Measuring Health System Performance Lecture 3 Health Service Outcomes Health Service Outcomes Reinhard Busse, Prof. Dr. med. MPH Department of Health Care Management Berlin University of Technology/ (WHO Collaborating Centre for Health Systems Research and Management) European Observatory on Health Systems and Policies
36
Embed
Measuring Health System Performance Lecture 3 › fileadmin › a38331600 › 2015.lectures … · Measuring Health System Performance Lecture 3 Health Service Outcomes Reinhard Busse,
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Measuring Health System Performance
Lecture 3
Health Service OutcomesHealth Service Outcomes
Reinhard Busse, Prof. Dr. med. MPHDepartment of Health Care Management
Berlin University of Technology/
(WHO Collaborating Centre for Health Systems Research and Management)
European Observatory on Health Systems and Policies
Outline
• Concept and background
– Population health vs. Health service indicators
– Health service outcomes = quality?
• Dimensions of quality
• Outcome vs process
• Different type of Health Service Outcome indicators• Different type of Health Service Outcome indicators
2001-03 (standardised for age, sex, urgency/readmission, LOS within 50 CCS
groups leading to 80% all deaths, excluding small hospitals and those with poor
data recording, using year 2000 standard)
100
120
140
HS
MR
s (9
5%
CIs
) 200
1-2
003
0
20
40
60
80
96
35
68
14
83
81
51
25
89
50
103 3
52
44
85 5
78
36
12
100
72
94
13
104
65
33
34
95
101
39
93
82
79
23
61
47
37
20
87
97
45
31
107
19
98
54
102
Hospital number (assigned by BJ)
HS
MR
s (9
5%
CIs
) 200
1-2
003
Mortality for particular conditions?
• Select conditions where quality of care is associated with
mortality (AMI, Stroke, Hip)
• Challenges:
– sample size
(ex. Dimick et al. 2004 – studies 7 operations that the Agency for Healthcare (ex. Dimick et al. 2004 – studies 7 operations that the Agency for Healthcare
Research and Quality in the US recommended surgical mortality as a quality
indicator – found that only for 1 (CABG surgery) was sample size large
enough to make quality assessments)
– generalizability of results
(Challenges include: International comparisons of mortality, timing,
measurement)
Source: Dimick et al. (2004) Surgical Mortality as an Indicator of Hospital Quality: The Problem with Small Sample Size. JAMA 292(7):847-851.
Patient Safety Indicators
• Common: infection rates (MRSA, C-difficile)
• Death among surgical inpatients with serious treatable
complications
• Adverse post-op outcomes (i.e. pressure ulcer, hip fracture,
hemorrhage etc)hemorrhage etc)
• Composite - Complication/patient safety for selected
indicators
• Never events, non events (wrong site surgery, wrong
• Various studies suggest Readmissions may not be always indicative of poor quality
• McClellan & Staiger (1999), Papanicolas & Mcguire (2011) – found some conditions had negative correlations between mortality and readmissions – USA and UK samples.
• Laudicella et al. (2013) hospitals’ performance in readmissions is • Laudicella et al. (2013) hospitals’ performance in readmissions is determined in part by their difference in the quality of care and in part by their difference in the share of unobservably sicker patients. (UK sample)
• Fischer et al. (2011): Systematic review of readmission indicators identified only 21 out of 486 studies addressed validity of indicator when using it as an outcome measure.
– Little consensus over time-frame, type of readmission and case-mix adjustment applied.
Ambulatory Care Sensitive Conditions:good ambulatory care should prevent or reduce hospitalisations
What are ACSCs?
Chronic conditions that include congestive heart failure, diabetes, asthma, angina, epilepsy and hypertension.
Actively managing patients Actively managing patients with ACS conditions – through vaccination; better self-management, disease-management or case-management; or lifestyle interventions – prevents acute exacerbations and reduces the need for emergency hospital admission.
