—“intensivists” —by predilection and trainingare generally believers in technological interven-tion (95). ICU-oriented physicians naturally are

THE NATURE OF ICU ILLNESSES

As was noted in chapter 5, diseases of the car-diovascular, respiratory, and neurological sys-tems, both medical and surgical, predominate inthe ICU. Failure of these systems often results inacute respiratory distress, which is manifested bysevere smothering or “air hunger, ” and circulatorycollapse or shock, which results in altered statesof consciousness. Even when the impulse on thepart of the medical professional is simply to makea desperately sick patient more comfortable andnot to initiate heroic measures in an attempt toreverse the illness, that impulse may require theuse of the full panoply of ICU technologies, par-ticularly the respirator. Some patients with can-cer and other chronic debilitating illnesses may

believers in the highly complex technology thatthey have mastered and often save lives thatwould have been lost under non-ICU conditions.Likewise, some nurses who choose ICU-basedcareers tend to be therapeutic activists, not proneto accepting the inevitability of a patient’s de-teriorating condition (278).

The highly technical nature of ICU care itselfaffects the way in which life and death decisionsare made. The most critically ill patients havemultiple organ systems failure and receive multi-ple interventions. The very exacting nature of thisform of patient management results in standardprotocols of treatment, perhaps at increased ex-pense (100), and in concentration on the detailsof treatment.

In such situations, the fundamental considera-tion of the long-term benefits to the patient re-ceiving care is often overlooked among the seem-ingly endless technical decisions that are madethroughout the course of an ICU stay. Yet, themost critically ill patients, who require the mostconcentrated focus on the details of day-to-daymanagement, are precisely those for whom fun-damental likelihood and quality of survival ques-tions are most appropriate.

be cared for outside the hospital, perhaps inhospices, with appropriate use of pain medicationand emotional support. Many terminal illnesses,however, produce symptoms that cause severedistress to the patient and that are frightening totheir families. The need for relief often results inhospitalization and treatment in the ICU. For ex-ample, symptoms of smothering from emphysemacannot be treated with medication alone—at least,not without the very real possibility of depress-ing the patient’s respiration to the point of risk-ing immediate death (102). “Naturalness,” there-fore, may have to be sacrificed for comfort, whichat times can only be achieved with ICU manage-ment and technologies (191).

Ch. 7–The ICU Treatment Imperative ● 57

“ICU diseases” often develop rapidly-some-times in seconds. When a severely ill, perhaps dy-ing patient is seen in an emergency room or ona medical floor, physicians, who are often not fa-miliar with the patient, naturally and appropri-ately attempt resuscitation (179). Frequently, thebasic physiological and other clinical data whichare necessary for a medical judgment on the sever-it y and likely outcome of an illness cannot be ac-quired before admission to the ICU (55).

With some terminal diseases there is time to an-ticipate and plan the degree and nature of in-

tervention in the event of a sudden deteriorationin the patient’s condition. Because end-stageemphysema, severe heart disease, or generalizedarteriosclerosis are, rightly or wrongly, not con-sidered terminal diseases in the same way that can-cer is, patients experiencing a sudden decompen-sation are routinely and responsibly treated withall available technology. Once initiated, however,treatment modalities that have been initiated pri-marily to respond to acute, disabling symptomsmay become difficult to stop for the reasons de-scribed below.

TRADITIONAL MORAL DISTINCTIONS IN MEDICINE

Along with the notion that physicians shouldnot “play God, ” the traditional medical ethic hasbeen to disregard subjective views of quality oflife in making life and death decisions. In termsof ICU care, this general ethic has been char-acterized as one in which “survival is being takenas equivalent to a life saved” (64).

Underlying the other considerations which playa part in ICU decisionmaking is the generallyactivist attitude of many physicians, who may em-body a fundamental and somewhat unique at-titude of American culture. The decision to pullback is frequently more difficult to make than thedecision to push ahead with aggressive support,using the complex and sophisticated medical tech-nology available (269). As other reviewers of in-tensive care have observed, this attitude has beencaptured in T. S. Elliot’s The Family Reunion (18):

Not for the good that it will doBut that nothing may be left undoneOn the margin of the impossible.

In situations where patients are in acute distressand where decisions must be made in seconds orminutes, there are powerful reasons initially toapply a lifesaving technology. Having done so,it is often quite difficult to reverse the courseof treatment in the light of new information orthoughtful judgment because of traditional moraldistinctions in medicine. Specifically, fundamentalmoral and ethical distinctions are frequently madebetween actively causing death and allowing it to

occur by declining to intervene; between with-holding and withdrawing treatment; and betweenordinary and extraordinary treatment (191).

Many primary decisionmakers in the ICU feel,for example, that having decided to put a patienton a respirator, one is committed to its continueduse and thus make a fundamental distinction be-tween intentionally withholding and activelywithdrawing the respirator. Likewise, while someprominent medical ethicists have abandoned thedistinction between “ordinary” and “extraordi-nary” obligations to dying patients, physiciansgenerally continue to use the distinction (160). Theordinary-extraordinary distinction has been af-firmed consistently in Catholic moral theology,notably in a major address by Pope Pius XII in1957 (185). Certainly, most of the public considerthat there is a difference between care that is com-mon and reasonably simple and care that isunusual, complex, expensive, and uses elaborate“unnatural” technology. Similarly, most physi-cians consider intravenous fluids to be ordinaryand standard therapy and resuscitation an ex-traordinary measure (160). Other physicians,however, consider even the use of respirators andother common ICU technologies to be routine, or-dinary treatment (191). Many physicians do notuse the ordinary-extraordinary distinction at all,but rather fundamentally consider whether anintervention, however invasive, is “medicallyindicated. ”

— —. —

58 ● Health Case Study 28: Intensive Care Units: Costs, outcome, and Decisionmaking

As the President’s Commission for the Studyof Ethical Problems in Medicine and Biomedicaland Behavioral Research has observed, invoca-tion of these moral distinctions is often so mechan-ical that it neither illuminates an actual case norprovides an ethically persuasive argument (191).Nevertheless, the moral distinctions cited abovehave great importance in ICU decisionmaking. Be-cause many different individuals typically partici-pate in ICU decisions, a range of moral attitudesare represented (278). There is a natural tendencyto defer to the individual, whether physician,nurse, or family member, who firmly holds atraditional moral view.

In addition, there is still uncertainty about thelegal interpretation of these moral distinctions. Forexample, a New Jersey appeals court recentlyreversed a lower court order allowing removal ofa feeding tube in an extremely ill, demented pa-tient with no hope of recovery.¹ The two courtsdiffered in their interpretation of whether naso-gastric feedings (nourishment) constituted ordi-nary or extraordinary treatment for this particu-lar patient in question.

A California appellate court, on the other hand,in vacating a lower court’s reinstatement of mur-der charges against two physicians for terminatingcertain treatments for a patient they diagnosed ashopelessly comatose, explicitly rejected the dis-tinction between ordinary and extraordinary care.The court, rather, invoked an ethical measure of“proportion,” writing:

Proportionate treatment is that which has atleast a reasonable chance of providing benefits to

‘See In re Conroy, 188 A/.}, Super. 523, 532 (Ch. Div. 1983).

the patient, which benefits outweigh the burdensattendant to treatment. Thus, even if a proposedcourse of treatment might be extremely painfulor intrusive, it would still be proportionate treat-ment if the prognosis was for complete cure orsignificant improvement in the patient’s condition.On the other hand, a treatment course which isonly minimally painful or intrusive may nonethe-less be considered disproportionate to the poten-tial benefits if the prognosis is virtually hopelessfor any significant improvement in condition.z

Ironically, as was pointed out by the President’sCommission, if there is any reason to draw amoral distinction between withholding and with-drawing treatment, it generally cuts the oppositeway from the usual formulation: greater justifica-tion ought to be required to withhold treatmentrather than to withdraw it (191). Whether a par-ticular intervention will have positive effects isoften uncertain until the therapy has actually beentried (50). If therapy is initiated and it then be-comes clear that the patient is not benefiting fromit, this is actual demonstration, rather than meresurmise, to support terminating that treatment(191). Yet, physicians who believe in a moraldistinction between withholding and withdraw-ing treatment, or who are concerned that anotherindividual or the courts would judge their actionsbased on this distinction, might choose not to uti-lize the lifesaving treatment in the first place outof concern that the treatment could not subse-quently be readily withdrawn.

‘Neil Leonard Barber, Robert Joseph Nedjl v. Superior Court ofthe State of California for the County of Los Angeles; Court of Ap-peals of the State of California, Second Appellate District, Civil No.60350; Oct. 12, 1983.

THE DIFFUSION OF DECISIONMAKING RESPONSIBILITY

Because of the nature of ICU care, many pro-fessionals become important decisionmakers, in-cluding nurses who attend to the patient full time,housestaff, consultants, and the patient’s personalphysician. In larger hospitals, there are frequentlyone or more ICU-based physicians in attendancewho also are involved in the decisionmaking proc-ess. In some ICUs, particularly in large, teaching

hospitals, the primary legal responsibility for apatient’s care is transferred to the ICU or to anICU-oriented specialist (165,244). often, the pa-tient’s personal physician feels intimidated by theclinical complexity and the bureaucracy perceivedin the ICU and gives up an active role in decision-making (136). As a result, the patient’s personalphysician, who often has the best understanding

Ch. 7—The ICU Treatment Imperative . 59

of the patient’s baseline medical condition, qualityof life, and personal values, goals and concerns,does not participate in important decisions aboutthe care the patient receives in the ICU (194). Atthe same time, physicians who do not treat manyICU patients may have unrealistic expectationsabout what ICUs can accomplish and do notknow how or when to address fundamental issuesabout terminating particular kinds of care (243,244).

Many ICU patients enter the hospital throughthe emergency room, often in a hospital wheretheir personal physician, if they have one, is noton staff. Victims of acute trauma or sudden severeillness may not have a previously established rela-tionship with the physician(s) who is caring forthem.

ICUs in large hospitals utilize a team approachto individual patients, which is felt to result ina higher quality of care (207,208). Some havewondered, however, whether a patient cared forby an ICU team in fact has a doctor (212). Withthe team approach, decisionmaking responsibilitymay be diffused, and the difficult issue of ter-minating special care is frequently deferred or

deflected. Families who are interested in address-ing this painful issue may not know how to engagea diverse team of busy professionals in discussion.

With multiple professionals in decisionmakingroles, there may well be different medical andmoral views expressed. Unanimity among profes-sionals is desirable, especially when the issue iswithdrawing life-support (233,243). In such situ-ations, there is a natural tendency to defer to amember of the group who holds a traditionalmoral view, such as the distinction between with-holding and withdrawing treatment.

Decisions not to treat a debilitated patient ina nursing home (27), not to transfer to the ICUa patient with end-stage cystic fibrosis (58) or can-cer (253), or to choose a hospice rather than a hos-pital, can often be addressed privately by patientsand their doctors. In the ICU, however, such im-portant decisions are more visible and often con-troversial (278). When the responsible physicianaddresses the issue of termination of special life-support with a family or patient, for example,counterpressures from other physicians and nursesmay make decisionmaking extremely difficult andemotionally charged.

PROBLEMS OF INFORMED CONSENT IN THE ICU

As the President’s Commission has emphasized,the voluntary choice by a competent and informedpatient should determine whether or not a life-sustaining therapy will be undertaken or con-tinued (191). Unfortunately, the ICU environmentis ill-suited to guaranteeing patient competenceand to providing the necessary flow of informa-tion to ensure fully informed consent.

Case studies have demonstrated that patientsin ICUs, as well as other seriously ill patients, donot always act or communicate in their own in-terest (124). As noted in chapter 5, ICU patientsmay undergo acute psychological reactions tosleep deprivation, sensory overstimulation, de-pendency, and nearness of death. This is true inother life-threatening situations as well, but unlike

the ICU, there is usually sufficient time and asatisfactory environment in such cases for work-ing with the patient before taking an irrevocabledecision.

Moreover, ICU patients suffer from subtle, butreal, metabolic disturbances which alter theirjudgment. They are frequently in severe, althoughoften reversible, pain or discomfort. Furthermore,a patient on a respirator may be reasonably com-petent to give informed consent but unable tosatisfactorily communicate his or her wishes. Thisis due to the extreme difficulty ICU patients ex-perience in communicating, often because theyhave an endotracheal tube in their throat or havehad a tracheotomy, which makes verbalizing im-possible (225).

60 ● Health Case Study 28: Jntensive Care Units: Costs, Outcome, and Decisionmaking

The natural response to some situations is todefer decisionmaking, particularly with respect toterminating care, until the patient is able to giveinformed consent. Indeed, physicians may haveto salvage the life of a critically ill patient in or-der to obtain his or her informed consent to stopcare (69). Medical professionals naturally have abias toward supporting patient survival until itcan be determined that a patient is competent andthat the choice to stop treatment is truly informed(245).

When the patient lacks the capacity to give in-formed consent, the family is normally recognizedas having the authority to make a decision on thepatient’s behalf. In practice, this procedure gen-erally works well (191). Yet, in some cases, familymembers may have motivations which do notnecessarily support the best interests of the pa-tient (152,244) or, they may disagree among them-selves. Again, because of the uncertainty of whoshould make life and death decisions on behalfof an incompetent patient, physicians naturallyadopt a policy of continuing intensive care untilresolution of disputes and roles occurs.

Finally, it has been noted that patients with seri-ous acute illnesses are generally more passive and

distant from treatment decisions than patientswith chronic stable diseases (150). For many med-ical decisions, patients and their families can par-ticipate in decisionmaking with full appreciationof the medical issues involved. Because of thehighly technical nature of ICU care, however, pa-tients and families may not fully understand theimplications of the many decisions that must beconfronted in the ICU and are more prone to de-fer to physicians (280). In the ICU, the doctor’sorientation toward the patient is to be active andin control of the situation, while the patient ispassive and dependent (280). Some even considerit to be the ICU physician’s responsibility to bringa family to the point where it can look at the pa-tient’s situation from the physician’s perspective(278).

Whether the physician adopts a controlling at-titude or not, it may nevertheless take some timefor patients or their families to accept the fact thatcontinued therapy is hopeless, and the process ofinforming them of the condition places the phy-sician in a difficult position. “It is extremely dif-ficult to tell a critically ill patient that all is notgoing well” (232).

LEGAL PRESSURES: DEFENSIVE

The past decade has seen an explosion in thenumber of malpractice lawsuits brought againstmedical professionals, particularly suits chargingthat a physician was negligent in his or her dutyto provide adequate medical care. For this casestudy, malpractice is defined as a wrongful act,committed by one or more parties upon anotherperson; the injured party may seek monetarydamages from the person(s) responsible as com-pensation for an injury. The injured party mustdemonstrate that the injury was caused by con-duct which failed to conform to the “standard ofcare” for that medical problem and that class ofprovider (199).

While many malpractice claims ultimately areunsuccessful, they have caused doctors and othermedical personnel to become more cautious and,

MEDICINE

in effect, to practice “defensive medicine,” whichinvolves taking or not taking certain action moreas a defense against potential legal liability thanfor the patient’s benefit (68). Although the extentto which defensive medicine is practiced is notknown, it clearly has contributed to the provi-sion of costly, unnecessary, and sometimes haz-ardous medical care.

Physicians, under certain circumstances, mayalso be subject to potential criminal prosecution,The criminal law confines people’s freedom of ac-tion in order to protect society, not simply in-dividuals, and therefore, consent is never acceptedas a defense against the crime of murder (191).Taking innocent human life is seen as a wrongagainst the entire society, not just against the deadperson. As such, criminal prosecution is the ex-

Ch. 7—The ICU Treatment Imperative ● 61

elusive prerogative of the State and may be broughtagainst a physician whose patient died becauseof the physician’s failure to perform the duty oftreating the patient according to accepted medi-cal standards (191). Reported criminal prosecu-tions of health care professionals for killing pa-tients are rare, and it is felt that merely the threatof prosecution provides appropriate protectionagainst abuse (191).

