A systematic review of longitudinal cohort studies on the health of migrant populations

Background: Interest in research on migrant health is increasing. The aim of this study is to review sample characteristics, study design, and outcomes (participation and retention rate) of longitudinal studies of the health of migrant populations, and to evaluate whether there are differences in outcomes related to study populations and methodology. Methods: A literature search of prospective longitudinal studies on migrants’ health was performed in Medline and Web of Science, with 545 articles retrieved. Key informants were contacted when needed. After identification, screening, and eligibility, nine articles were included. Results: The most commonly studied topics were occupational and mental health (44.4%). Two studies had sample sizes of >5000 subjects, and 4 studies recruited families. One study targeted undocumented workers. Study duration was 2 years in 4 studies with 1 follow up wave. Two studies collected biological samples, and 2 used incentives. Higher participation (PR) and retention (RR) rates were found in studies of families, studies of groups perceived to be at high risk, studies where the researchers had close community ties, and studies where complete contact information had been obtained by the researchers. Lower PR and RR were associated with large time delays between waves and targeting irregular workers. Respondent driven sampling (RDS) was successful in reaching hidden populations. Conclusions: Identification of documented migrants through governmental records, early follow up, use of a variety of strategies (including digital technologies) to locate participants and maintaining personal relationships are the main factors influencing PR and RR. It is essential to consider them when planning research and to foresee and plan for the difficulties that might arise during a longitudinal study.