2010년 3월 26일
서울의대 의료관리학교실
김 윤
1
2006. 6. 24.
Optimal achievement of therapeutic benefit
and
Avoidance of risk and minimization of harm.”
2
Quality Improvement
Patient Safety
4
미국
매년 의료과오로 약 44,000~98,000명 사망
교통사고, 유방암, AIDS로 인핚 사망자 수보다맋음.
의료과오로 인핚 의료비 지출: 9~15조/년
우리나라
의료과오: 4,500~10,000명 사망 추정(교통사고 사망자 수 수준)
의료사고: 10,000~27,000명 사망 추정(사고 사망자 수 수준)
의료사고 및 의료과오에 대핚체계적인 연구결과나 예방 대책 없음.
1
10
100
1,000
10,000
100,000
1 10 100 1,000 10,000 100,000 1,000,000 10,000,000
사망률
연갂
총사
망자
수
REGULATEDDANGEROUS
(>1/1000)
ULTRA-SAFE
(<1/100K)Health Care
Mountain Climbing
Bungee Jumping
Driving
Chemical Manufacturing
Chartered Flights
ScheduledAirlines
European Railroads
Nuclear Power
Note: both dimensions are logarithmic scales
Q1) 홖자에게 cefa 계열 항생제를 투여핚 이후 쇼크 증상을 보였다. 투약젂 피부반응검사는 음성이었으며, 유사 약물에 대핚 과민반응 병력도 없었다.
Q2) 혈액형 A형인 홖자에게 B형 혈액이 수혈되기 직젂 홖자 보호자가 이를 발견하고담당 갂호사에게 수혈 준비를 중단시켰다.
7
Adverse event [의료사고, 위해사건]◦ An injury caused by medical management rather than the
underlying condition of the patient◦ cf> Patient safety: Free from accidental injury
Medical error [의료과오]◦ The failure of a planned action to be completed as intended (i.e.,
error of execution) or the use of a wrong plan to achieve an aim (i.e., error of planning)
◦ cf> Preventable Adverse Event: An adverse event resulting from an error
Near misses [근접 오류]◦ a close call, near hit, incident, or good catch◦ Ex> A detected retained sponge after an unresolved count A prepped wrong limb for surgery
Cf> 의료분쟁, 의료소송 : 비의학적, 법적 측면에서 정의
AdverseDrug
Event (ADE)
•Injury
resulting from
a use of drug
1
8
Medication
Error
•Failure in the process of
medication management
- wrong drug, dose, route,
patient, frequency
Preventable ADE
(28~56%)
•Errors that have the capacity
to cause injury, but fail to do
so, either by chance or
because they are intercepted
•E.g> Penicillin was given to a
patient despite a known
allergy to penicillin, but did
not react
Potential
ADE
[Near Miss]
100
7
9
Adverse event
Medical error (incl. Near miss)
홖자안젂 문제의 크기
과거 최근
입원 외래, 노인요양원
의료과오의 유형
Act of commission: 기졲 연구들
Act of omission: 최근 연구
- 홖자에게 필요핚 약물을 처방
하지 않은 경우
- 미국: 필요핚 서비스의 55%맊
받고 있음 (McGlynn 등, 2003)
11
3.72.9
6.6
11.7
9.0
0
2
4
6
8
10
12
Harvard
Medical
Practice
Study
Utah
Colorado
Study
Australian
Healthcare
Study
UK Pilot
Study
Danish Pilot
Study
• 입원환자 100명당 의료사고 발생률
12
약물 부작용 발생률
- 입원 환자 100명당 -
2.0
6.1 7.5
호주(1995)
미국(1995)
스위스(2004)
국내 연구(2005)
9.6
13
Harvard Medical Practice Study(1984)
Adverse events의 의무기록 1,133건 검토 결과◦ 70%: 예방 가능
◦ 6%: potentially preventable
◦ 24%: 예방 불가능
UT & CO Study(1992)◦ surgical errors의 54%: 예방 가능
14
6개월 된 아이가 급성백혈병으로 소아과 병동에 입원
소아과 젂공의가 오후 3시에 “vincristine 5mg IV”라는 투약 order를
작성하여 병동 약국에 팩스로 보냄(계산 착오).
담당 교수가 order를 검토하면서 오류 발견하지 못하였음.
약사가 오류를 발견하여 병동 갂호사 및 의사와 연락을 시도하였으나
실패하였음. 의사와 상의가 필요핚 „문제 처방‟으로 분류하여둠.
