The effect of sex, obesity and smoking on health transitions: a statistical meta-analysis based on a systematic literature review
Rostock Center - Discussion Paper 7
Rostock, Rostocker Zentrum zur Erforschung des Demografischen Wandels (2007), unpublished
Context: There is a large and complex body of literature on the effect of various risk factors on health and mortality. This meta-analysis structures and summarizes the knowledge contained in this literature.
Objective: We calculate the average effect-sizes of 55 empirical studies on the impact of sex, obesity and smoking on four different transitions between the statuses of notdisabled, disabled and death. We assess which risk factors and transitions are studied most and whether the impact of theses risk factors differs between the transitions.
Data Source: Expert recommendations, the electronic databases Medline, PsycINFO and SOCA (Sociological Abstracts) and the references in the articles.
Study Selection: The search is confined to the years 1985–2005 and produced a total of 8016 articles. Finally, 55 articles were included in the detailed analysis, selected by two independent researchers according to the study population, a longitudinal design of at least one year, risk factors, the transition and the outcome measures.
Data Extraction: The study characteristics and results were extracted and entered by hand into Excel Tables.
Data Synthesis: The results are presented in groups by risk factor and transition. For each group, we present the average effect size as the result of the statistical metaanalysis, including a fixed effect and a random effect model, and a test for heterogeneity and publication bias. Additionally, a meta-regression was performed.
Conclusions: The results show which risk factors and transitions have not been studied sufficiently to provide significant evidence. They reveal the results expected for the risk factor sex: women have a higher morbidity and a lower mortality. For current smokers, the results show an increased risk, more than for former smokers. Regarding obesity, the evidence shows that for healthy people a higher BMI is associated with a higher risk of
disability, but – surprisingly – that for healthy as well as unhealthy people, a higher BMI is associated with a lower risk of dying.