Incident Diabetes and Mobility Limitations: Reducing Bias Through Risk-set Matching.

ABSTRACT:

Background

Increased prevalence of diabetes in the U.S. population could contribute substantially to increases in disability at older ages. Previous studies have examined the association between prevalent diabetes and various impairments and disabilities. Methods considering incident, rather than prevalent, diabetes as the exposure of interest can reduce bias in estimates of these associations.

Methods

Risk-set matching, a type of propensity score matching meant to handle time-varying exposures, was used to estimate the relationship between incident diabetes and mobility limitations among adults in the Health and Retirement Study. This approach ensures that covariates precede diabetes onset rather than follow it.

Results

Individuals who were diagnosed with diabetes du

SUBMITTER: Fishman EI 

PROVIDER: S-EPMC4481690 | biostudies-literature | 2015 Jul

REPOSITORIES: biostudies-literature