ABSTRACT
Garssen et al. provide meta-analytic evidence, from longitudinal studies with control for baseline outcomes, for protective associations of religious service attendance and religious importance with subsequent depression. The present commentary considers whether or not these associations might be causal, and what the practical relevance of these associations might be. Further discussion is given to the methodological considerations that are important for advancing our understanding of the relationship between religion and mental health. More meta-analyses of the type conducted by Garssen et al. (i.e. restricted to longitudinal studies with control for baseline outcome) should be carried out with other health and well-being outcomes. To better establish the evidence base, cross-sectional studies should be excluded from meta-analyses.
Disclosure statement
No potential conflict of interest was reported by the author.