Health Related Quality of Life and Health

Status
Robert M. Kaplan
University of California, Los Angeles
Dennis G. Fryback
University of Wisconsin (Retired)
Prepared for
Economic Analysis of Nutrition Interventions: Methods, Research and
Policy
National Institutes of Health
February 23, 2010

www.healthmeasurement.org
What Methods Can Be Used to Measure
Health?

www.healthmeasurement.org
 Summary Measures of Population Health

• Preference-based, generic measures of

health-related quality of life (HRQoL)

– Main focus of this talk:

• What these are
• Their role in a population health data system
• How they might be implemented in public health

www.healthmeasurement.org
 Measures of Health – a quick typology

• Mortality-based measures • All the familiar stuff

– death rates, life expectancies,
etc.

www.healthmeasurement.org
 Measures of Health – a quick typology

• Mortality-based measures • Indicators:

– death rates, life expectancies, etc. – Single, countable things
• Morbidity-based measures • TB rate
• C-section rates
– indicators
• % population who exercise

– Examples:
• Healthy People 2010 ―Leading
Indicators‖
• WHO ―Core Health Indicators‖
• America’s Health Rankings
• Wisconsin County Health
Rankings

www.healthmeasurement.org
 Measures of Health – a quick typology
Health Status Measures: point in time summaries
of state of a person’s health
• Mortality-based measures
– death rates, life expectancies, Disease-, organ-specific....
etc. Created to be sensitive to changes in symptoms or
functional impairment due to a particular
• Morbidity-based measures disease process
– indicators Examples:
– health status measures – Arthritis Impact Measurement System (AIMS)
– Vision Function Questionnaire (VFQ-25)
• disease-, organ-specific
– McGill Pain Questionnaire
– NY Heart Association Classification

Some physician-reported, others patient-reported

*Many of these are scored by summing across
questions in a questionnaire*

www.healthmeasurement.org
 Measures of Health – a quick typology

• Mortality-based measures Generic Health Status Measures

– death rates, life expectancies, Most famous: SF-36 health profile
etc. One questionnaire with 36 questions 
• Morbidity-based measures Several questions about each of 8 different
domains of health
– indicators
– health status measures KEY: multiple
multiple scales toscales to cover
cover broad scopebroad
of health,
not tied to one disease or organ system
scope of health, not tied to one disease
• disease-, organ-specific or organ system
• “Generic” health status Scoring:
Psychometric scales based on summing
responses to multiple questions
Separate scores for each subscale or
health concept
PF, RP, BP, GH, VT, SF, RE, MH

PCS MCS
www.healthmeasurement.org
 Measures of Health – a quick typology
HRQoL indexes
• Mortality-based measures
– death rates, life expectancies, Like generic health status – try to
etc. comprehensively cover conceptual
basis of health with multiple
• Morbidity-based measures questions about health
– indicators
– health status measures Scoring
• disease-, organ-specific – not simple sums--these are not
• “generic” psychometric scales!
– health-related quality-of-life – econometric methods used to elicit
utility weights (―preferences‖) for
(HRQoL) indexes
health states
– 0 = dead, 1= perfect health
– average preference weights from
community sample of people

www.healthmeasurement.org
 Data Pyramid for Population Health (after Wolfson)
more aggregated = summarization, HRQoL Indexes
evaluation
preference-weighted aggregate
scores summarizing overall health

Generic Health Status Profiles

Vector of health status domain scales

Disease-specific Scales
Do not necessarily cover all health
domains

more disaggregate = explanation, Multitude of health indicators

description

www.healthmeasurement.org
• Two areas of concern in population health —
Morbidity and Mortality

– “Morbidity”: how do people feel, how health problems affect them

• ability/disability
• functional capacity
• independence
• other aspects of health & well-being

– “Mortality”: how long people live

• mortality rates and survival rates
• life expectancy

• One summary measure, HRQoL, combines all the aspects of

morbidity
• A second summary measure, QALE, combines HRQOL and
mortality into a single number

www.healthmeasurement.org
How do we measure quality of life?

www.healthmeasurement.org
 Traditional

• LifeExpectancy
• Infant Mortality
• Disability Days

www.healthmeasurement.org
 Measuring Effectiveness

What is the
meaning of life?
www.healthmeasurement.org
www.healthmeasurement.org
www.healthmeasurement.org
www.healthmeasurement.org
www.healthmeasurement.org
www.healthmeasurement.org
If widely different interventions
are to be compared…..

