This document summarizes a presentation on robust methods for assessing health-related quality of life (HRQoL). It discusses using quality-adjusted life years (QALYs) to collapse multi-dimensional HRQoL data but notes this can result in biased estimates. A two-step methodology is proposed where HRQoL domains are first estimated separately, then coefficients are transformed to the QALY scale based on predicted values. Simulations show the two-step approach provides less biased estimates than existing single-stage methods. The methodology is applied to SF-6D HRQoL data.
Dehradun Call Girls Service {8854095900} ❤️VVIP ROCKY Call Girl in Dehradun U...
Robust Methods for Health-related Quality-of-life Assessment
1. Introduction
Methodology
Simulations
Application
Conclusion
Robust Methods for Health-related
Quality-of-life Assessment
Ian McCarthy
Baylor Scott & White Health
Center for Clinical Effectiveness
Utah Health Services Research Conference
April 30, 2014
This project was supported by grant number K99HS022431 from the Agency for Healthcare Research and
Quality. The content is solely the responsibility of the author and does not necessarily represent the official
views of the Agency for Healthcare Research and Quality.
Robust Methods for Health-related Quality-of-life Assessment
2. Introduction
Methodology
Simulations
Application
Conclusion
Background
Cost- and comparative-effectiveness studies becoming
increasingly important
Require assessment of health-related quality-of-life (HRQoL)
outcomes and quality-adjusted life-years (QALYs)
Common approach first collapses the multi-dimensional
HRQoL profile into a one-dimensional QALY (Drummond
et al., 2005; Brazier et al., 2002; Brazier & Ratcliffe, 2007)
EQ-5D
SF-6D
HUI
Robust Methods for Health-related Quality-of-life Assessment
3. Introduction
Methodology
Simulations
Application
Conclusion
Problem
Loss of information when reducing HRQoL profile into QALY, with
potentially biased and inconsistent marginal effects estimates
(Mortimer & Segal, 2008; Devlin et al., 2010; Parkin et al., 2010;
Gutacker et al., 2012):
1 Floor and ceiling effects not present in the underlying domains
but imposed by the scoring algorithm.
2 Nonlinearities in the relationship between the outcome and
independent variables which are difficult to approximate using
the summary score.
Robust Methods for Health-related Quality-of-life Assessment
4. Introduction
Methodology
Simulations
Application
Conclusion
Current Study
1 Monte Carlo study showing the bias of the estimated
coefficients when relying solely on QALYs or some other
summary score based on several ordered outcome variables.
2 Propose new two-step methodology that first estimates
coefficients in each HRQoL domain and then transforms the
coefficients and marginal effects into the QALY domain based
on predicted values from the first-stage regressions.
Robust Methods for Health-related Quality-of-life Assessment
5. Introduction
Methodology
Simulations
Application
Conclusion
Estimating QALYs
Marginal Effects: Standard Approach
Marginal Effects: Proposed Methodology
The SF-6D
Developed by John Brazier and other, the SF-6D is formed from a
subset of questions from the SF-36 or SF-12 and is a common
HRQoL outcome intended to provide a general measure of a
patient’s health status (Brazier et al., 2002; Brazier & Ratcliffe,
2007).
Six dimensions/domains of health: (Physical functioning, role
limitations, social functioning, pain, mental health, and
vitality)
Each domain characterized numerically with a range of
integers. Best value is 1, and worst value ranges from 4 to 6.
Scoring algorithm developed in Brazier et al. (2002) and
Brazier & Ratcliffe (2007) for calculating a population-based
index score from the SF-6D questionnaire
Robust Methods for Health-related Quality-of-life Assessment
6. Introduction
Methodology
Simulations
Application
Conclusion
Estimating QALYs
Marginal Effects: Standard Approach
Marginal Effects: Proposed Methodology
Scoring the SF-6D
Physical Functioning (PF)
PF=2 or PF=3 -0.035
PF=4 -0.044
PF=5 -0.056
PF=6 -0.117
Role Limitations (RL)
RL=2 or RL=3 or RL=4 -0.053
Social Functioning (SF)
SF=2 -0.057
SF=3 -0.059
SF=4 -0.072
SF=5 -0.087
Pain (P)
P=2 or P=3 -0.042
P=4 -0.065
P=5 -0.102
P=6 -0.171
Mental Health (MH)
MH=2 or MH=3 -0.042
MH=4 -0.100
MH=5 -0.118
Vitality (V)
V=2 or V=3 or V=4 -0.071
V=5 -0.092
Combination of Domains
“Most Severe” -0.061
Robust Methods for Health-related Quality-of-life Assessment
7. Introduction
Methodology
Simulations
Application
Conclusion
Estimating QALYs
Marginal Effects: Standard Approach
Marginal Effects: Proposed Methodology
Focus on QALYs
By far the most common methodology for estimating
coefficients and ultimately marginal effects is to first reduce
the multi-dimensional health profile to a one-dimensional
QALY (Austin et al., 2000; Austin, 2002; Richardson &
Manca, 2004; Manca et al., 2005; Basu & Manca, 2012)
Recent literature on how best to accommodate distributional
features somewhat specific to QALYs (Austin, 2002; Basu &
Manca, 2012), including a censored least absolute deviation
model and a Beta MLE approach
Robust Methods for Health-related Quality-of-life Assessment
8. Introduction
Methodology
Simulations
Application
Conclusion
Estimating QALYs
Marginal Effects: Standard Approach
Marginal Effects: Proposed Methodology
First Stage Regression
1 Estimate an ordered probit model separately for each domain,
d = 1, ..., 6, with the follow-up HRQoL response (yid,t1 )
modeled as a function of person-specific variables (xi ),
baseline HRQoL response (yid,t0 ), and treatment status (Ti ).
