Transcript Document

First Aid to an ailing health system?
Can you help?
Clinical Associate Professor T. Hannan
FRACP;FACHI;FACMI
July 2013
THEMES
• COSTS OF CARE-are national fundamentals different?
• HEALTH CARE IS AN INFORMATION MANAGEMENT BUSINESS-(this
is what clinicians do)
• INFORMATION OVERLOAD-technology driven
• CARE OUTCOMES-current measures
– $/Variation/Communication/Quality/Access/ Readmission rates
• PATIENTS AS CARE MANAGERS-challenges
• DO ANY TOOLS HELP US?
•
•
NEW PUBLICATIONS:
Hannan T, Celia C, Are doctors the structural weakness in the e-Health building? Intern
Med J; 2013; 43: October; **2013
•
Editorial JAMA July 2013: Will Physicians Lead on Controlling Health Care Costs?
Ezekiel J. Emanuel, MD, PhD; Andrew Steinmetz, BA
HEALTH CARE IS UNAFFORDABLE! [NEJM 2012]-WORLDWIDE
Health Expenditures as a Percentage of Gross Domestic Product (GDP) in Selected OECD Countries, 1960–2009.
AUSTRALIA
Fineberg HV. N Engl J Med 2012;366:1020-1027.
HEALTH COSTS PER CAPITA OF GDP-OECD
Australia in top 1/3
Australian Health Care
Estimates of health expenditure in Australia 2000–01 and 2010–11.
(After adjustment for inflation)
$77.5 billion in 2000–01
$87 billion 2005-06 9% of GDP
$122.5 billion in 2009–10
$130.3 billion in 2010–11
Estimated 16-20% of GDP by 2045
“Analysis of health inflation suggests that prices in the health sector
have grown quicker than in the broader economy over the past 10
years.”
http://www.aihw.gov.au/WorkArea/DownloadAsset.aspx?id=10737423003
7 July 2015
Health care is a service business
• What clinicians deliver…
–
–
–
–
–
7 July 2015
advice
medication
devices
surgery
physical therapy
Health care is an information business
7 July 2015
Health care is an information business
• What clinicians actually do…
–
–
–
–
–
find information (prior records)
“There is no healthcare without
gather
informationand
(history,
lab)
management,
therephysical,
is no
record
informationwithout
(notes,information.”
reports, etc.)
management
process information (risks/benefits
→ Neto
decisions)
Gonzalo Vecina
Head,
Brazilianorders,
National Health
transmit information
(advice,
letters)
Regulatory Agency
• The quality, efficiency, and effectiveness of care
depend on our ability to manage information
7 July 2015
→ Electronic Health Records
DATA/INFORMATION/KNOWLEDGE
TSUNAMI
7 July 2015
7 July 2015
10
A WORLDWIDE HUMAN LIMITED INFORMATION
MANAGEMENT AND KNOWLEDGE ACCESS CAPACITY
PROBLEM INDEPENDENT OF THE HEALTH CARE MODELLING
“We must remove ourselves from the ‘unscientific, non
data driven personal recommendations’ for care.”
Dr M Smith CHCF 2009
THREATS TO QUALITY OF CARE
1. OVERUSE-receiving treatment of no value
2. UNDERUSE –failing to receive needed treatment
3. MISUSE-errors and defects in treatment
L. Leape. Five Years After To Err Is Human. What Have We Learned? JAMA. 2005;293:2384-2390
“If we keep practicing medicine as we know it today,
healthcare will become an unbearable burden on all
modern societies in a very near future. We are in a real
race between healthcare innovation that can change this
worrisome trend and the resistance to change inherent
to our healthcare system.” E. Topol.”The Creative Destruction of Medicine.”
7 July 2015
12
MGH's Inpatient Mortality Rate (Brown) and Adjusted Cost per Patient Who Was Discharged Alive (in 2010
MGH's Inpatient Mortality Rate (Brown) and Adjusted Cost per Patient Who
Was Discharged Alive (in 2010 Dollars; Blue), 1821–2010.
Dollars; Blue), 1821–2010.
Meyer GS et al. N Engl J Med 2012;366:2147-2149.
