Artigos

Introduction
More than 280 million people worldwide have depression, a debilitating and common mental disorder with a prevalence of 5%1. Depression is characterized by symptoms such as depressed mood, loss of interest or pleasure in activities, changes in appetite and sleep patterns and persistent fatigue2. Previous studies have demonstrated that depression is associated with high medical comorbidities and an increased risk of overall mortality3.
The COVID-19 pandemic, caused by the SARS-CoV-2 virus, has resulted in an unprecedented global public health crisis, impacting various sectors of society and overwhelming healthcare systems worldwide. As of October 2025, there are over 778 million confirmed cases and over seven million COVID-19 deaths globally4, representing the second most common cause of death worldwide5. Since the onset of the pandemic, in March 2020, significant efforts have been directed towards identifying risk factors that may predispose individuals to more severe outcomes and fatalities associated with the virus6. Among these risk factors, pre-existing medical conditions such as cardiovascular diseases, diabetes, and obesity have been widely studied and recognized7. However, the relationship between mental health conditions, particularly depression, and COVID-19 mortality remains insufficiently understood.
Therefore, investigating whether prior depression is associated with a higher likelihood of death from COVID-19 is crucial for understanding the implications of mental health on the severity of COVID-19 outcomes and to inform intervention strategies that can reduce mortality among vulnerable populations. Thus, this study aimed to systematically review and meta-analyze studies on the association between pre-existing depression and COVID-19 mortality.


Methods
This systematic review and meta-analysis was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines8 and prospectively registered in PROSPERO (CRD42023473116).
We investigated whether pre-existing depression is a risk factor for mortality among individuals diagnosed with SARS-CoV-2 using the PECO strategy as follows: population = individuals with SARS-CoV-2; exposure = depression; control (not exposed) = no depression; outcome = mortality.

Eligibility criteria
Analytical observational studies (cohort or case–control) with the following information for evaluation were considered eligible for the qualitative synthesis: studies that evaluated mortality rates among individuals with and without depression diagnosis using standardized diagnostic criteria or symptomatology and documented before COVID-19 event or present as of the index date (ie, comorbid); with severe acute respiratory syndrome infection by SARS-CoV-2 confirmed by laboratory tests.

Information sources
We searched the following databases from inception to October 2025: PubMed, Embase, Virtual Health Library (Lilacs/Bireme/VHL Brazil), Scopus, Web of Science, PsycINFO and CINAHL. In addition, the reference lists of the included articles were cross-checked to identify articles that were not assessed in the search strings.



Search strategy
The complete search strategies used in each database have been included as supplementary material (Supplementary material 1).We searched by title and abstract. There was no restriction regarding language.

Selection process
The search results were imported into the Covidence online platform9, in which duplicates were identified and excluded; study selection was then performed. The first stage of selection consisted of reading the titles and abstracts to identify and remove studies that did not meet the objective of this systematic review. All articles that passed this first filtering process were read in full to apply the eligibility criteria. The selection process was conducted by six reviewers and each abstract/text was read independently by two reviewers (M.V., Y.A., I.V., F.M., T.R. and E.P.), with conflicts resolved by a seventh reviewer (D.M.).

Data collection process
Data were extracted by six investigators (M.V., Y.A., I.V., F.M., T.R. and E.P.) into online spreadsheets (Google Sheets). The following data were extracted: author; article title; year; country of publication; type of study; follow-up period (days); SARS-CoV-2 diagnostic method; method used to diagnose depression; confounding factors; comorbidities assessed; inclusion criteria; exclusion criteria; contingency table; and medical care.

Study risk of bias assessment
The methodological quality was evaluated by two previously trained independent reviewers (M.V. and Y.A.) using the Newcastle–Ottawa scale. The methodological quality score of cohort and case–control studies was estimated using the following three components: group selection (0–4 points); quality of adjustment for confusion (0–2 points) and exposure evaluation after outcome (0–3 points). The maximum score was 9, representing high methodological quality. Differences between the reviewers regarding the selection of articles and/or quality analysis were resolved by consensus, and in case of persistent disagreement a third reviewer evaluated the article.

