Abstract
BACKGROUND: During the Coronavirus Disease 2019 (COVID-19) pandemic, the World Health Organisation reported a 25% increase in anxiety and depression, and multiple studies indicated that COVID-19 experiences might increase the prevalence of mental illness with subsequent high demands on mental health (MH) services. However, few studies have focussed upon MH across the entire period of pandemic restrictions within England or considered implications for pandemic preparedness. METHODS: We conducted an interrupted time-series analysis of MH service utilisation across England’s National Health Service, including primary care consultations, emergency department attendances, and telephone advice line contacts. The study period was January 1st 2019 to April 20th 2022. Using data from before and after pandemic restrictions, negative binomial regression models generated expected MH utilisation if the pandemic had not occurred. Expected and observed MH utilisation were compared. MH service indicators were analysed both overall and stratified by age group. RESULTS: Early restrictions saw significant declines in access to MH services, telephone calls for MH advice reduced by 36.8% (95% CI -42.0, -31.9) and general practitioner (GP) in-hours consultations for depression decreased by 64.6% (95% CI -77.8, -53.3). Later restrictions revealed an increase in consultations in primary care for anxiety, with an increase of 41.8% (95% CI 38.7, 44.7) in GP out-of-hours. By the final period of restrictions, most MH indicators had either returned to expected levels or were significantly above expected presentations. Young people (15–24) exhibited MH utilisation differences —sharply reduced anxiety and MH during initial restrictions but increasing anxiety in later restrictions within primary care. CONCLUSIONS: COVID-19 restrictions were associated with overall decreases in the utilisation of MH services but increases from in person to remote services were observed. For future pandemic preparedness, remotely accessible MH services are important when in-person services are reduced and the surveillance sources used in this study offers the possibility of real-time decision making. TRIAL REGISTRATION: The data used in this study are based on patients accessing healthcare services in England and are therefore retrospectively registered. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12913-026-14362-z.