![Page 1: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/1.jpg)
Inst i tute for Cl in ical Evaluat ive SciencesInst i tute for Cl in ical Evaluat ive Sciences
Propensity Score Methods to Adjust for Bias in Observational Data
SAS HEALTH USERS GROUP
APRIL 6, 2018
![Page 2: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/2.jpg)
Overview
1.What is observational data?
2.What is the propensity score?
3.Statistical adjustment using the propensity score
a)Matching on the propensity score
b)Inverse probability of treatment weighting
![Page 3: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/3.jpg)
Follow-up
☺☺
☺ ☺ ☺☺ ☺ ☺☺
☺
☺☺☺ ☺☺☺ ☺☺☺☺ ☺☺☺☺ ☺☺
Treatment
Group
☺☺☺ ☺☺☺ ☺☺☺☺ ☺☺☺☺ ☺☺
Control
Group
☺☺☺ ☺☺☺ ☺☺☺☺ ☺☺☺☺
☺☺☺ ☺☺☺ ☺☺☺☺ ☺☺☺☺
☺☺☺ ☺☺☺ ☺☺☺☺ ☺☺☺☺
Inclusion/exclusion
criteria
Population Study
population
Baseline Outcome
Randomized Controlled Trials
(the “gold standard”)
![Page 4: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/4.jpg)
Characteristics of RCTs
• Randomization ensures subjects in both
treatment groups are equally matched on all
factors
• Allow causal inference
![Page 5: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/5.jpg)
But
• High cost
• Often short duration and/or underpowered.
• Problems with generalizability:
• Treatment is “ideal” (high compliance, careful follow-up means that any problems may be caught early).
• Many people who are given the treatments in “real life” are excluded from the trials
• Some situations cannot be randomized.
![Page 6: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/6.jpg)
What is Observational Data?
The choice of treatment is not under the control of the researcher - the researcher can only ‘observe’ what treatment was given.
Examples:• Data obtained using chart review• Electronic medical records• Survey data or health study data• Administrative data.
![Page 7: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/7.jpg)
The Study
Two medications used to treat chronic obstructive lung disease (COPD)
• Long-acting anticholinergic (LAAC)• Long-acting beta-agonist (LABA)
Compare overall mortality and risk of hospital admission related to COPD
Gershon et al. Annals of Internal Medicine, 2011
![Page 8: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/8.jpg)
Inst i tute for Cl in ical Evaluat ive Sciences
Ontario Drug
Benefit Plan
(which drug
is the person
taking)
Hospital Discharge
Database
(diagnoses)
Physician Billing
Database
(diagnoses)
Registered
Persons
Database
(age, sex, SES)
Hospital Discharge
Database
(for outcomes).
![Page 9: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/9.jpg)
Analysis of Our StudyExposure variable is choice of drug (LAAC vs.
LABA)
Outcome is time to hospitalization or death
survival analysis will be used
Bias is a concern
![Page 10: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/10.jpg)
Bias
Bias in confounders we can measure
And bias in confounders we can’t measure, e.g., smoking,
fitness
Covariate LAAC LABA
Specialist care in last year (%)
44.7 50.5
Prior lung functiontesting (%)
69.7 74.3
![Page 11: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/11.jpg)
Statistical Adjustment for Observational
Data
• Propensity score methods
• Instrumental variable analysis
• And others
![Page 12: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/12.jpg)
Propensity Score
Rosenbaum and Rubin (1983) realized the bias
from covariates can be eliminated by controlling for
a scalar-valued function (a “balancing score”)
calculated from the baseline covariates, i.e., the
propensity score
The propensity score is a way of summarizing the
information in all the prognostic variables
![Page 13: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/13.jpg)
What is the Propensity Score?
PS = probability that a person received one treatment
(rather than the other), given that person’s observed
covariates
Calculated using logistic regression to estimate the
propensity for a person to be prescribed a LAAC
(rather than a LABA)
proc logistic descending;
model LAAC = age sex diabetes hypertension
rural_res incquint ...;
output out = score predicted = ps;
run;
proc PSMATCH
![Page 14: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/14.jpg)
Calculating the Propensity Score
proc logistic descending;
model LAAC = age sex diabetes
hypertension rural_res incquint ....;
Patients predicted, based on their characteristics, to
be likely to be prescribed a LAAC will have a high
propensity score
Patients predicted to be unlikely to be prescribed a
LAAC (likely to be prescribed a LABA instead) will
have a low propensity score
![Page 15: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/15.jpg)
Variable Selection
1. All measured baseline covariates
2. Baseline covariates associated with treatment
choice
3. Baseline covariates associated with the
outcome
4. Baseline covariates associated with both
treatment assignment and outcome
![Page 16: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/16.jpg)
Propensity Score Methods
1. Covariate adjustment using the Propensity Score
2. Stratification on the PS
3. Matching on the PS.
4. Inverse probability weighting
![Page 17: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/17.jpg)
Inst i tute for Cl in ical Evaluat ive Sciences
Matching
![Page 18: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/18.jpg)
Matching
1. Create a matched sample based on logit(PS)
2. Assess balance between treated and untreated
subjects in the matched sample.