Statistical methods and analysis
A meta-analysis was performed using a random effects model with inverse variance weighting to obtain the summary measure using the metan command (Stata version 12.0 statistical software). Statistical heterogeneity was assessed using the I2 statistic and its respective 95% confidence intervals (95% CI), where I2 ? 50% was considered to indicate substantial heterogeneity. A sensitivity analysis excluding one study at a time ('leave one out') was conducted to assess the influence of individual studies on the pooled estimate. The results are shown as forest plots and tables. The pooled Odds Ratio (OR) of COVID-19 mortality between groups of depressives and non-depressives (reference) was calculated. Subgroups analyses were also performed considering: 1) the adjustment for confounding factors; and 2) medical care. For the random-effects model we used the DerSimonian and Laird method. Publication bias was assessed using Egger's test.







Results
Figure 1 shows the PRISMA flow diagram of the study search and selection process. A total of 4,068 articles were identified through database searches (PubMed = 699, Embase = 1,108, Virtual Health Library = 1,059, Scopus = 796, Web of Science = 322, PsycINFO = 55, and CINAHL = 29). After removing 1,285 duplicates, 2,783 studies remained for title and abstract screening. During this initial screening phase, 2,737 studies were excluded for not meeting the inclusion criteria. A total of 46 articles were then retrieved for full-text review by two independent researchers. After detailed assessment, 27 studies were excluded for absence of mortality data or lack of an appropriate control group. Finally, 19 studies met all eligibility criteria and were included in both the qualitative synthesis and meta-analysis.
Table 1 summarizes the main characteristics of the included studies. Among the 19 studies, one was conducted in a low- and middle-income country, and 18 were performed in high-income countries. Altogether, the studies comprised 2,151,552 individuals diagnosed with COVID-19, of whom 197,712 had depression. The mean age of participants ranged from 53 to 101 years across studies. Most studies used cohort designs.
All selected studies were assessed using the Newcastle-Ottawa Scale and had a score ? 6. The main sources of potential bias included lack of control for confounding variables in some analyses and incomplete adjustment for comorbidities. No major concerns regarding missing data or unclear outcome ascertainment were identified.

Overall Analysis
The individual effect estimates and 95% confidence intervals (CIs) for each study included in the meta-analysis are shown in Figure 2. The pooled analysis demonstrated that individuals with depression had a significantly higher odds of death from COVID-19 compared with those without depression (OR = 1.53; 95% CI: 1.29–1.81). This finding indicates that depression is associated with a 53% higher chance of COVID-19 mortality.
However, a high level of heterogeneity was observed among the included studies (I² = 97.4%; p < 0.001), suggesting variability in effect estimates possibly related to differences in study design, population characteristics, or adjustment strategies for confounding variables. Given the high observed heterogeneity (I² > 90%), we conducted a sensitivity analysis excluding one study at a time ('leave one out') to assess the influence of individual studies on the pooled estimate. The exclusion of any study resulted in significant variations in the effect estimate (Supplementary material 2).
Despite the heterogeneity, most of individual estimates pointed toward the same direction of association, consistently indicating increased mortality among depressed individuals.

Subgroup analyses considering the adjustment for confounding factors
To explore sources of heterogeneity, subgroup analyses were conducted according to whether studies adjusted for potential confounding factors. As shown in Figure 3, studies that did not adjust for confounders exhibited a slightly stronger association between depression and COVID-19 mortality (OR = 1.56; 95% CI: 1.28–1.91) and maintained a high degree of heterogeneity (I² = 97.4%). This pattern suggests that part of the variability across studies may stem from differences in the control of key covariates such as age, sex, and comorbidities. The magnitude of the association in unadjusted analyses underscores the potential confounding influence of these variables, but also supports the robustness of the observed relationship.



Subgroup analyses considering the type of medical care
A second subgroup analysis was conducted considering the type of medical care (Figure 4). Within the subgroup that evaluated hospitalized and/or intensive care unit (ICU) patients, heterogeneity was markedly reduced (I² = 72.3%), and the association between depression and COVID-19 mortality remained statistically significant, although of smaller magnitude (OR = 1.33; 95% CI: 1.07–1.64). This attenuation may reflect the more homogeneous clinical context of hospitalized populations and the influence of disease severity on outcomes. Nevertheless, the persistence of a significant association reinforces the role of depressive disorders as an important factor related to adverse COVID-19 prognosis, even among critically ill patients.

