Article Text

O21-2 What is the impact of temporary employment on self-assessed health? a panel study in catalonia 2002–2012
  1. Francesc Xavier Belvis Costes1,2,
  2. Mireia Julià1,2,
  3. Joan Benach1,2
  1. 1Health Inequalities Research Group (GREDS-EMCONET), Universitat Pompeu Fabra, Barcelona, Spain
  2. 2Johns Hopkins University – Universitat Pompeu Fabra Public Policy Centre, Barcelona, Spain


Background The consistent cross-sectional evidence of the association between flexible employment arrangements and (negative) health outcomes needs to be tested on longitudinal data to take into account healthy worker effect and individual heterogeneity. Although panel data research on this issue is yet not-so-scarce, few works have been performed on South European countries, which seems important taking into account that results are mixed and depend (among other factors) on labour market specific legislation and welfare state development. The aim of this research is to estimate the effect of temporal employment on self-rated health (SRH) in a year’s time span, using a longitudinal design.

Methods Data come from a panel conducted between 2002 and 2012 in Catalonia. A total of n = 15,769 non-missing data episodes of sample members between 25–59 years were selected. A dynamic ordinal logit random effects model was adjusted, where the dependent variable SRH was modelled as a function of one-year lagged SRH, labour market status, other independent variables thought to affect SRH, Mundlak means, initial health condition and a specific individual random effect.

Results Preliminary results indicate a high state dependence of self-rated health. When the latter and main health-related predictors are controlled, SRH of temporary workers does not differ significantly from those of permanent ones (OR = 0.95, 95% CI: 0.84–1.08). The only labour market status negatively affecting SRH is unemployment (OR = 1.17, 95% CI: 1.00–1.36).

Conclusions Results suggest that research on the health effects of flexible employment should move to consider a wider time span or other outcomes to detect significant negative effects on health.

Statistics from

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.