RESEARCH THEME 2: COVID-19

Electronic supplementary material:

The online version of this article contains supplementary material.

© 2023 The Author(s) Cite as: Cazé AB, Cerqueira-Silva T, Bomfim AP, de Souza GL, Azevedo ACA, Brasil MQA,
Santos NR, Cunha RK, Dan J, Bandeira AC, de Goes Cavalcanti LP, Barral-Netto M, Barral AMP,
Barbosa CG, Boaventura VS. Prevalence and risk factors for long COVID after mild disease: A
cohort study with a symptomatic control group. J Glob Health 2023;13:06015.

Prevalence and risk factors for long

COVID after mild disease: A cohort study
with a symptomatic control group
Ana B Cazé1* , Thiago Cerqueira-Silva2* , Background There is limited data on the preva-
Adriele P Bomfim1,2 , Gisley L de Souza3, Amanda lence and risk factors for long COVID and few pro-
CA Azevedo4, Michelle QA Brasil1,2 , Nara R Santos5, spective studies with appropriate control groups
Ricardo Khouri1,2,6 , Jennifer Dan7,8 , Antonio C and adequate sample sizes. We performed a pro-
Bandeira9,10 , Luciano PG Cavalcanti11,12 , Manoel spective study to determine the prevalence and
risk factors for long COVID.
Barral-Netto1,2 , Aldina Barral2,13 , Cynara
G Barbosa1† , Viviane S Boaventura1,2† Methods We recruited individuals aged ≥15 years
who were clinically suspected of having an acute
1
Federal University of Bahia, Salvador, Bahia, Brazil SARS-CoV-2 infection from September 2020 to
2
Gonçalo Moniz Institute, Fiocruz, Salvador, Bahia, Brazil April 2021. We collected nasopharyngeal swabs
Bahia School of Medicine and Public Health, Salvador, Bahia, Brazil
3
three to five days following symptom onset for
Santa Izabel Hospital, Santa Casa da Bahia, Salvador, Bahia, Brazil
4

Department of Health Surveillance/Epidemiological Surveillance,

5 analysing using reverse transcriptase polymerase
Campo Formoso, Bahia Brazil chain reaction (RT-PCR). We also collected clin-
Department of Microbiology, Immunology and Transplantation,
6
ical and sociodemographic characteristics from
Rega Institute for Medical Research, Laboratory of Clinical and both SARS-CoV-2 positive and negative partici-
Epidemiological Virology, Belgium pants using structured questionnaires. We fol-
Center for Infectious Disease and Vaccine Research, la Jolla Institute
7
lowed-up the participants via telephone interview
for Immunology (LJI), San Diego (UCSD), California, USA
Department of Medicine, Division of Infectious Disease and Global
8 to assess early outcomes and persistent symptoms.
Public Health, University of California, San Diego (UCSD), California, For COVID-19 cases, 5D-3L EuroQol question-
USA naire was used to assess the impact of symptoms
Faculty of Technology and Sciences of Salvador, Salvador, Bahia,
9
on quality of life.
Brazil
Central State Laboratory Lacen/BA, Salvador, Bahia, Brazil
10 Results We followed 814 participants (412
Federal University of Ceará, Ceará, Brazil
11
COVID-19 positive and 402 COVID-19 negative
Christus University Center, Fortaleza, Ceará, Brazil
12
persons). Most (n = 741/814) had mild symptoms.
Institute for Research in Immunology, São Paulo, São Paulo, Brazil
13
Both groups had similar sociodemographic and
*Joint first authorship.
clinical characteristics, except for the hospitaliza-
†Joint senior authorship.
tion rate (15.8% in the COVID-19 positive vs 1.5%
in the COVID-19 negative group). One month af-
ter disease onset, 122/412 (29.6%) individuals in
the COVID-19 positive (long COVID) and 24 (6%)
in the COVID-19 negative group reported residu-
al symptoms. In the long COVID group, fatigue,
Correspondence to: olfactory disorder, and myalgia were the most fre-
Viviane Sampaio Boaventura quent symptoms in the acute phase. Compared to
Federal University of Bahia, Salvador, Bahia, Brazil
121 Waldemar Falcão Street, Candeal, Salvador, Bahia, Brazil, recovered individuals, older age and having more
CEP 40296-710 than five symptoms during the acute phase were
Brazil risk factors for long COVID. Quality of life was
viviane.boaventura@fiocruz.br

www.jogh.org • doi: 10.7189/jogh.13.06015 1 2023 • Vol. 13 • 06015

 Cazé et al.
RESEARCH THEME 2: COVID-19

evaluated in 102 out of 122 cases of long COVID, with 57 (55.9%) reporting an impact in at least one
dimension of the European Quality of Life 5 Dimensions 3 Level (EQ-5D-3L) questionnaire.
Conclusions In this prospective study consisting predominantly of individuals with mild disease, the
persistence of symptoms after an acute respiratory illness was associated with a diagnosis of COVID-19.
Polysymptomatic acute disease and older age were risk factors for long COVID.

