CitaonMichelen, M., Cheng, V., Manoharan, L., Elkheir, N., Dagens, D., Hastie, C.,
O’Hara, M., Suett, J. C., Dahmash, D. T., Bugaeva, P., Rigby, I., Munblit, D., Harriss, E.,
Burls, A. ORCID: 0000-0001-9540-622X, Foote, C., Scott, J., Carson, G., Olliaro, P., Sigfrid,
L. and Stavropoulou, C. ORCID: 0000-0003-4307-1848 (2021). Characterising long term
Covid-19: a living systematic review. BMJ Global Health, 6, e005427. doi: 10.1136/bmjgh-
2021-005427
Permanent repository link: https://openaccess.city.ac.uk/id/eprint/26712/
Link to published versionhttp://dx.doi.org/10.1136/bmjgh-2021-005427
Copyright: !"
#$%$
#&'($
!"
Reuse:
)) #$
*$
"($$$
$#
!
1
MichelenM, etal. BMJ Global Health 2021;6:e005427. doi:10.1136/bmjgh-2021-005427
Characterising long COVID: a living
systematic review
Melina Michelen ,
1,2
Lakshmi Manoharan,
2
Natalie Elkheir ,
3
Vincent Cheng
,
4
Andrew Dagens,
2
Claire Hastie ,
5
Margaret O'Hara,
5
Jake Suett
,
6
Dania Dahmash,
2
Polina Bugaeva,
7
Ishmeala Rigby,
2
Daniel Munblit,
8,9,10
Eli Harriss,
11
Amanda Burls,
1
Carole Foote,
12
Janet Scott,
13
Gail Carson
,
2
Piero Olliaro,
2
Louise Sigfrid,
2
Charitini Stavropoulou
1
Original research
To cite: MichelenM,
ManoharanL, ElkheirN,
etal. Characterising long
COVID: a living systematic
review. BMJ Global Health
2021;6:e005427. doi:10.1136/
bmjgh-2021-005427
Handling editor Seye Abimbola
LM, NE, VC and AD contributed
equally.
LS and CS are joint senior
authors.
Received 17 February 2021
Accepted 19 August 2021
For numbered afliations see
end of article.
Correspondence to
Dr Charitini Stavropoulou;
C. Stavropoulou@ city. ac. uk
© Author(s) (or their
employer(s)) 2021. Re- use
permitted under CC BY.
Published by BMJ.
ABSTRACT
Background While it is now apparent clinical sequelae
(long COVID) may persist after acute COVID-19, their
nature, frequency and aetiology are poorly characterised.
This study aims to regularly synthesise evidence on long
COVID characteristics, to help inform clinical management,
rehabilitation strategies and interventional studies to
improve long- term outcomes.
Methods A living systematic review. Medline, CINAHL
(EBSCO), Global Health (Ovid), WHO Global Research on
COVID- 19 database, LitCovid and Google Scholar were
searched till 17 March 2021. Studies including at least 100
people with conrmed or clinically suspected COVID-19 at
12 weeks or more post onset were included. Risk of bias
was assessed using the tool produced by Hoy et al. Results
were analysed using descriptive statistics and meta-
analyses to estimate prevalence.
Results A total of 39 studies were included: 32 cohort, 6
cross- sectional and 1 case–control. Most showed high or
moderate risk of bias. None were set in low- income countries
and few included children. Studies reported on 10 951 people
(48% female) in 12 countries. Most included previously
hospitalised people (78%, 8520/10 951). The longest mean
follow- up time was 221.7 (SD: 10.9) days post COVID-19
onset. Over 60 physical and psychological signs and symptoms
with wide prevalence were reported, most commonly
weakness (41%; 95% CI 25% to 59%), general malaise (33%;
95% CI 15% to 57%), fatigue (31%; 95% CI 24% to 39%),
concentration impairment (26%; 95% CI 21% to 32%) and
breathlessness (25%; 95% CI 18% to 34%). 37% (95% CI
18% to 60%) of patients reported reduced quality of life; 26%
(10/39) of studies presented evidence of reduced pulmonary
function.
Conclusion Long COVID is a complex condition with
prolonged heterogeneous symptoms. The nature of studies
precludes a precise case denition or risk evaluation. There
is an urgent need for prospective, robust, standardised,
controlled studies into aetiology, risk factors and
biomarkers to characterise long COVID in different at- risk
populations and settings.
PROSPERO registration number CRD42020211131.
INTRODUCTION
SARS- CoV- 2 first emerged in December 2019
causing a widespread pandemic. Most people
experience asymptomatic or mild- to- moderate
acute COVID- 19 symptoms, while around
15% of people are estimated to progress to
more severe disease requiring hospitalisation
and approximately 5% become critically ill.
1
While the acute phase of the disease was
characterised early, there are still limited
data on long- term outcomes.
