1026195

research-article2021
SMO0010.1177/20503121211026195SAGE Open MedicineDemissie and Bitew

SAGE Open Medicine

Systematic Review

SAGE Open Medicine

Mental health effect of COVID-19 Volume 9: 1­–11

© The Author(s) 2021
Article reuse guidelines:
pandemic among women who are sagepub.com/journals-permissions
DOI: 10.1177/20503121211026195
https://doi.org/10.1177/20503121211026195

pregnant and/or lactating: A systematic journals.sagepub.com/home/smo

review and meta-analysis

Dereje Bayissa Demissie

and Zebenay Workneh Bitew

Abstract
Objectives: The purpose of this study is to examine the effects of coronavirus disease 2019 pandemic on the prevalence
of anxiety, depression, stress, insomnia, and social dysfunction among pregnant and/or lactating women and to measure the
global pooled prevalence of mental health effects among these populations in the era of coronavirus disease 2019 pandemic.
Methods: Comprehensive literature searching was conducted and studies published from 1 January 2020 to 30 September
2020 reporting the prevalence of anxiety, depression; stress, insomnia, and social dysfunctions were included. The pooled
prevalence of anxiety, depression, stress, insomnia, and social dysfunctions was estimated using a random-effect model. In
this study, all statistical analyses were performed using STATA (version 15) software.
Results: There were a total of 19 studies included in the meta-analysis, of which 16, 14, 4, 2, and 2 studies were included
in computing the pooled prevalence of anxiety, depression, stress, insomnia, and social dysfunction, respectively. The
pooled prevalence of anxiety was 33% (95% confidence interval: 50%−61%), with significant heterogeneity between studies
(I2 = 99.68%, p = 0.001). The pooled prevalence of depression was 27% (95% confidence interval: 9%−45%), with remarkable
heterogeneity between studies (I2 = 99.29%, p = 0.001). Likewise, the pooled prevalence of stress was 56% (95% confidence
interval: 30.07%−82.22%), with significant heterogeneity between studies (I2 = 98.8%, p = 0.0001). The pooled prevalence
of social dysfunction was 24.3% (95% confidence interval: 13.41%−62.03%), with significant heterogeneity between studies
(I2 = 97.5%, p = 0.0001) and finally, the pooled prevalence of insomnia was 33.53% (95% confidence interval: 3.05%−64.0%),
with significant heterogeneity between studies (I2 = 99.6%, p = 0.0001).
Conclusions: In this study, the mental health effects of the COVID-19 pandemic among pregnant and lactating women
were found to be significant. Stress was the most common mental health problem in these population groups. Therefore,
policymakers and health planners should give great emphasis to addressing maternal mental well-being during and after this
global health crisis. Maternal mental health must be one of the international and national public health priority agendas to
enhance the well-being of pregnant and lactating women. Besides, giving psychological support to pregnant and lactating
women may reduce the long-term negative effects of this pandemic.

Keywords
Mental health, Coronavirus disease 2019, pregnant, lactating women, anxiety, depression, stress, insomnia, social
dysfunctions

Date received: 15 December 2020; accepted: 26 May 2021

Introduction
School of Nursing, St. Paul’s Hospital Millennium Medical College, Addis
Coronavirus disease 2019 (COVID-19) is a disease caused Ababa, Ethiopia
by a novel coronavirus (2019-nCoV) that was first reported
Corresponding author:
in Wuhan, Hubei Province, China in December. Since then, Dereje Bayissa Demissie, School of Nursing, St. Paul’s Hospital Millennium
there have been over 1,203,459 cases of COVID-19 infec- Medical College, Paster, Gulele, Addis Ababa 1271, Ethiopia.
tions worldwide, with 64,754 deaths.1 Email: [email protected]

Creative Commons Non Commercial CC BY-NC: This article is distributed under the terms of the Creative Commons
Attribution-NonCommercial 4.0 License (https://creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use,
reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open
Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage).
2 SAGE Open Medicine

