Causal inference with longitudinal outcomes and non-ignorable dropout: Estimating the effect of living alone on cognitive decline
2016 (English)In: Journal of the Royal Statistic Society, Series C: Applied Statistics, ISSN 0035-9254, E-ISSN 1467-9876, Vol. 65, no 1, 131-144 p.Article in journal, Editorial material (Refereed) Epub ahead of print
In this paper we develop a model to estimate the causal effect of living arrangement (living alone versus living with someone) on cognitive decline based on a 15-year prospective cohort study, where episodic memory function is measured every five years. One key feature of the model is the combination of propensity score matching to balance confounding variables between the two living arrangement groups $-$in order to reduce bias due to unbalanced covariates at baseline, with a pattern mixture model for longitudinal data $-$in order to deal with non-ignorable drop-out. A fully Bayesian approach allows us to convey the uncertainty in the estimation of the propensity score and subsequent matching in the inference of the causal effect of interest. The analysis conducted here adds to previous studies in the literature concerning the protective effect of living with someone, by proposing a modeling approach treating living arrangement as an exposure.
Place, publisher, year, edition, pages
John Wiley & Sons, 2016. Vol. 65, no 1, 131-144 p.
Aging, Bayesian inference, Episodic memory, Non-ignorable missingness, Pattern-mixture model, Propensity score matching, Sensitivity
Psychology Probability Theory and Statistics
Research subject Statistics
IdentifiersURN: urn:nbn:se:umu:diva-107712DOI: 10.1111/rssc.12110ISI: 000367978400007PubMedID: 26839439OAI: oai:DiVA.org:umu-107712DiVA: diva2:849140
FunderSwedish Research Council