– The test of a good propensity score model is
how well it balances the measured variables
between treated and untreated subjects.
3. For unbalanced variables, add interactions or
higher order terms to the propensity score
logistic regression, recalculate the propensity
score and repeat the process.
![Page 19: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/19.jpg)
Inst i tute for Cl in ical Evaluat ive Sciences
Before Matching After Matching
BaselineCovariate
LAACN=28,563
LABAN=17,840
Standarddifference
LAACN=15,532
LABAN=15,532
Standard difference
Lung function testing (%)
69.7 74.3 10.2 72.4 73.0 1.3
Specialist care previous year (%)
44.7 50.5 11.6 49.0 49.1 0.2
Also using inhaledcorticosteroid
48.3 52.1 7.7 51.1 51.3 0.3
Co-diagnosis of CHF
40.2 38.2 4.1 39.0 39.2 0.4
Hospitalizedfor COPD in previous 6 months
8.0 7.3 2.5 7.8 7.8 0.1
![Page 20: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/20.jpg)
Analysis of Matched Data
Analysis of Matched Data Must Incorporate the Matching
Means Paired t-test
Proportions McNemar’s test
Survival models Stratify on matched pairs
Logistic regression GEE estimation to account for matched pairs
![Page 21: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/21.jpg)
Matched Analyses ...
• Compares patients who are all potential candidates for both treatments.
• Matching pairs patients who are similar with respect to their propensity score matches on many confounders simultaneously
• Unmatched individuals are discarded
• The resulting matched sample may not be representative of all patients receiving treatment
![Page 22: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/22.jpg)
Interpretation of a Matched Analysis
Estimates the Average Treatment Effect for the
Treated (ATT) – the average treatment effect for
those who ultimately received the treatment
![Page 23: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/23.jpg)
Inst i tute for Cl in ical Evaluat ive Sciences
Inverse Probability of Treatment
Weighting Using the Propensity Score
![Page 24: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/24.jpg)
The Weights
𝑊 =𝑍
𝑃𝑆+
1 − 𝑍
1 − 𝑃𝑆
where Z = 1 for the treatment group and 0 for the
control group
![Page 25: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/25.jpg)
The Weights
Recall that our PS is the probability of receiving a
LAAC (rather than a LABA)
𝑊 =𝐿𝐴𝐴𝐶
𝑃𝑆+
1 − 𝐿𝐴𝐴𝐶
1 − 𝑃𝑆
where LAAC is a 0/1 variable.
![Page 26: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/26.jpg)
The Weights
𝑊 =𝐿𝐴𝐴𝐶
𝑃𝑆+
1 − 𝐿𝐴𝐴𝐶
1 − 𝑃𝑆
For those who received LAAC (LAAC = 1), weight =
1 / (probability of receiving LAAC):
𝑊 =1
𝑃𝑆For those who received LABA (LAAC = 0), weight =
1 / (probability of receiving LABA):
𝑊 =1
1 − 𝑃𝑆
![Page 27: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/27.jpg)
The Weights
Similar to survey weights
Respondents from oversampled groups are
assigned low weights
– Selection probability = 1% weight = 1 / 0.01
= 100
Respondents from undersampled groups are
assigned high weights
– Selection probability = 0.2% weight = 1 /
0.002 = 500
![Page 28: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/28.jpg)
Data Set to Estimate the Outcome of
Treatment
𝑊 =𝑍
𝑃𝑆+
1 − 𝑍
1 − 𝑃𝑆
ID Ztreatment = 1
control = 0
PS Weight1/PS
Outcome under
treatment
1 treatment 0.33 1 / 0.33 = 3 Y1
2 control 0.33 0 ?
3 control 0.33 0 ?
4 treatment 0.67 1 / 0.67 = 1.5 Y4
5 treatment 0.67 1 / 0.67 = 1.5 Y5
6 control 0.67 0 ?
![Page 29: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/29.jpg)
Data Set to Estimate the Outcome of
Treatment Estimated average outcome of treatment = 1
𝑁σ𝑖=1𝑁 𝑤𝑖 × 𝑌𝑖 where 𝑤𝑖 =
1
𝑃𝑆for treated people
and 0 for controls.ID Z
treatment = 1control = 0
PS Weight1/PS
Outcome under treatment
1 treatment 0.33 1 / 0.33 = 3 Y1
2 control 0.33 0 ?
3 control 0.33 0 ?
4 treatment 0.67 1 / 0.67 = 1.5 Y4
5 treatment 0.67 1 / 0.67 = 1.5 Y5
6 control 0.67 0 ?
