Study cohort
A total of 12 371 319 individuals aged 1–24 years were included in the study cohort between 1 January 2010 and 31 March 2022. This consisted of10 362 294 individuals from 1475 practices in England and 2 009 025 individuals from 406 practices in Scotland, Wales, and Northern Ireland. Supplementary Figures S1 and S2 show how the two cohorts were delineated, respectively. Table 1 shows their demographic characteristics as of 1 March 2020.
Table 1. Demographic characteristics of the study population as of 1 March 2020.a
Psychiatric diagnoses
The monthly observed and expected incidence rates of psychiatric diagnoses for females and males between January 2019 and March 2022 are presented in Figures 1 and 2, respectively. Temporal trends for the whole study period January 2010 through March 2022 are shown in Supplementary Figures S3 and S4 for females and males, respectively. Across this observation period, incidence of ADHD, autism spectrum conditions, and substance misuse were consistently higher for males than for females, whereas anxiety disorders, depression, and personality disorders were more commonly diagnosed in females.
Coinciding with the implementation of initial COVID-19 restrictions, incidence rates across all diagnostic outcomes fell sharply in April 2020 for both sexes. Following this initial reduction, incidence rates of ADHD in females rose substantially, exceeding expected levels by February 2021. For the 2 years since the pandemic began, diagnoses of ADHD in females were 24.7% (95% confidence interval (CI) = 11.9 to 38.9%) greater than expected (Table 2). The increases were attributable to females aged 20–24 years and, to a lesser extent, those aged 17–19 years, with the incidence being 158.6% (95% CI = 83.8 to 264.3%) and 52.8% (95% CI = 4.8 to 122.4%), respectively, greater than expected (Supplementary Table S1). In the year before the pandemic, ADHD was more commonly diagnosed among females in the least deprived (quintile 1) than in the most deprived (quintile 5) areas, and this differential has widened since (Supplementary Figure S5 and Table S2).
Table 2. Difference between expected versus observed number of psychiatric diagnoses and prescription of psychotropic medications by sex in the UK (March 2020 to March 2022) Although ADHD incidence rate in males returned to the expected level a year after the initial lockdown (Figure 2), rates were 17.1% (95% CI = 12.1 to 21.9%) lower than expected across the period as a whole (Table 2). Similarly, diagnoses of autism spectrum conditions in females rose following the initial reduction in April 2020, surpassing the expected rates in March 2021, but there was no significant difference between the observed and expected rates in the 2 years following the pandemic’s onset.
This pattern of incidence rates remaining below or close to the expected levels during the pandemic’s first 2 years was found for all other diagnostic groups, for both sexes and across all age groups and deprivation quintiles (Figures 1 and 2,Table 2, and Supplementary Tables S1–S4). The reduction in incidence was more pronounced for males than females. Between March 2020 and March 2022, incidence of depression was 48.2% (95% CI = 45.0 to 51.3%) lower than expected for males and 31.8% (95% CI = 27.7 to 35.7%) lower for females. Supplementary Figure S6 shows that, for both sexes, the drop in depression diagnoses was largely attributed to the fall in the use of ‘inclusive’ coding. Diagnoses in females based on ‘stringent’ codes actually remained close to the expected levels since October 2020. However, the use of ‘stringent’ codes remained consistently below the expected values for males.
Psychotropic medication prescriptions
As with psychiatric diagnoses, there was a sharp fall in the incidence of psychotropic medication prescribing in April 2020 (Figures 3 and 4). In the following 2 years, incidence was 7.7% (95% CI = 3.5 to 11.6%) lower than expected for females and 22.8% (95% CI = 19.5 to 26.0%) lower for males (Table 2). The reductions in overall incident prescribing were found across almost all age groups and for all deprivation quintiles (Supplementary Tables S5–S8).
Psychotropic medications, except those for ADHD, were more commonly prescribed for females than for males. As with the substantial rise in incidence rates of ADHD observed among females, the incidence of ADHD medication prescribing rose for females and were 14.1% (95% CI = 0.9 to 28.9) greater than expected in the 2 years since March 2020. The increase was driven by those aged 20–24 years, and those from practices in less deprived areas (Supplementary Tables S5 and S6).
Incidence rates for all other types of psychotropic medications prescribing were below the expected levels, with the drop being greater for males than for females, except for benzodiazepines for which the difference between observed and expected rates were similar for both sexes (Table 2).
Supplementary Figures S7 and S8 show the sex-specific event rates of psychotropic medications prescribing. The rates since March 2020 were close to the expected levels for females, but significantly lower than expected for males, although the difference was smaller than for incident prescribing (Table S9 and Table 2). Unlike for diagnostic and incidence prescribing rates, there was no significant increase in the total number of ADHD medications issued for females.