Reporting biases and certainty of evidence
The intercept of the Egger test was 0.110 (p = 0.722), indicating no evidence of small-study effects and suggesting a low probability of publication bias.
Considering the high methodological quality of the included studies and the consistency of results across subgroups, the overall certainty of the evidence supporting an association between depression and COVID-19 mortality was judged as moderate. This classification takes into account the observational design of the studies and the high level of heterogeneity.






Discussion
Our results demonstrate that pre-existing depression is significantly associated with an increase in COVID-19 mortality (OR=1.53 [95% CI: 1.29–1.81]). Based on a systematic literature search and subsequent meta-analysis of 19 studies involving more than two million individuals diagnosed with COVID-19, among which approximately 9% had a history of depression, we observed that those with pre-existing depression had 53% more chances of dying from COVID-19 compared to those without depression.
Our findings are consistent with the meta-analysis conducted by Ceban et al. (2021), which also reported a significant association between preexisting mood disorders and increased COVID-19 mortality (OR = 1.55; 95% CI: 1.34–1.79). The similarity in magnitude across both analyses, despite methodological differences, supports the robustness of the association. However, the lack of distinction in several included studies as to whether an individual with a mood disorder had major depressive disorder or bipolar disorder, in addition to COVID-19 not being exclusively diagnosed through laboratory tests, affect inferences and interpretations of the results.
Depression is a mental health condition influenced by a combination of biological, psychological and social factors11. There are numerous known risk factors for depression. Family history of depression, neurochemical imbalances, such as changes in brain chemicals like serotonin, dopamine and norepinephrine, hormonal changes, such as those during pregnancy, menopause or thyroid disorders, stressful life events and low socioeconomic status can contribute to the development of depressive symptoms12.
Confounding factors that may impact the association between COVID-19 mortality and depression include several variables such as age, sex, hypertension, diabetes, cardiovascular diseases, chronic kidney disease13 and medication use, in addition to clinical and psychosocial characteristics14. For instance, older age is a known risk factor for both COVID-19 mortality13 and depression15, which may distort this relationship. In addition, socioeconomic factors, such as education level and income, can influence both mental health and access to COVID-19 treatment, potentially confounding the analysis of the relationship between depression and mortality16.
In subgroup analyses, studies that did not adjust for confounding variables (e.g., age, sex, comorbidities, socioeconomic status) showed a higher effect size (OR = 1.56; 95% CI: 1.28–1.91) compared to those that applied statistical adjustments (OR = 1.37; 95% CI: 1.16–1.62). This difference indicates that part of the unadjusted association may reflect the influence of overlapping risk factors that are independently associated with COVID-19 mortality, such as advanced age, cardiovascular disease, and diabetes. Nonetheless, even after adjustment, the association between depression and mortality persisted, supporting the independent contribution of depression to worse COVID-19 outcomes and reinforcing the robustness of the overall effect.
A high level of heterogeneity (I² = 97.4%) was observed among the studies, which was expected due to variations in population characteristics, clinical contexts, and adjustment criteria. In subgroup analyses considering the type of medical care, within the subgroup that evaluated hospitalized patients and/or those in the intensive care unit (ICU), heterogeneity was notably reduced (I² = 72.3%) and the association remained statistically significant, although of a smaller magnitude (OR = 1.33; 95% CI: 1.07–1.64). This finding suggests that the clinical context and disease severity influence the strength of the association between depression and mortality. Hospitalized cohorts tend to represent more homogeneous and severe cases, while population studies encompass a wider range of disease severity and socioeconomic diversity, which may increase variability. Therefore, the observed heterogeneity does not compromise the consistency of the association, but reflects the complex and multifactorial nature of both depression and COVID-19 outcomes.
The use of antidepressants may also influence the outcomes of the infection, either due to immunomodulatory effects or drug interactions17,18. Patients with a history of depression may be at greater risk due to the presence of comorbidities, such as cardiovascular disease or obesity19, which aggravate the effects of COVID-1920. Our results align with findings from Hoertel et al. (2021)21, who reported that the use of antidepressant medication (SSRI and non-SSRI classes) was associated with a reduced risk of intubation or death in hospitalized COVID-19 patients (HR = 0.56; 95% CI: 0.43–0.73; p < 0.001). While our meta-analysis did not aim to evaluate treatment effects, the consistency between these findings highlights that depression and its management may play a role in modulating COVID-19 outcomes. Differences across studies, including diagnostic criteria, data sources, and pandemic phases, may explain the variation in effect size estimates.
Collectively, these findings underscore that depression should be recognized as a clinically relevant comorbidity in infectious disease contexts and that mental health interventions may indirectly improve survival outcomes during health crises.
The WHO proclaimed the end of the Public Health Emergency of International Concern on May 5, 2023; however, this does not mean that COVID-19 is no longer a health hazard. The disease's global spread is still being referred to as a pandemic. This suggests that it is time for countries to transition from an emergency mode to managing COVID-1922 along with other infectious diseases like lower respiratory infections and tuberculosis, ranked, respectively, as the fifth and tenth leading causes of death worldwide5. Although the peak of the pandemic has passed, its influence will be significant and long-lasting. The SARS-CoV-2 virus, which caused the pandemic, is still killing one person every three minutes, while many survivors are suffering from the incapacitating effects of long COVID, which can leave them debilitated for months23.
During the COVID-19 pandemic, there has been a steady flow of synthesis research published on a wide variety of topics. While the need to disseminate information to the medical community and general public was paramount, concerns have been raised regarding the scientific rigor in published reports. Overall, the quality of these reviews fell short of what is expected for systematic reviews24.