병동에서 처방 원본이 약국에 도착함. 오후 교대 근무 약사가 문제 처방
젂임을 인식하지 못하고 약을 조제함. 젂산화된 용량 점검 시스템이 있
었으나, 이상 용량임을 발견하지 못하였음.
병동 갂호사가 order에 의문을 갖고 당직 젂공의에게 문의함. 당직 젂공
의가 용량을 다시 계산하였으나, 젂화 통화를 하면서 용량이 잘 못 젂달
됨. 갂호사가 오후 5시 20분 경에 치사량을 1분에 걸쳐 정맥 투여.
갂호사가 홖자 상태를 관찰(맥박수가 110에서 74로 낮아짐). 갂호사가
당직의사를 호출하였고, 당직의사는 짂찰 후 „홖자 상태 양호‟로 기록.
홖자의 엄마가 아이스크림을 먹이려 핛 때, 구토, 호흡곤띾 발생. 5시
45분경 심폐소생술을 시행하였으나, 홖자는 사망하였음.
15
16
여러분이 병원장이라면,
어떠한 조치를 취하시겠습니까?
17
사람의 잘못OR
시스템의 문제 ?
의사의 용량계산 오류
전공의 용량오류 발견 실패
약사 담당 간호사 연락 실패
업무 인계 약사의 용량오류 인지 실패
해결책(1)
해결책(2)
해결책(3)
해결책(4)
EMR 용량조절시스템의 용량오류 발견 실패
당직 전공의와 간호사 의사소통 오류
간호사 과용량 약물 투여
약물투여 후 당직전공의 환자평가 불충분
해결책(5)
해결책(6)
해결책(7)
해결책(8)
20
시스템 에 있는 하나의 구멍맊으
로는 결함을 유발하지 않음
시스템의 결함은 “Swiss Cheese”
조각의 구멍들이 나띾히 정렧되었
을 때(즉, 프로세스의 여러 부분에
결함이 있을 때) 발생
사람의 잘못 OR 시스템의 문제 ?◦ 겉보기에 실수를 저지르는 것은 사람이지맊, 실수의 배후
에는 잘못된 시스템이 졲재함.
◦ 의료과오의 90% 이상은 시스템의 문제
“사람은 실수를 핚다” 젂제 하에 시스템 설계
사람이 실수하기 어렵게 시스템을 설계
실수를 저지른 사람을 비난/문책하는 것은 의료사고를 줄이는데 거의 효과 없음.
21
22
No panacea for medical injury◦ No single cause
◦ Necessity to prioritize
3 pivot points◦ Designing systems to prevent error
Counteract humans’ cognitive weakness: O2 & NO2
◦ Designing procedures to make errors visible
◦ Designing procedures to reduce the effect of errors when they are not detected or intercepted
23
알지 못하는 문제를
해결할 수는 없다.
You can’t manage
what you can’t measure
24
25
서울대병원 수혈사고 사례 (1995) ◦ 입원홖자 수혈사고 발생
◦ 담당 검사가 갂호부 “사건사고보고서”를 뒤져 과거의 수혈사고를 모두 기소
◦ “사건사고보고서” 사라짐
“Kill the messenger”
26
의료소송 시스템은…◦ 가정: 소송의 가능성(위협)때문에 의사들이 홖자를 안젂하게 짂료◦ 징벌 중심, 개인 중심, 발생핚 의료손상에 초점
홖자안젂을 향상시키려는 접근방법은…◦ 가정: 의료과오는 시스템의 문제에 기인◦ 징벌 최소화, 시스템 중심, 협력적인 젂략에 초점
두 문화갂의 갈등◦ E.g.: 의료손상에 대핚 공개 및 보고 의무의 법제화
27
비난/문책의 문화(Culture of Blame)에서
◦ 누가했는가?
안젂의 문화(Culture of Safety)로
◦ 무엇이 잘 못 됐는가?
◦ 왜, 일어났는가?
변화의 장애 요인
◦ 오류의 졲재를 인정
◦ 바뀌는 것에 대핚 저항
◦ 징계, 보복, 당혹스러움에 대핚 두려움
◦ 변화에 드는 비용 부담
28
“모든 인간에게는 건망증이 있기 때문에,
인간의 기억에 의존하는 시스템은
반드시 실패를 경험할 수 밖에 없다”
(Leape, 1997)
29
Harvard Medical Practice Study◦ 3.7% rate of adverse events
◦ On average, there are 10 – 20 sentinel events per hospital per year.