• The measure of health must be able to

encompass not only differences in length
of life but differences in the quality of that
life, in symptoms and ability to function.

www.healthmeasurement.org
 Survival Analysis

•Alive 1.0
•Dead 0.0

www.healthmeasurement.org
 Problem with Survival Analysis

•Tennis player 1.0

•Man in coma 1.0

www.healthmeasurement.org
Purpose of Quality Adjusted
Survival Analysis

• To summarize
life expectancy
with
adjustments for
quality of life
www.healthmeasurement.org
 Example Case: 68 year old COPD patient
Description

• Shortness of breath
• Drove Car
• In Bed or Chair for Most of Day
• Performed No Major Role Activity, but did perform self-care
• Weight
• Peer Rating equals .605
• For each year in this state, the patient loses 1 - .605 = .395
well years

www.healthmeasurement.org
 What is a QALY?

• The QALY is a measure of health outcome which

assigns to each period of time a weight, ranging
from 0-1, corresponding to the quality of life during
that period, where 1 corresponds to perfect health
and a weight of 0 corresponds to a health state
judged equivalent to death (Peter Neumann, Tufts Medical Center)

www.healthmeasurement.org
 The concept of a QALY

1.0
0.9
0.7

0.3

time 0 2 3.5 4.5 7

QALYs = 2x1 + 1.5x0.7 + 1x0.3 + 2.5x0.9 = 5.6
from Peter Neumann, Tufts Medical Center

www.healthmeasurement.org
 Using QALYs to measure the impact of a
treatment

Quality Cost
of Life

Time Time

Without treatment:
With treatment: From Peter Neumann, Tufts Medical Center

www.healthmeasurement.org
 Quality-Adjusted Life Year
• Combines morbidity and mortality into a single
index
• Represents life expectancy with adjustments for
quality of life
• Is defined as a year of life free of all disabilities and
symptoms

www.healthmeasurement.org
 “life path” Continuum to
Excellent show time in
health states and
Overall overall length of
HRQOL
life

Dead
start death
time

Continuum of health state

weights

www.healthmeasurement.org
Cross-sectional samples of individuals HRQoL
at a point in time can be used for meaningful
population measures
• Community averages for HRQoL summarize health at a point
in time

• Cross-sectional HRQoL data can be combined with mortality

data
– old idea, attributed to Sullivan at Bureau of the Census in 1960s
– elaborated by others since

www.healthmeasurement.org
 Summary Measures of Population Health
(Molla et al, NCHS Statistical Note, 2001)

Mortality Component: Morbidity Component:

Age-specific death rates Population morbidity,
from Census disability, or HRQOL –
HP2000 used
HALex from NHIS

Life table technique

Summary measure of population health:

Healthy life expectancy or healthy life years

www.healthmeasurement.org
 NCHS Table of Health-adjusted Life Expectancy
(US females from 2000 census and NHIS)
Age bracket LE HALE HALE/LE
0–4 years ................. 79.5 69.8 87.8%
5–9 years ................. 75.0 65.4 87.2%
10–14 years ............... 70.1 60.5 86.3%
15–19 years ............... 65.1 55.7 85.6%
20–24 years ............... 60.3 51.0 84.6%
20.5 HALYs 25–29 years ............... 55.4 46.4 83.8%
30–34 years ............... 50.5 41.8 82.8%
-13.8 HALYs 35–39 years ............... 45.7 37.2 81.4%
6.7 HALYs 40–44 years ............... 40.9 32.8 80.2%
10-year HALE 45–49 years ............... 36.2 28.5 78.7%
for 55-59 yo 50–54 years ............... 31.6 24.3 76.9%
female 55–59 years ............... 27.1 20.5 75.6%
60–64 years ............... 22.9 17.0 74.2%
65–69 years ............... 18.9 13.8 73.0%
70–74 years ............... 15.2 10.8 71.1%
75–79 years ............... 11.8 8.1 68.6%
80–84 years ............... 8.7 5.8 66.7%
www.healthmeasurement.org
 What has held us back?

• Distractions
– Disagreements on which measure is best
– Disagreements on general philosophy of outcome
measurement
• Generic vs disease specific
• Psychometric vs. utility based
• Disciplinary differences – statistics, economics, medicine,
psychology, anthropology….

www.healthmeasurement.org
 We do agree on some of the core issues

• Most measures can be traced back to Sullivan

(1966)
– Sullivan rarely cited
• Content of items is remarkably similar
• Most measures combine measures of life length
and life quality
• Most quality of life measures are hybrid health
status/utility measures
– Health states and health weights (Erickson)

www.healthmeasurement.org
 John Ware

• Think of different approaches

as brand names of products
designed to measure the
same underlying construct…
health

www.healthmeasurement.org
 Preference and Utility Assessment

• Standard Gamble
• Time Trade-off
• Rating Scales
• Think scoring systems
www.healthmeasurement.org
Small set of potential HRQOL indexes available
today
• EQ-5D
• HUI2, HUI3
• SF-6D
• QWB-SA
• HALex
Each has an associated questionnaire varying from
5 to nearly 60 questions; varying times to complete
from 2 min to 15 min on average.

