CoChat
Sign Up
Cited by 381COVID-19 and Mental HealthLong-Term Effects of COVID-19Maternal Mental Health During Pregnancy and Postpartum

Long-term effects of COVID-19 on mental health: a systematic review

Open access
Bourmistrova, Nicole Wallbridge, Solomon, Tomas, Braude, Philip, Strawbridge, Rebecca, Carter, Ben·2022·Journal of Affective Disorders
DOI 10.1016/j.jad.2021.11.031PMC PMC8758130
Open this paper in CoChat — free account

Abstract

Background Acute effects of COVID-19 can be life-threatening. Alterations in mental health during the active infection have been documented, but the long-term consequences are less clear. Method A systematic review was undertaken to investigate the effect of COVID-19 infection on long-term mental health outcomes. Three databases [PubMed, Medline (Ovid) and Cochrane library] were searched between 1st October 2019 and 29th August 2021 with additional hand searching to identify all published studies reporting symptoms of generalised anxiety, depression, post-traumatic stress disorder (PTSD), or sleep disturbance in participants at least one month after COVID-19 infection. The prevalence and mean symptom score of each were assessed. Results Eight hundred and eighty five studies were found, of which 33 were included in the review involving a total of 6743 participants. The studies’ risk of bias were typically fair quality. The median study age of participants was 57.8 years (IQR 49.3–60.7), with 63.0% male (IQR 57.0%–73.0%). Participants typically experienced no or mild symptoms of long-term anxiety (GAD-7, STAI-S, HADS) and depression (PHQ-9, BDI, PHQ-2, HADS). Prevalence varied depending on the measurement tool. Sleep disturbances (primarily insomnia) were most commonly reported as mild. PTSD prevalence was similar to anxiety and depression. Conclusion The overall effect of the pandemic has been linked with worsening psychiatric symptoms. However, the long-term effect from direct COVID-19 infection has been associated with no or mild symptoms. Studies exhibited the long-term prevalence of anxiety, depression, PTSD, and sleep disturbances to be comparable to general population levels.

Integrity signals

Registry

Crossref · 10.1016/j.jad.2021.11.031

Cites

71 of 71 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 48 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    Participants in the included studies generally had no or mild long-term anxiety and depression symptoms.

    “Participants typically experienced no or mild symptoms of long-term anxiety (GAD-7, STAI-S, HADS) and depression (PHQ-9, BDI, PHQ-2, HADS).”
    • Also stated in: § Discussion · plain
      “The included participants typically exhibited no worse than mild symptoms of anxiety and depression.”
    Quote in the paperSubstantial claim
  • Findingown data

    Depression prevalence was about half as high in cohorts assessed after three months (10.36%) as in those assessed earlier (20.84%).

    “Cohorts followed up for more than three months had half the prevalence of depression compared to those that were followed up for less than three months, reporting a prevalence of 10.36% and 20.84% respectively.”
    Quote in the paperNot a substantial claim
  • Findingcitation

    A single GAD-7 study found no anxiety, with a mean of 1.88 (95% CI 0.68-3.08).

    “One study reported no anxiety with an average score of 1.88 (95% CI = 0.68, 3.08; (Kang et al., 2021)”
    Quote in the paperNot a substantial claim
  • Findingown data

    In mild to moderate COVID-19 cases, 17.52% of 3431 patients had at least mildly severe anxiety symptoms.

    “This included 3431 patients 17.52% of whom were experiencing at least mildly severe anxiety symptoms.”
    Quote in the paperSubstantial claim
  • Findingown data

    In severe COVID-19 cases, depression prevalence was 20.39% among 309 patients, averaging none to mild symptoms.

    “Depression prevalence in severe COVID-19 cases was 20.39% out of 309 patients, averaging none to mild symptoms of depression.”
    Quote in the paperNot a substantial claim
  • May combine two claims. 17
  • Classified as probably not a substantial claim. 12

  • May combine two claims.

    Findingown data

    The systematic review search identified 885 articles, of which 33 studies involving 6743 participants met inclusion criteria.

    “Eight hundred and eighty five studies were found, of which 33 were included in the review involving a total of 6743 participants.”
    • Also stated in: § Study selection and characteristics · plain
      “The search resulted in a total of 885 articles (Fig. 1 - PRISMA flow chart). After title, abstract and full text screening, thirty-three studies were included”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Findingown data

    The median age of study participants was 57.8 years (IQR 49.3-60.7) and median proportion male was 63.0% (IQR 57.0%-73.0%).