Long COVID has received less attention from the scientific community than acute COVID-19, representing
less than 2% of COVID-19 publications in PubMed Clinical Queries (as of August 31, 2022). Consequent-
ly, several important aspects remain undefined, such as its prevalence. Long COVID has been reported to
affect between 7.5% to 89% of patients [1-4]. However, most of these studies do not include well-matched
controls, which is essential because long COVID symptoms are non-specific and may be attributed to other
factors such as the physical and mental impact of having COVID-19 infection or hospitalization [5,6]. Pro-
spective studies enrolling symptomatic individuals with and without confirmed COVID-19 are necessary
to better estimate the prevalence of long COVID and its associated risk factors.
Identifying risk factors for developing long COVID is key to mitigating its anticipated impact on the health-
care systems [7]. Some large cohorts using previously hospitalized patients reported risk factors for devel-
oping long COVID [1,8,9]. However, few studies evaluated risk factors in individuals after mild disease [10].
We examined the clinical characteristics and diagnosis of all patients presenting with acute respiratory ill-
ness at three healthcare units in Bahia-Brazil and the prevalence of long COVID in the SARS-CoV-2 posi-
tive group. Additionally, we explored the risk factors associated with this condition and its impact on the
participants’ quality of life.

METHODS
Study population
We recruited individuals aged ≥15 years who were suspected of having a SARS-CoV-2 infection from
healthcare units in three municipalities of Bahia State, Brazil (Irecê, Campo Formoso, and Lauro de Fre-
itas). We excluded individuals who had difficulty reporting symptoms, who were at >20 days post-symp-
tom onset at the time of recruitment, or who previously reported a SARS-CoV-2 infection. The recruitment
period lasted from September 2020 to April 2021. The ancestral strain or gamma variant of SARS-CoV-2
was dominant in Brazil from November 2020 to August 2021 (Figure S1 in the Online Supplementary
Document) [11].

Exposure measurement
We collected nasopharyngeal swab samples from all patients to test for SARS-COV-2 using reverse tran-
scriptase polymerase chain reaction (RT-PCR). We excluded patients who had an RT-PCR ten days follow-
ing symptom onset. We collected the samples using either the BIOMOL OneStep/COVID-19 or the Allplex
SARS-CoV-2 Assay testing kit, with a sensitivity above 90% and specificity above 97% [12,13]. We reported
the results following the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE)
(Online Supplementary Document) [14].

Recruitment strategy and data collection

Consecutive in-person recruitment included the application of a structured survey (Questionnaire 1 and
Figure S2 in the Online Supplementary Document) using the Research Electronic Data Capture (RED-
Cap) software hosted by the Gonçalo Moniz Institute (IGM/FIOCRUZ) in Bahia, Brazil. We collected in-
formation on sociodemographic characteristics (age, gender, skin colour, body mass index, anthropomet-
ric data, municipality of residence), comorbidities (hypertension, diabetes, heart disease, chronic lung
disease, other), date of symptom onset, acute symptoms (fever, myalgia, arthralgia, cough, shortness of
breath, nasal congestion, sneezing, coryza, sore throat, sinus pain, retroocular pain, ear pain, headache,
chest pain, chest discomfort, rash, abdominal pain, loss of appetite, vomiting, diarrhoea (defined as two
or more episodes in the last 24 hours), weakness, fatigue, light-headedness or fainting, loss of smell, and
loss of taste).