Although all practitioners face the possibilityof a malpractice suit and, to a much lesser extent,criminal prosecution, concern is certainly greatamong those who work with ICU patients. Inlarge part, this is because ICU patients are criti-cally ill, and death, therefore, is a common oc-currence. Indeed, the physician may permit a pa-tient’s death by withholding or withdrawing aparticular treatment or technology, an action thatis likely to make the doctor feel vulnerable tosubsequent legal liability. In fact, however, thelegal problems with the treatment and nontreat-ment of terminally and critically ill patients ap-pears to have been exaggerated—there are noknown cases of liability in the United States con-cerning the withholding of medical care from aterminally or critically ill patient (272). Yet thephysician’s sense of vulnerability to malpracticelitigation is likely to increase, because decision-making in the ICU is unusually visible. The at-tending physician, the patient, and the next-of-kin have direct decisionmaking responsibility, butothers, including ICU staff, family members, andother physicians, who may have strong opinionson a life and death decision, are also involved,although less directly (278).

In the determination of standards of care bywhich to judge physician’s actions, malpracticecourts have traditionally imposed on physiciansthe duty to provide medical care at the level whichis considered usual practice within their own ora similar locality. However, with the advent ofstandardized medical training and rapid dissemi-nation of information, this “locality rule” has beenreplaced in many States by a standard of carebased on the usual practice of the national medi-cal community (199).

Thus, physicians are more likely to utilize anICU in the first place if its use is the prevailing

national standard of care for a particular medi-cal problem, Once the patient is in the ICU, thephysician’s actions in this highly specialized arenaare likely to be judged by often higher nationalstandards of care than by the standard of otherlocal practitioners. To avoid charges of negli-gence, physicians are likely to use ICUs more fre-quently and for longer periods of time than theymight otherwise feel is appropriate.

Although the threat of criminal prosecution isgenerally remote for most health professionals,it has arisen in the context of the delivery roomand the neonatal ICU, when State prosecutorsfound criminal intent to murder in cases involv-ing an abortion3 and the care of severely disablednewborns (197). Even in the adult ICU, criminalintent may be alleged by prosecutors who viewactions such as failing to resuscitate, “pulling theplug,” and “overdosing” with painkilling or sedat-ing medications as intentionally causing a person’sdeath. In 1982, for example, homicide chargeswere brought against two Los Angeles physicianswho withdrew intravenous fluids and nasogastricfeedings from a comatose patient, with the ap-proval of the patient’s family (5,133). Althoughthe California Court of Appeal ultimately ruledthat charges against the physicians be dropped(171), the case undoubtedly caused significantconcern in the medical community (169).

Defensive medicine is also a factor in decisionsto use the ICU in routine, monitor cases. Wherethe standard in the community is to use the ICUfor monitoring of specific conditions, such as post-operative neurosurgery cases or uncomplicatedmyocardial infarctions, individual practitionersmay put themselves out on a legal limb if theychoose to care for the patient outside the ICU(141).

Decisions to terminate life-support are seldomchallenged in court and would seem to be reason-ably well protected if the hospital has establishedexplicit criteria and procedures for reaching suchdecisions and if medical personnel follow the hos-pital’s guidelines (191). Some hospitals have for-mal policies for issuing “do not resuscitate” orders(191). These policies were initially adopted in the

3See Commonwealth v. Edelin, 359 N.E. 2014 (Mass. 1976).

— .


mid-1970s (154,193,243,253) in recognition thatnonresuscitation was appropriate when a patient’swell-being would not be served by an attempt toreverse a cardiac arrest. Yet a recent study of non-resuscitation decisionmaking suggested that phy-sicians frequently for-m opinions about a patient’sdesires for resuscitation without involving the pa-tient or the patient’s family, and often physicianstake actions which do not conform to the patient’spreference (15).

For difficult ICU decisions that do not involveresuscitation, however, physicians and staff maynot have formal procedures to follow and there-fore must speculate about their potential legalliability. The President’s Commission surveyfound that only 1 percent of hospitals in this coun-try, and just 4.3 percent of hospitals with morethan 200 beds, have ethics committees to help doc-tors and families reach decisions about withhold-ing or withdrawing life-support (191). That num-ber may have increased dramatically—to perhaps20 percent of hospitals—in the 2 years since thePresident’s Commission survey (21). Among theirother functions, these committees may provideinformation to the hospital’s medical staff on theirlegal responsibilities in certain situations.

In the absence of an institutional mechanismfor advising staff on the possible legal implica-tions of their actions, physicians, understandably,tend to adopt a cautious, “defensive” approachto decisionmaking. This is especially true in hos-pitals where legal responsibility for care of a pa-tient in the ICU is not turned over to an ICU-basedphysician. It is not unusual, for example, for pa-tients who are “brain dead” to remain on an ICUrespirator for days because of unfounded physi-cian or hospital concern about a possible malprac-tice suit or criminal prosecution. Physicians mayfeel more confident to disconnect the respiratorif hospital guidelines indicate when it is appro-priate to do so. Even when a hospital does estab-lish written guidelines and procedures for mak-ing life and death decisions, however, they arenecessarily cautious and conservative in content(111). Thus, whether physicians decline to with-hold or withdraw ICU technology for fear of legalliability or whether the institution provides guide-lines, the result is the same: the continuation ofICU care for a longer time than is often necessaryor desirable for the patient’s well-being.

PAYMENT AND THE TREATMENT IMPERATIVE

Methods of hospital and physician payment de-scribed in chapter 6 have tended to be permissivefactors for provision of excessive ICU care. It isunlikely that the care is performed primarily toreceive greater income. Rather, the payment sys-tem has not interfered with the factors describedin this section which do produce an ICU treat-ment imperative.

Until 1982, hospitals in States without prospec-tive rate-setting programs were reimbursed bysome insurers for actual costs of ICU care and byother insurers according to charge schedules which

permitted hospitals to recoup the costs of caringfor very high-cost, seriously ill ICU patients. Phy-sician payments based on a usual, customary, andreasonable charge system or even on a fixed feeschedule have tended to amply reward physicianswho provide technical ICU services. While the in-centives for the hospital have now changed, atleast for Medicare patients, physician paymentsystems still permit physicians to provide con-tinued, high-level ICU care without direct con-sideration of the costs to either the patient orsociety.

Ch. 7–The ICU Treatment Imperative ● 63

THE ABSENCE OF CLINICAL PREDICTORS

As with many chronic or terminal illnesses,there is an absence of data for the common ICUconditions on which to make predictions of anindividual ICU patient’s chances of immediate sur-vival, as well as the likelihood of his or her long-term survival. Probabilities based on quantitativeinformation for populations of similar patients areused as a reference point on which to base deci-sions about treatment of patients (277). This ap-proach is common for cancer, for example, wherethere are defined stages of disease and accumu-lated outcome data based on alternative modesof therapy.

Were it possible to predict which risk factorsconsistently yield poor outcomes, many patientsmight be considered unsalvageable at an earlierpoint in their ICU stay (247). With reliable predic-tors of ICU survival, many of the other factorsthat result in excessive ICU care would becomeless important. For example, physicians wouldhave less concern about legal liability if reliabledata were available to support their clinical judg-ment that special care should be terminated fora particular individual.

It has been argued that the use of predictivescores should have its greatest application to deci-sions involving groups of patients or on how toexpend societal resources and may have morelimited application to decisions involving individ-ual patients (157). Unfortunately, accurate quan-titative approaches to clinical decisions are diffi-cult. Collecting large, accurate data bases isexpensive and time-consuming; verifying theirrelevance to other patient populations is costlyand sometimes not feasible. Data bases can rapid-ly become obsolete for predictive purposes oncenew tests or procedures become available (157).Collecting data on heterogeneous ICU populationsin which diagnostic monitoring and therapeuticintervention often occur simultaneously is particu-larly problematic. Yet, unfortunately, as will bepointed out below, purely subjective prognostica-tion in the ICU is especially uncertain.

In recent years, work has begun on establish-ing quantitative predictive models which would

aid in predicting outcome of ICU care (143,223,236,247,270). Up to this point, no such model ofclinical predictions has been accepted for generalICU use (176). However, the Acute Physiologyand Chronic Health Evaluation (APACHE) scaledeveloped by Knaus and colleagues has begun toreceive particular attention as an objective meas-ure of the severity of illness of ICU patients forresearch and evaluation purposes, much as theTherapeutic Intervention Scoring System scale ofCullen (see ch. 5) has been used as an objectivemeasure of ICU resource use. A recent simplifica-tion of the APACHE model may make this ap-proach more widely useful to help physiciansmake more precise treatment decisions (138). Bydesign, however, the APACHE scale is moreappropriate for predicting outcomes of popula-tions of ICU patients rather than prognosticatingfor individual patients.

A generally reliable predictive model is avail-able in burn units, and has been used to makedecisions about individual patients (123). Its usein clinical decisionmaking, however, has not beengenerally accepted by experts in the field (263).

Recently, a scale of rating the likelihood of sur-vival for patients in coma (149) has been devel-oped and is used in some ICUs for individual deci-sionmaking. For the great majority of ICU patients,however, no predictive scale is available. Even ifsuch scales were available, it would be difficultto apply a population-based scale to individuals(229), especially where a “wrong” decision canhave such profound implications.

For a patient-care area that is as technologicallybased as the ICU, judgments on outcome havebeen remarkably subjective. Subjective prognosti-cation near the end of life is notoriously uncer-tain (201) and varied (177). Some feel that physi-cians tend to maintain overly pessimistic prognosesbecause patients with poor outcomes claim greaterphysician attention (70). Some physicians employa strategy that has been called the “hanging ofcrepe, ” i.e., predicting the worst so that anythingless dire will be viewed as a major achievement(229). Others feel that physicians remember the

64 ● Health Study 28:: ]ntensive Care Units: Costs, Outcome, and Decisionmaking

rare “miraculous” recovery, forget the more com-mon failures, and act on that faulty memory(280).

Other problems with ICU outcome predictioninclude the fact that recognition of terminal pa-tients during an acute admission is difficult (198);as noted earlier, an acute illness is often not seenin the context of the patient’s overall condition.Furthermore, in many community hospitals, onlya few physicians ever handle a significant num-ber of ICU patients. Most physicians have limitedexperience with the relative prognoses of thesevery sick patients (165). Very few hospitals rec-ognize an institutional responsibility to advisephysicians, patients, and families on likely out-comes of ICU care, even for the group of patientswho might be in a vegetative, nonrecoverablestate (191). Opinion on likely prognosis remainsan individual physician’s responsibility and, notinfrequently, dramatically different opinions areoffered by the various physicians involved in aparticular case.

Another major problem is the lack of mean-ingful predictors of the outcome of chronic illness(215). Many ICU patients suffer an acute, majorICU episode as part of a deteriorating chronic con-dition, e.g., emphysema, cancer, cirrhosis, or re-nal failure. Often the issue is not the likelihoodof surviving the acute episode, but rather whatthe natural course of the illness would be evenwith a favorable acute recovery. As was notedin chapter 5, it is generally accepted by ICU ex-perts that ICU care does not favorably affect thecourse of a chronic illness, but rather reverses anacute deterioration in the illness. Some patients,when given information about relatively poor life

expectancy and quality of life, choose not toundergo temporary lifesaving treatment. For ex-ample, cancer patients, relying on population-based outcome studies, sometimes choose not tosubmit to active cancer therapy. For the mostpart, prognostic indexes, stratified by disease andseverity of illness, do not exist for most other com-mon chronic conditions (158).

Physicians have demonstrated dramaticallydivergent predictions of life expectancy for pa-tients with “end-stage” diseases (177). In the ab-sence of data on acute or chronic outcome, phy-sicians can offer only imprecise, qualitativeassessments, i.e., survival is “unlikely,” “unusual,”or “possible,” rather than the quantitative assess-ments, which have probability ranges attached,i.e., “1O to 20 percent chance of one year survival”(277).

A fundamental dilemma is that the rare mirac-ulous recovery does occasionally occur. Describ-ing the dismal outcomes of 18 patients treated inan ICU for acute renal failure after rupture of anabdominal aneurysm, Morgan was one of the firstICU specialists to note the problem of high-cost,low-yield ICU care (162). The patients were el-derly (mean age 65.2 years), with a high incidenceof obesity, chronic pulmonary disease, and arte-riosclerotic heart disease. Despite energetic clini-cal efforts and dramatically high cost per patient,17 out of 18 died. Looked at another way, how-ever, one survived and was able to return to hisprevious functional level. A retrospective reviewof clinical records in these cases did not permitsuccess or failure of treatment to be predicted byany means other than actual trial.

8.Foregoing Life - Sustaining

Treatment

8.Foregoing Life-Sustaining Treatment

INTRODUCTION

Chapter 7 described the various, interrelatedfactors that help produce an environment in whichexcessive intensive care unit (ICU) care is some-times provided. The ICU treatment imperative isnow being moderated by two relatively recent de-velopments.

First, there has been increasing recognition ofthe emotional torment for the patient, the family,physician, and the hospital staff, of seeminglyendless ICU stays that ultimately end with thedeath of the patient (243,247,278). A growinghumanistic concern for the patient and his familysupports the need to preserve the dignity of a dy-ing patient, and may require earlier cessation ofactive life-support (18).

Second, there has been a growing recognitionthat the high costs of treating the most severelyill ICU patients may be too high, particularly ifthey obviously limit the resources available totreat moderately sick patients who are more likelyto benefit from intensive care (54,247).

These two developments were explicitly recog-nized by experts in critical care medicine at theCritical Care Consensus Development Conferenceconvened by the National Institutes of Health(NIH). The Consensus statement on Critical CareMedicine concludes:

It is not medically appropriate to devote limitedICU resources to patients without reasonableprospect of significant recovery when patientswho need those services, and who have a signifi-cant prospect of recovery from acute] y life-threat-ening disease or injury, are being turned away forwant of capacity. It is inappropriate to maintainICU management of a patient whose prognosishas resolved to one of persistent vegetative state,and is similarly inappropriate to employ ICUresources where no purpose will be served but aprolongation of the natural process of death (176).

The NIH statement is significant not only be-cause it recognizes the futility of ICU care in somesituations but also because it acknowledges thatICU care is, in fact, already being rationed tosome extent.

A full discussion of the difficult medical, ethi-cal, and legal issues involved in deciding to foregolife-sustaining treatment either because of the de-sire to permit death with dignity or because ofa need to ration ICU resources is clearly beyondthe scope of this case study. Readers are referredto the recently published report on this subjectby the President’s Commission for the Study ofEthical Problems in Medicine and Biomedical andBehavioral Research (191). In this chapter, a fewissues of particular relevance to ICUs are brieflydiscussed.

THE NATURAL PROCESS OF DEATH

ICUs are uniquely capable of interfering withthe natural process of death, since respirators andother ICU technologies are able to sustain vitalfunctions long after the patient has any chanceof recovery. As a consequence of these lifesav-ing technologies and the moral considerations in-volved in their use, many people today die “ICUdeaths” rather than natural deaths (135). For ex-

ample, once a patient has been placed on a respi-rator, death may occur only when the physiciansteps in and discontinues its use. Indeed, someICUs are developing policies and procedures for“terminal weaning” off of a respirator, which pro-vide for the withdrawal of a respirator in a humaneand efficient manner for the acknowledged pur-pose of permitting the patient to die (94).

67

68 ● Health Case Study 28: : ]ntensive Care Units: Costs, Outcome, and Decisionmaking

As chapter 7 pointed out, in some situations,such as when the patient’s prognosis and wishesconcerning treatment are not initially known,there may be greater moral justification for per-mitting a death by withdrawing life-sustainingtreatment, than for passively allowing death tooccur by withholding ICU care in the first place.The ability of ICU technologies to intervene inthe natural death process is also evident in pa-tients suffering from a severe, debilitating, chronicillness who can survive their acute illness but whocan never improve beyond their original unsatis-factory functional status.