COVID-19 has both mental/emotional and social impli- terms of anxiety, depression, stress, insomnia, and social
cations for pregnant and postpartum women who have been dysfunction in pregnant and lactating women. This system-
physically separated from families, relatives, and society all atic review and meta-analysis will focus on a growing array
around the world. Understanding the maternal mental health of articles on mental well-being and COVID-19 published
implications of the COVID-19 pandemic is becoming after the outbreak of the COVID-19 pandemic in different
increasingly necessary to best avoid the occurrence of severe countries. To provide a more global viewpoint, we conducted
mental disorders as a secondary consequence during the a comprehensive systematic review and meta-analysis of
postpartum period.2,3 available research findings examining anxiety, depression,
A study conducted in Toronto, Canada, revealed that 29% stress, insomnia, and social instability in pregnant and lactat-
of individuals had symptoms of posttraumatic stress disorder ing women during the COVID-19 pandemic.
and 31% had depression during severe acute respiratory syn-
drome (SARS) outbreak.4 During the current pandemic, a
study was done in China’s Wuhan region reported that 53.8% Methods
of respondents were in moderate or severe mental health prob-
Eligibility criteria and search strategies
lems, of which 17% and 29% had moderate to severe depres-
sion and anxiety, respectively.5 During pregnancy and the We followed the Preferred Reporting Items for Systematic
postpartum period are vulnerable times for mothers them- Reviews and Meta-analysis (PRISMA) guideline25 to pre-
selves to more likely have cognitive and behavioral issues, pare the whole document. National surveys, published and
while psychological distress can have negative consequences unpublished articles were explored from different databases.
for both the mother and the baby. Studies showed that symp- In addition, the reference lists of included articles were
toms of anxiety and depression were higher among pregnant cross-checked to identify articles that were not assessed in
women than men during the COVID-19 pandemic.5,6 A study the search strings. We sought comprehensive literature
undertaken among prenatal women indicates that “elevated research published on PubMed, CINAHL (EBSCOhost),
prenatal anxiety and depression symptoms” might increase the Global Health (CABI), Medline (EBSCOhost), and other
risk of postpartum depression as well as prenatal infection and sources (Google Scholar and Google) from 01 January 2020
illness rates.7,8 Besides, previous studies revealed that prenatal to 30 September 2020 that reported prevalence of anxiety,
anxiety and depression can cause changes in physical activity, depression; stress, insomnia, and social dysfunctions. Studies
nutrition, sleep pattern, maternal mood, and fetal health which conducted among pregnant and lactating women to validate
may increase the risk of miscarriage, preterm birth, lower birth the psychological consequences of COVID-19 have been
weight, and lower Apgar scores at birth.9–13 Children of moth- included in this systematic review and meta-analysis.
ers who have endured elevated stress are at greater risk of sub-
sequent mental health problems.13–17 Prenatal anxiety and
Inclusion and exclusion criteria
depression are also correlated with changes in brain develop-
ment and function in infants and children.18–21 These long- The inclusion criteria in this meta-analysis were as follows:
lasting psychological and neurological effects underscore the (1) population (pregnant women, lactating women, breast-
significance of alleviating prenatal discomfort for both preg- feeding women, women in antenatal care, and puerperal
nant women and their infants. A meta-analysis conducted in women); (2) exposure (novel coronavirus, COVID-19,
the general population reported that the pooled prevalence of nCoV, severe acute respiratory syndrome coronavirus 2, and
depression during the COVID-19 outbreak is 25% (95% con- SARS-CoV-2); (3) outcome (impact, effect, mental health,
fidence interval (CI): 18%–33%) with significant heterogene- psychology, anxiety, depression, stress, and social dysfunc-
ity between studies (I2 = 99.60%, p < 0.001).22 tion); (4) study design (cohort studies, cross-sectional stud-
It is also necessary to identify possible resilience factors ies, epidemiology, and observational studies); (5) study
that can help guard against high prenatal stress. Social rein- setting (community-based surveys, health institutions, and
forcement may minimize the impact of prenatal stress and has Web-based surveys). The data from each study were verified
been shown to alleviate the effects of prenatal anxiety and for eligibility using study area, study setups, assessment
depression symptoms of maternal and infant stress response methods, study designs, title, abstract, and full texts.
systems.23 Physical exercise is often correlated with decreased Eventually, observational studies reporting the magnitude
depressive and anxiety symptoms in pregnant women24 con- of different forms of psychological disorders among preg-
sidering the possible negative psychological effects of psy- nant or lactating women were included.
chological, health, and financial instability coupled with social Exclusion criteria: (a) the same patients were enrolled in
exclusion, there is an immediate need to evaluate the preva- different articles; (b) commentaries, editorials, case reports, let-
lence of psychological distress in pregnant women during this ters, and family-based studies; and (c) short communications.
pandemic and to establish a protective factor. Nonetheless, studies with incomplete or unclear diagnos-
To date, there is no systematic study or meta-analysis in tic methods and without full texts were excluded. Letters to
the mental health effects of the COVID-19 pandemic in editors, conference proceedings, and qualitative studies were
Demissie and Bitew 3