2023 • Vol. 13 • 06015 2 www.jogh.org • doi: 10.7189/jogh.13.06015

 Prevalence and risk factors for long COVID

RESEARCH THEME 2: COVID-19

Follow-up and outcome measurement
We followed-up on SARS-CoV-2 positive (COVID-19) and negative (non-COVID-19) patients who reported
COVID-19-like symptoms, recruiting both on the same day prior to nasopharyngeal swab sampling. We
considered the date of symptom onset as day zero. We conducted a second structured questionnaire (Ques-
tionnaire 2 and Figure S3 in the Online Supplementary Document) 30 to 60 days after disease onset via
telephone to document the RT-PCR result, acute symptoms, and clinical outcomes. We classified partici-
pants as non-hospitalised or hospitalised in the intensive care unit (ICU) or non-ICU. We applied Ques-
tionnaire 2 up to 40 days after recruitment (Questionnaire 1) to capture any symptoms in the acute phase.
We conducted a third structured questionnaire 60 to 250 days after disease onset by telephone to evaluate re-
sidual symptoms (Questionnaire 3 and Figure S4 in the Online Supplementary Document). The questions
regarded residual or new symptoms: cough, fatigue, shortness of breath, headache, chest pain, dysphonia,
dysphagia, loss of appetite, loss of smell, loss of taste, myalgia, arthralgia, fever, and other non-listed symp-
toms. We excluded patients reporting suspected or confirmed re-infection after the recruitment. Figure S2
in the Online Supplementary Document shows the flow of an individual after recruitment.

Outcomes of interest
We defined long COVID according to the Centers for Disease Control and Prevention classification [15] and
persistence of symptoms as the presence of at least one residual symptom >1 month post-disease onset. The
list of evaluated residual symptoms (Questionnaire 3) was based on previous studies [16,17].
We applied the European Quality of Life 5 Dimensions 3 Level (EQ-5D-3L) questionnaire via telephone to
assess health dimensions: mobility, self-care, daily activities, pain/discomfort, and anxiety/depression. We
also used the European Quality visual analogue scale (EQ-VAS), with 0 rated as the worst and 100 as the
best imaginable health state [18]. We applied the Portuguese-validated version of the EQ-5D-3L question-
naire [19].

Statistical analysis
We described categorical data using counts and percentages and continuous data (determined using the
Shapiro-Wilk test) using medians and interquartile ranges (IQRs). We constructed the correlation matrix
based on the frequency of reported residual symptoms. We performed multivariable logistic regression to
explore risk factors previously associated with long COVID in the literature: age, sex, and number of symp-
toms in the acute phase [10,20-22]. We also performed a sensitivity analysis, including on the comorbidi-
ties. The estimated 95% Wald confidence interval
(95% CI) was employed for measures of associa-
tion to interpret the findings. We used the R statis-
tical software (version 4.2.0) for all analyses [23].

Ethics
We obtained informed consent from all partici-
pants. The Ethics in Research Committee of the
Gonçalo Moniz Research Center approved this
study (approval No. 4.315.319/202).

RESULTS
We recruited 1268 patients with suspected SARS-
CoV-2 infection up to five days post-disease onset,
all of whom answered the first questionnaire. We
excluded 39 (3.0%) individuals who were tested
outside RT-PCR recommended period or reported
a second SARS-CoV-2 infection and 415 (32.8%)
who were lost to follow-up. A total of 814 (64.2%)
patients were considered eligible for the analysis
Figure 1. Flowchart of the study population. RT-PCR – reverse transcription of residual symptoms (Figure 1 and Figure S5 in
polymerase chain reaction. the Online Supplementary Document). Ques-

www.jogh.org • doi: 10.7189/jogh.13.06015 3 2023 • Vol. 13 • 06015

 Cazé et al.
RESEARCH THEME 2: COVID-19
tionnaires 2 and 3 were applied 40 (IQR = 31-59) days and 102 (IQR = 63-150) days post-symptom onset,
respectively. We observed no difference in sociodemographic data and frequency of positive RT-PCR be-
tween the included (n = 814) and excluded (n = 454) participants (Table S1 in the Online Supplementary
Document). Among the included participants, 412 tested positive (COVID-19 group) and 402 tested nega-
tive (non-COVID-19 group) for SARS-CoV-2. The COVID and non-COVID groups had a median age of 36
(IQR = 28-48) and 34 (IQR = 25-43) years, respectively. Furthermore, 55% of the participants in the COVID
group and 64% in the non-COVID group were female. Eighty-four per cent of the participants in the COVID
group and 99% in the non-COVID group had mild symptoms (Table 1 and Table S2 in the Online Supple-
mentary Document). More patients were hospitalised in the COVID-19 group (15.9%) than non-COVID-19
group (1.5%) (Table 1). Follow-up time was 3.3 months for the COVID group and 3.7 months for the non-
COVID group (Figure S6 in the Online Supplementary Document).