Poorly or Unsupported Clinical Decision Making
RESOURCE UTILISATION-CANADA
OVERUSE-5% CKD patients 25% Duplicate testing
Duplicate Lab Tests* by Group, BC, 2005.
0.45
2003
Number of Lab Tests (Millions)
0.4
# Duplicate Lab Tests in 2005 = 1.14M
COST = $4.55M
0.35
2004
2005
0.3
0.25
~$4.55 M (~$4.50/test)
0.2
0.15
0.1
0.05
0
CVD
DM-CVD
CKD-CVD-DM
CKD-CVD
DM
CKD
DM-CKD
* duplicate test defined as same test within 30 days
Dr. Adeera Levin, Director, Kidney Function Clinic, St. Paul's Hospital, University of British Columbia, Rm.
6010-A, 1081 Burrard St., Vancouver BC V6Z 1Y6; fax 604 806-8120; [email protected]
CANADA 2003-2005
7 July 2015
15
OVERUSE / INAPPROPRIATE USE OF RADIOLOGY /
PHARMACY RESOURCES
CANADA 1999-2009
• Prescriptions-community pharmacies• 272 million (1999) to 483 million (2009).
• Appropriate vs. Inappropriate use?
• CT scanners -198 to 465
• MRI scanners- 19 to 266 from federal investments.
• Number of Scans:
• 58% increase CT scans
• 100% increase MRIs. (Compared to 2003)
• Source: www.healthcouncilcanada.ca
Heather Dawson Director, Analysis and Reporting, Health Council of Canada Healthcare
Policy Vol.6 No.4, 2011
7 July 2015
16
UNITED KINGDOM:
RESOURCE UTILISATION- DEM AFTER HOURS
Resource Utilisation - 1998/99
87% Unnecessary out-of-hours tests
80% Diagnostic uncertainty
79% Medico-legal protection *
66% Avoid leaving work for colleagues
71% Prevent criticism from staff (especially Consultants)
76% Lessen anxiety and reduce stress levels
71% Agreed attempts should be made to reduce unnecessary testing
McConnell AA, Bowie P. Health Bull (Edinb). 2002 Jan;60(1):40-3.
Unnecessary out-of-hours biochemistry investigations--a subjective view of necessity.
7 July 2015
17
ACROSS THE CORRIDOR – LGH
Overuse/Underuse/Misuse
$m’s
DEM
•
•
•
•
•
•
•
Nurse automatic ordering directive
RMO / Registrar
Poor communication tools
Lines of responsibility blurred
Poor documentation
Appropriateness/Timeliness of
ordering
Post discharge results
AMU
PRACTISING UNDER THE FEAR OF
LITIGATION
Without individualized data, physicians assume
that they are performing at tolerable rates.
2000 Project HOPE—People-to-People Health Foundation, Inc. Health Affairs, March/April 2000
Medicare Pharmacy Coverage: Ensuring Safety Before Funding by Lee N. Newcomer
Is More Testing Better?
The “diagnosis of uncertainty”-effects on clinical
decision-making behaviour, costs and outcomes.
(Takes CDM further away from the Dx)
1. N Engl J Med. 1975 Jul 31;293(5):229-34. Therapeutic decision making: a cost-benefit analysis. Pauker SG,
Kassirer JP.
2. Johns RJ, Blum BI. The use of clinical information systems to control cost as well as to improve care. Trans Am Clin
Climatol Assoc. 1979;90:140-52.
7 July 2015
19
COMMUNICATION
Prof L. Weed 1989
• They are highly motivated, and if they are not, nothing works in the long run anyway
• They do not charge. They even pay to help
• There is one for every member of the population
1994-2007
COMMUNICATION IN HEALTH CARE.
GPs -CONSULTANT COMMUNICATION
3% HOSPITALISED PATIENTS.
