Prospective associations between depression and risk of hospitalisation for infection: Findings from the UK Biobank

BACKGROUND: Associations between depression and non-communicable disease have been well-described. However, the evidence for its role in the development of infectious disease is less understood. We aimed to examine prospective associations between depression and risk of hospitalisation for infection in middle-aged adults from the UK Biobank (linked with Hospital Episode Statistics) and assessed the role of several depression-related factors. METHODS: We assessed prospective associations between depression status at the baseline assessment (2006-2010) and hospitalisations for infection up to the end of March 2016 in 460,418 middle-aged adults enrolled in the UK Biobank (mean age = 56.23 ± 8.11 years, 53.5% female). Cox regression was used to assess associations between depression and subsequent hospitalisations for any infections, as well as infection subtypes, viral infections, and bacterial infections. Amongst those with depression, we also examined the role of depression duration, the age of onset, and the use of antidepressants in hospitalisation risk. RESULTS: Depression at baseline was prospectively associated with an increased risk of hospitalisation for infection (adjusted hazard ration (aHR) = 1.20, 95% confidence interval (CI) = 1.16 to 1.25). This association was found for all infection subtypes apart from infections of the central nervous system (p = 0.911) and the skin (p = 0.313). Receipt of a depression diagnosis in late adulthood and use of antidepressants (but only in those with none/mild depressive symptoms at baseline) increased the risk of hospitalisation for infection amongst those with depression. CONCLUSIONS: These findings suggest that depression might be a risk factor which could be used to identify those at risk of hospitalisation for infection. Future research is required to understand the underlying factors that might result in this increased risk, so that targeted interventions can be developed. FUNDING: AD and AR are funded by Guy's Charity grant number EIC180702 (MLTC Challenge Fund); AD and JAT are co-funded by MRC and NIHR through grant number MR/S028188/1. IB is supported by the NIHR Maudsley BRC and by the NIHR Collaboration for Leadership in Applied Health Research and Care South Londnoo at King's College Hospital NHS Foundation Trust, King's College London. The views expressed are those of the author[s] and not necessarily those of the ESRC, NIHR, the Department of Health and Social Care or King's College London.