Title: Transparency in the Use of Propensity Score Methods
1Transparency in the Use of Propensity Score
Methods
- John D. Seeger, PharmD, DrPH
- Chief Scientist, i3 Drug Safety
- Adjunct Assistant Professor, Harvard School of
Public Health - September 9, 2008
With thanks to Alec Walker, Tobias Kurth, Jeanne
Loughlin, Mona Eng, and Alex Cole
2Propensity Score Analysis When?
Thanks to S. Schneeweiss
3Motivation
- Assume matching when comparing 2 treatments
- For every drug user with given characteristics
- Find a comparator with identical characteristics
- Example Male, age 45, smoker, with HTN
- Matching fails
- Age (10 categories) x
- Sex (2 categories) x
- Prior diagnoses (5 _at_ 2 categories each)
- Prior drug therapy (5 _at_ 2 categories each)
- Preceding cost of care (5 categories)
- ?102,400 potential matching groups
4Propensity Score Collapses Exposure Predictors
- Single value
- Probability subject will receive therapy vs
comparator - Removes confounding by components of the score
- Patient characteristics that favor one therapy
over another - Permits
- Restriction
- Matching
- Stratification
- Modeling
- Weighting
5Should Propensity Scores Always be Used?
- More than 8 events per covariate leads to
unbiased estimates - So propensity score favored when
- Many more persons exposed to drug of interest
than study outcomes - Common exposure
- Rare outcome
- Allows for richer model (more predictors) of
exposure than outcome - Alternative hypotheses
Cepeda S, et al. Am J Epidemiol 2003158280-287.
6Estimate Propensity Score
- Predict treatment from baseline covariates within
database - Inclusion of predictors
- a-priori (what characteristics are used to
prescribe?) - Empiric (what differentiates initiators?)
- Generic (what patterns of healthcare predict
initiation?) - Coefficients of propensity score
- Interpretable and Informative
7Propensity Score Restriction
Sturmer T, et al. J Clin Epidemiol 200659437-47.
8Propensity Score Restriction
- Potential for serious adverse events from error
(name confusion) - Amaryl (glimepiride an oral hypoglycemic)
- Reminyl (galantamine for Alzheimers disease)
- 36,816 people with AD diagnosis (14,626 Reminyl
dispensings) - 236 Amaryl recipients
- 24 Amaryl recipients in the lowest decile of the
propensity score - 13 with a single dispensing of Amaryl or no
diabetes diagnoses - 2 with no diabetes-related claims across entire
claim history - Medical record review suggested no error
- Propensity score restriction may be used as a
screening method to identify unusual patterns of
healthcare for closer scrutiny - Possible medication dispensing errors
- Others
- Confirmation requires additional data, which
could be obtained through medical record review.
9Propensity Score Distribution and Strata
C-statistic 0.739
10Effect of Temazepam Relative to Zopiclone
Transparent analysis Within-stratum
balance Stratum-specific effect estimates as
well as pooled estimate Explicit evaluation of
potential for effect measure modification
11Matching on the Propensity Score
- Matching can be performed by
- Standard automated case-control matching programs
where the matching range is specified - Nearest available match based on the propensity
score - Greedy matching techniques(http//www2.sas.com.pr
oceedings/sugi26/p214-26.pdf)
12(No Transcript)
13(No Transcript)
14Characteristics Before Matching
15Balance Achieved by Matching
16Analysis by 2X2 Table
17MI Outcome (After Matching)
31 (7-48) Risk Reduction
HR0.69 (0.52-0.93)
Cumulative Incidence
Statin Non-Initiators
Statin Initiators
Months of Follow-Up
18Regression Adjustment with Propensity Scores
- Regression adjustment
- All study participants are used
- Still a two-step approach (exposure and outcome)
- More power compared to including all covariates
into the model, since degrees of freedom are
gained - However, assumes the underlying association
between the score and the outcome is modeled
appropriately
19Weighting
1 in treated (ê(X)/(1-ê(X)) in untreated
1/ê(X), in treated 1/(1- ê(X)), in untreated
IPTW
SMR
20Baseline Characteristics
21Cohort Results
22Are Divergent Results Possible?
Kurth T, et al. Am J Epidemiol 2006163262-70.
23What About Unmeasured Confounding?
- Obesity, Smoking, Exercise
24Accounting for Variables had Little Effect
25Conclusion
- Propensity score can be useful for addressing
confounding (by indication) - Allows for rich model of exposure to be developed
- Advantageous when number of people with a study
outcome is small relative to number of exposed
persons and number of potential confounders is
large - Drug effects (particularly adverse ones)
- Consider transparency
- When selecting propensity score
- When building propensity score
- When using propensity score
26John.Seeger_at_i3DrugSafety.com