Paid employment and common mental disorders in 50–64-year olds: analysis of three cross-sectional nationally representative survey samples in 1993, 2000 and 2007

Abstract

Associations between employment status and mental health are well recognised, but evidence is sparse on the relationship between paid employment and mental health in the years running up to statutory retirement ages using robust mental health measures. In addition, there has been no investigation into the stability over time in this relationship: an important consideration if survey findings are used to inform future policy. The aim of this study is to investigate the association between employment status and common mental disorder (CMD) in 50-64-year old residents in England and its stability over time, taking advantage of three national mental health surveys carried out over a 14-year period. Data were analysed from the British National Surveys of Psychiatric Morbidity of 1993, 2000 and 2007. Paid employment status was the primary exposure of interest and CMD the primary outcome - both ascertained identically in all three surveys (CMD from the revised Clinical Interview Schedule). Multivariable logistic regression models were used. The prevalence of CMD was higher in people not in paid employment across all survey years; however, this association was only present for non-employment related to poor health as an outcome and was not apparent in those citing other reasons for non-employment. Odds ratios for the association between non-employment due to ill health and CMD were 3.05 in 1993, 3.56 in 2000, and 2.80 in 2007, after adjustment for age, gender, marital status, education, social class, housing tenure, financial difficulties, smoking status, recent physical health consultation and activities of daily living impairment. The prevalence of CMD was higher in people not in paid employment for health reasons, but was not associated with non-employment for other reasons. Associations had been relatively stable in strength from 1993 to 2007 in those three cross-sectional nationally representative samples.