also excluded. The EndNote X8 reference manager was used The prevalence was calculated by dividing the total number
to manage articles. of mental disorders by the total sample size and multiplying
The appropriateness of the key terms was checked prior to it by 100. The binomial distribution formula was used to
conducting searches in each database. Example of search string compute the standard error for each original study. The
in PubMed: ((“Psychology”(Mesh) OR (“Mental Health”(Mesh) pooled estimates of depression, anxiety, and stress were
OR (“Anxiety”(Mesh) OR (“Depression”(Mesh)) OR computed using “meta pop” using a sample size as a weight
“Depression, Postpartum”(Mesh) OR (“well-being”) OR (wgt) variable. This was done due to significant variability in
(“social instability”) OR (“Stress, Psychological”(Mesh) AND the sample size of the included studies. The pooled estimates
(“COVID-19” (Supplementary Concept))) OR (“severe acute were presented with their 95% CIs. The effect sizes were
respiratory syndrome coronavirus 2” (Supplementary prevalence of each component of mental disorders.
Concept))) OR (novel coronavirus)) OR (nCoV)) AND All studies measured anxiety using standardized scales,
(“Pregnant Women”(Mesh) OR (prenatal)) OR (perinatal)) OR the most common being the Generalized Anxiety Disorder
(postpartum)) OR (antenatal)) OR (postnatal)) OR (puerperal)) 7-item (GAD-7) scale and Hospital Anxiety and Depression
OR (peurperal)) OR (lactating women) Journal Article, Scale (HADS). Similarly, all studies measured depression
Observational Study, in the last 1 year, Humans, English, Adult: among pregnant and lactating women using standardized
19–44 years. scales, the most common being the Edinburgh Postnatal
Depression Scale (EPDS) and HADS. Stress was also
assessed based on the Perceived Stress Scale (PSS), while
Data extraction process
insomnia and social dysfunction were diagnosed based on
A standardized data extraction checklist was prepared using standardized self-rating scale, respectively.
Microsoft Excel 2016, and the data were extracted by two
investigators (Z.W.B. and D.B.D.), independently. Name of
the author(s), publication year, study country, sample size, Statistical methods and analysis
study population, diagnostic methods, anxiety, depression, The pooled estimates were calculated using STATA Version
stress, insomnia, and social dysfunction were used in the 15 (STATA Corporation, College Station, Texas) software in
extraction process. There were no discrepancies between this meta-analysis. Both random- and fixed-impact methods
ZWB and DBD on the inclusion of studies. The extracted were used to measure the pooled estimates. The pooled esti-
data were cross-checked interchangeably by the two authors mates were computed using random-effects models and
and inconsistencies were solved accordingly. weighted using the inverse variance method in the presence
of high heterogeneity among studies. Subgroup analyses
Quality assessment of studies were done using different parameters (diagnostic methods
and study country). We verified the appropriateness of each
Critical appraisal of the included studies was performed by datum before the analysis of each datum. Forest plots, sum-
two authors (Z.W.B. and D.B.D.), independently using mary tables, and texts were used to present the findings of
Joanna Briggs Institute (JBI) Critical Appraisal Checklist for this study.
Observational Studies was used for quality assessment.26
The scores were added up and changed to percentages. The
minimum score was 0 and 8 was the maximum one since all Publication bias and heterogeneity
included studies were cross-sectional studies. Articles with The funnel plot Sterne and Egger’s28 regression test were
>50% quality scores were included in this systematic review used to measure publication bias at a 5% significant level. In
and meta-analysis (Supplemental Table 2). The inter-rater addition, heterogeneities among studies used to compute the
agreement was computed by an author (ZWB) before the pooled estimates in this meta-analysis were explored using
decision of inclusion in this study was made. We computed forest plot, I2 test, and the Cochrane Q statistics.29 The I2
inter-rater agreement using Cohen’s kappa coefficient (κ). values of 25%, 50%, and 75% were interpreted as low,
The findings revealed that there were substantial agreement27 medium, and high heterogeneity, respectively.30 In the cur-
between the two raters (κ = 0.652, p ⩽ 0.001). rent meta-analysis, significant heterogeneity was considered
when the I2 value was ⩾50%, with a p value < 0.05. The
Summary measures possible sources of significant heterogeneity were addressed
through sub-group and sensitivity analyses.
The primary outcome of this research was to determine the
mental health consequences of the COVID-19 pandemic in
pregnant and lactating women using various diagnostic Results
methods. The pooled prevalence was computed for depres-
sion, anxiety, stress, insomnia, and social dysfunction. Selection of studies
Subgroup analyses were also done using diagnostic methods In the initial search, 3007 studies were obtained from data-
and the country where the original studies were performed. bases and gray literature sources. Primarily, 586 studies were
4 SAGE Open Medicine

Identification
Studies identified through data base
Studies obtained from other sources=44
searching=2963

Screening Studies after duplicated studies were removed Duplicated Studies

(n=2421) (n=586)

Studies excluded by title

and abstract (n=2392)
Eligibility

Full texts of studies assessed for

eligibility Full texts studies excluded
(n=29) (n=10)
- Inconsistent outcomes (n=10)
Included

19 studies were included in this systematic

review & meta-analysis

Figure 1. PRISMA flow chart showing study selection process.

excluded due to duplication. Then, 2392 studies were funnel plot seems asymmetric despite the Egger’s regression
screened using titles and abstracts and 2421 were removed. test (p = 0.098) did not confirm the asymmetry of funnel plot
Finally, the full texts of 29 studies were assessed for eligibil- (Figure 3). The funnel plots report an upward pattern with no
ity. Of the total 29 studies, 10 were excluded due to incon- point falling within the predicted 95% CI for anxiety.
sistency of results.31–40 Eventually, 19 eligible studies were Sensitivity analysis was also performed to identify the pos-
used in the final analysis of the current systematic review sible source of heterogeneity among the included studies.
and meta-analysis41–59 (Figure 1). However, there was no single study having a remarkable
effect on the heterogeneity of the pooled estimate (Figure 4).
Finally, the funnel plots look asymmetric for anxiety (see
Study characteristics
Figure 3) which is congruency with Rosenthal’s method of
Of total of 19 studies included in the final analysis, four stud- finding of fail-safe-N analysis warrants acknowledgment of
ies were done in Canada,42,47,52,57 three studies in China,48,50,56 possible publication bias within the article, which revealed
two in Iran,41,45 six in European countries (Belgium,55 United that the current existed body of literature.
Kingdom,44 Turkey,58 Ireland,51 Bosnia and Herzegovina and
Serbia,46 and Italy49), and the rest were conducted in the Prevalence of depression among pregnant and
United States,59 Colombia,53 and Sri Lanka.54 All studies
were cross-sectional studies and critical appraisal of cross-
lactating women
sectional studies conducted revealed that about 95% of stud- A total of 18,335 pregnant and lactating women from 14
ies scored more than 75%, only one scored lowest scored studies were used to compute the pooled prevalence of
62.5%. The majority of the studies were conducted among depression41–43,46–48,52–55,57–59 was 27% (95% CI: 9%, 45%)
pregnant women alone, except studies done in United by using the DerSimonian and Laird random-effects model
States,59 Belgium,55 and Iran45 were conducted in both preg- (I2 = 99.29%, p = 0.001) (Figure 5).
nant and lactating women (Table 1). The possible source of higher heterogeneity among the
included studies was identified using funnel plot and Egger’s
Prevalence of anxiety among pregnant and lactating women. A regression test. Thus, the funnel plot pinpointed that no pub-
total of 16 studies were used to compute the pooled prevalence lication bias, which was confirmed by an objective test
of anxiety among pregnant and lactating women.41–43,45,47,49–57,59 (Egger’s test, p = 0.208; Figure 6).
A total of 16 627 pregnant and lactating women were used to Finally, sensitivity analysis was done to identify the pos-
compute the pooled estimate of anxiety, 33 cases per 100 preg- sible source of heterogeneity among the studies used in the
nant and lactating women (95% CI: 50%–61%) using the Der- pooled estimates. The figure showed that all studies contrib-
Simonian and Laird random-effects model (I2 = 99.68%, uted to the higher heterogeneity in the pooled prevalence of
p = 0.001; Figure 2). depression among pregnant and lactating women (Figure 7).
Publication bias was checked using funnel plot, and it was The funnel plots report a rightward shift with very few
objectively verified using Egger’s regression test. Hence, the points falling within the predicted 95%CI for depression
Demissie and Bitew 5