25% DISCHARGE SUMMARIES NEVER ARRIVED
75% DELAY IN DISCHARGE SUMMARY 25.3 DAYS
(20.8 DAYS TO TYPE SUMMARIES IN HOSPITAL)
60% STANDARD LETTERS ARE NOT READ
90% REFERRAL LETTERS CONTAIN NO INFORMATION
RELEVANT TO THE PROBLEMS RELATED TO
REFERRAL- MOST ILLEGIBLE
P.J. Branger, J.S.Duisterhout. Communication in Health Care. JAMIA. 1994;69-77
7 July 2015
1994-2007
COMMUNICATION-HOSPITALS TO PRIMARY CARE- Kriplani:JAMA 2007
• Direct communication Hosp-PCP 3-20%
• Availability of Discharge Summary
• 1st post discharge visit-12-34%
• 4 weeks-51-77%
• Affect on QOC of FU visits-25%
• PCP dissatisfaction HIGH
Communication lacking important information
Diagnostic test results missing
33-63%
Treatment or hospital course
7-22%
Discharge medications
20-40%
Test results pending at discharge
65%
Patient or family counselling
90-92%
Follow-up plans
2-43%
7 July 2015
ADVERSE EVENTS -IDENTIFICATION AND PREVENTION
2006 Institute of Medicine -nearly 1.5 million
preventable adverse drug events each year.
Hasan, S., G. T. Duncan, et al. "Automatic detection of omissions in medication lists." J Am Med
Inform Assoc 18(4): 449-58.
“Most hospitals rely on spontaneous voluntary
reporting to identify adverse events, but this method
overlooks more than 90% of adverse events detected by
other methods .........Retrospective chart review
improves the rate of adverse event detection but is
expensive and does not facilitate prevention.”
Potential identifiability and preventability of adverse events using information systems. D Bates et.al J Am Med Informatics Assoc. 1994;1:404-411
7 July 2015
23
Medication-related malpractice claims
Causes, Preventability, and Costs.
• ADE - 6.3% of claims
• Preventable 73%
• IP vs. OP 50% each
• 46% were life threatening or fatal.
Indemnity costs for preventable inpatient ADEs mean, $376,500
Indemnity costs for non-preventable ADE ~$36,000
ADE associated with malpractice claims -severe, costly, and
preventable.
Rothschild JM, Federico FA, Gandhi TK, Kaushal R, Williams DH, Bates DW. Analysis of medication-related
malpractice claims: causes, preventability, and costs. Arch Intern Med. 2002 Nov 25;162(21):2414-20
7 July 2015
Trends in 30-Day Readmission Rates, 2002–2009.
National Trends National
in 30-Day
Readmission Rates, 2002–2009.
Joynt KE, Jha AK. N Engl J Med 2012. DOI: 10.1056/NEJMp1201598
Mortality at 30 Days among All Hospitals, According to Pay-forPerformance Status 2002–2009among All Hospitals, According to Pay-forPerformance Status, 2002–2009.
No evidence that the largest hospital-based pay-for-performance program
led to a decrease in 30-day mortality.
Jha AK et al. N Engl J Med 2012. DOI: 10.1056/NEJMsa1112351
CURRENT HEALTH DATA MEASURMENT TOOLS
Case-Mix/DRGs/Activity-Based Funding
• Lack of a robust measurement program
• Take years to collect
• No nationally agreed-on methods for systematically
identifying, tracking, and reporting adverse events.
• A shortage of good patient-safety metrics
• Poor quality measures are plentiful.
• Current patient-safety indicators, which use billing data
have poor sensitivity and specificity- their utility varies
with hospitals’ billing practices.[Case-Mix, DRGs, ABF]
• INFLATIONARY to health care costs
Ashish K. Jha, David C. Classen, M.DGetting Moving on Patient Safety — Harnessing Electronic Data for
Safer
365;19 NEJM.org 1756 November 10, 2011
7 JulyCare..NEJM
2015
Australian PCEHR-WHAT IS ITS
FUNCTIONALITY?
SLOW LEARNERS-ON QUALITY AND PATIENT SAFETY
2000-To Err Is Human Building a Safer Health System. INSTITUTE OF MEDICINE.
2005 -Leape, L.L. and D.M. Berwick, Five years after To Err Is Human: what have we learned? JAMA.
2011- Health Information Technology Institute Of Medicine, Health IT and Patient Safety Building Safer
Systems for Better Care, The National Academies
Press: Washington D.C.
2011-Jha, A.K. and D.C. Classen, Getting moving on patient safety--harnessing electronic data for safer
care. N Engl J Med.
7 July 2015
Time to tackle unwarranted variations in practice
THE VARIATION PHENOMENON
“The variation phenomenon in modern medicine -the observation of
differences in the way apparently similar patients are treated from one health
care setting to another.”
