- Most people with severe coronavirus infections (ie, SARS, MERS, and COVID-19) appear to recover without experiencing mental illness.
- Delirium may feature in the acute stages of COVID-19, and clinicians should be aware of potential long-term mental health effects such as depression, anxiety, fatigue, and PTSD in the wake of the COVID-19 pandemic.
- With only 12 low-to-moderate quality COVID-19 studies (including seven non-peer reviewed preprints) to draw on, and no post-recovery data, continued research will be needed.
- Experts note that findings from past coronavirus outbreaks are useful, but may not perfectly predict prevalence of psychiatric complications from current pandemic.
Most people admitted to hospital with severe COVID-19 should recover without experiencing mental illness if infection with SARS-CoV-2 follows a similar course to the coronavirus epidemics of severe acute respiratory syndrome (SARS) in 2002 and Middle East respiratory syndrome (MERS) in 2012, according to the first systematic review and meta-analysis looking at the psychiatric consequences of coronavirus infections in over 3,550 patients hospitalised with SARS, MERS, and COVID-19, published in The Lancet Psychiatry journal.
Nevertheless, the findings suggest that delirium (a mental state characterised by changes in consciousness, behavioural disturbance, and sometimes hallucinations) may be common in hospitalised patients in the acute stages of SARS, MERS, and COVID-19 illness.
The study looks at severe cases in which individuals have been treated in hospital, and does not apply to milder cases or asymptomatic cases.
In the longer-term, the analysis suggests that SARS and MERS survivors may be at risk for mental illnesses such as depression, anxiety, fatigue, and post-traumatic stress disorder (PTSD) in the months and years following discharge from hospital. While there are no available data on these diagnoses in COVID-19 patients, the authors say that the medical community should be aware of the potential for high rates of these common mental problems in the aftermath of the current pandemic.
The authors caution that given that the majority (68/72) of studies included in the analysis were of either low or medium quality, and mainly involved data on patients with SARS and MERS treated in hospital, the true mental health effects of COVID-19 infection cannot be precisely predicted and will require ongoing research.
"Our analysis of more than 3,550 coronavirus cases suggests that most people will not suffer from mental health problems following coronavirus infection", says Dr Jonathan Rogers from University College London, UK, who co-led the research. "While there is little evidence to suggest that common mental illnesses beyond short-term delirium are a feature of COVID-19 infection, clinicians should monitor for the possibility that common mental disorders such as depression, anxiety, fatigue, and PTSD could arise in the weeks and months following recovery from severe infection, as has been seen with SARS and MERS." 
He continues, "With few data yet for COVID-19, high quality, peer-reviewed research into psychiatric symptoms of patients infected with SARS-CoV-2 as well as investigations to mitigate these outcomes is needed. Monitoring for the development of symptoms should be a routine part of the care we provide." 
There are several reasons why severe coronavirus infections might have psychiatric consequences, including possible direct effects of viral infection (including on the central nervous system), the degree of physiological compromise (eg, low blood oxygen), the immune response, and medical interventions. Other reasons relate to the wider social impact, including social isolation, the psychological impact of a novel severe and potentially fatal illness, concerns about infecting others, and stigma.
Although the COVID-19 pandemic has affected a large proportion of the world's population, relatively little is known about its potential effects on mental health.
To provide more evidence, the authors of the new study conducted a systematic review and meta-analysis of all studies and preprint articles (reporting data on the psychiatric and neuropsychiatric features of individuals with suspected or laboratory-confirmed coronavirus infection (SARS, MERS, or SARS-CoV-2) .
In total, 65 peer-reviewed studies up to March 18, 2020, and seven preprints between January 1 and April 10, 2020, reporting outcomes for patients admitted to hospital were included in the analyses. Pooled point prevalence (the proportion of people affected at a given time) from seven articles were included in the meta-analysis.
Analysis of data from two studies that systematically assessed common symptoms of patients admitted to hospital with SARS and MERS found that confusion occurred in 28% (36/129) of patients, suggesting delirium was common during acute illness (table 2). There were also frequent reports of low mood (42/129; 33%), anxiety (46/129; 36%), impaired memory (44/129; 34%), and insomnia (34/208; 12%) during the acute stage.