2
Symptoms
of long- lasting COVID- 19 sequelae and
complications, termed long COVID by
people living with long COVID,
3
have been
Key questions
What is already known?
► A signicant number of people continue to describe
ongoing symptoms long after the acute phase of
COVID- 19, often referred to as long COVID.
► Long COVID is a heterogeneous condition with an
uncertain prevalence, for which there is currently no
precise case denition.
What are the new ndings?
► The breadth of reported symptoms suggests a com-
plex, heterogeneous condition affecting both those
who were hospitalised and those managed in the
community.
► Our review identies weakness (41%; 95% CI 25%
to 59%), general malaise (33%; 95% CI 15% to
57%), fatigue (31%; 95% CI 24% to 39%), concen-
tration impairment (26%; 95% CI 21% to 32%) and
breathlessness (25%; 95% CI 18% to 34%) as the
most common symptoms reported.
What do the new ndings imply?
► The current evidence base of the clinical spectrum of
long COVID is limited, based on heterogenous data,
and vulnerable to biases, hence caution should be used
when interpreting or generalising the results.
► Our review identies areas where further long COVID
research is critically needed to help characterise
long COVID in different populations and dene its
aetiology, risk factors and biomarkers, as well as the
impact on variants of concern and vaccination on
long- term outcomes.
on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from
2
MichelenM, etal. BMJ Global Health 2021;6:e005427. doi:10.1136/bmjgh-2021-005427
BMJ Global Health
reported worldwide. Yet the underlying aetiology behind
prolonged or fluctuating symptomatology is limited and
there is no widely accepted uniformed case definition.
4
Instead, long COVID has been defined pragmatically as
‘not recovering for several weeks or months following the
start of symptoms’.
4
Others have distinguished between
postacute COVID- 19, referring to symptoms beyond
3 weeks, and chronic COVID- 19, referring to symptoms
beyond 12 weeks,
5
while the National Institute for Health
and Care Excellence distinguishes between ongoing
symptomatic COVID- 19 lasting from 4 to 12 weeks and
post COVID-19 syndrome continuing for over 12 weeks.
6
The number of people living with long COVID is
unknown. Attempts to quantify the prevalence of long
COVID use different methods, including national surveys
and patient- led studies, making it difficult to compare
across studies. The UK’s Office for National Statistics has
estimated that on average 1 in 5 people have symptoms
beyond 5 weeks, while 1 in 10 have symptoms persisting
over 12 weeks.
7
A patient- led survey found that in survival
analysis, the chance of full recovery by day 50 was smaller
than 20%
8
and a COVID- 19 symptom app study found
that 13.3% (558/4182) patients had symptoms lasting 28
days or more, 4.5% (189/4182) patients had symptoms
for 8 or more weeks and 2.3% (95/4182) patients had
symptoms lasting over 12 weeks.
9
The symptoms of long COVID are equally ill- defined,
with patients describing it as a fluctuating illness of dispa-
rate symptoms.
8 10
Indeed, the National Institute for
Health Research has suggested that postacute COVID- 19
may consist of several distinct clinical syndromes
including: a postintensive care syndrome, chronic fatigue
syndrome, long- term COVID- 19 syndrome and disease
from SARS- CoV- 2 inflicted organ damage.
11
Addition-
ally, even with an expanding knowledge of risk factors in
the acute phase, little is currently known on predictive
factors for developing long COVID.
9
Despite suggested
classifications, there is yet no clear consensus.
Our early understanding of long COVID has been
accumulated from case reports and cross- sectional
online survey studies as the pandemic global research
focus has largely been on studies of hospitalised patients
during the acute phase. As the pandemic progresses,
emerging studies have followed up patients to present
the fluctuating multiorgan sequelae of acute COVID- 19,
yet evidence is still scarce. There continues to be a call
to further understand and acknowledge this condition
by incorporating patient knowledge and experiences,
together with standardised studies, exploring underlying
aetiologies behind different syndromes.
12 13
Given the enormous number of people worldwide who
have suffered from COVID- 19, it is essential to establish
a precise categorisation of long COVID. Such categori-
sation will not only help people better understand their
symptoms but also direct research into prevention, treat-
ment and support, ultimately allowing us to understand
and prepare to respond to the long- term consequences
inflicted by the COVID- 19 pandemic. Our review seeks
to synthesise and continually update the evidence on the
character and prevalence of long COVID.
METHODS
Systematic reviews conducted early during the COVID- 19
pandemic soon became redundant due to the rapidity
with which new research was released. In recognition of
this, many reviewers have moved towards the concept of
a ‘living systematic review’ (LSR), which compared with
traditional systematic reviews has in- built mechanisms
for regular update and renewal.
