Five-year illness trajectories across racial groups in the UK following a first episode psychosis

Abstract

PurposePsychosis disproportionally affects ethnic minority groups in high-income countries, yet evidence of disparities in outcomes following intensive early intervention service (EIS) for First Episode Psychosis (FEP) is less conclusive. We investigated 5-year clinical and social outcomes of young people with FEP from different racial groups following EIS care.MethodData were analysed from the UK-wide NIHR SUPEREDEN study. The sample at baseline (n = 978) included White (n = 750), Black (n = 71), and Asian (n = 157) individuals, assessed during the 3 years of EIS, and up to 2 years post-discharge (n = 296; Black [n = 23]; Asian [n = 52] and White [n = 221]). Outcome trajectories were modelled for psychosis symptoms (positive, negative, and general), functioning, and depression, using linear mixed effect models (with random intercept and slopes), whilst controlling for social deprivation. Discharge service was also explored across racial groups, 2 years following EIS.ResultsVariation in linear growth over time was accounted for by racial group status for psychosis symptoms—positive (95% CI [0.679, 1.235]), negative (95% CI [0.315, 0.783]), and general (95% CI [1.961, 3.428])—as well as for functioning (95% CI [11.212, 17.677]) and depressive symptoms (95% CI [0.261, 0.648]). Social deprivation contributed to this variance. Black individuals experienced greater levels of deprivation (p < 0.001, 95% CI [0.187, 0.624]). Finally, there was a greater likelihood for Asian (OR = 3.04; 95% CI [2.050, 4.498]) and Black individuals (OR = 2.47; 95% CI [1.354, 4.520]) to remain in secondary care by follow-up.ConclusionFindings suggest variations in long-term clinical and social outcomes following EIS across racial groups; social deprivation contributed to this variance. Black and Asian individuals appear to make less improvement in long-term recovery and are less likely to be discharged from mental health services. Replication is needed in large, complete data, to fully understand disparities and blind spots to care.