Article Text

Download PDFPDF
How to use sequence analysis for life course epidemiology? An example on HIV-positive Sub-Saharan migrants in France
  1. Anne Gosselin1,2,
  2. Annabel Desgrées du Loû1,2,
  3. Eva Lelièvre3
  4. The PARCOURS Study Group
    1. 1 IRD, Institut de Recherche pour le Développement, Marseille, France
    2. 2 SAGESUD Team, CEPED (Université Paris Descartes, IRD, Inserm), Paris, France
    3. 3 Mobilité, Logement et Entourage Unit, National Institute for Demographic Studies (INED), Paris, France
    1. Correspondence to Dr Anne Gosselin, CEPED – Paris Descartes University, Paris, France; anne.gosselin{at}


    Background Life course epidemiology is now an established field in social epidemiology; in sociodemography, the quantitative analysis of biographies recently experienced significant trend from event history analysis to sequence analysis. The purpose of this article is to introduce and adapt this methodology to a social epidemiology question, taking the example of the impact of HIV diagnosis on Sub-Saharan migrants’ residential trajectories in the Paris region.

    Methods The sample consists of 640 migrants born in Sub-Saharan Africa receiving HIV care. They were interviewed in healthcare facilities in the Paris region within the PARCOURS project, conducted from 2012 to 2013, using life event history calendars, which recorded year by year their health, family and residential histories. We introduce a two-step methodological approach consisting of (1) sequence analysis by optimal matching to build a typology of migrants’ residential pathways before and after diagnosis, and (2) a Cox model of the probability to experience changes in the residential situation.

    Results The seven-clusters typology shows that for a majority, the HIV diagnosis did not entail changes in residential situation. However 30% of the migrants experienced a change in their residential situation at time of diagnosis. The Cox model analysis reveals that this residential change was in fact moving in with one’s partner (HR 2.99, P<0.000) rather than network rejection.

    Conclusion This original combination of sequence analysis and Cox models is a powerful process that could be applied to other themes and constitutes a new approach in the life course epidemiology toolbox.

    Trial registration number NCT02566148.

    • communicable diseases
    • demography
    • epidemiological methods
    • life course epidemiology
    • longitudinal studies

    Statistics from

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


    • Contributors AG, ADL and EL conceived the study and research questions. AG performed the statistical analysis and wrote the original draft version of the paper. EL and ADL critically revised the draft. All authors have read and approved the final manuscript.

    • Funding This study was supported by the French National Agency for Research on AIDS and Viral Hepatitis (ANRS) and the General Direction of Health (DGS, French Ministry of Health). The sponsor of the study had no role in study design, data collection, data analysis, data interpretation or writing of the paper.

    • Competing interests None declared.

    • Patient consent Obtained.

    • Ethics approval The Advisory Committee on Data Collection in Health Research (CCTIRS) and the French Data Protection Authority (CNIL) both approved this study.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Data sharing statement All data are available for researchers upon request; please contact the Corresponding Author.

    • Collaborators The Parcours Study Group included: Annabel Desgrées du Loû, F Lert, R Dray Spira, N Bajos, N Lydié (scientific coordinators), J Pannetier, A Ravalihasy, A Gosselin, E Rodary, D Pourette, J Situ, P Revault, P Sogni, J Gelly, Y Le Strat, N Razafindratsima.