14 15
We conducted a
‘living’ systematic review to provide frequently updated
evidence on the symptoms and complications of long
COVID. This review was developed in collaboration with
infectious disease clinicians, public health professionals,
information specialists, review methodologists with expe-
rience in clinical epidemic research and members of
the global Long COVID Support Group, which includes
people living with long COVID. This is the first version
of this LSR, which will be updated approximately every 6
months as new evidence emerges, using the established
protocol and review platform. The updates will be led by
the International Severe Acute Respiratory and emerging
Infection Consortium (ISARIC) systematic review team
in collaboration with members of Long COVID Support.
Previous versions will be archived in online supplemental
materials. The findings will be disseminated via BMJ Global
Health and on a dedicated webpage with infographics and
a brief summary for lay people and professionals.
Protocol registration
This report was structured according to the Preferred
Reporting Items for Systematic Reviews and Meta-
Analyses statement guidelines.
16
The protocol was regis-
tered with PROSPERO and published in a peer- reviewed
journal.
17
Search strategy
The following databases were searched: Medline and
CINAHL (EBSCO), Global Health (Ovid), WHO Global
Research Database on COVID- 19 and LitCovid from 1
January 2020 to 17 March 2021. Additionally, we searched
Google Scholar on 17 March 2021, screening the first
500 titles. A ‘backwards’ snowball search was conducted
of the references of systematic reviews. Full search terms
are included in online supplemental file 1. The search
terms and inclusion criteria have, for this first version,
been designed to cast a wide net and will be modified in
line with new evidence, research priorities and clinical
and policy needs.
Eligibility criteria
Peer- reviewed studies were considered eligible if they
included at least 100 people with laboratory confirmed
and/or clinically diagnosed COVID- 19. Without
a clear, internationally agreed case definition, we
included studies that reported symptoms or outcomes
assessed at 12 or more weeks post COVID- 19 onset.
6
on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from
MichelenM, etal. BMJ Global Health 2021;6:e005427. doi:10.1136/bmjgh-2021-005427
3
BMJ Global Health
There were no language restrictions. Reviews and
opinion pieces were excluded. Studies were excluded
if they included fewer than 100 participants, to avoid
small study effects,
18
or the follow- up was unclear or
less than 12 weeks post onset.
Screening
Screening was performed independently by two system-
atic reviewers. Any disagreements were resolved via
consensus or a third reviewer. Non- English articles
were translated using Google Translate and assessed
by a systematic reviewer with good knowledge of the
language. The data were managed using the review soft-
ware Rayyan.
19
Data extraction
Data extraction was performed using Microsoft Excel. A
data extraction template informed by a previous review
20
was reviewed, updated and piloted before being finalised.
Data extracted included study design, population char-
acteristics, outcomes, prevalence, duration of symptoms
and risk factors. Data extraction was performed by one
systematic reviewer and checked by a second reviewer.
Disagreements were resolved through consensus. To
avoid duplication of data in future updates and ensure
robustness, data extraction was not performed for non-
peer- reviewed preprints.
Risk of bias assessment
The included studies were assessed for risk of bias using
the tool produced by Hoy et al
21
(online supplemental
file 2). This assessment checklist is a validated tool for
assessing risk of bias in prevalence studies. The checklist
has 10 domains for assessing risk of bias, used to calculate
a cumulative overall risk of bias for the whole study.
Data analysis
We undertook individual descriptive analysis for each
study. We presented symptom proportions by different
settings, as presented in the individual studies: hospi-
talised, non- hospitalised or a mix of both populations if
no subset data were available. Symptoms were broadly
grouped into physiological clusters through discussion
with clinicians. Proportion of symptoms and its 95% CIs
were estimated using the exact method.
22
If there were
two or more studies for each symptom, a meta- analysis
was performed using a random intercept logistic regres-
sion model with Hartung- Knapp modification due to the
heterogeneity and skewed sample sizes.
23 24
Heterogeneity
between estimates was assessed using the I
2
statistic.
25
Additional subgroup analysis was conducted to explore
the modification of the following factors on proportion
of symptoms: hospitalisation, settings, continents and
follow- up timing. We also conducted meta- regression
analysis on the percentage of females and intensive care
unit (ICU) patients where there were more than 10 studies
for the symptom. Sensitivity analyses were conducted
to examine the impact of high risk of bias studies and
statistical methods, Freeman- Tukey double arcsine trans-
formation using inverse variance meta- analysis, on the
estimates. Funnel plots were plotted using proportion
of the symptom against the precision and sample sizes
22
where there were more than 10 studies for the symptom
to explore risk of publication bias. All analysis and data
presentation were performed using metaprop
26
and
ggplot2
27
in R (V.4.0.5) via RStudio (V.1.3.1093).
28
The
data are presented using a combination of infographics,
prepared by a design company (Design Science
29
) and
scientific tables to facilitate interpretation by different
stakeholders, including non- specialists.
Figure 1 Map of study distribution.
on October 1, 2021 by guest. Protected by copyright.http://gh.bmj.com/BMJ Glob Health: first published as 10.1136/bmjgh-2021-005427 on 27 September 2021. Downloaded from