In the past, pneumonia was known as “the oldman’s friend, ” because it often provided a rela-

tively quick and painless death for those facingyears of disability (170). Medical technology,however, has largely removed this escape hatch.As a result, extremely elderly patients and patientswith a severe chronic illness frequently survivetheir ICU stays, often at a great expense, only tobe restored, at best, to their previous state of illhealth. Similarly, patients who suffer a devastat-ing injury and illness that in the past would havebeen fatal are now frequently saved by excellentmedical skills and modern ICU technology, onlyto exist in a permanent state of profound physi-cal or mental impairment (51).

FUNDAMENTAL ETHICAL, MORAL, AND LEGAL CONSIDERATIONS

For certain categories of patients, there has beenconsiderable discussion in medical circles aboutthe extent to which physicians and hospitals shouldbe obligated initially to provide and then to con-tinue ICU care. Many ICU physicians have takenthe position, for example, that a necessary pre-requisite to admission to an ICU is the potentialsalvageability of the patient (51,176,200,238).Some feel that in cases where the patient is clearlymoribund and has no chance of improving, thephysician’s duty is to make the patient comfort-able and not to impose intensive care (51,152).Although patients’ families may attempt to pres-sure physicians into using the ICU, the physicianand the institution probably would be on safelegal ground in denying such care, assuming thefacts of the case sustain their position and thatthe decisionmaking process was reasonable (51).

A more difficult situation arises when a patientis terminally or irreparably ill, but is consideredto have a chance of surviving the present acutedeterioration (51). In such situations, the fun-damental decision on whether to use life-sustain-ing technology should, if possible, be made bythe patients after they have been fully informedof their options and understand their implications(191). A terminally ill patient’s right to forego ordiscontinue life-sustaining treatment has beenestablished, and is usually protected by the con-

stitutional right to privacy (191). An immediateproblem is that the term “terminal” is not a stand-ard technical term with clear and precise criteriafor its definition (12). Physicians may not agreeon when an illness is terminal, and some do noteven use the term.

In addition, as noted in chapter 7, critically illpatients are frequently not competent to make aninformed decision. This is particularly true of ICUpatients who suffer subtle alterations of conscious-ness and develop psychological reactions to theirillness or to the ICU itself. If the patient’s in-capacity to consent is temporary, the decision toforego the use of life-sustaining treatment mayhave to be postponed. If the condition is perma-nent, however, the question arises as to whoshould make the decision on the patient’s behalfand on what basis (151).

The President’s Commission recommended thatwhen a patient lacks the capacity to make a de-cision—a common ICU occurrence—a surrogatedecisionmaker should be designated. Ordinarily,this will be the patient’s next-of-kin (191). Prob-lems arise when family members disagree, arethemselves incapable of good decisionmaking, ordemonstrate family interests that conflict with thepatient’s interest (191). In many instances, an in-capacitated patient may have no family or even

Ch. 8–Foregoing Life-Sustaining Treatment ● 69

close friends who can act as a surrogate on maki-ng decisions about life-sustaining treatment.

At times, there maybe a fundamental disagree-ment between physicians and the patient’s next-of-kin on the appropriate treatment for an in-competent, seriously ill patient. When such dis-agreements cannot be resolved through discussionor through a hospital-based forum such as anethics committee, or when the patient is a wardof the State, the issue may have to be resolvedin court. Sometimes, a physician may agree witha surrogate’s decision to forego life-sustainingtreatment, but, nevertheless, seek a judicial rul-ing for fear of criminal prosecution or civil liability(191). It should be emphasized that cases whichmandate specific procedures for determiningwhether to continue medical treatment for an in-capacitated patient have been decided by Statecourts. Therefore, these court-ordered remedies,which sometimes have differed in significantways, apply only to the State in which the casewas brought, unless courts in other States specif-ically adopt the same analysis. A discussion ofthe decisionmaking procedures mandated or ap-proved by the courts in situations where a patientcannot choose for himself is beyond the scope ofthis case study.

It should be noted, moreover, that there is legalconfusion even over when a patient ought to beconsidered terminally ill and over what constitutes“medical” treatment. A New Jersey appeals court,for example, overruled a trial judge’s decision thatwould have permitted removal of a life-sustainingfeeding tube from an 84-year-old woman consid-ered to be terminally ill but not facing imminentdeath (241). The appeals court found that removalof a feeding tube would have inflicted new suf-fering from dehydration and starvation on the pa-tient. The court found that the State has a “sub-stantial and overriding” interest in preserving thelives of patients who are not moribund. It alsoseems to have found a legal difference between“nourishment” and “medical treatment .“ “We holdonly that when nutrition will continue the life ofa patient who is neither comatose, brain dead, norvegetative, and whose death is not irreversibly im-minent, its discontinuance cannot be permittedon the theory of a patient’s right to privacy, or,

indeed, on any other basis. ”1 A New Jersey Su-preme Court review of this finding is pending.

The California appellate court decision in thecriminal case of People v. Nejdl and Barber, inwhich homicide charges were brought againstphysicians who withdrew nutrition in the formof intravenous fluids and nasogastric feedingsfrom a comatose patient, significantly departsfrom the reasoning used by the New Jersey Ap-peals Court in the case cited above. The Califor-nia case did not distinguish between “ordinary”and “extraordinary” care and instead defined theconcept of “proportionate treatment .“ The courtwrote:

Proportionate treatment is that which . . . hasat least a reasonable chance of providing benefitsto the patient, which benefits outweigh the bur-dens attendant to the treatment. Thus, even if aproposed course of treatment might be extremelypainful or intrusive, it would still be proportionatetreatment if the prognosis was for complete cureor significant improvement in the patient’s con-dition.z

The reasoning of the New Jersey Appeals Courtdecision, which has been appealed to the State Su-preme Court, and that of the California AppealsCourt would appear to be irreconcilable. Thus,considerable legal uncertainty remains over pre-cisely which medical therapies, if any, are con-sidered routine or ordinary and in which clinicalsituations they must be provided. Treatmentsmight be considered mandatory for some patientsbut not for others. Weaning a terminal patientwho is brain dead off a respirator would appearto be permissible, for example, but removing afeeding tube or intravenous line might not. Like-wise, physicians might be legally required to pro-vide different treatment for patients who are seri-ously ill and have no chance for sustained recoverythan for patients who are permanently comatoseor who face imminent death.

ISee lrJ re Ccmmy, 188 N.]. Super. 523, 532 (ch. Div. 1983 ) .‘See Neil Leonard Barber, Robert Joseph Nedjl v. Superior Court

of the State of California for the County of Los Angeles; Court ofAppeals of the State of California, Second Appellate District, CivilNo. 60350; Oct. 12, 1983.

25-338 0 - 84 - 6

70 ● Health Case Study 28: ]ntensive Care Units: Costs, Outcome, and Decisionmaking

The use of life-sustaining technology has alsobeen questioned for the patient who is not ter-minally ill but who finds the quality of life unac-ceptable and without any reasonable chance forimprovement (51,225). Judgments about qualityof life obviously reflect the values and biases ofthe person making the judgment (152) and there-fore are relevant only if they represent the viewsof the patient (148). While some courts have ven-tured into this area (26), there is much less legalprecedent on which to guide physicians about theobligation to provide ICU care for such patients,particularly where the patients are incompetentto decide for themselves (170). Because it tookyears for even the current level of consensus todevelop regarding the possibility of foregoing carefor terminally ill patients, one should expect asimilar evolutionary process on the issue of fore-going life-sustaining care for those with an unac-ceptable quality of life.

Beginning with the enactment of the Califor-nia Natural Death Act in 1976, 15 States and theDistrict of Columbia have enacted statutory au-

thorization for competent individuals to write an“advance directive” which directs their physiciansto forego life-sustaining treatment under circum-stances in which they are both incompetent andsuffering from a terminal condition (273). A “proxydirective” designates a surrogate of the patient’schoice to make decisions for the patient if he orshe is unable to do so; it maybe accompanied byan “instruction directive” which specifies the typeof care the person wants to receive. In addition,42 States have enacted “durable power of attor-ney” statutes, which provide authority to appointa proxy to act after a person becomes incompe-tent. Although developed in the context of prop-erty law, these statutes may be used to providelegal authority for an advance directive.

There are a number of unresolved issues abouthow advance directives should be drafted, givenlegal effect, and used in clinical practice. Never-theless, the President’s Commission recommendedtheir use as a way of honoring patient self-deter-mination (191)0

PROCEDURES FOR REVIEW OF DECISIONMAKING

The models of decisionmaking procedures forincompetent patients derived from court opinionsare quite different. The New Jersey SupremeCourt in the Quinlan³ case invoked the presenceof hospital “ethics committees” to provide con-sultation to an incompetent patient’s guardian andspecifically rejected judicial review of such deci-sions (191). By contrast, the Supreme JudicialCourt of Massachusetts in the Saikewicz 4 case ap-peared to explicitly reject the New Jersey methodof decisionmaking and instead has establishedjudicial review of these decisions as the rule ratherthan the exception (191). However, in followupdecisions, the Massachusetts court has seeminglymodified its Saikewicz opinion such that only cer-tain categories of cases would appear to requirejudicial review, such as when the family or thefamily and doctors are in disagreement or when

31n re Quinlan, 70 N.J. 10, 35sA. 2d 647, 699, cert. denied, 429U.S. 422 (1976).

4Superintendent of Belchertown State School v. Saikewicz, 370N.E. 2cI 417,434-3s (Mass. 1977).

the physicians needs, but cannot otherwise ob-tain, consent to a course of treatment when thepatient is a ward of the State (272).

The President’s Commission recommended thatresorting to courts should be reserved for occa-sions when adjudication is clearly required byState law or when concerned parties have dis-agreements over matters of substantial importancethat they cannot resolve. The Commission statedthat ethics committees and other institutional re-sponses can function more rapidly and sensitivelythan judicial review (191).

As was noted earlier, relatively few hospitalshave such ethics committees, and those in ex-istence serve various functions, ranging from for-mulating policy and guidelines and serving as aforum for considering difficult ethical problems,to consulting on prognosis in individual cases and,finally, to reviewing or even making treatmentdecisions (191). Because of the lack of general ex-perience with ethics committees, the Commission

Ch. 8—Foregoing Life-Sustaining Treatment ● 71

called for additional evaluation of various formsof formal and informal institutionally based com-mittees before general adoption in all hospitals(191).

It should be noted that over the past 15 years,Institutional Review Boards (IRBs) have been setup to review in advance the ethical considerationsof specific research involving human subjects. Al-though initially controversial, IRBs are now gen-erally accepted in the biomedical research com-munity (191).

RATIONING ICU CARE

Up to now, discussion of withholding andwithdrawing ICU care has focused primarily onthe perceived or actual interests of the patient.However, the NIH panel has acknowledged thefact that patients who might benefit from treat-ment in an ICU are denied admission because bedsare occupied by patients who do not have a rea-sonable prospect of “significant” recovery. Earlyanalysts recognized that a few individuals con-sumed a dramatically disproportionate share ofICU resources and suggested that those resourcesbe increased so as to avoid difficult choices aboutaccess (162). The President’s Commission advisedagainst limitations on access to life-sustaining careas an initial part of any cost-containment strat-egy (191). It argued, instead, that the first stepshould be the control of “small ticket” tests andtreatments, such as routine blood test and X-rays,which are believed by some to be less cost effec-tive than more dramatic forms of therapy (153),and which can be discussed in relatively dispas-sionate terms. Unfortunately, because marginalcosts of ancillary services are much less than aver-age costs, cutbacks on these services are not likelyto have a major impact on hospital costs (1).

In addition, the care of a typical high-cost ICUpatient is, to a large extent, an accumulation ofsmall ticket items. While some efficiencies in ICUcare can be achieved (227), the fact remains thatthe decision to initiate and continue ICU care forpatients for whom recovery is unlikely, but pos-sible, is one of the major causes of the increasingproportion of the Nation’s health costs accounted

IRBs or ethics committees have been identifiedin the so-called “Baby Doe” controversy, as a pos-sible approach to aiding decisionmaking aboutdetermining medical treatment for severely hand-icapped newborns. While the situation of severelyhandicapped newborns is somewhat differentfrom that of seriously ill adults, in part becauseit involves consideration of parental rights andobligations, it may be that the Baby Doe con-troversy will generate a heightened general interestin the role of ethics committees.

for by ICUs. Despite current efforts to makehealth care more efficient, it seems clear that fur-ther attempts at cost-containment will encounterthe reality that a large amount of medical careis consumed by patients with highly unfavorableprognoses (219).

“It is a basic tenet of our society that we willnot give up a life to save dollars, even a greatmany dollars” (111). Yet, to some extent, this“lifesaving imperative” is a myth, since society’sdevotion to saving lives is greatest where thethreat is to identifiable individuals, such as trapped “miners or the victims of catastrophic disease.Society, however, accepts the loss of many “sta-tistical” lives (111), whether from the results oftoxic waste or inadequate preventive health care.

Physicians usually follow the same lifesavingimperative. Many health professionals, lawyers,and philosophers have warned that while societymay choose to limit medical treatments for eco-nomic reasons, it is not appropriate for physiciansto do it for individual patients (17,83,152,266).They argue that the doctor-patient relationshiprequires an absolute commitment to do everythingpossible for the individual patient, regardless ofthe effect on society’s resources.

However, most physicians by training andpractice accept the fact that there are limits to theresources that society can expend on any one in-dividual, and in some circumstances they act associety’s agent in balancing the needs of their pa-tient against the needs of other patients and so-

72 ● Health Case Study 28:: Intensive Care Units: Costs, Outcome, and Decisionmaking

ciety as a whole. For example, a patient with in-tractable gastrointestinal hemorrhage does notreceive limitless supplies of blood (152). At somepoint, a physician makes a decision, sometimesimplicitly, that society's interest in having a supplyof blood available for the community outweighsthe patient’s need for continued transfusions. Thethreshold for the decision to discontinue transfu-sions obviously varies, depending on factors suchas the patient’s underlying medical problem, age,and perceived life expectancy and the physician’spoint of view.

In less dire circumstances, physicians commonlyweigh the value of a marginal benefit to their pa-tients against the general cost to society. An ex-ample is the support of preventive health screen-ing based on population-related, cost-effectivenessdata. For instance, differences in the recom-mended intervals for screening for cervical can-cer through use of the PAP test (103) are essen-tially based on different views of how manymissed cases are acceptable on a cost-effectivenesscalculation. Physicians who choose one standardover others do so with an implicit acceptance thatthere is some level of risk that is acceptable foran individual patient.

Physicians in health maintenance organizations(HMOs) may practice a somewhat different styleof medicine, based on the reality of a fixed poolof resources. HMOs face “either-or” choices andmust decide whether particular treatments, suchas for catastrophic diseases, are better investmentsthan others, such as for prenatal care (111). While

the constraint of limited resources is imposed ex-ternally, HMO physicians, perhaps unconsciously,may alter their decisionmaking for individual pa-tients in accordance with the reality of limitedresources. While the HMO may exclude or limitcertain benefits explicity in its contract withsubscribers, it also counts on physicians to prac-tice “cost-effective” medicine, often at a small butmeasurable risk to certain individual patients (22).

The clear bias in ICU decisionmaking is to ini-tiate and continue ICU care even when it is ex-tremely unlikely that the patient will benefit fromsuch care. Nevertheless, because of limited ICUresources, decisions are made every day to cur-tail the care provided to individual ICU patientsand to restrict access to ICUs. In public hospitals,difficult decisions to ration limited ICU beds havebecome commonplace (186). Even in nonpublichospitals, rationing of ICU resources has occurred,particularly where there has been a shortage ofnurses (220,230).

Up to now, shortages in ICU capacity to treatpatients who might benefit from intensive carehave resulted primarily from internal hospitaldecisions on allocation of beds and other resourcesbetween the ICU and general floors, and not fromexternal economic restraints. As discussed inchapter 6, that will no longer be the case, how-ever, under the Medicare’s DRG hospital paymentsystem. In the next few years, therefore, muchmore attention will have to be given to how ICUcare should be rationed.

EXPLICIT OR IMPLICIT RATIONING OF ICU CARE?