D. Blumenthal. Editorial NEJM 331:1994;1017-8
Much of the variation in use of healthcare is accounted for by the willingness
and ability of doctors to offer treatment rather than differences in illness or
patient preference.
Variation that cannot be explained on the basis of illness, patient preferences
or the dictates of evidence-based medicine.
Identifying and reducing such variation should be a priority for providers.
(John Wennberg 2011-Dartmouth Institute)
Effective/Preference-sensitive/Supplysensitive Care
• Dartmouth Atlas Project researchers have distinguished between three
types of services:
• (1) “Effective Care”: interventions that are viewed as medically necessary
on the basis of clinical outcomes evidence and for which the benefits so
outweigh the risks that virtually all patients with medical need should
receive the them. E.g. #NOF
• (2) “Preference-sensitive Care”: treatments, such as discretionary
surgery, for which there are two or more valid treatment alternatives,
and the choice of treatment involves trade-offs that should be based on
patients’ preferences.
• (3) “Supply-sensitive Care”: services such as physician visits, referrals to
• specialists, hospitalizations and stays in intensive care units involved in
the medical (non-surgical) management of disease. In Medicare, the
large majority of these services are for patients with chronic illness.
Consumers: demand more services when they pay a lower share
of the costs (e.g. National Health Insurance schemes).
Higher-spending regions:
• do not have higher life expectancies or better health outcomes .
• Variation in treatments is greatest for the types of care for which
we lack evidence about relative effectiveness.
Despite the high cost of the U.S. health care system (and other
health systems) the degree to which it promotes the population’s
health remains unclear-substantial evidence exists that more
expensive care doesn’t always mean higher quality care.
The Challenge of Rising Health Care Costs — A View from the Congressional Budget
Office Peter R. Orszag, Ph.D., and Philip Ellis, Ph.D. N Engl J Med 2007; 357:17931795November 1, 2007
7 July 2015
Per capita Medicare spending varies considerably from
region to region-2003
Medicare spending on quality of care and access.
Using end-of-life care spending as an indicator of Medicare spending
Geographic regions into five quintiles of spending and examined costs and
outcomes of care for;
•hip fracture
•colorectal cancer
•acute myocardial infarction.
Outcomes:
Residents of high-spending regions received 60% more care but
did not have better quality or outcomes of care.
The Implications of Regional Variations in Medicare Spending. Part 1: The Content, Quality, and Accessibility of Care. Elliott S.
Fisher, MD, MPH; David E. Wennberg, MD, MPH; The´ re` se A. Stukel, PhD; Daniel J. Gottlieb, MS; F.L. Lucas, PhD; and ´ Etoile
L. Pinder, MS. Ann Intern Med. 2003;138:273-287
7 July 2015
.
GEOGRAPHIC VARIATIONS IN PHARMACY SPENDING
Using this more complete measure of spending reveals that arealevel variation in total spending is not driven primarily by patient
characteristics.
Geographic Variation in the Quality of Prescribing Yuting Zhang, Ph.D., Katherine Baicker, Ph.D., and Joseph P. Newhouse, Ph.D.
N Engl J Med 2010; 363:1985-1988November 18, 2010
7 July 2015
Does more $ per capita improve care?
Greater spending and higher supply-sensitive
care-beneficial?
•
•
•
•
Greater spending with higher use of supply-sensitive care: i.e. more doctors, etc.
High spending vs. Low Spending-69% more days in hospital pp than low spending.
154% MORE physician visits
Patients see MORE physicians in the last 6/12 of life.
IS MORE $ ON HEALTH BETTER WITH
CURRENT MODELS? THE ANSWER -NO
Better
Health
Worse
Health
7 July 2015
Individual US
States
Less state
spending
Less state
spending
Regional differences in Medicare spending are largely
explained by the more inpatient-based and specialist oriented
pattern of practice observed in high-spending regions. Neither
quality of care nor access to care appear to be better for
Medicare enrolees in higher-spending regions.
The Implications of Regional Variations in Medicare Spending. Part 1:The Content, Quality, and Accessibility
of Care Elliott S. Fisher, MD, MPH; Ann Intern Med. 2003;138:273-287.