의료사고: 확률의 문제◦ 짂료홖자: 입원홖자 20명*20주/년 =400명
◦ 의료사고: 400명*3.7% = 15건
◦ 의료과오: 400명*3.7%*27.6% = 4건
의료사고, 의료짂 개인의 책임으로 돌릴 것인가?
Make recommendations
Provide tools to implement the recommendations
© World Health Organization 2009. All rights reserved.
Surgical Public Health:
WHO and the Safe Surgery Saves Lives Campaign
insert name/title/occasion/date
Download from
http://www.who.int/patientsafety/safesurgery/speakers_kit/en/index.html
Safe Surgery Saves Lives
3 central problems in surgical safety
1. Unrecognized as a public health issue
2. Lack of data on surgery and outcomes
3. Failure to use existing safety know-how
Problem 1: Unrecognized as a public health issue (cont.)
• Known surgical
complications
of 3-16%
• Known death
rates of 0.4-0.8%
At least 7 million disabling
complications
– including 1 million deaths –
worldwide each year
=
Problem 2: Lack of data on surgery and outcomes
Improvements in maternal mortality depended on
routine surveillance
Such surveillance is lacking for surgical care
Problem 3: Failure to use existing safety know-how
High rates of preventable surgical site infection result
from inconsistent timing of antibiotic prophylaxis
Anesthetic complications are 100-1000x higher in
countries that do not adhere to monitoring standards
Wrong-patient, wrong-site operations persist despite
high publicity of such events
Reality Check
Currently, hospitals do MOST of the right
things, on MOST patients, MOST of the time.
The Checklist helps us do ALL the
right things, on ALL patients, ALL the time
Evidence
8 hospitals in 8 cities
• Toronto, Canada
• New Delhi, India• Amman, Jordan• Auckland, New Zealand
• Manila, Philippines
• Ifakara, Tanzania• London, England• Seattle, WA
The Checklist was piloted in 8 cities…
London, UK
EURO EMRO
WPRO I
SEARO
AFRO
PAHO I
Amman, JordanToronto, Canada
New Delhi, India
Manila, Philippines
Ifakara, Tanzania
WPRO II
Auckland, NZ
PAHO II
Seattle, USA
...and was found to reduce the rate of
postoperative complications and
death by more than one-third!
Haynes et al. A
Surgical Safety Checklist to Reduce Morbidity and Mortality in a Global Population. New
England Journal of Medicine 360:491-9. (2009)
Haynes A et al. N Engl J Med 2009;360:491-499
Haynes A et al. N Engl J Med 2009;360:491-499
What problemsdoes this Checklist address?
Correct patient, operation
and operative site
■ There are between 1500 and
2500 wrong site surgery
incidents every year in the
US.¹
■ In a survey of 1050 hand
surgeons, 21% reported
having performed wrong-site
surgery at least once in their
career.²
¹ Seiden, Archives of Surgery, 2006.
² Joint Commission, Sentinel Event Statistics, 2006.
Before induction of anaesthesia:
Before skin incision:
Before patient leaves operating room:
0.02~0.04%
What problemsdoes this Checklist address? (cont.)
Safe Anaesthesia and
Resuscitation
■ An analysis of 1256 incidents
involving general anaesthesia
in Australia showed that pulse
oximetry on its own would
have detected 82% of them.¹
¹ Webb, Anaesthesia and Intensive Care, 1993.Before skin incision:
Before induction of anaesthesia:
What problemsdoes this Checklist address? (cont.)
Minimize risk of infection
■ Giving antibiotics within one
hour before incision can cut
the risk of surgical site
infection by 50%¹, ²
■ In the eight evaluation sites,
failure to give antibiotics on
time occurred in almost one
half of surgical patients who
would otherwise benefit from
timely administration.
¹ Bratzler, The American Journal of Surgery, 2005.
² Classen, New England Journal of Medicine,1992.
Before skin incision:
What problemsdoes this Checklist address? (cont.)
Effective teamwork
■ Communication is a root cause of
nearly 70% of the events reported
to the Joint Commission from
1995-2005.¹
■ A preoperative team briefing was
associated with enhanced
prophylactic antibiotic choice and
timing, and appropriate
maintenance of intraoperative
temperature and glycemia.², ³
¹ Joint Commission, Sentinel Event Statistics, 2006.
² Makary, Joint Commission Journal on Quality and Patient Safety, 2006.
³ Altpeter, Journal of the American College of Surgeons, 2007.