Explicit Rationing

Provision of ICU care can involve both explicitand implicit forms of rationing. Explicit ration-ing of medical care generally involves direct ad-ministrative decisions on such issues as exclusionof certain types of services from insurance cov-erage, limitations on the availability of specificmethods of care, preauthorized and concurrentreview and approval for expensive treatments and

procedures, required intervals between provisionof specified services, and limitations on totalbenefits (159). In the context of the ICU, explicitrationing might include the establishment of med-ical criteria for treatment based on predictors ofoutcome for ICU care as they become available.In addition to predominately medical considera-tions, factors such as life expectancy, family role,and social contribution could also be formallyconsidered (196), although the experience of al-

Ch. 8—Foregoing Life-Sustaining Treatment ● 73

locating rare renal dialysis machines and selectingpatients for kidney transplants in the 1960s on thebasis of social factors was nearly intolerable tothose involved (2,14) and might not be accept-able to society.

The ethical considerations of how to decidewho should receive lifesaving treatment and whoshould not has received attention by bioethicists(13). It is relevant here to note that to avoid ex-plicit rationing for lifesaving treatments, healthplanners and policymakers have tended either toapprove facilities or financing mechanisms thatwill assure treatment for nearly everyone with aparticular illness, e.g. end-stage renal disease, orthey make a decision not to facilitate treatmentfor anyone suffering from a certain condition, e.g.patients needing heart transplants (111) exceptperhaps on an experimental basis (122). SinceICUs are not disease-specific, explicit rationing onthe basis of disease would not seem to be anappropriate means of limiting ICU care.

Explicit rationing of ICU care might also includelimits on covered benefits beyond a certain amount,or in certain clinical situations, where patientscould have to bear the costs of ICU care directly.Currently, most patients have insurance cover-age for most ICU costs. Many without coveragehave been subsidized. In public hospitals, ration-ing of limited ICU beds has been based largelyon a combination of medical factors, such aslikelihood of successful intervention, and demo-graphic factors, such as age, and not on considera-tions of ability to pay (186). There is the real ques-tion of whether society would tolerate explicitdenial of “life and death” ICU care on the basisof insurance coverage or personal wealth. In re-cent years, Congress has considered several pro-posals for national health insurance that wouldextend coverage to everyone for catastrophic ill-ness in order to avoid denial of care on the onehand and the possibility of extreme financial hard-ship and bankruptcy on the other (72).

Implicit Rationing

Implicit rationing involves limitations on theresources available to health care providers, suchas fixed budgets and restrictions on sites of careor hospital beds (159). These limitations are im-

plicit because they do not specify what servicesshould be provided to whom or what assessmentsphysicians should make. Instead, they achievetheir effect by placing greater pressures on phy-sicians and hospitals to make hard allocationchoices. Simple reliance on the price mechanismcan also be a rationing device, since everyone’sability to pay is limited at some point; for almostall resources in our society, price does “ration”access to goods and services. Cost-based paymentfor insured medical services has been a notableexception. The new DRG payment system forMedicare is a form of implicit rationing since thetotal payments allowed under the system arefixed, regardless of the level of services provided.

Other forms of payment limits could also re-quire rationing. Indeed, because many people lackinsurance altogether or have less than full-cost,open-ended coverage, implicit rationing occurs formany medical services today, particularly non-hospital care. It would be possible to limit totalsocial spending on ICUs (or anything else) throughthe implicit rationing device of patient cost-shar-ing, which does not require administrative deci-sions. Such price-based allocation of resources canbe troublesome, however, when applied to cata-strophic medical care for a variety of reasons (see111).

The cost of care for the sickest patients in theICU is currently being subsidized to a great ex-tent by those who are not as ill, and by the hos-pital. DRG fixed payments, which are not ad-justed according to the severity of the illness, willoften make high-cost Medicare ICU patients sig-nificant financial “losers” for the hospital. In thissituation, physicians will likely feel institutionalpressures not only to alter the style of ICU carethey provide to reduce costs, but also to recon-sider the thresholds for withholding and with-drawing ICU care from specific individuals. In ad-dition, hospitals may limit or even reduce thenumber of ICU beds, thus reducing access for pa-tients who would have received higher cost ICUcare. This form of implicit rationing of ICU careraises a number of questions:

● What protections will patients require toavoid arbitrary decisionmaking to limit care?Will certain categories of patients, such as theelderly, the retarded, or otherwise chronically

74 ● Health Case Study 28: : ]ntensive Care Units: Costs, Outcome, and Decisionmaking

●

●

dependent persons who might benefit fromICU care, be systematically excluded onpurely economic considerations?Will the potential threats of criminal prosecu-tion and malpractice suits act as a sufficientcountervailing force to the new incentivesthat DRGs will bring? More specifically, willthere be a fundamental conflict betweentraditional malpractice standards and newnorms of practice that may involve limitingcare more strictly? Malpractice law has tradi-tionally judged the behavior of medical careproviders almost exclusively by the custom-ary practice of their peers, rather than by anindependently determined standard of so-cially appropriate care (22). Malpractice lawgenerally does not recognize varying stylesof care to suit varying available resources.It remains to be seen whether courts will rec-ognize limited available resources as a fac-tor in determining negligence. In fact, hos-pitals and physicians may have new incentivesnot to treat very sick ICU-type patients inthe first place, not only because of the di-rectly negative economic consequences, butalso because it may place them in legal jeop-ardy under existing malpractice standards.Once care has been initiated, the primaryresponsibility of the provider is to meet ahigh standard of care that may not be reim-bursed sufficiently under the DRG paymentscheme. Hospitals may decide systematicallyto avoid the responsibility in the first place bydiverting and transferring patients elsewhere.Will society tolerate different levels of ICUcare based on willingness and ability to pay?Medicare will prohibit hospitals for the mostpart from seeking direct payments from itspatients above the allowable DRG payments

(Social Security Act Amendments of 1983,Public Law 98-21). Can a Medicare patientin a life or death situation be denied the con-tinued ICU care he or she desires and is will-ing to pay for personally, primarily throughprivate insurance, because Medicare pro-hibits patient payments above the DRGlimit? If not, it is likely that different typesof ICUs will develop, based largely on theability to pay.

● Finally, what procedures should be used toassist ICU decisionmaking in an era in whichat least some patients become financial“losers” for the hospital? A number of pro-cedural safeguards have been proposed toprotect the interest of patients who have in-sufficient capacity to make particular deci-sions on their own behalf, including: 1) nam-ing an appropriate surrogate to act on thepatient’s wishes or in the patient’s interest;2) establishing administrative arrangements,such as ethics committees for review and con-sultation of different decisions; and 3) per-mitting advance directives, such as livingwills, through which people designate so-meone to make health care decisions on theirbehalf, and/or give instructions about theircare (191). While initially proposed in thecontext of protecting the interests of in-competent patients, these or other proceduralsafeguards also appear necessary to protectthe interests of competent patients who mightotherwise be rationed out of the ICU. ICUdecisionmaking has been difficult when therewas no theoretical conflict between the in-terests of patient, physician, and institution.Under a prospective payment system, patients,physicians, and hospitals may have differentinterests.

9 ❑

Conclusions andPossible Future Steps

9Conclusions and Possible Future Step;

Until passage of the Social Security Act Amend-ments of 1983 (Public Law 98-21), intensive careunit (ICU) expansion was able to proceed with-out major consideration of costs because of thefavorable payment environment. Indeed, tight-ened section 223 limits on costs of routine hospi-tal beds in 1979 and 1980 may have even stimu-lated ICU expansion. It would seem clear thatMedicare’s inpatient hospital prospective diagno-sis-related group (DRG) payment system willcause hospital administrators and ICU directorsto look differently at the costs of ICU care. Un-fortunately, they will find no easy solutions tothe cost problem, particularly if Medicare allowsonly relatively low rates of annual spending in-creases.

Under DRG payment, some savings may begenerated by better organization and managementof ICUs, perhaps by centralizing separate ICUsinto larger, more general ICUs (212). Arguably,additional savings may be gained by substitutinglower paid health personnel for nurses or physi-cians to provide certain ICU functions (162,212).There may be new efforts to find cost-saving tech-nologies that can substitute for expensive ICUlabor. One ICU, for example, has demonstrateda significantly decreased ICU length of stay, at-tributable in part to the use of computer-assisteddecision algorithms (227).

In addition, it maybe possible in the near futureto predict more accurately which monitored pa-tients do not need to be in the ICU at all. In-termediate care units or other arrangements couldbe developed to care for these patients, probably,at a somewhat lower cos t (141) .

At the same time, however, it is now being rec-ognized that some ICU patients are dischargedprematurely from the ICU. One can argue thatlonger stays in the ICU for these patients wouldnot only represent a more appropriate use of theICU but also might even save the hospital moneyby reducing the costs of subsequently treating forthese prematurely discharged patients (246).

Nevertheless, the fact remains that relativelyfew ICU patients are responsible for a substan-tial portion of ICU costs. This case study has at-tempted to demonstrate the clinical, moral, legal,and economic factors which currently make it dif-ficult to decide not to treat even those patientswho show little promise of benefiting from ICUcare. The high-cost subgroup is spread among allages, diagnostic groups, and disability classes (40).There are as yet no demographic identifiers or ac-cepted general prognostic indicators which per-mit systematic exclusion of any of the high-costgroup from ICU care. Public programs, privateinsurers, perhaps the public at large, but almostcertainly hospital managers and providers, willface increasingly difficult decisions about whoshould be given ICU care and in what manner.The process of ICU decisionmaking will becomeeven more important when economics may dic-tate curtailing or even denying care to seriouslyill patients.

A number of steps might improve the environ-ment for intensive care decisionmaking:

Research on developing accurate predictorsof survival for patients with acute andchronic illnesses could be expanded in orderto permit better informed decisions based onthe likelihood of short- and long-term sur-vival. Since the results of outcome data willalways be incomplete and subject to differ-ing interpretations, especially in relation toan individual patient, hospitals might con-sider formalizing an institutional “prognosiscommittee” whose function would be to ad-vise physicians, families, and patients on thelikely survival with ICU care in individualsituations. Such a committee or hospitalfunction, perhaps utilizing a routinely up-dated national data base, obviously couldalso provide a similar function for non-ICUpatients.

The suitability of the current DRG methodof payment for ICUs should be tested. If, infact, the DRG scheme takes insufficient ac-

77

7$ ● Health Case Study 28: Intensive Care Units: Costs, Outcome, and Decisionmaking

●

●

●

count of severity of illness, it is likely thatsome hospitals and, consequently, some ICUpatients may face a degree of rationing thatCongress did not envision.The legal system, including legislators andthe courts, may need to recognize the possi-ble conflict between malpractice standardswhich assume quality of care that meets na-tional expert criteria, and a decisionmakingenvironment in which resources may be se-verely limited. At the same time, it must bekept in mind that the threat of both malprac-tice suits and criminal prosecution maybecome an even more important protectionagainst arbitrary or unfair denial and ter-mination of ICU care.Health professionals who are involved inmaking decisions regarding critically ill pa-tients might benefit from more education onmedical ethics and relevant legal proceduresand obligations. In recent years, the journalCritical Care Medicine, published by theSociety of Critical Care Medicine, has in-cluded articles and editorials on specific ethi-cal and legal issues. Likewise, new textbookson critical care medicine (224) have devotedchapters to specific ethical and legal issuesthat frequently arise in the ICU. More for-mal education at the graduate and postgrad-uate level for all health professionals whowork with critically ill patients might be con-sidered.

The actual decisionmaking process for criti-cally ill patients may need greater attention.

At a time when the interests of the ICU pa-tient, physician, and hospital were theoreti-cally the same, i.e., under a full-cost reim-bursement system, the need for formal rulesand procedures for life and death decisionsmight not have been necessary. Even so,many hospitals found the need to establishformal procedures for “Do Not Resuscitate”orders. With a payment system that sets theinterests of at least some very sick ICU pa-tients against the immediate financial inter-ests of the hospital, however, it may benecessary to impose additional formal pro-tections on the decisionmaking process. Hos-pitals might explore formalizing decision-making committees or mandating secondopinions to lessen the burden on individualsfaced with excruciatingly difficult choicesabout terminating life-support. Hospitalscould consider formally separating the ICUtriage function from the direct patient carefunction, particularly with regard to the ICUMedical Director, in order to minimize po-tential conflicts of interest. More generally,society will need to decide how it wishes con-flicts over decisions on terminating life-support to be resolved—in courts, throughformal hospital committees such as ethicscommittees, through government-imposedutilization review procedures which can fol-low fixed rules and-regulations, or other,haps more decentralized, mechanisms.

per-

Appendixes

Appendix Am —Acknowledgments andHealth Program Advisory Committee

ACKNOWLEDGMENTS

The development of this case study was greatly aided by the advice of a number of people. The authorand OTA staff would like to express their appreciation to the Medical Technology and Costs of the MedicareProgram Advisory Panel, to the Health Program Advisory Committee, and especially to the following individuals:

Randy Bovbjerg Richard LiebermanThe Urban Institute SysteMetricsWashington, DC Bethesda, MD

Martha Brooks Joanne LynnSpringfield, VA Cliften, VA

Mark Chassin John E. MarshallThe Rand Corp. National Center for Health Services Research

J. Richard Crout Michael McMullanNational Institutes of Health Health Care Financing Administration

Palmer Dearing Lois P. MyersBlue Cross/Blue Shield Association University of California, San Francisco

Patricia Donovan Joseph OnekW a s h Onek, Klein, & Farr. . / ,Betty Draper Washington, DC

The George Washington University Joseph E. ParrilloSchool of Medicine -

Mark GoodhartAmerican Hospital AssocAndria GrenedierAlexandria, VAAke GrenvikUniversity Health CenterStephanie Hadley

National Institutes of HealthSteven A. Schroeder

ation University of California, San FranciscoDouglas WagnerThe George Washington UniversitySchool of Medicine

of Pittsburgh Donald A. YoungProspective Payment Assessment Commission

National Institutes of Health Rebekah Zanditan

William Knaus Takoma Park, MD

The George Washington UniversitySchool of Medicine

81

82 ● Health Case Study 28:: Intensive Care Units: Costs, Outcome, and Decisionmaking

HEALTH PROGRAM ADVISORY COMMITTEE

Sidney S. Lee, Committee ChairPresident, Milbank Memorial Fund

New York, NY

Stuart H. Altman*DeanFlorence Heller SchoolBrandeis UniversityWaltham, MAH. David BantaDeputy DirectorPan American Health OrganizationWashington, DCCarroll L. Estes**ChairDepartment of Social and Behavioral SciencesSchool of NursingUniversity of California, San FranciscoSan Francisco, CARashi FeinProfessorDepartment of Social Medicine and Health PolicyHarvard Medical SchoolBoston, MAHarvey V. FinebergDeanSchool of Public HealthHarvard UniversityBoston, MAMelvin A. Glasser***DirectorHealth Security Action CouncilCommittee for National Health InsuranceWashington, DCPatricia KingProfessorGeorgetown Law CenterWashington, DCJoyce C. LashofDeanSchool of Public HealthUniversity of California, BerkeleyBerkeley, CA

Alexander LeafProfessor of MedicineHarvard Medical SchoolMassachusetts General HospitalBoston, MAMargaret Mahoney****PresidentThe Commonwealth FundNew York, NYFrederick MostellerProfessor and ChairDepartment of Health Policy and ManagementSchool of Public HealthHarvard UniversityBoston, MANorton NelsonProfessorDepartment of Environmental MedicineNew York University Medical SchoolNew York, NYRobert OseasohnAssociate DeanUniversity of Texas, San AntonioSan Antonio, TXNora PioreSenior AdvisorThe Commonwealth FundNew York, NYMitchell Rabkin’PresidentBeth Israel HospitalBoston, MA

● Until April 1983.● ’Until March 1984.● **Until October 1983.● ***Until August 1983.

