Abstract:
|
Objective: The goal is to estimate the causal effect of antidepressant among participants with incident MCI on time-to-dementia with death as a competing event. We used the Marginal Structural Fine-Gray Model, given the possibility of indication bias. Methods: A retrospective cohort study was conducted using the National Alzheimer's Coordinating Center Uniform Data Set. Records from September 2005 to December 2017 were included. Eligible participants (N=960) were incident MCI cases aged at least 65 years with at least one visit following the MCI diagnosis. A Marginal Structural Fine-Gray Model with Inverse Probability Weighting was used to assess the causal effect of antidepressant usage on time-to-dementia with death as a competing event, adjusting for the effects of identified confounders. Indication bias was addressed by this method as well. Results: Among all participants, mean age was 80.5, 60.5% were females, and 19.9% were taking antidepressants at their MCI diagnosis visit. The adjusted subdistribution hazard ratio of dementia for participants who reported using antidepressants at MCI diagnosis visit is 1.75 (1.33, 2.31) times that for participants who did not.
|