Table 1. Detailed description of the included studies for computing the prevalence of stress, anxiety, depression, and social
dysfunctions among pregnant women and lactating women 2020.

Author Study Study area Sample Diagnostic Anxiety Depression Stress Insomnia Social Quality
pop size method dysfunction
Parra-Saavedra et al.53 PW Colombia 946 SAQ 50.4 25 49.1 6/8 (75%)
Patabendige et al.54 PW Sri Lanka 257 HADS 17.5 19.5 28.4 6/8 (75%)
Ceulemans et al.55 PW Belgium 2421 EDS & GAD-7 42.6 25.3 5/8 (62.5%)
Ceulemans et al.55 LW Belgium 3445 EDS & GAD-7 42.4 23.6 8/8 (100%)
Liu et al.50 PW China 1947 SAS 17.15 8/8 (100%)
Lebel et al.57 PW Canada 1987 EDPS 59 37 6/8 (75%)
Gharagozloo et al.45 LW & Iran 403 CDAS 5.7 5.5 8/8 (100%)
PW
Durankuş and Aksu58 PW Turkey 260 EPDS 35.4 6/8 (75%)
Dib et al.44 PW The United 1329 SAQ 71 18 6/8 (75%)
Kingdom
Tutnjević and Lakić46 PW Bosnia and 152 SAS 44.1 38.2 6/8 (75%)
Herzegovina
and Serbia
Alijanpour et al.41 PW Iran 261 HADS and CDAS 10.29 11.3 8/8 (100%)
Milne et al.51 PW Ireland 70 SAQ 14 44 8/8 (100%)
Wu et al.48 PW China 4124 EPDS 34.2 8/8 (100%)
Farewell et al.59 PW & The United 27 PHQ-2 & GAD-7 60 12 88 8/8 (100%)
LW States
Berthelot et al.47 PW Canada 1754 EPDS 10.9 6 6/8 (75%)
Saccone et al.49 PW Italy 100 STAI 68 7/8 (87.5%)
Davenport et al.52 PW Canada 900 EPDS & STAI 72 40.7 8/8 (100%)
Yue et al.56 PW China 308 SAS 14.3 6/8 (75%)
Cameron et al.42 LW Canada 641 CESD & AMHDC. 36.27 33.16 6/8 (75%)
Chen et al.43 PW China 1160 SAS & SDS 10.34 28.62 6/8 (75%)

HADS: Hospital Anxiety and Depression Scale; GAD: generalized anxiety disorder; SAS: Self-Rating Anxiety Scale; EPDS: Edinburgh Postnatal Depression
Scale; PHQ: Patient Health Questionnaire; SDS: Self-Rating Depression Scale.
Keynotes.
BDI-I = Beck Depression Inventory—I.
CES-D = Center for Epidemiological Studies—depression.
DASS-21 = Depression, Anxiety, and Stress scales.
EPDS = Edinburgh Postnatal Depression Scale.
GAD-7 = Generalized Anxiety Disorder 7-items.
HADS = Hospital Anxiety and Depression Scale.
PHQ-9 = Patient Health Questionnaire.
PSS = Perceived Stress Scale.
SAS = Self-Rating Anxiety Scale.
SDS = Self-Rating Depression Scale.

(Figure 6). But an objective test (Egger’s test, p = 0.208) for (I2 = 98.8%, p = 0.0001). Similarly, the pooled prevalence of
depression confirmed that no publication bias. Finally, the insomnia was computed from two eligible articles with a
funnel plots look asymmetric for depression (see Figure 2) total of 2275 study subjects.44,53 Thus, 34% (95% CI: 3.05%–
which is congruency with Rosenthal’s method of finding of 64.0%) of pregnant and lactating women were found to have
fail-safe-N analysis warrants acknowledgment of possible insomnia in the DerSimonian and Laird random-effects
publication bias within the article, which revealed that the model (I2 = 99.6%, p = 0.0001). Besides, the pooled preva-
current existed body of literature. lence of social dysfunction among pregnant and lactating
women was estimated from two eligible studies with a total
Prevalence of stress, insomnia, and social dysfunction among sample of 473.45,51 The pooled prevalence of social dysfunc-
pregnant and lactating women. Out of four eligible studies, a tion was found to be 24.3% (95% CI: 13.41%–62.03%)
total of 1765 pregnant and lactating women were used to using the DerSimonian and Laird random-effects model
compute the pooled estimate of stress.44,46,54,59 The pooled (I2 = 97.5%, p = 0.0001).
prevalence of stress was 56% (95% CI: 30.07%–82.2%) Subgroup analysis based on country and diagnostic meth-
using the DerSimonian and Laird random-effects model ods was not found plausible result due to scant reports in
6 SAGE Open Medicine

Figure 2. Forest plot showing the pooled prevalence of anxiety per 100 pregnant and lactating Women.