App. A—Acknowledgments and Health Program Advisory Committee • 83

Dorothy P. RiceRegents LecturerDepartment of Social and Behavior SciencesSchool of NursingUniversity of California, San FranciscoSan Francisco, CARichard K. RiegelmanAssociate ProfessorGeorge Washington UniversitySchool of MedicineWashington, DCWalter L. RobbVice President and General ManagerMedical Systems OperationsGeneral ElectricMilwaukee, W1

Frederick C. RobbinsPresidentInstitute of MedicineWashington, DCRosemary StevensProfessorDepartment of History and Sociology of ScienceUniversity of PennsylvaniaPhiladelphia, PA

Appendix B. —Cost Estimates

As emphasized in chapter 3, there are significanttechnical problems in estimating the actual or even therelative costliness of intensive care unit (ICU) care, Itis essential to recognize some of the most importantdata problems that have had to be confronted. First,only charge data is generally available. Assumptionsabout the relation of charge to cost have been madeseparately for room and board and for ancillary serv-ices. Second, national data on the amount of inpatientICU care provided is available for Medicare, but notfor the general population. In addition, there are con-cerns about the reliability of the MEDPAR data base(254). The national estimates have necessarily had tobuild up from this Medicare data base.

Third, standardized national data exists for ICUbeds but not for ICU days. Usually, bed occupancyrates in ICUs are comparable to hospital bed occupan-cy rates in general. We assume, then, that the propor-tion of ICU days to total hospital days is nearly thesame as ICU beds to total hospital beds.

Fourth, the relevant data bases combine ICU andcoronary care unit (CCU) care. No attempt, therefore,is made to distinguish ICU and CCU costs. Further-more, the assumptions underlying cost estimating forICU and CCU care may not hold for other types ofspecial units, such as pediatric, neonatal, and burnICUs. A data base for intermediate care units is simplynot available at all. Therefore, the estimates presentedhere are for adult ICU/CCU costs which understatethe costs of more broadly defined special care units.As was noted in chapter 2, adult ICU/CCU beds in1982 made up 5.9 percent of hospital beds, while sep-arate pediatric, neonatal, and burn ICUs togethermade up another 1 percent of beds.

Definition 1—8 to 10 percent: The percentage ofhospital costs represented by the direct and indirectcost of running the ICU, as reflected in charges for ICUroom and board. The Health Care Financing Admin-istration (HCFA) has analyzed the use of and chargesfor accommodation and ancillary services in short-stayhospitals for Medicare beneficiaries based on a 20-percent sample of Medicare beneficiaries—the MED-PAR data base (112). In 1980, HCFA’S sample showedthat charges for ICU/CCU care constituted 7 percentof total hospital charges. Since Medicare patients’ uti-lization of ICUs is roughly in the same proportion asnon-Medicare patients (see ch, 4), we assume then thatabout 7 percent of all hospital charges were forICU/CCU room and board charges. As discussed inchapter 3, charges generally underestimate actual costsof operating ICUs. In one careful study from a singlehospital, the hospital charge for special care room andboard was found to be only 65 percent of the marginal

cost of maintaining the bed. In contrast, the marginalcost for general floor beds was less than the establishedcharge by approximately one-third (110). Thus, basedon this and other anecdotal reports, one can conserv-atively estimate that ICU/CCU costs represented 8 to10 percent of hospital costs in 1980. The proportionof hospital beds devoted to intensive care has, how-ever, increased since 1980. It is likely that the propor-tion of ICU bed days has increased as well. Therefore,today, the estimate would be at the high end of the8- to 10-percent range or even slightly higher.

Definition 2—14 to 17 percent: The percentage oftotal hospital costs consumed by patients when in theICU. This includes room and board and ancillaryservices.

Method A: The simple approach to this estimate isto double the room and board charges—room andboard makes up about 50 percent of total hospitalcharges—and then make a charge-to-cost adjustment.As noted in chapter 3, in general, hospitals mark upcosts for ancillary services by almost a third to deter-mine charges. Thus, it would not be appropriate tosimply double the cost estimate derived from the cal-culations in Definition 1 above. We simply do notknow precisely the appropriate charge-to-cost adjust-ments to make for ICU room and board charges andfor ancillary service charges. In addition, data suggestthat ICU patients use more ancillary services per daythan non-ICU patients (see ch. 3). The extent of thisadditional utilization is not precisely known.

If one assumes that the markup for the ancillaryservices and the markdown for ICU room and boardwere roughly the same and that ICU patients use thesame amount of ancillary services as non-ICU pa-tients—conservative assumptions—the estimate forpercentage of hospital costs consumed by patientswhen in the ICU would be 14 percent, relying on theMEDPAR data for 1980 presented above. If it is assumedthat ICU patients used 20 percent more ancillary serv-ices than non-ICU patients, the estimate rises to 15 per-cent. The recent expansion in ICU beds since 1980might add another 1 to 2 percent. The estimated range,then, is 14 to 17 percent.

Method B: Louise Russell provided a method forestimating the total costs of ICU care by relating thepercentage of the total hospital beds that were ICU/CCU beds to the relative costs per day in an ICU andin a general hospital ward (205). This method assumedthat days of care are proportional to the number ofbeds. Russell also used a 3:1 ratio for relative costlinessof an ICU day compared to a regular bed day. Hermethod, when applied to 1976 American HospitalAssociation (AHA) bed data, provides a conservative

84

App. B—Cost Estimates ● 85

estimate that adult ICU/CCU costs represented about13 percent of total hospital costs at that time. Updatingfor 1982 AHA data that 5.9 percent of beds in non-Federal, short-term hospitals are ICU or CCU bedswould give an estimate of about 1S percent, assumingthe same 3:1 cost ratio.

As noted in the discussion under Method A above,critical assumptions are used to generate the 3:1 rela-tive costliness ratio, i.e., that the markup for ancil-lary services is roughly comparable to the markdownfor ICU room and board, and that ICU patients useancillary services in the same proportion as non-ICUpatients. The 3:1 ratio may well be too conservative.A 3.5:1 ratio would give an overall estimate of about17 percent, using Russell’s method. Russell herselfusing 1979 AHA bed data estimated that almost 20 per-cent of all hospital costs are accounted for by inten-sive care (206). This estimate included costs of neonataland, presumably, pediatric ICU and burn unit beds.Thus, our estimates of percentage cost, 15 to 17 per-cent, using Russell’s method, is consistent with her ownestimate. This estimate also agrees with the estimatecalculated according to Method A above.

Definition 3—28 to 34 percent: The total hospitalcosts for patients who spend any time in the ICU.Some authors have utilized this concept to demonstratethe high proportion of total hospital costs accountedfor by intensive care patients (175). This calculationis relatively easy to obtain from hospital accountingreports. Reports from two large hospital ICUS showthat approximately 50 percent of the total hospitalcosts incurred by ICU patients occurs when patientsare on regular medical floors (54,175). Similarly, HCFA’SMEDPAR data demonstrates that the average roomand board charge for routine bed stay and for an ICU/CCU bed stay were roughly the same (112). Therefore,a user of both an ICU/CCU bed and a regular bedwould have charges two times the charge of the ICU/CCU stay. If by Definition 2, it was estimated that 14to 17 percent of total hospital costs are incurred bypatients while in the ICU, then about twice that per-centage—between 28 to 34 percent of hospital costs—probably is expended on patients who spend any timeduring their hospitalization in the ICU or CCU. Theestimate agrees with the findings in one large commu-nity hospital in which patients spending any time inthe ICU represented 9.5 percent of total hospital ad-missions and, yet, incurred nearly 30 percent of totalhospital charges (175). Unfortunately, while relativelyeasy to calculate, this cost definition is not very rele-vant to consideration of ICUS as a separate technology.

Definition 4—cannot be estimated: The incrementalcost generated by ICUS above the cost that a hospitalwould have to absorb for treating ICU-type patients

if the ICU did not exist. This definition tests whetherthe ICU is a cost generator independent of the patientsit treats. Certainly, some amount of the fixed ICU costswould be saved if the ICU did not exist. However,some of these costs, e.g., depreciation of ICU equip-ment, would be generated in any case since the costswould be transferred to regular medical and surgicalfloors. To the extent that efficiencies are achievableby aggregating equipment and personnel in separateareas, an initial impetus to development of ICUS, ICUSconceivably could reduce hospital costs. In fact, thescant data available suggests that costs of running aconventional medical floor did not decrease with de-velopment of the ICU (97).

Experts in provision of ICU care maintain that somepatients require ICU care to have a chance at survival(50). The sickest ICU patients simply would not sur-vive without the coordinated and concentrated careprovided in the ICU. For practical and ethical reasonsthat were discussed in chapter 5, this hypothesis can-not be directly tested. To the extent that these expertsare correct, ICUS do generate a large incremental costto the hospital, but with substantial benefits to sur-vivors. These very sick patients may consume as muchas 40 to 50 percent of ICU costs in some institutions(54,175).

ICUS, however, also generate increased incrementalcosts for patients who are likely to survive hospitaliza-tion whether they are cared for in the ICU or not.Griner followed the experience of patients admitted toa general hospital with the diagnosis of acute pul-monary edema for the year before and the year afterthe opening of an ICU (98). While the mortality rateof 8 percent did not change, the average hospital billfor patients admitted during the year after opening ofthe ICU was 46 percent greater than for those admit-ted the year before (99). His sample size, unfortu-nately, was quite small.

Griner’s study is essentially the only one of its kindwhich gives an estimate of the incremental cost of anICU for treating similar patients with similar medicaloutcomes. Difficulties from generalizing the results ofthis study for the purposes of this case study include:1) the patient population studied represents a smallsubpopulation of ICU patients; 2) the study is a dec-ade old; and 3) the observational period of ICU carewas the first year of its operation, a period duringwhich care may be the least efficiently provided.

In 1981, Cromwell’s group (49) attempted to isolatethe role of various factors which might explain varia-tions in inpatient charges using a complex regressionequation. One finding was that both hospital routineand ICU bed stays were significant explainers of ancil-lary use. They found that ICU bed days are associ-

25-338 0 - 84 - 7

86 ● Health cue Stucfy 28: Intensive care Units: Costs, Outcome, and Decisionmaking

ated with a greater use of ancillary services than rou-tine bed days. Using the regression, they found thatICU days on average cost about 56 percent more inancillary services than regular days, holding case mix,surgery, insurance status, and other variables constant.While the case mix measure used (diagnosis and ur-gency of admission) may not be a precise measure ofseverity of illness, the regression did confirm that theICU days are associated with additional costs in ancil-lary services above those that can be explained by pa-tient characteristics. Again, it is possible that very sick,“ICU-type” patients would have greater ancillary serv-

ices used for their care regardless of their bed location.The 56-percent increment, however, is substantial and,at least, suggests that the ICU itself may have beenpartly responsible for the greater use of ancillaryservices.

Griner’s and Cromwell’s work together suggest thatICUS generate incremental hospital costs both in ad-ditional direct ICU costs and in greater use of ancil-lary services to achieve similar outcomes as care onregular medical and surgical floors. An estimate of theamount of this cost cannot be provided.

—

References

References

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

Aaron, H. F., and Schwartz, W. B., The PainfulPrescription—Rationing I-fospital Care (Washing-ton, DC: The Brookings Institution, 1984).Abram, H. S., and Wadlington, W., “Selectionof Patients for Artificial and Transplanted Or-gans,” Ann. Intern. Med. 69:615, 1968.Abramson, N. S., Wald, K. S., Grenvik, A. N.A., et al., “Adverse Occurrences in Intensive CareUnits,” ]. A.M.A. 244(14):1582, 1980.American Hospital Association, Hospital Statis-tics, 1977 to 1983 editions (Chicago, IL: Ameri-can Hospital Association, 1977 to 1983).American Medical News, “Court Vacates Mur-der Against Two MDs,” Oct. 14, 1983, p. 1.Arthur D. Little, Inc., “Planning for GeneralMedical and Surgical Intensive Care Units: ATechnical Assistance Document for PlanningAgencies, ” prepared for the U.S. Department ofHealth, Education, and Welfare, publication No.(HRS) 79-14020 (Washington, DC: U.S. Govern-ment Printing Office, 1979).Avorn, J., “Benefit and Cost Analysis in GeriatricCare: Turning Age Discrimination in Health Pol-icy,” N. Engl. J. A4ed. 310(20):1294, 1984.Ayres, S. M., “Critical Care Medicine, ” introduc-tion in Major Issues in Critical Care Medicine,J. E. Parrillo and S. M. Ayres (eds. ) (Baltimore,MD: Williams & Wilkins, 1984).Baker, R., Knaus, W. A., Draper, E. A., et al.,“Initial Evaluation of No-Resuscitation Deci-sions, ” manuscript in preparation, 1983.Bartlett, R. H., Gazzaniga, A. B., Wilson, A. F.,et al., “Mortality Prediction in Adult RespiratoryInsufficiency,” Chest 67(6):680, 1975.Bates, D, V., “Workshop on Intensive CareUnits, ” comments of the National Academy ofSciences, National Research Council, Committeeon Anesthesia, Anesthesiology 25:192, 1964.Bayer, R., Callahan, D., Fletcher, J., et al., “TheCare of the Terminally 111: Mortality and Eco-nomics,” N. Engl. ]. Med. 309(24):1490, 1983.Beauchamp, T. L., and Childress, J. F., Principlesof Biomedical Ethics (New York: Oxford Univer-sity Press, 1979).Becker, E. L., “Finite Resources and MedicalTriage, ” Am, J, Med. 66:549, 1979.Bedell, S., and Delbanco, T. L., “Choices AboutCardiopulmonary Resuscitation in the Hospital–When Do Physicians Talk With Patients?” N.Engl. ]. Med. 310(17):1089, 1984.Bell, J. A., et al., “Six Years of MultidisciplinaryIntensive Care, ” Brit, M e d . ]. 2:483, 1974.

17.

18.

19.

20.

21,

22.

23<

24.

25.

26.

27.

28.

29.

30,

31,

Bendixen, H. H., Egbert, L. D., Hedley-White, J.,et al., Resp. Care (St. Louis, MO: The C. V.Mosby Co., 1965).Bendixen, H. H., “The Cost of Intensive Care,”ch. 22 in Costs, Risks, and Benefits of Surgery,J. P. Bunker, B.A. Barnes, and F. Mosteller (eds.)(New York: Oxford University Press, 1977).Bicknell, W. J., and Walsh, D. C., “Certification-of-Need: The Massachusetts Experience, ” N.Engl, ]. Med. 292:1054, 1975.Biles, B., Schramm, C. J., and Atkinson, J. G.,“Hospital Cost Inflation Under State Rate-SettingPrograms,” N, Engl. ], Med. 303(12):664, 1980.Bolsen, B., “MDs, Lawyers Probe Ethical, LegalIssues in Ending Treatment, ” American MedicalNews, p. 17, Apr. 6, 1984.Bovbjerg, R., “The Medical Malpractice Stand-ard of Care: HMOS and Customary Practice, ”Duke Law ~. 1975(6), 1976.Boyd, R., “Workshop on Intensive Care Units, ”comments of the National Academy of Sciences,National Research Council, Committee on Anes-thesia, Anesthesiology 25:192, 1964.Bradburn, B. G,, and Hewitt, P. B., “The Effectsof the Intensive Therapy Ward Environment onPatients’ Subjective Impressions: A FollowupStudy, ” Intensive Care Med. 7:15, 1980.British Medical Association, Planning Unit, re-port of the Working Party on Intensive Care,(1):5, 1967.Brooks, T. A., “Withholding Treatment andOrders Not to Resuscitate, ” ch. 10 in Legal andEthical Aspects of Treating Critically and Ter-minally 111 Patients, A. E. Doudera and J. D.Peters (eds. ) (Ann Arbor, MI: Association ofUniversity Programs in Health Administration,1982).Brown, N. K., and Thompson, D. J., “Nontreat-ment of Fever in Extended-Care Facilities, ” N.Engl. ], Med. 300(22):1246, 1979.Bulkley, B. H., “The Coronary Care Unit, ” ch.1 in Major Zssues in Critical Care Medicine, J. E.Parrillo and S. M. Ayres (eds.) (Baltimore, MD:Williams & Wilkins, 1984).Byrick, R. J., Mindorff, C., McKee, L., et al.,“Cost-Effectiveness of Intensive Care for Respi-ratory Failure Patients, ” Crit. Care Med. 8(6):332,1980.Cadmus, R. R., “Special Care for the CriticalCase,” Hospitals 20:65, 1954,Callahan, J. A., Spiekerman, R. E., Broadbent,J. C., et al., “St. Mary’s Hospital-Mayo Clinic

89

90 ● Health Cme Study 28 : Inte~ive care l.lnits: Costs, Outcome, and Decisionmaking

32

33.