2 Form predicted probabilities of every possible response, j, in
each domain, d, denoted ˆpd
j .
The regression results provide a predicted (marginal) probability for
each of 31 possible outcomes for each person.
Robust Methods for Health-related Quality-of-life Assessment
9. Introduction
Methodology
Simulations
Application
Conclusion
Estimating QALYs
Marginal Effects: Standard Approach
Marginal Effects: Proposed Methodology
“Most Severe” Category
1 Defined as any one of the following (Brazier et al., 2002): 4
or more in the physical functioning, social functioning, mental
health, or vitality domains; 3 or more in the role limitation
domain; or 5 or more in the pain domain
2 Since the probabilities, Pd
ij , are potentially correlated across
domains, the probability of a “most severe” health status can
be calculated following the principle of inclusion and exclusion
for probability:
P (A1 ∪ A2 ∪ ... ∪ AN) = P (A1) + ... + P (AN) +
N
n=2
(−1)n+1
P (∩ n events) .
Robust Methods for Health-related Quality-of-life Assessment
11. Introduction
Methodology
Simulations
Application
Conclusion
Marginal Effects on QALYs
Treatment Effects with Selection
Data Generating Processes
The D × 1 vector of latent HRQoL values, y∗
i , is simulated as
follows:
y∗
i = γ + βxi + εi , where
ε ∼ N (0D×1, ID×D) ,
x ∼ U[0, 1],
γ = ID×1, and
β = 1.5 × ID×1.
Discrete HRQoL values are generated based on the value of the
latent value, y∗
id , relative to the Jd × 1 vector of threshold values in
each domain.
Robust Methods for Health-related Quality-of-life Assessment
12. Introduction
Methodology
Simulations
Application
Conclusion
Marginal Effects on QALYs
Treatment Effects with Selection
Simulated QALY Distributions
01020304050
Frequency
.4 .6 .8 1
SF-6D Index Score
010203040
Frequency
.2 .4 .6 .8 1
SF-6D Index Score
01020304050
Frequency
.4 .6 .8 1
SF-6D Index Score
020406080
Frequency
.3 .4 .5 .6 .7 .8
SF-6D Index Score
050100150200
Frequency
.4 .6 .8 1
SF-6D Index Score
Robust Methods for Health-related Quality-of-life Assessment
18. Introduction
Methodology
Simulations
Application
Conclusion
Data Summary
Results
Data
Data collected prospectively on adult scoliosis patients from over
10 participating members of the International Spine Study Group
(ISSG).
Variable Mean Standard
Deviation
Age 56.76 14.51
BMI 26.59 5.84
Baseline SF-6D 0.61 0.12
Follow-up SF-6D 0.66 0.12
Count Percent
Operative 193 53%
Female 309 85%
Robust Methods for Health-related Quality-of-life Assessment
22. Introduction
Methodology
Simulations
Application
Conclusion
Intuition
Collapsing multi-dimensional profile into a single summary
measure introduces floor/ceiling effects and nonlinearities that
are difficult to accommodate in a single equation framework.
With selection into treatment (whether on observables or
unobservables), standard methods relying only on QALYs
provide biased estimates of true treatment effect.
An alternative approach is to estimate coefficients based on
the full health profile and then re-interpret effects in the
QALY domain based on predicted probabilities in the
first-stage regressions.
Robust Methods for Health-related Quality-of-life Assessment
24. Introduction
Methodology
Simulations
Application
Conclusion
Bibliography I
Austin, P.C. 2002. A comparison of methods for analyzing health-related quality-of-life measures. Value in Health,
5(4), 329–337.
Austin, P.C., Escobar, M., & Kopec, J.A. 2000. The use of the Tobit model for analyzing measures of health
status. Quality of Life Research, 9(8), 901–910.
Basu, A., & Manca, A. 2012. Regression Estimators for Generic Health-Related Quality of Life and
Quality-Adjusted Life Years. Medical Decision Making, 32(1), 56–69.
Brazier, J., & Ratcliffe, J. 2007. Measuring and valuing health benefits for economic evaluation. Oxford University
Press, USA.
Brazier, J., Roberts, J., & Deverill, M. 2002. The estimation of a preference-based measure of health from the
SF-36. Journal of health economics, 21(2), 271–292.
Devlin, N.J., Parkin, D., & Browne, J. 2010. Patient-reported outcome measures in the NHS: new methods for
analysing and reporting EQ-5D data. Health economics, 19(8), 886–905.
Drummond, M.F., Sculpher, M.J., & Torrance, G.W. 2005. Methods for the economic evaluation of health care
programmes. Oxford University Press, USA.
Gutacker, N., Bojke, C., Daidone, S., Devlin, N., & Street, A. 2012. Analysing Hospital Variation in Health
Outcome at the Level of EQ-5D Dimensions.
Manca, A., Hawkins, N., & Sculpher, M.J. 2005. Estimating mean QALYs in trial-based cost-effectiveness analysis:
the importance of controlling for baseline utility. Health economics, 14(5), 487–496.
Mortimer, D., & Segal, L. 2008. Comparing the incomparable? A systematic review of competing techniques for
converting descriptive measures of health status into QALY-weights. Medical decision making, 28(1), 66.
Parkin, D., Rice, N., & Devlin, N. 2010. Statistical analysis of EQ-5D profiles: does the use of value sets bias
inference? Medical Decision Making, 30(5), 556–565.
Richardson, G., & Manca, A. 2004. Calculation of quality adjusted life years in the published literature: a review of
methodology and transparency. Health economics, 13(12), 1203–1210.
Robust Methods for Health-related Quality-of-life Assessment