50%–61%) with high heterogeneity between studies

(I2 = 99.68%, p = 0.001). Sensitivity analysis was also per-
formed to identify the possible source of heterogeneity
among the included studies. However, there was no single
study having a remarkable effect on the heterogeneity of the
pooled estimate. This suggested that the source of this high
heterogeneity in the prevalence rates of anxiety among the
studies included in this meta-analysis was maybe the scale
used for its diagnoses, with the highest prevalence rates in
studies used the Anxiety GAD-7 scale and HADS and the
lowest in those using the PHQ-9 and Self-Rating Anxiety
Scale. This may be due to the use of self-reported online sur-
vey data may imply the presence of social desirability bias.
This pooled prevalence is higher than a study done in China’s
Figure 3. Funnel plot showing the distribution of included Wuhan region reported that 29% anxiety symptoms.5 This
studies for prevalence of anxiety among pregnant and lactating difference may be due to present pooled prevalence and
women. pregnant women during the COVID-19 pandemic may be
particularly affected, prolonged, increased prenatal anxiety
each diagnostic method. The performed subgroup analysis symptoms increase the risk of prenatal infection and illness
based on country and diagnostic methods with null results rates.7,8
for anxiety and depression. The current meta-analysis of 14 studies revealed that the
pooled prevalence of depression among pregnant and lactat-
ing women during the COVID-19 outbreak is 27% (95% CI:
Discussion 9%–45%) with heterogeneity between studies (I2 = 99.29%,
In this meta-analysis, 16 studies were used to compute the p = 0.001), which is higher than the pooled prevalence in the
pooled prevalence of anxiety among pregnant and lactating general population 25% (95% CI: 18%–33%).59 This differ-
women during the COVID-19 outbreak is 33% (95% CI: ence may be due to study population difference, this study
Demissie and Bitew 7

Figure 4. Sensitivity analysis of studies used to compute pooled prevalence of anxiety.

Figure 5. Forest plot showing the pooled prevalence of depression among pregnant and lactating women in the era of COVID-19.
8 SAGE Open Medicine

stress and has been found to minimize the effects of prenatal

anxiety and depressive symptoms on maternal and child
stress response systems.23 Similarly, the pooled prevalence
of insomnia was computed from two eligible articles with a
total of 2275 study subjects.44,53 Thus, 34% (95% CI: 3.05%–
64.0%) of pregnant and lactating women were found to have
insomnia in the DerSimonian and Laird random-effects
model (I2 = 99.6%, p = 0.0001). Besides, the pooled preva-
lence of social dysfunction among pregnant and lactating
women was estimated from two eligible studies with a total
sample of 473.45,51 The pooled prevalence of social dysfunc-
tion was found to be 24.3% (95% CI: 13.41%–62.03%) by
using the DerSimonian and Laird random-effects model
(I2 = 97.5%, p = 0.0001). Despite the likely negative psycho-
Figure 6. Funnel plot showing the distribution of included logical sequelae of psychological, health, and financial insta-
studies for prevalence of depression among pregnant and
lactating women.
bility combined with social alienation, physical exercise is
often correlated with decreased depressive and anxiety
symptoms in pregnant individuals24 Pregnancy is an espe-
among pregnant and lactating women and pregnant women cially sensitive period when psychiatric depression may
in particular, prolonged, intensified prenatal depression have detrimental effects on both mother and child.
symptoms may raise the risk of postpartum depression, as Considering recent meta-analysis conducted in general pop-
well as prenatal infection and disease rates during the ulation revealed that the overall global prevalence of depres-
COVID-19 pandemic.7,8 sive disorders is 25% in the general population22 which is
On the present meta-analysis, the possible source of lower, as compared to 27% among pregnant women and lac-
higher heterogeneity among the included studies was identi- tating, in addition, this, the present meta-analysis findings
fied using funnel plot and Egger’s regression test. Thus, the identified higher pooled proportion of anxiety 34% and
funnel plot pinpointed that no publication bias, which was stress 56% during the COVID-19 outbreak. This implies a
confirmed by an objective test (Egger’s test, p = 0.208). substantial impact of the current pandemic situation on men-
Finally, sensitivity analysis was done to identify the possible tal health among pregnant and lactating women that should
source of heterogeneity among the studies used in the pooled be targeted during antenatal care and postpartum care-level
estimates. The figure showed that all studies contributed to strategies.
the higher heterogeneity in the pooled prevalence of depres- This is the first systematic review and meta-analysis
sion among pregnant and lactating women. This suggested undertaken of all available studies of anxiety, depression,
that the source of this high heterogeneity in the prevalence stress, and social dysfunctions among pregnant and lactating
rates of depression among the studies included in this meta- women during the COVID-19 outbreak. Meta-analysis
analysis was maybe the scales used for its diagnoses, with revealed that greater power than individual studies to esti-
the highest prevalence rates in studies used the EPDS scale mate more accurate rates of anxiety, depression, stress, and
and HADS and the lowest in those using the CES-D (Center social dysfunctions, by considering a much larger population
for Epidemiological Studies—depression and Self-Rating drawn from different countries. However, limitations should
Depression Scale. This may be due to the use of self-reported be considered when interpreting these results. For example,
online survey data may suggest the presence of social desir- randomization of the sample was not available in certain
ability bias. situations, and data had to be gathered by Internet surveys,
Out of four eligible studies, a total of 1765 pregnant and which may have led to selection biases, such as oversam-
lactating women were used to compute the pooled estimate pling of younger and more qualified individuals.60 Ultimately,
of stress.44,46,54,59 The pooled prevalence of stress was 56% the latest meta-analysis relies on studies that involve preg-
(95% CI: 30.07%–82.2%) with heterogeneity between stud- nant and lactating women only. The effect of COVID-19 on
ies (I2 = 98.8%, p = 0.0001). The pooled prevalence of stress the psychological well-being of disadvantaged populations,
among pregnant and lactating women is the highest of all such as health staff, outpatients, or elderly persons, is pro-
psychological impact of COVID 19 among these popula- jected to be high. Future epidemiological trials performed in
tions as identified in this meta-analysis which the pooled these subpopulations, as well as systemic analyses of evi-
prevalence of anxiety 33%, depression 27%, and social dys- dence pooling, are also especially required to adjust public
function was 24.3% (95% among pregnant and lactating health strategies.
women). It is also important to search for possible resistance Limitation of the study Letters to editors, short communi-
factors that can help to defend against high prenatal stress. cations, and preprints were not considered in this study
Social reinforcement can alleviate the impact of prenatal which may be the limitation of this study. Another weakness
Demissie and Bitew 9