www.healthmeasurement.org
 Health Domains addressed by these HRQoL indexes
QWB-SA SF-6D (from SF-36 questionnaire) EuroQol EQ-5D
– Mobility – Physical function – Mobility
– Physical activity – Role limitation – Self-care
– Social activity – Social function – Usual activities
– Symptoms – Pain – Pain/discomfort
– Mental health – Anxiety/depression
– Vitality

HALex HUI2 HUI3

– Self-rated health – Sensation – Vision
– Physical activity – Mobility – Hearing
limitations – Emotion – Speech
– Cognition – Ambulation
– Self-care – Dexterity
– Pain – Emotion
– Cognition
– Self-care
– Pain
www.healthmeasurement.org
 Scale ranges for these HRQoL indexes
QWB-SA SF-6D (from SF-36 questionnaire) EuroQol EQ-5D
– Mobility – Physical function – Mobility
– Physical activity – Role limitation – Self-care
– Social activity – Social function – Usual activities
– Symptoms – Pain – Pain/discomfort
– Mental health – Anxiety/depression
0 ... [0.09 ... 1.0] – Vitality
[-.11 ... 0 ... 1.0]
0 ... [0.31 ... 1.0]
HALex HUI2 HUI3
– Self-rated health – Sensation – Vision
– Physical activity – Mobility – Hearing
limitations – Emotion – Speech
– Cognition – Ambulation
0 ... [0.10 ... 1.0]
– Self-care – Dexterity
– Pain – Emotion
[-.02 ... 0 ... 1.0] – Cognition
– Self-care
– Pain
www.healthmeasurement.org [-.36 ... 0 ... 1.0]
 Preference weights source for the HRQoL indexes

QWB-SA SF-6D (from SF-36 EuroQol EQ-5D

questionnaire)

San Diego, CA England national US national

sample sample

HALex HUI2 HUI3

(ad hoc US for Ontario, Canada Ontario, Canada

Healthy People
2000)

www.healthmeasurement.org
www.healthmeasurement.org
 Data from the NHMS –population sample (n=3844) in US
EQ-5D

0.90
Women 0.90
Men HUI2
HUI3
SF-6D
QWB-SA
0.85 0.85

0.80 0.80
Group Mean Index Score

0.75 0.75

0.70 0.70

0.65 0.65

0.60 0.60

35-44 45-54 55-64 65-74 75-89 35-44 45-54 55-64 65-74 75-89

Age Group Age Group

www.healthmeasurement.org
 US data sets with HRQoL index
• One-time data sets
– National Health Measurement Study
• EQ-5D, HUI2, HUI3, SF-6D, QWB-SA, HALex
– US Valuation of the EQ-5D (USVEQ)
• EQ-5D, HUI2, HUI3
– Joint Canada US Survey of Health (JCUSH)
• HUI3
• Continuing data sets
– Medical Expenditure Panel Study (MEPS)
• EQ-5D (2000-2003)
• SF-12 (a reduced form of SF-36 that can still be used to compute the SF-6D
HRQoL index) (2000-present)
– National Health Interview Survey (NHIS)
• data subsuming HALex

www.healthmeasurement.org
 NHMS: Relation between summary health and 6 HRQoL indexes
EQ-5D HUI2 HUI3

1.0 1.0 1.0

0.8 0.8 0.8

0.6 0.6 0.6

Index Score

0.4 0.4 0.4

0.2 0.2 0.2

0.0 0.0 0.0

-0.2 -0.2 -0.2

-2 0 2 -2 0 2 -2 0 2

X Axis Title X Axis Title

QWB-SA SF-6D HALex

1.0 1.0 1.0

0.8 0.8 0.8

0.6 0.6 0.6

Index Score

0.4 0.4 0.4

0.2 0.2 0.2

0.0 0.0 0.0

-0.2 -0.2 -0.2

-2 0 2 -2 0 2 -2 0 2

www.healthmeasurement.org
 Summary

• Utility-based measures are available to estimate the

impact of nutrition at the population level and
nutritional interventions at the individual level
• Generic methods allow the comparison of
investments in nutrition with investments in other
aspects of health care
• There are very few applications at present
• We look forward to the development of these
methods for studies in comparative effectiveness.

www.healthmeasurement.org