Evaluating the shift in psychiatric care: Associations between remote consultation use and clinical outcomes in a large longitudinal cohort

The shift to remote consultations represents a profound change in mental health service delivery, comparable to the historic closure of long-stay institutions, yet it occurred without formal evaluation. We aimed to identify diagnostic groups for whom remote care is safe and those for whom it may carry risk. We conducted a retrospective study assembling five annual cohorts using electronic health records from South London, including adults with psychiatric diagnoses (ICD-10 "F" codes) receiving community/outpatient care between March 2019 and March 2024. We examined associations between 10% increments in the proportion of remote consultation use and five outcomes: psychiatric hospitalisation, psychiatric assessments in emergency care, home treatment team contacts, and clinical mentions of improvement or deterioration. Logistic regression models were adjusted for demographic, social, and clinical factors, and stratified by diagnosis. Based on 106,153 observations across five cohort-years, higher use of remote consultations was not associated with worse outcomes for most psychiatric diagnoses, suggesting that remote care is non-inferior to in-person care. However, greater remote care use in patients with schizophrenia-related disorders (F20-F29) was associated with increased odds of hospitalisation (OR: 1.06 [1.03-1.09]), and emergency assessment (OR: 1.04 [1.01-1.07]). While remote mental health consultations are generally safe, but clinical caution is warranted for schizophrenia-related disorders.