Figure 7. Sensitivity analysis of studies used to compute pooled prevalence of depression.

may be revealed that a small fail-safe N suggests that the agendas to enhance the well-being of pregnant and lactating
conclusion of the meta-analysis may be susceptible to publi- women. Besides, giving psychological support to pregnant
cation bias. Lack of reporting metrics that are known to and lactating women may reduce the long-term negative
influence anxiety and depression symptoms is a limitation to effects of this pandemic.
the study.
Acknowledgements
Conclusion The authors would like to address their gratitude to the authors of the
included studies for this systematic review and meta-analysis and St.
The mental health effects of COVID-19 among pregnant and Paul’s Hospital Millennium Medical College.
lactating women were determined in terms of anxiety,
depression, stress, social dysfunction, and insomnia. This Author contributions
systematic review and meta-analysis determined the overall
global pooled prevalence of anxiety 33%, depression 27%, D.B.D. and Z.W.B. conceived and designed the review. Z.W.B.
prepared the draft of the manuscript. The final version of the manu-
stress 56%, insomnia 34%, and social dysfunction 24% dur-
script was approved by all the authors.
ing the COVID-19 pandemic outbreak in the globe among
pregnant and lactating women. This implies a substantial
Declaration of conflicting interests
impact of the current pandemic situation on mental health
among pregnant and lactating women that should be targeted The author(s) declared no potential conflicts of interest with respect
during antenatal care and postpartum care-level strategies. to the research, authorship, and/or publication of this article.
This embryonic condition involves collaborative efforts by
the research community to add to the monitoring of pregnant Funding
and lactating women during the COVID-19 epidemic and to The author(s) received no financial support for the research, author-
examine the short- and long-term detrimental effects on the ship, and/or publication of this article.
mental health well-being of mothers and infants. Therefore,
policymakers and health planners should give great empha- Ethical approval
sis to addressing maternal mental well-being during and after Ethical approval for this study was obtained from INSTITUTIONAL
this global health crisis. Maternal mental health must be one REVIEW BOARD (APPROVAL NUMBER/ID) (SPHMMC
of the international and national public health priority 202/21/2019) from the ethical review board of St. Paul’s Hospital
10 SAGE Open Medicine