Twelve studies focusing on COVID-19 seemed to show a similar picture, with evidence of delirium (confusion in 26/40 intensive care unit patients, 65%; agitation in 40/58 ICU patients, 69%; and altered consciousness in 17/82 patients who subsequently died, 21%) while acutely ill (table 5).
Six studies looking at SARS and MERS patients after recovery from initial infection found frequent reports of low mood (35/332 patients, 11%), insomnia (34/208, 12%), anxiety (21/171, 12%), irritability (28/218, 13%), memory impairment (44/233, 19%), fatigue (61/316, 19%), and frequent recall of traumatic memories (55/181, 30%) over a follow-up period ranging from 6 weeks to 39 months (table 2).
The researchers estimate that the prevalence of PTSD among SARS and MERS survivors was 33% at an average of 34 months after the acute stage of illness (121/402 cases in four studies), whilst rates of depression and anxiety disorders was around 15% at an average of 23 months (77/517 cases from five studies) and one year (42/284 cases from three studies) after the acute stage respectively (figure 2).
However, the authors warn that these may be overestimates of the true mental health burden resulting from these outbreaks.
"It is likely that the apparently high rates of anxiety disorders, depression, and PTSD seen in SARS and MERS patients overestimate the actual burden", says co-lead author Dr Edward Chesney from King's College London, UK. "The lack of adequate comparison groups or assessment of patients' previous psychiatric history means that it is hard to separate the effects of coronavirus infections from pre-existing conditions, the impact of an epidemic on the population as a whole, or that selection bias (the possibility that patients were recruited into studies on the basis of factors that were associated with subsequent development of psychiatric illness) led to high prevalence figures."
The authors note several limitations in the methodology and reporting of studies in the analysis, including the use of preprint articles that had not been subject to peer review; the exclusion of non-English-language articles; and the small sample size of several studies. Additionally, systematic assessment of psychiatric symptoms was rare, and the use of self-reported data (which might not be accurate) was common, whilst few studies included objective biological measures, such as blood markers of genetic, inflammatory, and immune function, or brain imaging. Finally, follow-up time for the post-illness studies varied between 60 days and 12 years, which makes direct comparison between studies difficult.
Writing in a linked Comment, lead author Dr Iris Sommer (who was not involved in the study) from the University Medical Centre Groningen in the Netherlands, says, "Findings from previous coronavirus outbreaks are useful, but might not be exact predictors of prevalences of psychiatric complications for patients with COVID-19. The warning from Rogers and colleagues that we should prepare to treat large numbers patients with COVID-19 who go on to develop delirium, posttraumatic stress disorder, anxiety, and depression is an important message for the psychiatric community."
She goes on to explain, "Treatment of patients admitted to the hospital for SARS-CoV-2 infection seems to be different from treatment of those admitted for SARS-CoV and MERS-CoV infections. Furthermore, the social situation to which COVID-19 survivors return is completely different from that of SARS and MERS survivors. These differences are relevant for the prevalence of psychiatric disorders in both acute and post-illness stages."
Peer-reviewed / Systematic review and meta-analysis / People
NOTES TO EDITORS:
The study was funded by Wellcome Trust, UK National Institute for Health Research (NIHR), UK Medical Research Council, NIHR Biomedical Research Centre at University College London Hospitals NHS Foundation Trust and University College London. It was conducted by researchers from University College London, South London and Maudsley NHS Foundation Trust, UK; King's College London, UK; University of Pavia, Italy; and University College London Hospitals NHS Foundation Trust.
 Quotes direct from authors and cannot be found in the text of the Article.
 The authors excluded studies of neurological complications without specified neuropsychiatric presentations and those investigating the indirect effects of coronavirus infections on the mental health of people who are not infected, such as those mediated through physical distancing measures such as self-isolation or quarantine
The labels have been added to this press release as part of a project run by the Academy of Medical Sciences seeking to improve the communication of evidence. For more information, please see: http://www.