    “The median study age of participants was 57.8 years (IQR 49.3–60.7), with 63.0% male (IQR 57.0%–73.0%).”
    • Also stated in: § Study selection and characteristics · plain
      “The median study age of patients was 57.8 years old (interquartile range [IQR] 49.3 to 60.7 years old) and the median study proportion of male participants was 63.0% male (IQR 57.0% to 73.0%).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingreasoning

    D'Cruz et al. found 57% sleep disturbance, but the authors left it out of the pooled estimate because the measure was not validated.

    “Finally, D'Cruz et al. (2021) reported a prevalence of 57%, however, since a validated tool was not used, this was excluded from the statistical summary.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites D'Cruz et al. (2021)

      Not checked in the free run.

  • May combine two claims.

    Findingcitation

    Among 30 severe COVID-19 cases, 40% had poor sleep quality, though most reported no poor sleep.

    “the prevalence of sleep quality disturbance was 40% out of 30 cases, with most participants indicating no poor sleep.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Nine studies removed individuals with prior psychiatric disorders, and seventeen did not state whether they did so.

    “Nine studies excluded patients with a history of psychiatric disorders (Albu et al., 2021, Guo et al., 2020, Mazza, 2021, Mazza et al., 2020, Mendez et al., 2021, Negrini et al., 2020, Ortelli et al., 2021, Taquet et al., 2021, van den Borst et al., 2021) , with the other seventeen studies not specifying.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingcitation

    Two PHQ-9 studies reported no depression overall with average scores of 2.75 (95% CI 1.21-4.29) and 5.31 (95% CI 4.23-6.38).

    “Kang et al. (2021) and Wong et al. (2020) reported no depression overall, with average scores of 2.75 (95% CI = 1.21, 4.29) and 5.31 (95% CI = 4.23, 6.38).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Seven included studies reported participants with a lifetime psychiatric history, with prevalences of 40%, 13.9%, 30%, 15.4%, 10%, and 8%, and one study reported 4.8% anxiety and 1.3% depression history.

    “Seven studies included participants with a lifetime psychiatric history (de Graaf et al., 2021, Horn et al., 2020, Kang et al., 2021, Park et al., 2020b, Poyraz et al., 2021, Tarsitani et al., 2021) with a prevalence of 40%, 13.9%, 30%, 15.4%, 10% and 8% respectively, and (Noviello et al., 2021) included 4.8% and 1.3% patients with anxiety and depression history respectively.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    In COVID-19 survivors, long-term rates of anxiety, depression, PTSD, and sleep problems were similar to those in the general population.

    “Studies exhibited the long-term prevalence of anxiety, depression, PTSD, and sleep disturbances to be comparable to general population levels.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: General-population mental health levels during the pandemic provide a valid benchmark for long-term mental health after COVID-19 infection.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    In prior SARS, MERS, and Ebola outbreaks, about half of survivors had at least one of the psychiatric conditions reviewed here.

    “Past epidemics caused by SARS, MERS, or ebola have demonstrated that half of the survivors suffered from at least one of the psychiatric conditions examined in this review (Chua et al., 2004; Keita et al., 2017).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Chua et al., 2004 source

      Not checked in the free run.

    • cites Keita et al., 2017 source

      Not checked in the free run.

  • May combine two claims.

    Interpretationreasoning

    GAD-7 studies yielded the highest anxiety prevalence, probably because they used a low threshold.

    “Anxiety prevalence percentage was highest in GAD-7 studies, likely due to the low cut-off score used”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    The review protocol was published in PROSPERO (CRD42021239505).

    “The protocol for this study was published in the PROSPERO register (CRD42021239505, 2021).”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    The authors judged meta-analysis unsuitable given anticipated clinical differences among studies.

    “Meta-analysis, as outlined in the protocol, was not deemed appropriate due to the expected clinical heterogeneity between studies.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationcitation

    Long-term anxiety was below one study's control group and similar to pandemic-era general-public samples, suggesting it was relatively limited.