Millennium Medical College (Or) Ethical approval for this study prematurity and intrauterine growth retardation. Eur J Clin
was waived by INSTITUTIONAL REVIEW BOARD* because Nutr 2003; 57(2): 266–272.
*REASON FOR WAIVER*. For printing and secretarial services 13. Stein A, Pearson RM, Goodman SH, et al. Effects of perina-
only. tal mental disorders on the fetus and child. The Lancet 2014;
384(9956): 1800–1819.
Informed consent 14. Glover V. Maternal depression, anxiety and stress during
pregnancy and child outcome; what needs to be done. Best
Written informed consent was obtained from all subjects before the
Pract Res Clin Obstet Gynaecol 2014; 28(1): 25–35.
study. Not applicable.
15. MacKinnon N, Kingsbury M, Mahedy L, et al. The association
Informed consent was not sought for this study because this
between prenatal stress and externalizing symptoms in child-
study was a systematic review and meta-analysis.
hood: evidence from the Avon Longitudinal Study of Parents
and Children. Biol Psychiatry 2018; 83(2): 100–108.
ORCID iD 16. Van den Bergh BR, Van den Heuvel MI, Lahti M, et al.
Dereje Bayissa Demissie https://orcid.org/0000-0003-1006-4318 Prenatal developmental origins of behavior and mental health:
the influence of maternal stress in pregnancy. Neurosci
Supplemental material Biobehav Rev 2020; 117: 26–64.
17. Van den Bergh BR, Dahnke R and Mennes M. Prenatal stress
Supplemental material for this article is available online. and the developing brain: risks for neurodevelopmental disor-
ders. Dev Psychopathol 2018; 30(3): 743–762.
References 18. Adamson B, Letourneau N and Lebel C. Prenatal maternal
1. World Health Organization (WHO). Clinical management anxiety and children’s brain structure and function: a system-
of severe acute respiratory infection when novel coronavi- atic review of neuroimaging studies. J Affect Disord 2018;
rus (2019-nCoV) infection is suspected: interim guidance. 241: 117–126.
Geneva: WHO, 2020, p. 21. 19. Lebel C, Walton M, Letourneau N, et al. Prepartum and post-
2. Cullen W, Gulati G and Kelly BD. Mental health in the partum maternal depressive symptoms are related to children’s
COVID-19 pandemic. QJM 2020; 113(5): 311–312. brain structure in preschool. Biol Psychiatry 2016; 80(11):
3. Da Silva AG, Miranda DM, Diaz AP, et al. Mental health: why 859–868.
it still matters in the midst of a pandemic. Braz J Psychiatry 20. Sandman CA, Buss C, Head K, et al. Fetal exposure to mater-
2020; 42: 229–231. nal depressive symptoms is associated with cortical thickness
4. Hawryluck L, Gold WL, Robinson S, et al. SARS control and in late childhood. Biol Psychiatry 2015; 77(4): 324–334.
psychological effects of quarantine, Toronto, Canada. Emerg 21. Qiu A, Rifkin-Graboi A, Chen H, et al. Maternal anxiety and
Infect Dis 2004; 10(7): 1206–1212. infants’ hippocampal development: timing matters. Transl
5. Wang C, Pan R, Wan X, et al. Immediate psychological Psychiatry 2013; 3(9): e306.
responses and associated factors during the initial stage of the 22. Bueno-Notivol J, Gracia-García P, Olaya B, et al. Prevalence
2019 coronavirus disease (COVID-19) epidemic among the of depression during the COVID-19 outbreak: a meta-analysis
general population in China. Int J Environ Res Public Health of community-based studies. Int J Clin Health Psychol 2021;
2020; 17(5): 1729. 21: 100196.
6. Al-Rabiaah A, Temsah M-H, Al-Eyadhy AA, et al. Middle 23. Thomas JC, Letourneau N, Campbell TS, et al. Social buffering
East Respiratory Syndrome-Corona Virus (MERS-CoV) asso- of the maternal and infant HPA axes: mediation and modera-
ciated stress among medical students at a university teaching tion in the intergenerational transmission of adverse childhood
hospital in Saudi Arabia. J Infect Public Health 2020; 13: experiences. Dev Psychopathol 2018; 30(3): 921–939.
687–691. 24. Demissie Z, Siega-Riz AM, Evenson KR, et al. Physical activ-
7. Bayrampour H, Tomfohr L and Tough S. Trajectories of ity and depressive symptoms among pregnant women: the
perinatal depressive and anxiety symptoms in a community PIN3 study. Arch Womens Ment Health 2011; 14(2): 145–157.
cohort. J Clin Psychiatry 2016; 77(11): e1467–e1473. 25. Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting
8. Coussons-Read ME. Effects of prenatal stress on pregnancy items for systematic reviews and meta-analyses: the PRISMA
and human development: mechanisms and pathways. Obstet statement. Ann Intern Med 2009; 151(4): 264–269.
Med 2013; 6(2): 52–57. 26. Munn Z, Moola S, Lisy K, et al. Methodological guidance for
9. Accortt EE, Cheadle AC and Schetter CD. Prenatal depression systematic reviews of observational epidemiological studies
and adverse birth outcomes: an updated systematic review. reporting prevalence and cumulative incidence data. Int J Evid
Matern Child Health J 2015; 19(6): 1306–1337. Based Healthc 2015; 13(3): 147–153.
10. Grigoriadis S, Graves L, Peer M, et al. Maternal anxiety dur- 27. Belur J, Tompson L, Thornton A, et al. Interrater reliability in
ing pregnancy and the association with adverse perinatal out- systematic review methodology: exploring variation in coder
comes: systematic review and meta-analysis. J Clin Psychiatry decision-making. Sociol Method Res 2021; 50: 837–865.
2018; 79(5): 17r12011. 28. Sterne JA and Egger M. Funnel plots for detecting bias in
11. Qu F, Wu Y, Zhu Y-H, et al. The association between psycho- meta-analysis: guidelines on choice of axis. J Clin Epidemiol
logical stress and miscarriage: a systematic review and meta- 2001; 54(10): 1046–1055.
analysis. Sci Rep 2017; 7(1): 1731. 29. Rücker G, Schwarzer G, Carpenter JR, et al. Undue reliance
12. Rondó PHC, Ferreira RF, Nogueira F, et al. Maternal psycho- on I2 in assessing heterogeneity may mislead. BMC Med Res
logical stress and distress as predictors of low birth weight, Methodol 2008; 8(1): 79.
Demissie and Bitew 11

