Obesity effects on depression: systematic review of epidemiological studies
Atlantis, E. & Baker, MK. (2008). Obesity effects on depression: systematic review of epidemiological studies. International Journal of Obesity,32(6), 881-891. United Kingdom: United Kingdom. Retrieved from https://doi.org/10.1038/ijo.2008.54
Background: Obesity is a well-known cause of cardiovascular disease burden and premature death, but effects on psychological morbidity remain uncertain. This article reports findings following a systematic review of epidemiological studies to determine whether obesity causes depression. Methods: Multiple databases were searched for English-language studies of etiology of obesity (exposure variable, analyzed as an ordered category) on depression outcomes (dependent variables, continuous or categorical). Studies in children and in women during pregnancy or postpartum were excluded, as were nonrepresentative cross-sectional studies. Searches and identification of studies for inclusion were performed by EA, whereas a descriptive synthesis of important study characteristics was undertaken independently by us. Results: We reviewed 24 out of approximately 4500 potentially relevant studies; 4 were prospective cohort studies and 20 were cross-sectional studies (10 from the United States). Effect measures reported in all prospective cohort studies were consistent and suggested that obesity may increase the odds of future depression outcomes (symptoms or nonclinical diagnosis of depression). Effect measures reported in most cross-sectional studies from the United States supported the hypothesized association between obesity and prevalence of depression outcomes for women but not men, in contrast most cross-sectional studies from populations other than the United States consistently failed to find such associations. Conclusions: Overall, there is a weak level of evidence supporting the hypothesis that obesity increases the incidence of depression outcomes. Few high-quality prospective cohort studies exist, and cross-sectional studies account for the vast body of published evidence, and therefore firm conclusions for causality cannot yet be drawn. Our finding warrants additional high-quality etiological research on this topic.