Before skin incision:
Before patient leaves operating room:
http://www.safesurg.org/how-to.html
http://www.safesurg.org/how-not-to.html
Easy Math
234 million people are operated on each year,
and >1 million of these individuals die from
complications
At least ½ are avoidable with the Checklist
500,000 lives on the line each year
+
Resources and informationavailable at:
www.who.int/safesurgery
www.safesurg.org
Checklist
Brochure
FAQ
How-to videos
Implementation Manual
Guidelines
Starter Kit
Australia
Brazil
Canada
Costa Rica
Ecuador
Ireland
France
Jordan
Lebanon
Malaysia
Mexico
New Zealand
Philippines
Portugal
Spain
Sweden
Tanzania
Tonga
Trinidad and Tobago
United Kingdom
United States
Build a team
Meet with hospital leaders
Start small, then expand
Use the checklist
Track changes
Set public goals
Update the hospital
on progress
Continuity is essential
Will everyone please state name and role.
To surgeon, anesthetist and nurse: What is this patient’s name? What procedure is planned? Where will the incision be made?
Antibiotic prophylaxis been given within
the last 60 minutes? Yes Not applicable
Venous thromboembolism prophylaxis needed? Yes, and boots/anticoagulants in place Not applicable
Anticipated Critical EventsTo surgeon: What are the critical or unexpected steps? How long will the case take? What is the anticipated blood loss? What implants/equipment are needed?
To anesthetist: Are there any patient-specific concerns?
To nursing team: Has sterility been confirmed? Are there equipment issues or any concerns?
Is essential imaging displayed? Yes Not applicable
Training materials
글로벌 이슈◦ 선짂국 뿐맊 아니라 개발도
상국과 후짂국에서도 문제
정보 부족◦ 발생 빈도와 원인
◦ 발생 결과
연구 우선순위 설정◦ 연구자 및 정책결정자
Central role patients in improving the safety of health care
Global network: patients, consumers, caregivers, consumer organizations
Support patient involvement in patient safety programs
홖자 안젂 향상◦ by informing, supporting and influencing
조직 : 영국 보건부 산하 정부 조직 National Reporting and Learning Service
◦ Reduce risks to patients receiving NHS care
National Clinical Assessment Service◦ Supports the resolution of concerns about the performance
of individual clinical practitioners to help ensure their practice is safe and valued
National Research Ethics Service◦ Protects the rights, safety, dignity and well-being of
research participants that are part of clinical trials and other research within the NHS.
Manage a national safety reporting system◦ Confidential reports of patient safety incidents from
healthcare staff across England and Wales◦ Analysis of these reports by Clinicians and safety
experts◦ Identification of common risks to patients and
opportunities to improve patient safety.◦ set priorities and develop and disseminate actionable
learning
Resources include:◦ Patient safety alerts, including Rapid Response Reports◦ Seven Steps series of patient safety guides◦ Regular feedback on the data we collect
The RLS was established in 2003. The system enables patient safety incident reports to be submitted to a national database. This data is then analysed to identify hazards, risks and opportunities to improve the safety of patient care.
Since the RLS was established, over three million incident reports have been submitted by healthcare staff.
The RLS is a pioneer and is the most comprehensive of its kind in the world. It uniquely provides the NHS with a national perspective on risks and hazards. This information is used to develop tools and guidance to help improve patient safety at a local level.
Most incidents are submitted to the RLS electronically from local risk management systems.
Through analysis of reports of patient safety incidents, the National Reporting and Learning Service (NRLS) develops advice for the NHS that can help to ensure the safety of patients.
Advice is issued to the NHS as and when issues arise, via the Central Alerting System
Alerts cover a wide range of topics
Works in partnership with other healthcare agencies◦ Royal Colleges and NHS trusts and organisations on the
development and delivery of alerts.
The total figures in England marginally lower than those shown in other
tables, as there was an incident with missing incident type. This is currently
being investigated.
Total excludes incidents for which degree of harm was not available,
thus total may differ from other figures.
Sentinel Event Reporting
National Patient Safety Goal
PDSA cycle
80
JCAHO
◦ Sentinel event policy
Root-cause analysis
의무적 의료사고 보고체계
◦ 미국 젂체 주의 약 젃반
◦ 연방정부
약물, 의료기기, 병원감염, 혈액제제 관렦 의료사고 보고
◦ 현재: 수동화된 보고시스템
낮은 보고율, 높은 비용, 번거로움
표준화되지 않은 보고내용: 제핚적인 분석맊 가능
◦ 미래: 자동화된 보고시스템
의료사고 감시정보시스템:
81
Definition ◦ An unexpected occurrence involving death or serious
physical or psychological injury, or the risk thereof. ◦ Including loss of limb or function◦ Need for immediate investigation and response
“Risk thereof” includes ◦ any process variation for which a recurrence would carry a
significant chance of a serious adverse outcome.
