A Retrospective Longitudinal Analysis of Mental Health Admissions: Measuring the Fallout of the Pandemic

The study investigates the longer-term impact of COVID-19 on psychiatric inpatient admissions, an area less explored compared with the well-studied early pandemic period. Early literature reported heterogeneous effects of lockdowns and public health measures on mental health service utilization, with many centers observing reduced admissions during lockdown, while some reported increases. Elevated anxiety, depression, and suicidal ideation were noted in certain settings, although non-suicidal self-injury sometimes fell during lockdowns. Policies and restrictions varied internationally and likely influenced admission patterns. The authors aim to evaluate how psychiatric hospital admissions at Malta’s main psychiatric facility (Mount Carmel Hospital) changed over the first two pandemic years (2020–2021) compared with a pre-pandemic reference year (2019), and to relate observed changes to Malta’s pandemic timeline and societal mobility trends.

Prior studies reported mixed impacts of COVID-19 on psychiatric service utilization: many centers saw decreased admissions during lockdowns, while others (e.g., some UK centers) reported increases. Studies highlighted rises in anxiety, depression, and suicidal ideation, but some documented reductions in non-suicidal self-injury presentations during strict lockdowns. Early findings were often based on short-term, small-scale observational snapshots, contributing to conflicting conclusions. There has been a noted paucity of data on inpatient psychiatry specifically. The literature also discusses indirect effects of the pandemic (healthcare access disruptions, patient avoidance, telemedicine adoption and inequalities) and potential direct neuropsychiatric sequelae of COVID-19 via inflammatory and neuroendocrine mechanisms, which could affect depression, anxiety, and psychosis.

Design: Retrospective longitudinal analysis of all psychiatric inpatient admissions to Mount Carmel Hospital (Malta) from 2019–2021. Sample and inclusion/exclusion: 5760 presentations screened. Exclusions: 415 returns from leave (not formally discharged); 431 returns after transfer to general medical hospital for non-psychiatric reasons; 50 miscellaneous (e.g., double entry or no clear psychiatric disorder); 572 with insufficient clinical records. Final analytic sample: 4292 psychiatric admissions (2019: 1348; 2020: 1378; 2021: 1566). Variables: Patient gender, age group, nationality; Mental Health Act (MHA) status on admission (voluntary vs. involuntary under MHA); clinical diagnosis by ICD-10 mental health diagnostic blocks (F00–F99). All F50–59 were eating disorders; F80–89 were autism spectrum disorders (ASD). Self-injurious behavior categories recorded: non-suicidal self-injury (NSSI), suicidal ideation (SI), and suicidal self-injury (SSI). For patients with multiple diagnoses, only those directly contributing to the presentation were recorded. Data sources: Electronic case summaries (equivalent to discharge letters) and Emergency Department records for first presentations via ED. Statistical analysis: Normality assessed with Shapiro–Wilk. As most distributions were normal, comparisons used Student’s paired t-tests of mean monthly admissions: 2019 (pre-COVID reference) vs 2020; 2019 vs 2021; and 2020 vs 2021. This approach was chosen over yearly proportion comparisons to avoid type II error given absolute increases across multiple diagnostic blocks in 2021. For societal context, mean monthly percentage changes from baseline in Google Community Mobility Reports (retail/recreation, grocery/pharmacy, parks, transit, workplaces, residences) were correlated with total monthly admissions using Pearson’s correlation for Feb 2020–Dec 2021 (23 months; Feb 2020 included from 15th onward). Analyses performed in IBM SPSS v26. Institutional Review Board approval obtained from Mount Carmel Hospital.

