Prevalence of frailty in severe mental illness: findings from the UK Biobank

BACKGROUND: Severe mental illness (SMI) is associated with significant morbidity. Frailty combines biological ageing, comorbidity and psychosocial factors and can predict adverse health outcomes. Emerging evidence indicates that frailty is higher in individuals with SMI than in the general population, although studies have been limited by sample size. AIMS: To describe the prevalence of frailty in people with SMI in a large cohort using three different frailty measures and examine the impact of demographic and sociodemographic variables. METHOD: The UK Biobank survey data, which included individuals aged 37-73 years from England, Scotland and Wales from 2006 to 2010, with linked in-patient hospital episodes, were utilised. The prevalence of frailty in individuals with and without SMI was assessed through three frailty measures: frailty index, physical frailty phenotype (PFP) and Hospital Frailty Risk Score (HFRS). Stratified analysis and dichotomous logistic regression were conducted. RESULTS: A frailty index could be calculated for 99.5% of the 502 412 UK Biobank participants and demonstrated greater prevalence of frailty in women and an increase with age. The prevalence of frailty for those with SMI was 3.19% (95% CI 3.0-3.4), 4.2% (95% CI 3.8-4.7) and 18% (95% CI 15-23) using the frailty index, PFP and HFRS respectively. The prevalence ratio was between 3 and 18 times higher than in those without SMI. CONCLUSIONS: As a measure, frailty captures the known increase in morbidity associated with SMI and may potentially allow for earlier identification of those who will benefit from targeted interventions.