Does hospital admission risk for depression vary across social groups? A population-based register study of 231,629 middle-aged Finns.

PURPOSE: Evidence on social differentials in depression outcomes remains inconsistent. We assess social predictors of psychiatric admission for depression in a community setting.
METHODS: A register-based 14% sample of community-dwelling Finns aged 40-64 at the end of 1997 was assessed for depression and psychiatric comorbidity, using register data on psychiatric hospital care and medication purchases in 1996-1997. Those with inpatient treatment for unipolar depression (n = 846), those with antidepressant treatment (n = 8,754), and those with neither (n = 222,029) were followed for psychiatric admission with a diagnosis of unipolar depression in 1998-2003. Differentials in admission rates by socioeconomic position, employment status, and living arrangements were studied using Cox proportional hazards modelling.
RESULTS: Among those with prior inpatient or antidepressant treatment, the material aspects of socioeconomic position increased admission risk for depression by 20-40%, even after controlling for baseline depression severity and psychiatric comorbidities, whereas education and occupational social class were unrelated to admission risk. Among inpatients, also having no partner, and among antidepressant users, being previously married and living without co-resident children increased admission risk. However, among inpatients few excess risks reached statistical significance. Among those with no inpatient or antidepressant treatment, all measures of low social position and not living with a partner predicted admission, and the factors had more predictive power in admission than among those with prior treatment.
CONCLUSIONS: Further studies should disentangle the mechanisms behind the higher admission risk among those with fewer economic resources and no co-resident partner.