    “Symptoms were lower than the prevalence reported in a control group from one of the included studies (Noviello et al., 2021) and comparable to similar studies in the general public during the pandemic (Salari et al., 2020; Zhang et al., 2020; Krishnamoorthy et al., 2020), indicating relatively limited long-term anxiety in COVID-19 patients.”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Noviello et al., 2021

      Not checked in the free run.

    • cites Salari et al., 2020

      Not checked in the free run.

    • cites Zhang et al., 2020

      Not checked in the free run.

    • Rests on an assumption: General-population mental health levels during the pandemic provide a valid benchmark for long-term mental health after COVID-19 infection.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    The review is reported in adherence to PRISMA guidelines.

    “We report the findings in adherence to the PRISMA guidelines.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    The authors pre-specified in their protocol that long-term meant assessment one month or more after COVID-19 diagnosis.

    “The definition of ‘long-term’ was pre-specified in our protocol”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation

    An earlier systematic review (search date 18 March 2020) found that psychiatric disorders following COVID-19 exposure had a low prevalence and a trajectory similar to other coronavirus epidemics.

    “It found similarities in the trajectory of other coronavirus epidemics with a low prevalence of mental health conditions (Rogers et al., 2020).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Rogers et al., 2020 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    Discrepancies in risk-of-bias assessment were resolved by consensus.

    “Discrepancies were resolved by consensus.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    Some GAD-7 means exceeded pre-pandemic German general-population values, while others fell below them, leaving the comparison uncertain.

    “Moreover, the average GAD-7 scores from Hasan et al., 2021 and Gou et al. (2020) were higher than the general public before the COVID-19 pandemic in a German population (Löwe et al., 2008) while the average scores from Park et al., (2020b)) and Kang et al. (2021) were lower indicating uncertainty.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Löwe et al., 2008

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    Two investigators independently screened articles for inclusion.

    “Two investigators (TS and NWB) independently screened articles for inclusion.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationcitation

    Even allowing for over-sensitive diagnosis, anxiety was less common than in SARS, MERS, and Ebola outbreaks and than in short-term symptom studies.

    “Even when disregarding the potential for over-sensitivity of diagnosis in the included studies, the overall prevalence for anxiety was lower compared with SARS, MERS, and ebola epidemics (Chua et al., 2004; Keita et al., 2017; Jeong et al., 2016) than studies that looked at short-term symptoms (Krishnamoorthy et al., 2020).”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Chua et al., 2004 source

      Not checked in the free run.

    • cites Keita et al., 2017 source

      Not checked in the free run.

    • cites Jeong et al., 2016

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data

    A sensitivity analysis examined prevalence at three months or more after diagnosis and omitted severe COVID-19 cases.

    “A sensitivity analysis was conducted comparing outcome prevalence from at least 3 months post-diagnosis (excluding severe COVID-19 cases).”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductown data

    Risk of bias was assessed by two authors working independently with the Newcastle-Ottawa Scale.

    “Two independent authors (NWB and TS) assessed the risk of bias (RoB) using the Newcastle-Ottawa Scale (NOS) (Wells et al., 2000).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Wells et al., 2000

      Not checked in the free run.

  • May combine two claims.

    Interpretationreasoning

    The lowest anxiety prevalence came from Xiong et al., probably because interviews overestimate less than self-report scales.

    “Conversely, Xiong et al. (2021) had the lowest prevalence which is likely to be due to clinical interviews which are less likely to result in overestimation compared to self-rated questionnaires.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Xiong et al. (2021) source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing). May combine two claims.

    Study design and conductnone given

    Longer-term assessment could yield more reliable estimates and account for psychiatric history, COVID-19 severity, and lack of comparable controls.

    “Longer-term assessment may also offer more reliable estimates and may also account for participants’ psychiatric history, the severity of COVID-19, and the lack of comparable controls.”
    Quote in the paperNot a substantial claim
24 of 24

  • The paper relies on this without support.

    The findings from adult hospital or treatment-centre samples can be generalised to COVID-19 survivors outside those settings.

    “All studies examined an adult sample from treatment centres or hospitals.”

  • The paper relies on this without support.

    Mental health symptoms measured at follow-up are attributable to direct COVID-19 infection rather than to indirect pandemic effects.

    “However, the long-term effect from direct COVID-19 infection has been associated with no or mild symptoms.”