Bardsley et al (2012) Is secondary preventive care improving? Observational study of 10-year
trends in emergency admissions for conditions amenable to ambulatory care
http://bmjopen.bmj.com/content/3/1/e002007?cpetoc
Patient Reported Outcome Measures
(PROMs)
• Instruments which gain information about health, illness and
the effects of health care interventions from the perspective
of the patient (Fitzpatrick et al, 1997)
• Four types of PROM:
– Generic– Generic
– Utility
– Disease specific
– Individual
Generic proms are applicable to the
widest possible range of health problems
Utility instruments assign utilities to
respondent’s health states – the EQ5D
Disease-specific proms are tailored to the
specific disease for which they are intended
Individual level proms are tailored to the
specific disease for which they are intended
PROM results in England 2013/14
Participation and
Coverage: There have
been 128,759 PROMs-
eligible procedures
carried out in hospitals
and 98,695 pre-operative
questionnaires returned
so far, a participation rate so far, a participation rate
of 76.7%.
For the 98,695 pre-
operative questionnaires
returned, 44,460 post-
operative questionnaires
were sent out, of which
11,423 have been
returned so far - a return
rate of 25.7% (73.3%).
Yet, how effective are PROMs?
• Fitzpatrick (2009) assess PROMs with respect to 7 criteria:– Reliability
– Validity
– Responsiveness
– Precision
– Interpretability
– Acceptability– Acceptability
– Feasibility
• PROMs are increasingly being used as performance instruments
• Hold much promise – but realising their potential requires: – Credible data collection, instruments and analysis
– Good reporting so that information is useful for different users and for different types of decisions
– Recognitions of limitations
– An open mind to this type of information
Outcome vs Process Measures
The case for measuring outcomes of care
• Central indicator of the success of health care
• Essential for determining “what works” in health care
• Can nurture innovation
The case for measuring processes of care
• Certain aspects of process (such as waiting
times or patient experience) are often valued
by patients
• Certain processes are known to be
associated with desired health outcomes.
• Measuring outcomes can be difficult, costly
and takes a long time -
• Can nurture innovation
• Are universal and do not become easily obsolete
• Clinical attention is focused on securing improved health rather than ‘checklists of activities’
• Harder to manipulate than process measures
Measuring outcomes can be difficult, costly
and takes a long time -
• Process measures are almost instantaneous
and can be acted on quickly.
• Process measures are usually readily
attributable to the provider of care and so
more easily interpretable (as opposed to
outcome measures which display a lot of
random noise).
• It is easier to devise incentive schemes
associated with process rather than
outcomes.
Process Measures
• Process Measures (waiting times) vs Clinical Process Measures (Measuring blood pressure for Hypertensive patients)
• Advantages: fast to collect, easier to attribute • Advantages: fast to collect, easier to attribute directly to health services, reflect compliance with good practice.
• Disadvantages: may be less relevant when considered alone, not always applicable, may become dated.
Developing Clinical Process Measures
• Selecting topics
• Reviewing clinical evidence
• Identifying quality indicators
• Constructing process measures
• Creating scoring methods
Process indicators (examples)
Measure Set Measures
Acute Myocardial Infarction (AMI)
■ Aspirin on Arrival
■ Aspirin Prescribed at Discharge
■ ACEI or ARB for LVSD
■ Beta Blocker Prescribed at Discharge
■ Beta Blocker on Arrival
■ Thrombolytic agent received within 30 minutes of hospital arrival
■ Percutaneous Coronary Intervention within 120 minutes of hospital arrival
■ Adult Smoking Cessation Advice/Counseling
■ LVF Assessment ■ Discharge Instructions
Heart Failure (HF)
■ LVF Assessment
■ ACEI for LVSD
■ Discharge Instructions
■ Adult Smoking Cessation Advice/Counseling
Pneumonia (PN)
■ Initial antibiotic received within 4 hours of hospital arrival
■ Oxygenation Assessment
■ Pneumoccoccal Screening and/or Vaccination
■ Blood Cultures
■ Adult Smoking Cessation Advice/Counseling
■ Appropriate initial antibiotic selection
Surgical Infection Prevention (SIP)
■ Prophylactic antibiotic received within 1 hour prior surgical incision
■ Prophylactic antibiotic discontinued within 24 hours after surgical infection
Patient Safety Indicators (PSI)
■ Postoperative Septicemia
■ Postoperative PE/DVT
■ Infection due to medical care
■ OB trauma without instruments
Interpretation issues: what measures of
quality can and cannot tell you
• MOST indicators require further investigation or
validation before one can be confident that it
indicates ‘good’ or ‘bad’ quality.
• Often our assessment depends on where an Often our assessment depends on where an
organization/physician/unit is placed in relation to
others rather than to an absolute standard, but this
can also be influenced by OTHER factors:
– Data issues, differences in clinical practice, external