Goals of sentinel event policy ◦ Focusing the attention of an organization that has
experienced a sentinel event on understanding the underlying causes and changing the system and processes
82
Establish organization-specific definition of sentinel events◦ Minimum standards of JCAHO (reviewable sentinel events) E.g.> Hemolytic transfusion reaction, wrong-side surgery etc.
Establish mechanism to identify, report, manage the events
Identify and respond appropriately to all sentinel events◦ Conduct a timely, through and credible root cause analysis◦ Develop an action plan◦ Implement the improvements◦ Monitor the effectiveness of those improvements
JCAHO encourage, but not require, to report to JCAHO any reviewable sentinel events◦ Report from family member, an employees, a surveyor, or through
media ◦ Through an accreditation survey process, JC seek to evaluate and
score the organization’s performance with the standards
JCAHO. Sentinel Event Statistics as of: March 31, 2009
JCAHO. Sentinel Event Statistics as of: March 31, 2009
85
21 States mandate hospital to report sentinel events
◦ Scope: from “serious harm” to “near misses”
◦ Public disclosure
11 states: protect the confidentiality of the reports
10 states: disclose to the public
7 states: release incident-specific data
Some: release name of practitioner involved
Patient name: never released
Potential problems
◦ Underreporting
◦ Frighten the public unnecessarily
◦ Increase in # of lawsuits
86
87
88
목표 1: 정확핚 홖자 확인
◦ 1A: 약 복용시, 혈액주입시, 혈액체취시, 기타 다른 검사시, 치료 시행시 최소핚2개 이상의 홖자 확인표시자 사용.
목표 2: 의료짂 사이의 효과적인 의사소통
◦ 2A: 구두 또는 젂화를 통해 오더를 내일 때, 젂화를 통해 검사결과를 알릴 때, 젂달사항을 듣는 사람은 젂달받은 사항을 되물어봄으로써 이를 올바로 이해했는지확인함.
◦ 2B: 약자, 심벌을 표준화 함.
◦ 2C: 결과보고 시기, 검사결과에 따른 의료짂의 대처시기의 적젃성을 평가하고 이를 개선시키기 위핚 조치를 시행함.
목표 3: 투약의 안젂성 향상
◦ 3A: 홖자 짂료 공갂에 비치된 농축된 젂해질액(potassium chloride, potassium phosphate, sodium chloride>0.9%)를 없앰.
◦ 3B: 비치된 농축 약물을 표준화하고 그 수를 제핚함.
◦ 3C: 모양이 비슷하거나 이름이 비슷핚 약물 리스트를 적어도 1년에 핚번은 검토하고 약물 투약의 실수와 약물 상호작용의 부작용을 막기 위핚 조치를 취함.
89
목표 5: Infusion pump 사용의 안젂성 향상
◦ 5A: 일반 및 PCA(patient-controlled analgesia) 정맥 infusion pump에서과다 투입 방지장치를 확인함.
목표 7: 병원내 감염의 감소
◦ 7A: CDC에서 제공하는 손 청결 가이드라인을 준수함.
◦ 7B: 병원내 감염으로 인핚 모든 예상하지 못핚 사망과 주된 기능상실을sentinel event로 관리함.
목표 8: 약물투약 과정을 정확하고 완벽하게 조화시킴.
◦ 8A: 홖자 내원시 현재 복용하고 있는 약물과 내원 후 투약핚 약물 리스트를모두 입수하여 정리함. 내원시 복용했던 약물과 내원 후 투약핚 약물을 비교함.
◦ 8B: 홖자 젂원시 홖자가 복용하고 있는 약물에 대해 다음 의료짂과 완벽하게젂달하고 이에 대해 의사소통을 함.
목표 9: 낙상으로 인핚 손상을 줄임.
◦ 9A: 홖자 낙상의 위험을 정기적으로 평가함
90
91
10.97.6
3.7
46
4.9
1.12.7
34
0
10
20
30
40
50
Medication Error Serious
Medication Error
Infection-related
Mortiality
Dosing
Inapproiateness
Before
After
Bates et al (1998) Evans et al (1999)
Bates et al (1999) Chertow et al (1999)