![Page 30: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/30.jpg)
Data Set to Estimate the Outcome for
Controls
𝑊 =𝑍
𝑃𝑆+
1 − 𝑍
1 − 𝑃𝑆
ID Ztreatment = 1;
control = 0
PS Weight Outcome under control
1 treatment 0.33 0 ?
2 control 0.33 =1 / (1 – 0.33) = 1.5 Y2
3 control 0.33 =1 / (1 – 0.33) = 1.5 Y3
4 treatment 0.67 0 ?
5 treatment 0.67 0 ?
6 control 0.67 1 / (1 – 0.67) = 3 Y6
![Page 31: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/31.jpg)
Data Set to Estimate the Outcome for
ControlsEstimated average effect for controls =1
𝑁σ𝑖=1𝑁 𝑤𝑖 × 𝑌𝑖 where 𝑤𝑖 =
1
1 − 𝑃𝑆for people in the
control group and 0 for people in the treated group
ID Ztreatment = 1;
control = 0
PS Weight Outcome
1 Treatment 0.33 0 ?
2 control 0.33 1 / (1 – 0.33) = 1.5 Y2
3 control 0.33 1 / (1 – 0.33) = 1.5 Y3
4 treatment 0.67 0 ?
5 treatment 0.67 0 ?
6 control 0.67 1 / (1 – 0.67) = 3 Y6
![Page 32: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/32.jpg)
Estimating the Treatment Difference
Estimated difference (treatment A – treatment B) =
1
𝑁σ𝑖=1𝑁 𝑍𝑖×𝑌𝑖
𝑃𝑆-1
𝑁σ𝑖=1𝑁 1 − 𝑍𝑖 ×𝑌𝑖
1 −𝑃𝑆
Estimate of the variance
– Robust sandwich type variance estimators
– Bootstrapping
May trim very large weights (propensity score < 1st
percentile or > 99th percentile)
![Page 33: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/33.jpg)
Interpretation of an Inversely Weighted
Analysis
Estimates the Average Treatment Effect (ATE): an
estimate of the treatment effect, if it were applied to
the entire population
![Page 34: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/34.jpg)
It’s Magic
![Page 35: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/35.jpg)
Well, Not Quite
The analyses make no claims to balance
unmeasured covariates
The analyses remove hidden biases only to the
extent that the unmeasured variables are correlated
with the available covariates
Sensitivity analyses can help quantify the possible
effects of unmeasured confounders.
![Page 36: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/36.jpg)
Drawbacks to the Propensity Score
Available data is probably missing key covariates
(e.g., living arrangements, smoking history)
Definition of the baseline time may be difficult (it
should be the time at which the decision about
treatment was made).
Does not eliminate the need to think about patient
identification and selection
![Page 37: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/37.jpg)
Advantages of the Propensity Score
Reduced dimensionality of covariates (important for
rare outcomes)
Can demonstrate that the two groups are similar on
all measured covariates
Like an RCT, does not predict the outcome for a
person with a given set of characteristics
Like an RCT, does not tell you the role of the
covariates in predicting the outcome
Like an RCT, can build planned sub-analyses into
the design
![Page 38: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/38.jpg)
Advantages of Observational Studies
Useful when it is not feasible to use an RCT
– Unethical to withhold treatment
– Exposure believed to be harmful
– Patients will not agree to be randomized
– RCT too expensive
Generalizable (all patients, all providers)
Allows studies of rare events, and studies with long
follow-up times
![Page 39: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/39.jpg)
Disadvantages of Observational Studies
Researcher has no control over assignment of
subjects to treatments
Researcher often has no control over what
covariates are available, their definitions, or the
quality of their measurement
![Page 40: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/40.jpg)
ReferencesAustin PC. An introduction to propensity score methods for reducing the
effects of confounding in observational studies. Multivariate Behavioural
Research. 2011; 46: 399 - 424.
Austin PC, Stuart EA. Moving towards best practice when using inverse
probability of treatment weighting (IPTW) using the propensity score to
estimate causal treatment effects in observational studies. Statistics in
Medicine. 2015; 34: 3661 – 3679.
Anything else written by Peter Austin
Introducing the PSMATCH procedure for propensity score analysis:
https://www.youtube.com/watch?v=JM2uu39zEAs (a very good
introduction to both propensity scores and matching as well as the
PSMATCH procedure)
![Page 41: Propensity Score Methods to Adjust for Bias in Observational Data Group... · 2018-04-06 · Stratification on the PS 3. Matching on the PS. 4. Inverse probability weighting. Institute](https://reader030.vdocuments.net/reader030/viewer/2022041123/5d255e3f88c993cd7d8d45ff/html5/thumbnails/41.jpg)
Inst i tute for Cl in ical Evaluat ive Sciences 41
Thank You