- Total admissions: 2019=1348; 2020=1378; 2021=1566 (notable increase in 2021 beyond pre-pandemic levels). - Baseline (2019) mean monthly presentations: NSSI 5.6 (SD 2.54); SI 23.2 (SD 3.79); SSI 7.8 (SD 2.08). Highest diagnostic blocks: substance use, mood, schizophrenia-spectrum, anxiety, personality disorders. - 2020 vs 2019: - Overall admissions similar. - Non-Maltese admissions increased: 22.7/month to 28.7/month (p=0.005). - Childhood behavioural/emotional disorders decreased: 2.8/month to 0.8/month (p=0.002). - SI increased: 23.2/month to 36.1/month (p=0.004). - SSI borderline increase: 7.8/month to 11.3/month (p=0.05). - No significant differences by sex or MHA status. - 2021 vs 2019: - Significant surges in multiple diagnoses: mood disorders 27.2 to 42.5/month (p<0.001); schizophrenia, schizotypal and delusional disorders 17.9 to 23.8/month (p=0.002); personality disorders 6.9 to 12.6/month (p=0.012); anxiety disorders 15.8 to 20.3/month (p=0.011); ASD 1.5 to 3.0/month (p=0.032). - Involuntary admissions increased: 31.5 to 46.6/month (p=0.001). - Maltese admissions increased: 89.7 to 102.8/month (p=0.028); non-Maltese stable vs 2020. - Self-injurious behavior increased: NSSI 5.6 to 9.0/month (p=0.03); SI 23.2 to 44.8/month (p<0.001); SSI 7.8 to 13.3/month (p=0.003). - Sex distribution shift driven by increased female admissions. Age groups: proportionate increases most pronounced in 60–69 years (e.g., 2019 vs 2021: +5.6/month; p=0.004; 2020 vs 2021: +5.2/month; p=0.01). - 2021 vs 2020: - Significant additional increases for mood and anxiety disorders; childhood behavioural/emotional disorders returned to pre-pandemic levels. - Among self-injury metrics, only SI showed a significant increase vs 2020. - Mobility correlations (Feb 2020–Dec 2021): Strong positive correlations between monthly admissions and mobility in parks (r=0.800, p<0.001), transit (r=0.755, p<0.001), retail/recreation (r=0.699, p<0.001), grocery/pharmacy (r=0.577, p=0.004), workplaces (r=0.607, p=0.002). Negative correlation with residential mobility (r=−0.699, p<0.001). Patterns indicate admissions rose with greater time spent in public spaces and fell with increased time at home.

The findings address the research aim by demonstrating that the pandemic’s impact on inpatient psychiatry at Malta’s main psychiatric hospital evolved over time: 2020 showed relatively modest changes overall but clear signals of deteriorating mental health (increased SI and borderline SSI), while 2021 saw a marked and broad-based surge in admissions across multiple diagnostic categories, increased involuntary hospitalizations, and increased self-injurious behaviors, indicating greater frequency and severity of presentations. These patterns align with emerging longer-term observational data showing initial reductions or fluctuations followed by rebounds and sustained increases in psychiatric presentations. The authors discuss multiple contributory factors: disruptions to healthcare access and reduced face-to-face care likely delayed presentations; adoption of telemedicine, while beneficial, may have exacerbated inequalities (e.g., older adults’ lower digital engagement); and potential direct neuropsychiatric effects of COVID-19 through inflammatory and neuroendocrine pathways may have influenced increases in depression, anxiety, and psychosis. Age-specific trends (notably increases in 60–69 years) and changes in child and adolescent presentations reflect the impact of school closures and service disruptions, with ASD populations particularly vulnerable due to loss of structured supports and therapies. Strong correlations between admissions and community mobility suggest that restrictive measures and pandemic-related behaviors influenced service utilization: admissions increased when mobility in public domains rose and decreased when residential time increased, consistent with delayed help-seeking during stricter restrictions or heightened COVID-19 concerns. Overall, the results underscore the sustained, multifactorial, and inequitable effects of the pandemic on psychiatric inpatient care, with implications for policy and service planning.

The study concludes that the long-term increase in psychiatric inpatient admissions during the pandemic stems from three interacting factors: (1) direct pathogenic effects of the virus; (2) secondary psychosocial, occupational, and economic impacts; and (3) delays in accessing care driven by anxiety and systemic barriers. The evidence, including correlations with mobility, suggests that the latter two exerted greater influence than direct viral effects. The authors recommend that future pandemic policy planning prioritize mitigation of mental health harms, ensure equitable and timely access to psychiatric care (including telepsychiatry), and strengthen community-based strategies, screening, and suicide prevention.

- Generalizability is limited by variation in pandemic responses across regions; local policies may yield different mental health outcomes. - Hospital admissions may not fully capture mental health burden in populations less likely to be hospitalized (very old or very young). - The study did not distinguish new admissions from readmissions, limiting interpretation of incident vs recurrent illness. - Personality disorders may be underrepresented due to diagnostic focus in discharge summaries, potentially underestimating their true burden.