Strengths and limitations
The main strengths of this study include the comprehensive literature search across major biomedical databases, the inclusion of only high-quality studies, and the large pooled sample size, which increases the precision and reliability of the findings. Moreover, subgroup analyses enhanced the interpretability of the results by identifying sources of heterogeneity and demonstrating the robustness of the association across different analytical contexts.
However, several limitations should be acknowledged. Most included studies were conducted in high-income countries, potentially limiting generalizability to low- and middle-income settings, which account for over 80% of the global population. Differences in healthcare access, socioeconomic conditions, and diagnostic resources may influence the observed association. Additionally, variation in the assessment of depression (clinical diagnosis vs. screening instruments) and the lack of uniform adjustment for confounders may have contributed to residual heterogeneity. Future research should prioritize studies in diverse populations and incorporate standardized diagnostic and analytical procedures to better delineate causal mechanisms.




Conclusion
This systematic review and meta-analysis provide evidence that pre-existing depression is independently associated with a higher risk of COVID-19 mortality. Even after accounting for confounding factors, individuals with depression showed a consistent and significantly elevated risk of death compared with those without depression. These findings highlight the urgent need to integrate mental health screening and management into public health strategies during global health emergencies. Systematic identification and treatment of depressive symptoms, especially among individuals at high risk for severe infections, should be prioritized as part of preparedness and response measures.
Beyond the COVID-19 context, this evidence reinforces the importance of addressing mental health as a determinant of physical health outcomes. Future studies should explore whether early diagnosis and appropriate treatment of depression can mitigate mortality risk during infectious disease outbreaks and other large-scale health crises.

Declaração de Disponibilidade de Dados
As fontes dos dados utilizados na pesquisa estão indicadas no corpo do artigo.