7 July 2015
EFFECTIVE CARE-CAP management-Variation in Quality
(1) had evidence of being vaccinated;
(2) received antibiotics in timely fashion;
(3) had their level of oxygenation measured.
PREFERENCE SENSITIVE CARE
“Elective surgery” problem
DOES THE SUPPLY OF SURGEONS HELP?
These relationships have NOT changed over time 1992-93 to 2000-2001 and applies to
many other procedures.
SHARED DECISION MAKING
• After the implementation of shared decision making, the populationbased rates of prostatectomy fell 40%, providing a measure of
demand under shared decision making. (Rates in the control group,
Group Health Cooperative’s Tacoma site, did not change.) J.
Wennberg 2005
SUPPLY-SENSITIVE CARE-non surgical
The behavioural basis of this association must rest in Roemer’s law -- the long
held hypothesis that hospital beds, once built (and staffed), tend to be filled.
The impact of beds per capita on clinical decision making is subliminal, in the
sense that clinicians are unaware of differences in practice style associated with
bed capacity.
SUPPLY-SENSITIVE CARE-non surgical (ctd)
Unwarranted variation in the rates of supply-sensitive care.
Of concern here is:
• The frequency of physician visits (and revisits)• In the absence of evidence-based guidelines on the appropriate interval
between revisits, available capacity governs the frequency of revisits
• Hospitalizations,
• Stays in intensive care units,
• Referrals to specialists,
• Use of imaging and other diagnostic tests.
ICU UTILISATION
There is considerable variation in the rates of use of intensive care in managing chronic
illness among America’s best hospitals. During the last six months of life we observed
more than a five-fold variation.
Reasons for practice variation
Complexity: How many factors can the human mind simultaneously
balance to
optimize an outcome?
Lack of valid and poor access to clinical knowledge
-(poor evidence)
Subjective judgment / uncertainty
Subjective evaluation is notoriously poor across groups or over time
and enthusiasm for unproven methods
Human error- -- humans are inherently fallible information
processors- -- Clem MacDonald, PhD
Technology is NOT the problem. RMRS 2012
(est. 1976)
Regenstrief Institute: April 2012: 18 hospitals
• >32 million physician orders entered by CPOE
• Data base of 6 million patients
• 900 million on-line coded results
• 20 million reports-diagnostic studies,
procedure results, operative notes and
discharge summaries
• 65 million radiology images
• CLINICAL DECISION SUPPORT- BLINK TIMES
(CCDSS-through iterative Dbase analysis)
7 July 2015
CCDSS SYSTEMS IN CLINICAL MEDICINE
THAT IMPROVE CLINICAL INFORMATION MANAGEMENT
(2000)
•
•
•
•
•
•
•
•
•
•
•
Alerting
Reminding
(Pryor TA, Clayton PD. -1990s)
Interpretation
Assisting
Critiquing
Diagnosing
Managing
Knowledge Access /Coupling (L.Weed-1990s)
Summarization (Fries/Whiting-O’Keefe-1980s)
Quality improvement using clinical data (B. James-1990s)
Variation in Care (J. Wennberg-1970s-2010)
7 July 2015
54
CCDSS IMPROVING RESOURCE UTILISATION,
OUTCOMES
$3 million per year savings ~$64bUS(1995) (? 2013-$tr)
0
-2
-4
-6
-8
-10
-12
-14
-10.5
-12.7 -11.9 -12.5
TOTAL
BED
TEST
DRUG
OTHER
LOS
-15.3 -15.2
-16
Physician inpatient order writing on microcomputer workstations-effects on resource
7 July 2015 WM Tierney and others. JAMA 1993;269:379-383
55
utilisation.