  • The paper relies on this without support.

    General-population mental health levels during the pandemic provide a valid benchmark for long-term mental health after COVID-19 infection.

    “Studies exhibited the long-term prevalence of anxiety, depression, PTSD, and sleep disturbances to be comparable to general population levels.”

  • The paper relies on this without support.

    The COVID-19 pandemic overall has been associated with worsening psychiatric symptoms.

    “The overall effect of the pandemic has been linked with worsening psychiatric symptoms.”

  • The paper relies on this without support.

    The PCL-5, TSQ, DTS, and IES-R instruments provide comparable measures of PTSD.

    “As for PTSD, overall PCL-5 and TSQ scores were lower compared to DTS and IES-R scores.”
5 of 5

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • PTSD follow-up trend reverses between results and discussion

    Major

    The results report higher PTSD prevalence after more than three months than before three months. The discussion instead says the sensitivity analysis found a decrease with longer follow-up.

  • Anxiety instrument counts total 14 studies, not 15

    Minor

    The listed instrument groups contain five GAD-7 studies, one STAI-S study, seven HADS studies and one clinical-interview study. These add to 14, rather than the stated 15.

    Reported 15 · recomputed 14

  • Mendez's PTSD score is assigned to conflicting instruments

    Minor

    Mendez is listed among studies using DTS, but its reported mean is subsequently described as a PCL-5 score.

  • HADS anxiety study count changes from seven to six

    Minor

    The anxiety results first list seven studies using HADS, then describe six studies using it.

  • The sole STAI-S anxiety study is attributed to different authors

    Minor

    The instrument list identifies Mazza as the one STAI-S study, but the subsequent description identifies Mendez as the only study using STAI-S.

  • Competing-interest declarations disagree

    Minor

    One declaration discloses an honorarium received by RS, while the later declaration says all authors have no conflicts of interest.

6 of 6

Cites

Flagged first, then by citations

8 of the 79 references couldn’t be matched to a known paper.

Research works printed but not linked: 3

  1. Darley, 2021, Persistent symptoms up to four months after community and hospital-managed SARS-CoV-2 infection, The Medical Journal of Australia, 10.5694/mja2.50963 · DOI
  2. Hasan, 2021, Mental Health of the COVID-19 Patients in Bangladesh. Mymensingh medical journal, MMJ, 30, 189
  3. Yu, 2020, Prevalence of sleep disturbances during COVID-19 outbreak in an urban Chinese population: a cross-sectional study, Sleep Med., 74, 18, 10.1016/j.sleep.2020.07.009 · DOI

Cited by

Sorted by citations

Related

Authors
Bourmistrova, N. W., Solomon, T., Braude, P., Strawbridge, R., & Carter, B.
Affiliations
  • King's College London (GB)
  • North Bristol NHS Trust (GB)
Corresponding author
Carter, Ben
ORCID
Venue
Journal of Affective Disorders
Publisher
Elsevier BV
Published
2022-02
Volume
299
Pages
118-125
ISSN
0165-0327 (ISSN-L)
DOI
10.1016/j.jad.2021.11.031
PMC
PMC8758130
OpenAlex
W3213906311
Semantic Scholar
822529366cca7ca42b197910253d2232de588965
Cited by
369
References
79
Citations by year
2026: 32 · 2025: 68 · 2024: 79 · 2023: 121 · 2022: 69
FWCI
8.93
Type
journal article
Language
en
Open access
Licence
Locations
  • Journal of Affective Disorders (publishedVersion)
  • PubMed Central (publishedVersion)
  • PubMed (publishedVersion)
Topics
  • Social Sciences › Psychology › Clinical Psychology › COVID-19 and Mental Health
  • Health Sciences › Medicine › Neurology › Long-Term Effects of COVID-19
  • Health Sciences › Medicine › Public Health, Environmental and Occupational Health › Maternal Mental Health During Pregnancy and Postpartum
Keywords
COVID-19 pandemic, long-term psychological effects, post-COVID-19 mental health, generalized anxiety disorder, depression, post-traumatic stress disorder, sleep disorders, insomnia
MeSH terms
COVID-19, SARS-CoV-2, Anxiety, Depression, Female, Humans, Male, Mental Health, Middle Aged, Pandemics

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-07.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-07.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review