References

1. Global Health Data Exchange (GHDx). Institute of Health Metrics and Evaluation. 2021 [cited 2024 Nov 05]. Available from: http://ghdx.healthdata.org/gbd-results-tool?params=gbd-api-2019-permalink/d780dffbe8a381b25e1416884959e88b
2. Marx W, Penninx BWJH, Solmi M, Furukawa TA, Firth J, Carvalho AF, et al. Major depressive disorder. Nat Rev Dis Primers 2023; 9(1)44.
3. Lee SY, Lee JP, Lee J, Park JY, Kim EY. Association between depressive symptoms and the risk of all-cause and cardiovascular mortality among US adults. Prog Neuropsychopharmacol Biol Psychiatry 2023; 125:110755.
4. World Health Organization (WHO). WHO Coronavirus Disease (COVID-19) Dashboard. 2024 [cited 2024 Nov 05]. Available from: https://covid19.who.int
5. World Health Organization (WHO). The top 10 causes of death. 2024 [cited 2024 Nov 05]. Available from: https://www.who.int/news-room/fact-sheets/detail/the-top-10-causes-of-death
6. Wang C, Ramasamy A, Verduzco-Gutierrez M, Brode WM, Melamed E. Acute and post-acute sequelae of SARS-CoV-2 infection: a review of risk factors and social determinants. Virol J 2023; 124
7. Monye IN, Makinde MT, Oseni TIA, Adelowo AB, Nyirenda S. Covid-19 and Pre-Morbid Lifestyle-Related Risk Factors-A Review. Health Serv Insights 2023; 16:11786329231215049
8. Page MJ, Moher D, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. PRISMA 2020 explanation and elaboration: updated guidance and exemplars for reporting systematic reviews. BMJ 2021; 372:n160.
9. Covidence systematic review software (Covidence). Veritas Health Innovation, Melbourne, Australia. Available from: https://www.covidence.org
10. Ceban F, Nogo D, Carvalho IP, Lee Y, Nasri F, Xiong J, et al. Association Between Mood Disorders and Risk of COVID-19 Infection, Hospitalization, and Death: A Systematic Review and Meta-analysis. JAMA Psychiatry 2021; 78(10):1079–1091
11. Remes O, Mendes JF, Templeton P. Biological, Psychological, and Social Determinants of Depression: A Review of Recent Literature. Brain Sci 2021; 11(12):1633.
12. Alberti A, Araujo Coelho DR, Vieira WF, Moehlecke Iser B, Lampert RMF, Traebert E, et al. Factors Associated with the Development of Depression and the Influence of Obesity on Depressive Disorders: A Narrative Review. Biomedicines 2024; 12(9):1994.
13. Zhang JJ, Dong X, Liu GH, Gao YD. Risk and Protective Factors for COVID-19 Morbidity, Severity, and Mortality. Clin Rev Allergy Immunol 2023; 64(1):90-107.
14. Aguayo GA, Fischer A, Elbéji A, Linn N, Ollert M, Fagherazzi G. Association between use of psychotropic medications prior to SARS-COV-2 infection and trajectories of COVID-19 recovery: Findings from the prospective Predi-COVID cohort study. Front Public Health 2023; 11:1055440.
15. Schaakxs R, Comijs HC, Lamers F, Kok RM, Beekman ATF, Penninx BWJH. Associations between age and the course of major depressive disorder: a 2-year longitudinal cohort study. Lancet Psychiatry 2018; 5(7):581-590. Erratum in: Lancet Psychiatry 2018; 5(10):e24.
16. Wang Y, Kala MP, Jafar TH. Factors associated with psychological distress during the coronavirus disease 2019 (COVID-19) pandemic on the predominantly general population: A systematic review and meta-analysis. PLoS One 2020; 15(12):e0244630.
17. Costa LHA, Santos BM, Branco LGS. Can selective serotonin reuptake inhibitors have a neuroprotective effect during COVID-19? Eur J Pharmacol 2020; 889:173629.
18. Mas M, García-Vicente JA, Estrada-Gelonch A, Pérez-Mañá C, Papaseit E, Torrens M, Farré M. Antidepressant Drugs and COVID-19: A Review of Basic and Clinical Evidence. J Clin Med 2022; 11(14):4038.
19. Zhang S, E L, Lu Z, Yu Y, Yang X, Chen Y, et al. The Chain-Mediating Effect of Obesity, Depressive Symptoms on the Association between Dietary Quality and Cardiovascular Disease Risk. Nutrients 2023; 15(3):629.
20. Sanchis-Gomar F, Lavie CJ, Mehra MR, Henry BM, Lippi G. Obesity and Outcomes in COVID-19: When an Epidemic and Pandemic Collide. Mayo Clin Proc 2020; 95(7):1445-1453.
21. Hoertel N, Sánchez-Rico M, Vernet R, Beeker N, Jannot AS, Neuraz A, et al. Association between antidepressant use and reduced risk of intubation or death in hospitalized patients with COVID-19: results from an observational study. Mol Psychiatry 2021; 26(9):5199-5212.