34.

35.

36.

37,

38,

39,

40.

41.

42.

43.

44

45,

46,

47

Medical Intensive Care Units: II. Patient Popula-tion,” Mayo Clin. Proc. 42:332, 1967.Campbell, D., Reid, J. M., Tefler, A. B. M., etal., “Four Years of Respiratory Intensive Care, ”Brit. Med. ]. 4:255, 1967.Campion, E. W., Mulley, A. G., Goldstein, R.L., et al., “Intensive Treatment for the Elderly, ”reply to Letter to the Editor, }. A.M.A. 247(23):3186, 1982.Campion, E. W., Mulley, A. G., Goldstein, R.L., et al., “Medical Intensive Care for the Elderly:A Study of Current Use, Costs, and Outcomes,”J. A.M.A. 246(18):2052, 1981.Caroline, N. L., “Quo Vadis Intensive Care:More Intensive or More Care?” guest editorial inCrit. Care Med. 5(5), 1977.Carroll, D. G., “Patterns of Medical Care in aMunicipal Hospital Intensive Care Unit: Conven-ience or Necessity?” Maryland State Med. J.20:89, 1971.Casali, R., et al., “Acute Renal InsufficiencyComplicating Major Cardiovascular Surgery,”Am. Surg. 181:370, 1975.Cassem, N. H., “When to Disconnect the Respi-rator,” Psychiatric Ann. 9:38, 1979.Cassem, N. H., and Hackett, T. P., “PsychiatricConsultation in a Coronary Care Unit,” Ann. Zn-tern. Med. 75:9, 1971.Chassin, M. R., “Costs and Outcomes of Medi-cal Intensive Care, ” Medical Care 20(2):165,1982.Civetta, J. M., “The ICU Milieu: An Evaluationof the Allocation of a Limited Resource, ” Resp,Care 21(6):501, 1976.Civetta, J. M., ‘The Inverse Relationship BetweenCost and Survival,” ], %rg. Res. 14(3):265, 1973.Civetta, J. M., “Selection of Patients for Inten-sive Care, ” in Recent Advances in Intensive Ther-apy, I. M. Ledingham (cd. ) (New York: Chur-chill Livingston, 1977).Clark, T. J. H., Collins, J. V., Evans, T. R., etal., “A Review of Experience Operating a Gen-eral Medical Intensive Care Unit, ” l?rit. Med. J.1:158, 1971.Coelen, C., and Sullivan, D., “An Analysis ofthe Effects of Prospective Reimbursement Pro-grams on Hospital Care, ” Health Care Finan.Rev. 2:3, 1981.Cohen, C. B., “Ethical Problems of IntensiveCare, ” Anesthesiology 47:217, 1977.Crockett, G. S., and Barr, A., “An Intensive CareUnit: Two Years Experience in a Provincial Hos-pital,” Brit. Med. ]. 2:1173, 1965.

48,

49,

50.

51.

52.

53<

54,

55.

56.

57.

58.

59.

60.

61.

Cromwell, J., and Kanak, J. R., “The Effects ofProspective Reimbursement Programs on Hospi-tal Adoption and Service Sharing, ” Health CareFinan. Rev. 4(2):67, 1982.Cromwell, J., Mitchell, J. B., and Windham, S.R., “The Cost Dynamics of Critical Illness, ” pre-pared by Health Economics Research, Inc., forthe National Center for Health Services Research,Office of the Assistant Secretary for Health, U.S.Department of Health and Human Services, grantNo. HS 04026, August 1981.Cullen, D. J., “Results, Charges, and Benefits ofIntensive Care for Critically Ill Patients, ” pres-entation at the National Institute of Health Con-sensus Development Conference, Critical CareMedicine, Mar. 7, 1983.Cullen, D. J., “Results and Costs of IntensiveCare,” Anesthesiology 47:203, 1977.Cullen, D. J., “Surgical Intensive Care: CurrentPerceptions and Problems,” Crit. Care Med.9:295, 1981.Cullen, D, J., Civetta, J. M., Briggs, B. A., et al.,“Therapeutic Intervention Scoring System: AMethod for Quantitative Comparison of PatientCare,” Crit. Care Med. 2:57, 1974.Cullen, D. J., Ferrara, L. C., Briggs, B. A., et al.,“Survival, Hospitalization Charges and Follow-up Results in Critically 111 Patients, ” N. EngZ. ].Med. 294(18):982, 1976.Cullen, D. J., Ferrara, L. C., Gilbert, J., et al.,“Indicators of Intensive Care in Critically 111 Pa-tients, ” Crit, Care Med. 5:173, 1977.Cullen, D. J., Keene, R., Kunsman, J. M., et al.,“Results, Charges and Benefits of Intensive Carefor Critically 111 Patients—Update 1983, ” Crit.Care Med. 12:102, 1984.Davidson, I. A., Bargh, W., Cruickshank, A. N.,et al., “Crush Injuries of the Chest: A FollowupStudy of Patients Treated in an Artificial Ven-tilation Unit, ” Thorax. 24:563, 1969.Davis, P. B., and di Sant’Angnese, P. A., “As-sisted Ventilation for Patients With Cystic Fi-brosis,” J, A.M.A. 239(18):1851, 1978.Day, H. W., “Effectiveness of an Intensive Cor-onary Care Area, ” Am, ]. Cardiology 1s:s1,1965.Derzon, R. A., “Influences of ReimbursementPolicies on Technology,” in Critical Issues inMedical Technology, B. J. McNeil and E, G.Cravalho (eds.) (Boston: Auburn House Publish-ing Co., 1982).Detsky, A. S., Stricker, S, C., Mulley, A. G.,et al., “Prognosis, Survival and the Expenditure

References . 91

62.

63.

64.

65.

66.

67.

68,

69.

70.

71.

72.

73.74.

75.

76.

77.

of Hospital Resources for Patients in an IntensiveCare Unit,” N. Eng], ), Meal, 305(12):667, 1981.Dlin, B., Rosen, H., Dickstein, K., et al., “TheProblems of Sleep and Rest in the Intensive CareUnit, ” Psychosomatic 12(3):155, 1971.Donowitz, L. G., Wenzel, R. P,, and Hoyt, J. W.r

“High Risk of Hospital-Acquired Infection in theICU Patient,” Crit. Care Med. 10(6):355, 1982.Dornhorst, A. C., comments on Intensive CareUnits in Section of Anesthetics, Proceedings ofthe Royal Society of Medicine 59:1293, 1966.Downs, J. B., “Intensive Care and the Adult Res-piratory Distress Syndrome, ” ch. 14 in MajorIssues in Critical Care Medicine, J. E. Parrillo andS. M. Ayers (eds. ) (Baltimore, MD: Williams &Wilkins, 1984).Drake, W. E., Jr,, et. al., “Acute Stroke Man-agement and Patient Outcome: ‘The Value ofNeurovascular Care Units (NCU),” Stroke 4:933,1973.Draper, E. A., Wagner, D. P., and Knaus, W.A., “The Use of Intensive Care: A Comparisonof a University and Community Hospital, ” HealthCare Finan. Rev. 3:49, 1981.Duke Law ]ournal, “The Medical MalpracticeThreat: A Study of Defensive Medicine, ” 1971:939.Engelhardt, H. T., Jr., “Case Studies in Bioethics:Case No. 228, A Demand to Die, ” The HastingsCenter Report 5:9, 197s.Epstein, F. H., “Responsibility of the Physicianin the Preservation of Life, ” Arch. Intern, Med.139:919, 1979.Equitable Life Assurance Society of the UnitedStates, l-lospital Daily service Charges, 1982,cata}og No. 104283 (New York: Equitab~e LifeAssurance Society, 1982).Feder, J., Hadley, J., and Holohan, J., “Insuringthe Nation’s Health: Market Competition, Cata-strophic and Comprehensive Approaches” (Wash-ington, DC: The Urban Institute, 1981).Federal Register, 45:21582-21588, 1980.Federal Register, 48(171):39752-39890, Sept. 1,1983.Federal Register, 49(129):27422-27463, Jan. 3,1984.Fedullo, A. J., and Swinburne, A. J., “Relation-ship of Patient Age to Cost and Survival in aMedical ICU,” Crit. Care Med. 11(3):155, 1983.Feldman, R., Sloan, F., Paringer, L., et al., “AnAnalysis of Hospitals’ Methods of CompensatingPhysicians, ” The Urban Institute, working paperNo. 1302-02, supported by the Health Care Fi-nancing Administration grant No. 95-P-97176/3-o2, December 1979.

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

88.

89.

90.

91.

92.

93.

94.

Feller, I., Tholen, D., and Cornell, R. G., “Im-provements in Burn Care, 1965 to 1979,” J, A,M,A,244:2074, 1980.Fineberg, H. V,, Scadden, D., and Godman, L.,“Care of Patients With a Low Probability ofAcute Myocardial Infarction-Cost Effectivenessof Alternatives to Coronary Care Unit Admis-sion,” N. Engl. ]. Med. 310(20):1301, 1984.Finkler, S. A., “The Distinction Between Costand Charges, ” Ann, Intern, Med. 96(1):102,1982.France, D., and Bismuth, H., “Five Years of Ex-perience With an Intensive Care Unit Specializ-ing in Abdominal Surgery, ” F’rog. Surg, 15:37,1977.Franklin, C., and Jackson, D., “Discharge De-cision-Making in a Medical ICU: Characteristicsof Unexpected Readmissions, ” Crit. Care Med.11:61, 1983.Freid, C., “Rights in Health Care—Beyond Eq-uity and Efficiency, ” N. Engl. ~. Med. 293;241,1975.Fuchs, V. R., Who Shall Live? Health, Economicsand Social Choice (New York: Basic Books, Inc.,1974).Fuchs, R., and Scheidt, S., “Improved Criteriafor Admission to Cardiac Care Units,” ]. A.M.A.246:2037, 1981.Galbally, B. P., “The Hospital Service-Targetsfor Tomorrow: Intensive Care—Productivityand Cost Control, ” National Hospital 16:15,1972.Gelber, R., director of Capital Hill Hospital ICU,personal communication, July 22, 1983.Gibson, R. M., Waldo, D. R., and Levit, K. R.,“National Health Expenditures, 1982, ” HealthCare Finan. Rev. 5(1):1-32, fall 1983.Gilligan, J. E., McCleave, D. J., Worthley, L.,et al., “The Role of Intensive Care Units, ” Aust.N.Z.], Surg. 46(4):301, 1976.Goldman, L., Weinberg, M., Weisberg, M., etal., “A Computer-Derived Protocol to Aid in theDiagnosis of Emergency Room Patients WithAcute Chest Pain,” N, Eng/. ]. Med. 307(10):588,1982.Greenbaum, D. M., and Holbrook, P. R., “Fel-lowship Programs in Critical Care Medicine—1982,” Crit. Care Med. 10(5):347, 1982.Greenbaum, D. M., and Holbrook, P. R., “Fel-lowship Programs in Critical Care Medicine,1982, Addendum,” Crit. Care Med. 10(11):785,1982.Grenvik, A., personal communication, Aug. 24,1983.Grenvik, A., “ ‘Terminal Weaning,’ Discontin-

.

92 ● Health Case Study 28: : Intensive Care Units: Costs, Outcome, and Decisionmaking

95.

96.

97.

98.

99.

100.

101

102,

103.

104.

105.

106.

107.

108.

1094

uance of Life-Support Therapy in the Terminally111 Patient,” Crit. Cm-e Med. 5:394, 1983.Grenvik, A., Leonard, J. J., Arens, J. F,, et al.,“Critical Care Medicine: Certification as a Multi-disciplinary Subspecialty, ” Crit. Care Med.9:117, 1981.Grenvik, A., Pouner, D. J., Snyder, J. V., et al.,“Cessation of Therapy in Terminal Illness andBrain Death,” Crit. Care Med. 6(4):284, 1978,Griner, P. F., “Medical Intensive Care in theTeaching Hospital: Costs Versus Benefits, Arm.intern. Med. 78(4):581, 1973.Griner, P. F., “Pulmonary Edema and the Inten-sive Care Unit, ” ch. 13 in Major issues in Criti-cal Care Medicine, J. E. Parrillo and S. M. Ayers(eds. ) (Baltimore, MD: Williams & Wilkins,1984).Griner, P. F., “Treatment of Acute PulmonaryEdema: Conventional or Intensive Care?” Ann.Intern. Med. 77(4):501, 1972.Griner, P. F., “Use of Laboratory Tests in aTeaching Hospital: Long-Term Trends: Reduc-tions in Use and Relative Costs, ” Ann. Intern.Med. 90:243, 1979.Griner, P. F., and Liptzin, B., “Use of the Lab-oratory in a Teaching Hospital: Implications forPatient Care, Education, and Hospital Costs, ”Ann. Intern. Med. 75(2):157, 1971.Guenter, C. A., and Welch, M. H., PubnonaryMedicine (Philadelphia: J. B. Lippincott Co.,1977).Gunby, P., “Compromise Reached on SuggestedIntervals Between Pap Tests,” ). A.M.A. 13(244):1411, 1980.Hackett, T. P., “The Psychiatrist’s View of theICU: Vital Signs Stable But Outlook Guarded,”Psychiatric Ann. 6:10, 1976.Hackett, T. P., Cassem, N. H., and Wishnie, H.A., “The Coronary-Care Unit: An Appraisal ofIts Psychologic Hazards,” N. Engl. ], Meal, 279(25):1365, 1968.Hadley, J., The Urban Institute, personal com-munication, June 8, 1983.Halbritter, R., Haider, M., Rackwitzf R., et al.,“Zur Prognose der Langzeitbeadmung auf einerinternen intensiv Station, ” Intensive Med. 16:233,1979.Haley, R. W., Schaberg, D. R., and Crossley, K.B., “Extra Charges and Prolongation of Stay At-tributable to Nosocomial Infections: A Perspec-tive in Our Hospital Comparison, ” Am. ]. Med.70:51, 1981.Harris, J., “Price Rules for Hospitals, ” Bell ]our-rzal of Economics 10:224, 1979.

110.

111.

112.

113.

114.

115.

116.

117.

118.

119.

120.

121

122,

123.

124.

Hauser, M. J., Tabak, J., and Baier, H., “Sur-vival of Patients With Cancer in a Medical Criti-cal Care Unit,” Arch. Zntern. Med. 142:527, 1982.Havighurst, C. C., Blumstein, J. F., and Bovb-jerg, R., “Strategies in Underwriting the Costs ofCatastrophic Disease, ” Law & ContemporaryProb/ems 40(4):122, autumn 1976.Helbing, C., “Medicare: Use of and Charges forAccommodation and Ancillary Services in Short-Stay Hospitals, 1979,” U.S. Department of Healthand Human Services, Office of Research, HealthCare Financing Administration, mimeographundated.Helbing, C., Supervisory Statistician, Office ofResearch, Health Care Financing Administration,U.S. Department of Health and Human Services,personal communication, June 6, 1983.Hewitt, P. B., “Subjective Follow up of PatientsFrom a Surgical Intensive Therapy Ward,” Brit.Med. ]. 4:669, 1970.Hilberman, M., “The Evolution of Intensive CareUnits,” Crit. Care Med. 3(4):159, 197s.Hook, E. W., Horton, C, A., and Schaberg, D.R., “Failure of Intensive Care Unit Support to In-fluence Mortality From Pneumococcal Bacteremia,”]. A,M.A. 249(8):1055, 1983.Hopkins, C. C., “Identification of Infection Prob-lems in Intensive Care Units, ” Crit, Care Quar-terly 3(4):1, 1981.Horn, S. D., “Does Severity of Illness Make aDifference in Prospective Payment?” HealthcareFinancing Management, 5:49, 1983.Howell, J. R., “Regulating Hospital Capital In-vestment: The Experience in Massachusetts, ” Na-tional Center for Health Services Research, Re-search Summary Series, U.S. Department ofHealth and Human Services publication No.(PHS) 81-3298 (Hyattsville, MD: U.S. Govern-ment Printing Office, 1981).Hubner, P. J. B., Goldberg, M. J., and Lawson,C. W., “Value of Routine Cardiac Monitoring inthe Management of Acute Myocardial InfarctionOutside a Coronary Care Unit, ” Brit. Med. }.1:815, 1969,Ibsen, B., “The Anaesthetist’s Viewpoint on theTreatment of Respiratory Complications in Po-liomyelitis During the Epidemic in Copenhagen,1952, ” Proc. Roy. Soc. of Med. 47:6, 1954,Iglehart, J. K., “The Politics of Transplantation, ”N. Engl. ], Med. 310(13):864, 1984.Imbus, S. H., and Zawacki, B. E., “Autonomyfor Burned Patients When Survival Is Unprece-dented,” Al. Engl, ]. Med. 297(6):308, 1977.Jackson, D. L., and Younger, S., “Patient Au-

References ● 93

125.

