How are Rural Hospitals

Using Hospitalists?

Michelle Casey, MS

Peiyin Hung, MSPH

Ira Moscovice, PhD

AcademyHealth | June 2013

Supported by the Office of Rural Health Policy,

Health Resources and Services Administration,

PHS Grant No. U1CRH03717

Background and Objectives

• Paucity of research with rural context– Unclear whether prior research findings are

generalizable to smaller rural facilities

– Hospitalists may be part-time with additional

responsibilities

• Study objectives were to examine:

– How hospitalists are being used in rural hospitals

– Which rural hospitals are using hospitalists

Trends in Hospitalist Use 2005-2010

8%11%

14%16% 18%

21%26%

32%

41%44%

56%61%63%

69%75%

79% 81%85%

2005 2006 2007 2008 2009 2010

CAHs All Other Rural Hospitals Urban Hospitals

Data Source: AHA Annual Surveys

Data Sources

• Primary data from a national phone survey of

rural hospitals May-August 2011

– Survey sample: Rural hospitals with <100 beds using

hospitalists in AHA annual survey

– Respondents: Hospital CEOs (2/3) and

clinical/administrative staff (1/3)

– Response rate: 86.4% (N=402)

– Statistical analysis of quantitative survey data and

qualitative analysis of open-ended responses

• Secondary data from American Hospital

Association Annual Surveys FY 2010

Primary Reasons for Using Hospitalists

Requests from

physicians,

26.6%

Improve care

quality /

continuity,

19.3%

Cover

unassigned

patients, 6.9%

Alleviate physician

shortage, 6.6%

Cover call/reduce

workload for

physicians, 16.4%

Allow physicians

to focus on clinical

practice, 10.3%

Recruit & retain

physicians, 8.6%

Hospitalist Specialties

82%

57%

24%

6%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

Internal

Medicine

Family Practice PAs and/or NPs Emergency

Medicine

Perc

en

t o

f su

rvey h

osp

itals

Additional Care Provided by Hospitalists

Care Settings by Hospitalists Percent of

hospitals

Hospital Outpatient Department 30%

Primary care in clinic or physician office 30%

Emergency Department 17%

Perceived Quality Impacts

Positive, 84.4%

Positive &

negative,

9.8% Hospitalists:

• Are available when needed & quick to

respond, spend more time with

patients

• Improve quality, patient safety

measures, communication with nurses,

teamwork

• Possess expertise, ability to handle

more acute patients

• Provide consistent, standardized care;

use evidence-based medicine

No change,

5.8%

Perceived Financial Impacts

No change, 5.8%

Positive,

44.6%

Both

positive &

negative,

16.9%

Negative,

32.6%

• Increased admissions

• Improved primary care

physician productivity,

ability to treat higher-

acuity patients

• Reduction in patient

complications, avg.

length of stay, transfers

• Costs more than

revenue generated

• Hospital has to

subsidize hospitalist

program

• High costs of

hospitalists’ salaries or

contracts

Perceived Financial Impacts

No change, 5.8%

Positive,

44.6%

Both

positive &

negative,

16.9%

Negative,

32.6%

Perceived Recruitment / Retention

Impacts

Easier,

74.4%

No

change,

25.4%

• PCPs don’t want to do

inpatient care or want

flexibility in doing it

• Reduced call, more

work/life balance for PCPs

• New candidates are only

interested in places with

hospitalists

Harder

0.6%

Which Facilities Are Using Hospitalists?

• Data

– AHA Annual Survey 2010

– Medicare payment classification data from University of

North Carolina at Chapel Hill

• Sample: Rural hospitals (n=1,462)

• Multivariate logistic regression model calculated

probabilities of hospitalist use given a hospital

characteristic

Measures• Dependent variable

– Binary hospitalist use variable

– 27% of rural hospitals had missing value

• Explanatory variables

– Inpatient days

– Medicare payment classification

– Total primary care physicians with admitting privileges

– System membership

– Ownership

– Census Divisions

Probability of Hospitalist Use by Small

Rural Hospitals

Variables Est. Std. Err. P-value

Medicare Payment Classification

[Reference = Rural PPS]

Critical Access Hospital (CAH) -9.2% 3.7% 0.012

Medicare Dependent Hospital (MDH) -8.6% 4.4% 0.054

Sole Community Hospital (SCH) -3.7% 3.8% 0.331

Rural Referral Center (RRC) 5.0% 4.6% 0.271

Total Inpatient Days [Reference = Quartile 1]

Quartile 2 (2,188 – 4,212) 15.6% 3.4% <.001

Quartile 3 (4,213 – 9,259) 24.3% 3.8% <.001

Quartile 4 (>9,259) 37.2% 5.5% <.001

Regression Results cont.

Variables Est. Std. Err. P-value

Total Primary Care Physicians 3.8% 1.1% <.001

Total Primary Care Physicians2 -0.1% 0.1% 0.048

System Member 4.6% 2.2% 0.039

Ownership

(Reference = Public/Government)

Private Non-Profit 7.5% 2.5% 0.002

For-Profit 6.8% 4.2% 0.107

Conclusions

• Hospitalist use by rural hospitals increased

threefold, 2005-2010

• In rural hospitals, hospitalists:

– Are family physicians and non-physician

providers as well as internists

– Frequently play multiple roles – also providing

outpatient, emergency, and/or primary care

• Hospitalists can help address workforce

shortages in rural areas

Conclusions cont.

• Financial impact of hospitalist use is more

complex than costs vs. revenue:

– Enhance recruitment, retention and efficiency of PCPs

– Care for unassigned and uninsured patients

• Likelihood of hospitalist use varies by:

– Type of Medicare reimbursement (prospective

payment vs. cost-based)

– Inpatient volume

Peiyin Hung

612-623-8317

hungx068@umn.edu

http://www.hsr.umn.edu/rhrc/

Medicare Payment

Classification• Critical Access Hospitals

– 25 or fewer beds

– 101% of reasonable costs for inpatient, outpatient and swing bed

care.

• Sole Community Hospitals

– located either 35 miles from similar hospitals

– receive the higher of the federal PPS rate or an updated hospital-

specific rate based on historical costs.

• Medicare Dependent Hospitals

– fewer than 100 beds and more than 60% of inpatient discharges or

days covered by Medicare

– received the PPS rate plus 75% of the difference between the PPS

rate and an updated rate based on their historical costs.

• Rural Referral Centers

– have a combination of high case mix intensity and specialist supply

– more than 275 beds

– reimbursed using urban PPS rates.

Survey Respondents and Non-Respondents

Hospital Characteristics

Survey Respondents with

Hospitalist Programs

(N = 329)

Non-

Respondents

(N = 73)

Staffed inpatient hospital beds

25 or fewer beds* 140 (42.6%) 13 (25.0%)

26 – 50 beds 72 (21.9%) 14 (26.9%)

51 – 100 beds 117 (35.6%) 25 (48.1%)

Ownership

Government, non-federal* 71 (21.6%) 19 (36.5%)

Private not-for-profit*** 229 (69.6%) 24 (46.2%)

For-profit 29 (8.9%) 9 (17.3%)

Census Region

Northeast 53 (16.1%) 5 (9.6%)

Midwest* 94 (28.6%) 7 (13.5%)

South*** 116 (35.3%) 35 (67.3%)

West 66 (20.1%) 5 (9.6%)

System Member* 164 (50.0%) 34 (65.4%)*p<0.05 **p<0.01 ***p<0.01