22. World Health Organization (WHO). Statement on the fifteenth meeting of the IHR (2005). Emergency Committee on the COVID-19 pandemic. 2023 [cited 2024 Nov 05]. Available from: https://www.who.int/news/item/05-05-2023-statement-on-the-fifteenth-meeting-of-the-international-health-regulations-%282005%29-emergency-committee-regarding-the-coronavirus-disease-%28covid-19%29-pandemic
23. The Guardian. Is Covid really over? WHO’s announcement sounds more like surrender than victory. 2023 [cited 2024 Nov 05]. Available from: https://www.theguardian.com/world/2023/may/07/is-covid-really-over-whos-announcement-sounds-more-like-surrender-than-victory
24. Baumeister A, Corrin T, Abid H, Young K, Ayache D, Waddell L. The quality of systematic reviews and other synthesis in the time of COVID-19. Epidemiology & Infection 2021; 149:E182.
25. Azar KMJ, Shen Z, Romanelli RJ, Lockhart SH, Smits K, Robinson S, et al. Disparities In Outcomes Among COVID-19 Patients In A Large Health Care System In California. Health Aff (Millwood) 2020; 39(7):1253-1262.
26. Bayrak M, Çadirci K. The associations of life quality, depression, and cognitive impairment with mortality in older adults with COVID-19: a prospective, observational study. Acta Clin Belg 2022; 77(3):588-595.
27. Berenguer J, Calvo-Alcántara MJ, Alvaro-Meca A, Estévez JC, Basanta M, Ruiz S, et al. Three Years of the Coronavirus Disease 2019 Pandemic in a European Region: A Population-Based Longitudinal Assessment in Madrid Between 2020 and 2022. Open Forum Infect Dis 2023; 11(1):ofad635.
28. Clouston SAP, Luft BJ, Sun E. Clinical risk factors for mortality in an analysis of 1375 patients admitted for COVID treatment. Sci Rep 2021; 11(1):23414.
29. Hassan L, Peek N, Lovell K, Carvalho AF, Solmi M, Stubbs B, et al. Disparities in COVID-19 infection, hospitalisation and death in people with schizophrenia, bipolar disorder, and major depressive disorder: a cohort study of the UK Biobank. Mol Psychiatry 2022; 27(2):1248-1255.
30. Janc J, Wo?niak A, Le?nik P, ?ysenko L. Does cognitive function impairment affect the duration of hospitalization and in-hospital mortality in geriatric patients hospitalized for COVID-19? PLoS One 2023; 18(4):e0284977.
31. Jegede O, Anand Raman A, Tiongson B, Garlapati PR, Hershberger J, Gayam V. Clinical characteristics, hospital course, and outcomes among COVID-19 positive patients with mental illness in a community hospital in New York City. International Journal of Mental Health 2020; 50(1):4–15.
32. Kostev K, Hagemann-Goebel M, Gessler N, Wohlmuth P, Feldhege J, Arnold D, et al. Is there an association between depression, anxiety disorders and COVID-19 severity and mortality? A multicenter retrospective cohort study conducted in 50 hospitals in Germany. J Psychiatr Res 2023; 157:192-196.
33. Kundi H, Çetin EHÖ, Canpolat U, Aras S, Celik O, Ata N, et al. The role of Frailty on Adverse Outcomes Among Older Patients with COVID-19. J Infect 2020; 81(6):944-951.
34. Orlando V, Rea F, Savaré L, Guarino I, Mucherino S, Perrella A, et al. Development and validation of a clinical risk score to predict the risk of SARS-CoV-2 infection from administrative data: A population-based cohort study from Italy. PLoS One 2021; 16(1):e0237202.
35. Schwarzinger M, Luchini S, Teschl M, Alla F, Mallet V, Rehm J. Mental disorders, COVID-19-related life-saving measures and mortality in France: A nationwide cohort study. PLoS Med 2023; 20(2):e1004134.
36. Seon JY, Kim S, Hong M, Lim MK, Oh IH. Risk of COVID-19 diagnosis and death in patients with mental illness: a cohort study. Epidemiol Psychiatr Sci 2021; 30:e68.
37. Wang Q, Xu R, Volkow ND. Increased risk of COVID-19 infection and mortality in people with mental disorders: analysis from electronic health records in the United States. World Psychiatry 2021; 20(1):124-130.
38. Wang Y, Yang Y, Ren L, Shao Y, Tao W, Dai XJ. Preexisting Mental Disorders Increase the Risk of COVID-19 Infection and Associated Mortality. Front Public Health 2021; 9:684112.
39. Yang H, Chen W, Hu Y, Chen Y, Zeng Y, Sun Y, et al. Pre-pandemic psychiatric disorders and risk of COVID-19: a UK Biobank cohort analysis. Lancet Healthy Longev 2020; 1(2):e69-e79.