30. Higgins JP and Thompson SG. Quantifying heterogeneity in a 47. Berthelot N, Lemieux R, Garon-Bissonnette J, et al. Uptrend in
meta-analysis. Stat Med 2002; 21(11): 1539–1558. distress and psychiatric symptomatology in pregnant women
31. Kotabagi P, Fortune L, Essien S, et al. Anxiety and depression during the coronavirus disease 2019 pandemic. Acta Obstet
levels among pregnant women with COVID-19. Acta Obstet Gynecol Scand 2020; 99: 848–855.
Gynecol Scand 2020; 99: 953–954. 48. Wu Y, Zhang C, Liu H, et al. Perinatal depressive and anxiety
32. Ali NA and Feroz AS. Maternal mental health amidst the symptoms of pregnant women during the coronavirus disease
COVID-19 pandemic. Asian J Psychiatr 2020; 54: 102261. 2019 outbreak in China. Am J Obstet Gynecol 2020; 223(2):
33. Diamond RM, Brown KS and Miranda J. Impact of COVID- 240.e1–240.e9.
19 on the perinatal period through a biopsychosocial systemic 49. Saccone G, Florio A, Aiello F, et al. Psychological impact of
framework. Contemp Fam Ther. Epub ahead of print 20 July coronavirus disease 2019 in pregnant women. Am J Obstet
2020. DOI: 10.1007/s10591-020-09544-8. Gynecol 2020; 223: 293–295.
34. Corbett GA, Milne SJ, Hehir MP, et al. Health anxiety and 50. Liu X, Chen M, Wang Y, et al. Prenatal anxiety and obstetric
behavioural changes of pregnant women during the COVID-19 decisions among pregnant women in Wuhan and Chongqing
pandemic. Eur J Obstet Gynecol Reprod Biol 2020; 249: 96–97. during the COVID-19 outbreak: a cross-sectional study.
35. Zanardo V, Manghina V, Giliberti L, et al. Psychological BJOG 2020; 127(10): 1229–1240.
impact of COVID-19 quarantine measures in northeastern 51. Milne SJ, Corbett GA, Hehir MP, et al. Effects of isolation on
Italy on mothers in the immediate postpartum period. Int J mood and relationships in pregnant women during the covid-
Gynaecol Obstet 2020; 150(2): 184–188. 19 pandemic. Eur J Obstet Gynecol Reprod Biol 2020; 252:
36. Zhang Y and Ma ZF. Psychological responses and lifestyle 610–611.
changes among pregnant women with respect to the early stages 52. Davenport MH, Meyer S, Meah VL, et al. Moms are not ok:
of COVID-19 pandemic. Int J Soc Psychiatry 2021; 67: 344–350. COVID-19 and maternal mental health. Front Glob Women’s
37. Suzuki S. Psychological status of postpartum women Health. Epub ahead of print 19 June 2020. DOI: 10.3389/
under the COVID-19 pandemic in Japan. J Matern Fetal fgwh.2020.00001.
Neonatal Med. Epub ahead of print 18 May 2020. DOI: 53. Parra-Saavedra M, Villa-Villa I, Pérez-Olivo J, et al. Attitudes
10.1080/14767058.2020.1763949. and collateral psychological effects of COVID-19 in pregnant
38. Taubman-Ben-Ari O, Chasson M, Abu Sharkia S, et al. women in Colombia. Int J Gynaecol Obstet 2020; 151: 203–
Distress and anxiety associated with COVID-19 among 208.
Jewish and Arab pregnant women in Israel. J Reprod Infant 54. Patabendige M, Gamage MM, Weerasinghe M, et al.
Psychol 2020; 38(3): 340–348. Psychological impact of the COVID-19 pandemic among preg-
39. Franchi M, Del Piccolo L, Bosco M, et al. Covid-19 and men- nant women in Sri Lanka. Int J Gynaecol Obstet 2020; 151:
tal health in the obstetric population: a lesson from a case of 150–153.
puerperal psychosis. Minerva Ginecol 2020; 72: 355–357. 55. Ceulemans M, Hompes T and Foulon V. Mental health status
40. Chandra PS, Shiva L, Nagendrappa S, et al. COVID 19 related of pregnant and breastfeeding women during the COVID-19
Psychosis as an interface of fears, socio-cultural issues and pandemic: a call for action. Int J Gynaecol Obstet 2020; 151:
vulnerability- case report of two women from India. Psychiatry 146–147.
Res 2020; 290: 113136. 56. Yue C, Liu C, Wang J, et al. Association between social sup-
41. Alijanpour M, Sadatmahalleh SJ, Samaneh Y, et al. Influence port and anxiety among pregnant women in the third trimester
psycho-sexual factors on the quality of life in pregnant women during the coronavirus disease 2019 (COVID-19) epidemic in
during the COVID-19 pandemic: a path analysis, 2020, https:// Qingdao, China: the mediating effect of risk perception. Int J
www.researchsquare.com/article/rs-45196/v2 Soc Psychiatry 2021; 67: 120–127.
42. Cameron EE, Joyce KM, Delaquis CP, et al. Maternal psy- 57. Lebel C, MacKinnon A, Bagshawe M, et al. Elevated
chological distress & mental health service use during the depression and anxiety symptoms among pregnant individu-
COVID-19 pandemic. J Affect Disord 2020; 276: 765–774. als during the COVID-19 pandemic. J Affect Disord 2020;
43. Chen S, Zhuang J, Chen Q, et al. Psychological investigation 277: 5–13.
on pregnant women during the outbreak of COVID-19, 2020, 58. Durankuş F and Aksu E. Effects of the COVID-19 pan-
https://www.researchsquare.com/article/rs-28455/v1 demic on anxiety and depressive symptoms in preg-
44. Dib S, Rougeaux E, Vázquez-Vázquez A, et al. Maternal nant women: a preliminary study. J Matern Fetal
mental health and coping during the COVID-19 lockdown Neonatal Med. Epub ahead of print 18 May 2020. DOI:
in the UK: data from the COVID-19 New Mum Study. Int J 10.1080/14767058.2020.1763946.
Gynaecol Obstet 2020; 151: 407–414. 59. Farewell CV, Jewell J, Walls J, et al. A mixed-methods pilot
45. Gharagozloo M, Sadatmahalleh SJ, Khomami MB, et al. study of perinatal risk and resilience during COVID-19. J
Mental health and marital satisfaction changes of pregnant Prim Care Community Health. Epub ahead of print 16 July
and lactating women during the COVID-19 pandemic, 2020, 2020. DOI: 10.1177/2150132720944074.
https://www.researchsquare.com/article/rs-49590/v1 60. Wang Y, Di Y, Ye J, et al. Study on the public psychological
46. Tutnjević S and Lakić S. Psychological impact of the COVID- states and its related factors during the outbreak of coronavirus
19 pandemic on pregnant women in Bosnia and Herzegovina disease 2019 (COVID-19) in some regions of China. Psychol
and Serbia, 2020, https://psyarxiv.com/su3nv/ Health Med 2021; 26: 13–22.