| Autor: |
Shook-Sa BE; Department of Biostatistics, University of North Carolina at Chapel Hill., Hudgens MG; Department of Biostatistics, University of North Carolina at Chapel Hill., Knittel AK; School of Medicine, University of North Carolina at Chapel Hill., Edmonds A; Department of Epidemiology, University of North Carolina at Chapel Hill., Ramirez C; School of Medicine, University of North Carolina at Chapel Hill., Cole SR; Department of Epidemiology, University of North Carolina at Chapel Hill., Cohen M; Stroger Hospital., Adedimeji A; Albert Einstein College of Medicine., Taylor T; SUNY Downstate Medical Center., Michel KG; Department of Infectious Diseases, Georgetown University., Kovacs A; Keck School of Medicine, University of Southern California., Cohen J; Department of Medicine, University of California, San Francisco., Donohue J; Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health., Foster A; Department of Medicine, Emory University., Fischl MA; Division of Infectious Diseases, University of Miami Miller School Medicine., Long D; The University of Alabama at Birmingham., Adimora AA; School of Medicine, University of North Carolina at Chapel Hill. |
| Abstrakt: |
Causal inference methods can be applied to estimate the effect of a point exposure or treatment on an outcome of interest using data from observational studies. For example, in the Women's Interagency HIV Study, it is of interest to understand the effects of incarceration on the number of sexual partners and the number of cigarettes smoked after incarceration. In settings like this where the outcome is a count, the estimand is often the causal mean ratio, i.e., the ratio of the counterfactual mean count under exposure to the counterfactual mean count under no exposure. This paper considers estimators of the causal mean ratio based on inverse probability of treatment weights, the parametric g-formula, and doubly robust estimation, each of which can account for overdispersion, zero-inflation, and heaping in the measured outcome. Methods are compared in simulations and are applied to data from the Women's Interagency HIV Study. |