126.

127.

128.

129.

130.

131.

132.

133.

134.

135.

136.

137.

138.

139.

140.

141.

tonomy and ‘Death With Dignity,’ “ N. Engl, ].Med. 301(8):404, 1979.Jennett, B., “Resource Allocation for the SeverelyBrain Damaged,” Arch. Neurol, 33:595, 1976.Joint Commission on Accreditation of Hospitals,Accreditation Manual for Hospitals (Chicago, IL:1980).Jones, J., Hoggart, B., Withey, J., et al., “Whatthe Patients Say: A Study of Reactions to the In-tensive Care Unit, ” Intensive Care Med. 5:89,1979.journal of the American Medical Association,“Guidelines for Organization of Critical CareUnits,” 222(12):1532, 1972.Kalla, A. H., and Voss, E. C., Jr., ‘% IntensiveCare Unit Cost Effective?” The West VirginiaMed. ]. 76(7):151, 1980.Keene, A. R., and Cullen, D. J., “Therapeutic In-tervention Scoring System: Update 1983, ” Crit.Care Med. 11:1, 1983.Kiely, W., “Critical Care Psychiatric Syndromes, ”Heart & Lung 2(1):54, 1973.Kiely, W. F., and Prucci, W. R., “PsychiatricAspects of Critical Care, ” ch. 4 in Mosby’s Com-prehensive Review of Critical Care, D.A. Zschoche(cd.) (St. Louis, MO: The C. V. Mosby Co.,1981).Kirsch, J., “A Death at Kaiser Hospital,” Califor-nia 79-175, 1982.Klarman, H. E., Francis, J. D., and Rosenthal,A. D., “Cost-Effectiveness Analysis Applied tothe Treatment of Chronic Renal Disease, ” Medi-cal Care 6:48, 1968.Knaus, W. A., “Changing the Cause of Death, ”editorial, J. A.M.A. 249(8):1059, 1983.Knaus, W. A., and Thibault, G. E., “IntensiveCare Units Today, “ in Cn”tical Issues in MedicalTechnology, B. J. McNeil and E. G. Cravalho(eds. ) (Boston: Auburn House Publishing CO.,1982).Knaus, W. A., and Wagner, D. P,, “IntensiveTreatment for the Elderly: Comment” letter to theeditor, ]. A.M.A. 47:3185, 1982.Knaus, W. A., Draper, E. A., Wagner, D. P.,et al., “APACHE II: A Severity of Disease Clas-sification System for Acutely 111 Patients, ” sub-mitted for publication.Knaus, W. A., Draper, E. A., and Wagner, D.P., “Evaluating Medical-Surgical Intensive CareUnits,” ch, 4 in Major Issues in Critical CareMedicine, J. E. Parrillo and S. M. Ayers (eds. )(Baltimore, MD: Williams & Wilkins, 1984).Knaus, W. A., Draper, E. A., Wagner, D. P.,et al., “Evaluating Outcome From Intensive Care:A Preliminary Multihospital Comparison,” Crit.Care Med. 10:491, 1982.Knaus, W. A., Draper, E. A., and Wagner, D.

142.

143.

144.

145.

146.

147,

148.

149.

150.

151.

152.

153.

154.

155.

156.

157.

P., “The Use of Intensive Care: New Research Ini-tiatives and Their Implications for NationalHealth Policy, ” in Milbank Mere. Fund Q./Health & Society 61(4):561, fall 1983.Knaus, W. A., LeGall, J. R., Wagner, D. P., etal., “A Comparison of Intensive Care in theU.S.A. and France,” I.ancet ii(8299):642, 1982.Knaus, W. A., Zimmerman, J. E., Wagner, D.P., et al., “APACHE-Acute Physiology andChronic Health Evaluation: A PhysiologicallyBased Classification System,” Crit. Care Med.9:591, 1981.Knaus, W. A,, Wagner, D. P., Draper, E. A.,et al,, “The Range of Intensive Care ServicesToday,” J. A.M,A, 246(23):2711, 1981.Kornfeld, D. S., “Psychiatric View of the Inten-sive Care Unit, ” Brit. Med. ~. 1:108, 1969.LeGall, J. R., Brun-Buisson, C., Trunet, P., et al.,“Influence of Age, Previous Health Status, andSeverity of Acute Illness on Outcome From In-tensive Care, ” Crit. Care Med. 10(9):575, 1982.Lepore, M. J., and Grace, W. J., “Role of the In-tensive Care Unit in Gastroenterology, ” Am. ].Gastro. 51:493, 1969.Levenson, S, A., List, N. D., and Zaw-Win, B.,“Ethical Considerations in Critical and TerminalIllness in the Elderly, ” J, Am, Geriatr. Soc,29:563, 1981.Levy, D. E., Bates, D., Cardonna, J. J., et al.,“Prognosis in Nontraumatic Coma,” Ann. Intern.Med. 94:293, 1981.Lidz, C. W., Meisel, A., Osterweis, M., et al.,“Barriers to Informed Consent,” Ann. Intern.Med. 99:539, 1983.Linn, B. S., “Costs of Inpatient Burn Care, ” inCritical Issues in Medical Technology, B. J.McNeil and E. G. Cravalho (eds. ) (Boston, MA:Auburn House Publishing Co., 1982).Lo, B., and Jensen, A. R., “Clinical Decisions ToLimit Treatment, ” Ann. Intern. Med. 93:764,1980.Maloney, T. W., and Rogers, D. E., “MedicalTechnology—A Different View of the Conten-tious Debate Over Costs,” N. Engl. J, Med. 301(26):1413, 1979.Massachusetts General Hospital, Critical CareCommittee, “Optimum Care for Hopelessly 111Patients,” N. Engl. ~. Med. 295(7):362, 1976.McCleave, D. J., Gilligan, J. E., and Worthley,L. 1. G., “The Role and Function of an AustralianIntensive Care Unit, ” Crit. Care Med. 5(5):245,1977.McKegney, F. P., ‘The Intensive Care Syndrome:The Definition, Treatment and Prevention of aNew Disease of Medical Progress,” Corm. Med.30(9):633, 1966.McNeil, B. J,, and Hanley, J. A., “Statistical Ap-

—

94 ● Health Case Study 28: : lntensive Care Units: Costs, Outcome, and Decisionmaking

158,

159.

160,

161.

162.

163.

164.

165.

166.

167.

168.

169.

170.

171.172.

preaches to Clinical Predictions, ” editorial, N.Engl, J. Med. 304(21):1292, 1981,McPeek, B., Gilbert, J. P,, and Mosteller, F.,“The Clinician’s Responsibility for Helping ToImprove the Treatment of Tomorrow’s Patients, ”N. Engl, ]. Med. 302(11):630, 1980.Mechanic, D., “Approaches To Controlling theCosts of Medical Care: Short-Range and Long-Range Alternatives,” N. EngL J. Med. 298(5):249,1978.Micetich, K. C., Steinecker, P, H., and Thomasma,D. C., “Are Intravenous Fluids Morally Requiredfor a Dying Patient?” Arch. Intern. Med. 143:975,1983.Moore, F. D., and Zook, C. J., “High-Cost Usersof Medical Care,” N. Eng2. ]. Med. 302(18):996,1980.Morgan, A,, Daly, C., and Murawsi, B. J.,“Dollar and Human Costs of Intensive Care, ” ].%rg. Res. 14(5):441, 1973.Mulley, A, G., Thibault, G. E., Hughes, R. A.,et al., “The Course of Patients With SuspectedMyocardial Infarction: The Identification of Low-Risk Patients for Early Transfer From IntensiveCare,” N. Engl. ]. Med. 302(17):943, 1980.Murata, G. H., and Ellrodt, A. G., “Medical In-tensive Care in a Community Teaching Hospi-tal,” West. ~, Med. 136(5):462, 1982.Myers, L. P., and Schroeder, S, A., “PhysicianUse of Services for the Hospitalized Patient: AReview, With Implications for Cost Contain-ment,” Milbank Mere. Fund Q./Health & Society59:481, 1981.Myers, L. P., Schroeder, S. A., Chapman, S. A.,et al., “What’s So Special About Special Care?”]nquiry 21:113, 1984.Nadelson, T., “Psychiatric Aspects of IntensiveCare of Critically Ill Patients,” in Intensive Care,J. J. Skillman (cd. ) (Boston: Little, Brown& Co.,1975).Nahum, L. H., “Madness in the Recovery RoomFrom Open-Heart Surgery, or ‘They Keep Wak-ing Me Up, ‘ “ editorial, Corm. Med. 29:771, 1965.Nelson, H., “Life-Support Court Edict LeavesPhysicians Cautious, ” Los Angeles Times, Oct.31, 1983.Nelson, L., “Questions of Age–Doctors DebateRight to Stop ‘Heroic’ Effort to Keep ElderlyAlive,” The Wall Street ]ournal, Sept. 7:1(c01 1),1982.New York Times, Oct. 14, 1983, p. B5.Nightingale, F., Notes on Hospitals, 3d ed.Longman, Green, Longman, Roberts & Green,p. 89, 1863.

173.

174.

175.

176.

177.

178

179.

180.

181.

182.

183,

184.

185.

186,

187,

188.

189.

Northey, D., Adess, M, L., Hartsuck, J. M., etal., “Microbial Surveillance in a Surgical Inten-sive Care Unit, ” Surgery 139:321, 1974.Nunn, J. F., Milledge, J. S., Singaraya, J., “Sur-vival of Patients Ventilated in an Intensive Ther-apy Unit,” Brit. Med. ~. 1:1525, 1979.Parno, J. R., Teres, D., Lemeshow, S., et al.,“Hospital Charges and Long-Term Survival ofICU Versus Non-ICU Patients,” Crit. Care Med.10(9):569, 1982.Parrillo, J. E., and Ayres, S. M. (eds. ), MajorIssues in Critical Care Medicine (Baltimore, MD:Williams & Wilkins, 1984).Pearlman, R. A., Inui, T. S., and Carter, W. B.,“Variability in Physician Bioethical Decision-Making: A Case Study of Euthanasia, ” Ann. 2n-tem. Med. 97:420, 1982.Pessi, T. T., “Experiences Gained in IntensiveCare of Surgical Patients: A Prospective Clini-cal Study of 1,003. Consecutively Treated Patientsin a Surgical Intensive Care Unit, ” Ann. Chir.Gynecol. Fenn, 62: Suppl. 185:3, 1973.Petty, T. L., “Don’t Just Do Something—StandThere,” Arch. Intern. Med. 139:920, 1979.Petty, T. L., and Nett, L. M., “The RespiratoryCare Department,“ in Intensive & RehabilitativeCare, 2d ed. (Philadelphia: Lea& Febiger, 1974).Petty, T. L., Lakshminarayan, S., Sahn, S. A.,et al., “Intensive Respiratory Care Unit: Reviewof Ten Years Experience,” J, A,M.A, 34:233, 1975.Piettre, A,, “Aspects of Economics, Ethics andCivilization,” lntes. Care Med. 3:253, 1979.Piper, K. W., and Griner, P. F., “Suicide At-tempts With Drug Overdose; Outcomes of Inten-sive vs. Conventional Floor Care, ” Arch. Jntern.Med. 134:703, 1974.Pitner, S. E., and Mance, C. J., “An Evaluationof Stroke Intensive Care: Results in a MunicipalHospital,” Stroke 4:737, 1973.Pius XII, “The Prolongation of Life,” The PopeSpeaks, Vatican City, 1958.Platt, R., Medical Director of Bronx MunicipalHospital, personal communication, July 8, 1983.Plunkett, O., comments on Intensive Care Unitsin section of Anesthetics, Proc. Roy. Soc. ofMed. 59:1293, 1966.Policy Analysis Inc., Evaluation of Effects ofCertificate-of-Need Programs, prepared for theHuman Resources Administration, U.S. Depart-ment of Health and Human Services, contractNo. 231-77-0114, 1981.Pozen, M. W., D’Agostino, R, B., Mitchell, J, B.,et al., “The Usefulness of a Predictive InstrumentTo Reduce Inappropriate Admissions to the Cor-

References ● 95

onary Care Unit, ” Ann. Intern. Meal, 92:238,1980.

190. Pozen, M. W,, D’Agostino, R. B., Selker, H. P,,et al., “A Predictive Instrument To Improve Cor-onary Care Unit Admission Practices on AcuteIschemic Heart Disease— A Prospective Multi-center Clinical Trial,” N. Engl. ]. Med. 310(120):1274, 1984.

191. President’s Commission for the Study of EthicalProblems in Medicine and Biomedical and Be-havioral Research, Deciding To Forego Life-Sustaining Treatment: Ethical, Medical, andLegal Issues in Treatment Decisions (Washington,DC: U.S. Government Printing Office, 1983).

192. Preston, G. A., Larson, E. L., and Stamm, W. E.,“The Effect of Private Isolation Rooms on PatientCare Practice, Colonization and Infection in anIntensive Care Unit,” Am. ]. Med. 70:641, 1981.

193. Rabkin, M. T., Gillerman, G., and Rice, N. R.,“Orders Not To Resuscitate, ” N. Engl. ~. Med.295(7):364, 1976.

194. Reichel, W., “The Continuity Imperative, ” edi-torial, J, A.M.A. 246(18):2065, 1981.

195. Relman, A. S., “Intensive Care Units: Who NeedsThem?” N. Engl, ]. Med. 302(17):965, 1980.

196. Rescher, N., “The Allocation of Exotic MedicalLifesaving Therapy,” Ethics 79(3):173, 1969.

197. Robertson, J. A., “Legal Aspects of WithholdingMedical Treatment From Handicapped Chil-dren,” ch. 22 in Legal and Ethica~ Aspects ofTreating Critically and Terminally 111 Patients,A. E. Doudera and J. D. Peters (eds.) (Ann Ar-bor, MI: Association of University Programs inHealth Administration, 1982).

198. Robin, E. D., “A Critical Look at Critical Care,”guest editorial, Crit. Care Med. 11(2):144, 1983.

199. Robinson, D., Abramson, N. S., Grenvik, A,,et al,, “Medicolegal Standards for Critical Care, ”Crit. Care Med. 8:524, 1980.

200. Robinson, J. S., comments on Intensive CareLJnits in Section of Anesthetics, Proc. Roy. Soc.of Med. 59:1293, 1966.

201. Rodman, G. H., Etling, T., Civetta, J. M., et al.,“How Accurate Is Clinical Judgment?” Crit. CareMed. 6:127 (abstract), 1978.

202. Roe, B. B., “The UCR Boondoggle: A DeathKnell for Private Practice, ” N. Engl. ]. Med.305(1):41-45, July 2, 1981.

203. Rogers, R. M., Weiler, C., and Ruppenthal, B.,“Impact of the Respiratory Intensive Care Unitin Survival of Patients With Acute RespiratoryFailure,” Chest 77:501, 1972.