CCDSS(EHR) 1996
COSTS/QUALITY/OUTCOMES/RESEARCH
“The plural of anecdote is not data”
160,000 patient over 4 years
Overall antibiotic use:
Mortality rates:
Antibiotic-associated ADE:
Antibiotic resistance:
Appropriately timed preoperative a/biotics:
Antibiotic costs per treated patient:
Acquisition costs for antibiotics:
decreased 22.8%
decreased from 3.65% to 2.65%
decreased 30%
remained STABLE
40% to 99.1%
decreased $122.66 to $51.90
fell 24.8% to 12.9%
($987,547) to ($612,500)
Our Case-Mix index which measures patient acuity levels
INCREASED during this period, meaning we were treating
sicker and sicker patients while better utilizing the delivery of
antibiotics. (******WENNBERG 2012)
Pestotnik, S. L. Classen, D. C. Evans, R. S. Burke, J. P. Implementing antibiotic practice guidelines through
computer-assisted decision support: clinical and financial outcomes.Ann Intern Med 1996 May 15
7 July 2015
TAKING CONTROL-Patient Centered Computing
“The Wisdom of Crowds”-James Surowiecki
2015-every adult in the world will have a mobile phone-(WHO)
SYSTEM DESIGN/USER INTERFACE
Eleanor: “Clayton, the Internet is a corner I will not turn.”
Eleanor: “There is no mail here!”
Eleanor: “What’s a scroll bar?”
Permissions: Clayton Lewis, Professor of Computer Science, University of Colorado NAS. Fostering Independence,
Participation, and Healthy Aging Through Technology: Workshop Summary
Gregg Vanderheiden, Director Trace R&D Center, Professor Industrial & Systems Engineering
and Biomedical Engineering University of Wisconsin-Madison
2,500+ emails later
Permissions: Clayton Lewis, Professor of Computer Science, University of Colorado NAS.
Fostering Independence, Participation, and Healthy Aging Through Technology: Workshop
Summary Gregg Vanderheiden, Director Trace R&D Center, Professor Industrial &
ystems Engineering and Biomedical Engineering University of Wisconsin-Madison
CLINICIANS - SLOW TO CHANGE /RESISTANCE
CULTURE OF MEDICINE
Five Years After To Err Is Human. L. Leape. What Have We Learned? JAMA. 2005;293:23842390
***
• Deeply rooted customs and training-”Doctor knows all”
• High standards of autonomous individual performance
• A commitment to progress through research• advances in biomedical science and cures to millions.
• Financial
• Loss of productivity
• Usability concerns
• Apprehension
• EHR-associated unintended consequences/New errors
[Good vs. Bad HIT]
• Workflow inefficiencies
•
Jennifer S Love, et al, Are physicians’ perceptions of healthcare quality and practice satisfaction
affected by errors associated with electronic health record use. JAMIA Dec.2011
7 July 2015
DESIGN GOALS FOR COMPLEX E-HEALTH SYSTEMS
•
•
•
•
•
•
•
COLLABORATION:
SCALABILITY:
FLEXIBILITY:
RAPID FROM DESIGN:
USE OF STANDARDS:
SUPPORT HIGH QUALITY RESEARCH:
WEB-BASED AND SUPPORT INTERMITTENT
CONNECTIVITY:
• LOW COST: preferably free/open source
• CLINICALLY USEFUL: feedback to providers and
caregivers is critical. If the system is NOT CLINICALLY
USEFUL it will not be used.
•
AMPATH Medical Record System (AMRS): Collaborating Toward An EMR for Developing Countries Burke W. Mamlin,
M.D. and Paul G. Biondich, M.D., M.S. Regenstrief Institute, Inc. and Indiana University School of Medicine, Indianapolis, IN
MEDICINE IN DENIAL. L. Weed
Deep disorder pervades medical practiceDisguised in euphemisms like "clinical judgment" and
"evidence-based medicine," disorder exists because
medical practice lacks a true system of care.
The missing system has two core elements:
1. Standards of care for managing clinical information
2. Electronic information tools designed to implement
those standards.
Electronic information tools are now widely discussed,
but the necessary standards of care are still widely
ignored.
THE “SYSTEM’ TO BE “HEALED”
"We have to disrupt ourselves or we
will be disrupted."
Craig Bradley, MD, Chief Medical Information Officer, Texas Tech School of Medicine, in a reference to healthcare
systems in his testimony before the HIT Standards Committee . July 2013
“The biggest information repository in most
organisations sits within the heads of those who
work there, and the largest communication
network is the web of conversations that binds
them. Together, people, tools, and conversations –
that is the “system”.”
ENRICO COIERA [UNSW/AIHI] (Int. J. Med. Inform 69(2-3):2003,205-222)
7 July 2015
63