A Calibration Approach to Transportability with Observational Data

08/15/2020 ∙ by Kevin P. Josey, et al. ∙ 0

An important consideration in clinical research studies is proper evaluation of internal and external validity. While randomized clinical trials can overcome several threats to internal validity, they may be prone to poor external validity. Conversely, large prospective observational studies sampled from a broadly generalizable population may be externally valid, yet susceptible to threats to internal validity, particularly confounding. Thus, methods that address confounding and enhance transportability of study results across populations are essential for internally and externally valid causal inference, respectively. We develop a weighting method which estimates the effect of an intervention on an outcome in an observational study which can then be transported to a second, possibly unrelated target population. The proposed methodology employs calibration estimators to generate complementary balancing and sampling weights to address confounding and transportability, respectively, enabling valid estimation of the target population average treatment effect. A simulation study is conducted to demonstrate the advantages and similarities of the calibration approach against alternative techniques. We also test the performance of the calibration estimator-based inference in a motivating real data example comparing whether the effect of biguanides versus sulfonylureas - the two most common oral diabetes medication classes for initial treatment - on all-cause mortality described in a historical cohort applies to a contemporary cohort of US Veterans with diabetes.



There are no comments yet.


page 1

page 2

page 3

page 4

This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.