204. Rose, H. D., and Babcock, J, B., “Colonizationof Intensive Care Unit Patients With Gram-

Negative Bacilli,” A m . ]; Epidemio2. 101:495,1975,

205. Russell, L. B., “Intensive Care, ” ch. 3 in Tech-nology in Hospitals: Medical Advances and TheirDiffusion (Washington, DC: The Brookings In-stitution, 1979).

206. Russell, L. B., “The Role of Technology Assess-ment in Cost Control, ” in Critical issues in Med-ical Technology, B. J. McNeil and E. G. Cravalho(eds.) (Boston: Auburn House Publishing Co.,1982).

207. Safar, P., “The Critical Care Medicine Contin-uum From Scene to Outcome, ” ch. 7 in MajorIssues in Critical Care Medicine, J. E. Parrillo andS. M. Ayres (eds.) (Baltimore, MD: Williams &Wilkins, 1984).

208. Safar, P., and Grenvik, A., “Critical Care Medi-cine: Organizing and Staffing Intensive CareUnits,” Chest 59(5):535, 1971.

209. Safar, P., and Grenvik, A., “Organization andPhysician Education in Critical Care Medicine, ”Anesthesiology 47:82, 1977.

210. Safa.r, P., Dekornfeld, T. J., Pearson, J. M., et.al., “The Intensive Care Unit,” Anesthesia 16(3):275, 1961.

211. Salkever, D. S., and Bice, T. W., ‘The Impactof Certificate-of-Need Controls on Hospital In-vestment,” Milbank Mere. Fund Q./Health &Society 185, 1976.

212. Sanders, C. A., “Hospital Management of Criti-cal Care—I, ” presentation at the National Insti-tute of Health Consensus Development Confer-ence, Critical Care Medicine, Mar. 8, 1983.

213. Schelling, T. C., “The Life You Save May BeYour Own,” Problems in Public ExpenditureAnalysis, S. Chase (cd. ) (Washington, DC: TheBrookings Institution, 1968).

214. Scheffler, R. M., Knaus, W. A., Wagner, D. P.,et al., “Severity of Illness and the RelationshipBetween Intensive Care and Survival, ” Am. },Publ. Health 725:449, 1981,

215. Schmidt, C. D., Elliott, C. G., Carmelli, D., etal., “Prolonged Mechanical Ventilation for Res-piratory Failure: A Cost-Benefit Analysis, ” Crit.Care Med. 11(6):407, 1983.

216. Schoenfeld, M. R., “Terror in the ICU, ” Forum1:14, 1978.

217. Schroeder, H. G., “Psycho-Reactive Problems ofIntensive Therapy,” Anesthesia 26(1):28, 1971.

218. Schroeder, S. A., Showstack, J. A., and Robert,H. E., “Frequency and Clinical Description ofHigh-Cost Patients in 17 Acute-Care Hospitals,N. Engl. ]. Med. 300:1306, 1979.

219. Schroeder, S. A., Showstack, J. A., and Schwartz,


220.

221,

222,

223.

224.

225.

226.

227.

228.

229.

230.

231.

232.

233.

234.

235.

236.

J., “Survival of Adult High-Cost Patients, Reportof a Follow-Up Study From Nine Acute-CareHospitals,” ~. A.M.A. 245(14):1446, 1981.Schwartz, S., and Cullen, D. J., “How Many In-tensive Care Beds Does Your Hospital Need?”Crit. Care Med. 9(9):625, 1981.Shapiro, A. R., “The Evaluation of ClinicalPredictions,” N, Engl. ]. Med. 296(26):1509,1977.Shoemaker, W. C., “Effectiveness of the Inten-sive Care Unit for Management of Accidental,Traumatic and Hemorrhagic Shock,” ch. 9 in Ma-jor Issues in Critica~ Care Medicine, J. E. Parrilloand S. M. Ayers (eds. ) (Baltimore, MD: Williams& Wilkins, 1984).Shoemaker, W. C., Appel, P. L., Bland, R., etal., “Clinical Trial of an Algorithm for OutcomePrediction in Acute Circulatory Failure, ” Crit.Care Med. 10(6):390, 1982.Shoemaker, W. C., Thompson L., and Hol-brook, P. (eds. ), Textbook on Critica2 CareMedicine (Philadelphia: W. B. Saunders, 1983).Shragg, T. A., and Albertson, T. E., “Moral,Ethical and Legal Dilemmas in the Intensive CareUnit,” Crit. Care Med. 12:62, 1984.Shubin, H., and Weil, M. H., “Touchstones inCritical Care Medicine-An Introduction,” Crit.Care Med. 21:281, 1974.Siegel, J. H., Cerra, F. B., Moody, E. A., et al.,“The Effect on Survival of Critically 111 and In-jured Patients of ICU Teaching Service Orga-nized About a Computer-Based PhysiologicCARE System,” ]. Trauma 20(7):558, 1980.Siegler, M., “Critical Illness: The Limits of Au-tonomy,” Hastings Center Report 7, 1977.Siegler, M., “Pascal’s Wager and the Hanging ofCrepe,” N. Engl. ]. Med. 293(17):853, 1975.Singer, D., Carr, P., Mulley, A., et al., “Ration-ing Intensive Care—Physician Responses to a Re-source Shortage, “N, Engl. ]. Med. 309(19):1155,1983.Skidmore, F. D., “A Review of 460 Patients Ad-mitted to the Intensive Therapy Unit of a Gen-eral Hospital Between 1965 and 1969, ” Brit. ~.Surg. 60:1, 1973.Skillman, J. J. (cd.), Intensive Care (Boston: Lit-tle, Brown & Co., 1975).Skillman, J. J., “Terminal Care in Patients WithChronic Lung Disease, “ Arch. intern. Med. 139,1979.Smits, H. L., “The PSRO in Perspective,” N.Engl. J. Med. 305(5):253, 1981.Snyder, J. V., McGuirk, M., Grenvik, A., et al.,“Outcome of Intensive Care—An Application ofa Predictive Model, ” Crit. Care Med. 9(8):598,1981.Society of Critical Care Medicine, Committee on

237.

238.

239.

240.

241.

242.

243.

244.

245.

246.

247.

248.

249.

250.

Training Guidelines, P. M. Winter (Chin.),“Guidelines for Training of Physicians in Criti-cal Care Medicine, ” Crit. Care Med. 1(1):39,1973.Soler, N. G., Bennett, M. A., Fitzgerald, M. G.,et al., “Intensive Care in the Management ofDiabetic Ketoacidosis, ” Lancet 4:951, 1973.Spagnolo, S. V., Hershberg, P. I., and Zimmer-man, H. J., ‘?vfedical Intensive Care Unit: Mor-tality Rate Experience in Large Teaching Hospi-tal,” NY State ~. Med. 73:754, 1973.Steel, K., Gertman, P. M., Crescenzi, C., et al.,“Iatrogenic Illness on a General Medical Serviceat a University Hospital,” N. Engl. ]. Med. 304(11):638, 1981.Stoughton, W. V., “Medical Costs and Technol-ogy Regulation: The Pivotal Role of Hospitals, ”in Critical Issues in Medical Technology, B. J.McNeil and E. G. Cravallo (eds. ) (Boston: Au-burn House Publishing Co., 1982).Sullivan, R., “Court in Jersey Forbids RemovingFeeding Tube From Terminally Ill,” New YorkTimes July 9:1 (co1. 1), 1983.Swan, H. J. C., “Invasive Hemodynamic Mon-itoring in Critical Care Units, ” ch. 12 in MajorIssues in Critical Care Medicine, J. E. Parrillo andS. M. Ayres (eds.) (Baltimore, MD: Williams &Wilkins, 1984).Tagge, G. F., Adler, D., and Bryan-Brown,C. W., “Relationship of Therapy to Prognosis inCritically Ill Patients, ” Crit. Care Med. 2(2):61,1974.Tagge, G. F,, Salness, G., Thams, J., et al., “Ex-perience With a Multidisciplinary Critical CareCenter in a Community Hospital,” Crit. CareMed. 3(6):231, 1975.Tait, M., and Winslow, G., “Beyond Consent:The Ethics of Decision Making in EmergencyMedicine,” West. ]. Med. 126(2):156, 1977.Teplick, R., Caldera, D. L., Gilbert, J. P., et al.,“Benefit of Elective Intensive Care AdmissionAfter Certain Operations, ” Anesth. Analg. 62:572, 1983.Teres, D., Brown, R, B., and Lemeshow, S.,“Predicting Mortality of Intensive Care Unit Pa-tients: The Importance of Coma, ” Crit. CareMed. 10(2):86, 1982.Thibauk, G. E., “The Medical Intensive CareUnit: A Five-Year Perspective,” ch. 2 in MajorIssues in Cn”tica2 Care Medicine, J. E. Parrillo andS. M. Ayers (eds. ) (Baltimore, MD: Williams &Wilkins, 1984).Thibault, G. E., Mulley, A. G., Barnett, C. O.,et al., “Medical Intensive Care: Indications, In-terventions, and Outcomes, ” N. Eng/. ]. Med.302(17):938, 1980.Thomas, L., “Notes of a Biology-Watcher: The

References ● 97

Technology of Medicine,” N. Engl. ~, Med. 285(24):1366, 1971.

251. Tomlin, P. J., “Intensive Care—Medical Audit,”Anesthesia 33:710, 1978.

252. Turnbull, A. D., Carlon, G., Baron, R., et al.,“The Inverse Relationship Between Cost and Sur-vival in the Critically Ill Cancer Patients, ” Crit,Care Med. 7(1):20, 1979.

253. Turnbull, A. D., Goldmeir, P,, Silverman, D.,et al., “The Role of an Intensive Care Unit in aCancer Center: An Analysis of 1035 Critically 111Patients Treated for Life-Threatening Complica-tions, ” editorial, Cancer 37:82, 1976.

254. U.S. Congress, Office of Technology Assessment,Diagnosis Related Groups (DRGs) and the Medi-care Program: Implications for Medical TechnoLogy, OTA-TM-H-17 (Washington, DC: U.S.Government Printing Office, July 1983).

255. U.S. Congress, Office of Technology Assessment,Health Technology Case Study #24: Variationsin Hospital Length of Stay: Their Relationship toHealth Outcomes, OTA-HCS-24 (Washington,DC: U.S. Government Printing Office, August1983).

256, U.S. Congress, Office of Technology Assessment,The Implications of Cost-Effectiveness Analysisof Medical Technology, OTA-H-126 (Washing-ton, DC: U.S. Government Printing Office,August 1980).

257. U.S. Congress, Office of Technology Assessment,Medical Technology and Costs of the MedicareProgram, OTA-H-227 (Washington, DC: U.S.Government Printing Office, July 1984).

258. U.S. Department of Commerce, Bureau of theCensus, State and Metropolitan Area Data Book1982 (Washington, DC: U.S. Department ofCommerce, 1982).

259. U.S. Department of Health and Human Services,Health Care Financing Administration, Office ofResearch and Demonstration, Bureau of DataManagement and Strategy, “List of DiagnosisRelated Groups (DRGs) With Relative Values,Average Length of Stay (ALOS), and Percent ofAdjusted Charges Per Discharge for AncillaryServices, ” 1979 data.

260, U.S. Department of Health and Human Services,Health Care Financing Administration, Office ofResearch and Demonstration, Bureau of DataManagement and Strategy, “Routine and SpecialCare Arithmetic Mean Length of Stay for Outlierand Non-Outlier Cases, ” Oct. 27, 1983.

261. U.S. Department of Health and Human Services,Health Care Financing Administration, Office ofResearch and Demonstration, Bureau of Data

Management and Strategy, “20 Percent SampleDischarges by Diagnosis Related Group AdjustedSample,” June 21, 1983.

262. U.S. Department of Health and Human Services,National Institutes of Health, Consensus Devel-opment Conference Statement, “Critical CareMedicine,” Mar. 7-9, 1983.

263. U.S. Department of Health and Human Services,National Institutes of Health, Consensus Devel-opment Conference statement, “Supportive Ther-apy in Burn Care, ” Nov. 10-11, 1978.

264. U.S. Department of Health and Human Services,Report to Congress: Hospital Prospective Pay-ment for Medicare (Washington, DC: U.S. De-partment of Health and Human Services, Decem-ber 1982).

265. Vanholder, R., and Colardyn, F., “Prognosis ofIntensive Care Patients: Correlation of Diagno-sis and Complications to Patient Outcome, ” ActaClinica Belgica 35(5):279, 1980.

266. Veatch, R, M., A Theory of Medical Ethics (NewYork: Basic Books, 1981).

267. Wagner, D. P., “National Costs of IntensiveCare,” unpublished manuscript, undated.

268. Wagner, D. P., Knaus, W. A., and Draper,E. A., “Case Mix, Resource Costs, and Severityof Illness, ” unpublished manuscript draft, 1983.

269. Wagner, D. P., Knaus, W. A., Draper, E. A.,et al., “Identification of Low-Risk Monitor Pa-tients Within a Medical-Surgical Intensive CareUnit,” Medical Care 21:425, 1983.

270. Wagner, D. P., Knaus, W. A., and Draper,E. A., “Statistical Validation of a Severity of Ill-ness Measure, ” Am. ~. Pub. Health 73(8):879,1983.

271. Wagner, D. P., Wineland, T. D., and Knaus, W.A., “The Hidden Costs of Treating Severely 111Patients: Charges and Resource Consumption inan Intensive Care Unit, ” Health Care Finan.Rev.5(1):81, 1983.

272. Wallace-Barnhill, G. L., Roth, M. D., Arm-strong, C. J., et al., “Health Care Law Update:Legal Protection for Critical Care Physicians:State of the Art in Termination of Life Supportand Living Will Legislation, ” Crit. Care Med.12:56, 1984.

273. Wanzer, S. H., Adelstein, S. J., Cranford, R. E.,et al., “The Physician’s Responsibility TowardHopelessly Ill Patients, “N, Engl. }. Med. 310(7):455, 1984.

274. Weil, M. H., “The Society of Critical Care Medi-cine, Its History and Its Destiny, ” Cn”t. Care Med.1(1):1, 1973.

275. Weil, M. H., and Rackow, E. C., “Critical Care

98 ● Health Case Study 28: : Intensive Care Units: Costs: Outcome, and Decisionmaking

276<

277<

278.

279.

Medicine—Caveat Emptor,” Arch. Intern. Med.143:1391, 1983.Weinberg, S. L., “Intermediate Coronary Care:Observations on the Validity of the Concept,”Chest 73:154, 1978.Weinstein, M. C., and Fineberg, H. V., ClinicalDecision Arudysis (Philadelphia: W. B. Saunders,1980).Weiser, B., “AS They Lay Dying, ” WashingtonPost [Apr. 17: Part I, 1 (co]. 1), 1983; Apr. 18:Part II, 1 (coI. 2), 1983; Apr. 19: Part III, 1 (co1.1), 1983; Apr. 20: Part IV, 1 (coI. 4), 1983; Apr.21: Part V, 1 (co1, 5), 1983. ]Williams, S. V., “Can Randomized Trials of In-tensive Care Meet Ethical Standards?” in Criti-cal Issues in Medical Technology, B. J. McNeiland E. G. Cravalho (eds. ) (Boston: AuburnHouse Publishing Co., 1982).

280.

281.

282.

283.

Wilson, J. B., “Medical Dilemmas in TerminalCare,” in Death by Decision (Philadelphia: West-minster Press, 1975).Wilson, R. F., “Development and Cost Effective-ness of Surgical Intensive Care Units, ” in Crit.Care Meal,, program abstract from the NationalHealth Institute of Health Consensus Develop-ment Conference, Mar. 8, 1983.Young, W. R., “It’s a Miracle That We Save Anyof Them,” Life Magazine, Dec. 2, 1966.Zimmerman, J. E., “Administrative Structure ofa Critical Care Unit, ” ch. 25 in Major Issues inCritical Care Medicine, J. E. Parrillo and S. M.Ayers (eds.) (Baltimore, MD: Williams & Wil-kins, 1984).