Table 1. Clinical, sociodemographic characteristics and frequency of residual symptoms for COVID-19 and non-
COVID-19 cases
COVID-19 cases Non-COVID-19 cases
Characteristics P-value*
(n = 412), n (%) (n = 402), n (%)
Age in years, median (IQR) 36 (28-48) 34 (25-43) <0.001
Age group in years <0.001
15-30 126 (30.6) 167 (41.5)
31-40 123 (29.9) 106 (26.4)
41-50 75 (18.2) 82 (20.4)
>50 88 (21.4) 47 (11.7)
Female sex 228 (55.3) 259 (64.4) 0.01
Number of acute symptoms, median (IQR) 11 (8-14) 10 (7-13) 0.01
Any comorbidities 148 (35.9) 131 (32.5) 0.3
Number of comorbidities 0.6
0 264 (64.1) 271 (67.4)
1 101 (24.5) 90 (22.4)
>1 47 (11.4) 41 (10.2)
BMI in kg/m2, median (IQR) 27 (23.5-29.8) 26.1 (23.5-29.1) 0.05
Obesity 102 (24.8) 86 (21.4) 0.7
Normal 310 (75.2) 316 (78.6)
Obese (BMI = 30-34) 21 (5.1) 17 (4.2)
Overweight (BMI = 35-39) 79 (19.2) 67 (16.7)
Extremely obese (BMI>40) 2 (0.5) 2 (0.5)
Type of hospitalization <0.001
Outpatient (mild) 346 (84.2) 395 (98.5)
Hospitalization, non-ICU (moderate) 59 (14.4) 6 (1.5)
Hospitalization, ICU (severe) 6 (1.5) 0 (0)
At least one residual symptom >1 month after disease onset 122 (29.6) 24 (5.8) <0.001
Number of residual symptoms <0.001
0 292 (70.9) 378 (94.0)
1 57 (13.8) 9 (2.2)
>1 63 (15.3) 15 (3.7)
Length of follow-up in months, median (IQR) 3.35 (2.1-5.2) 3.65 (2.1-4.9) 0.5
BMI – body-mass index, IQR – interquartile range
*Pearson χ2 test or Wilcoxon rank sum test.

After one-month post-disease onset, 122/412 (29.6%) individuals from the COVID group reported at least
one residual symptom, meeting the CDC’s definition of long COVID [15]. Residual symptoms were report-
ed only by 24 (6%) individuals (Table 1) from the non-COVID group. Among 122 long COVID cases, 57
(46.7%) presented with more than two residual symptoms.
Within the COVID group, fatigue (n = 56, 13.7%), olfactory disorder (n = 42, 9.9%), myalgia (n = 36, 8.7%), gus-
tatory disorder (n = 27, 6.6%), and headache (n = 26, 6.1%) were the most frequently reported residual symp-
toms (Figure 2), starting during the acute phase. Other symptoms not reported during the acute phase, such
as memory loss and hair loss, were reported by patients with long COVID (Table S3 in the Online Supple-
mentary Document). Compared to patients who recovered, the long COVID group had more women (63.1%
vs 52.1%) and older individuals (40 (IQR = 32-51) vs 35 (IQR = 28-47) years; (Table 2)). Hospitalisation for
disease was comparable between long COVID (n = 21/122, 17.2%) and recovered patients (n = 44/290, 15.2%).

2023 • Vol. 13 • 06015 4 www.jogh.org • doi: 10.7189/jogh.13.06015

 Prevalence and risk factors for long COVID

RESEARCH THEME 2: COVID-19

Figure 2. Frequency of acute and residual symptoms for COVID-19 and non-COVID-19 cases.

Table 2. Clinical and sociodemographic characteristics for COVID cases with or without long COVID
No long COVID (n = 290), Long COVID (n = 122),
Characteristics P-value*
n (%) n (%)
Age in years, median (IQR) 35 (28-47) 40 (32-51) 0.01
Age group in years 0.07
15-30 99 (34.1) 27 (22.1)
31-40 85 (29.3) 38 (31.1)
41-50 51 (17.9) 24 (19.7)
>50 55 (19.0) 33 (27.0)
Female sex 151 (52.1) 77 (63.1) 0.04
BMI in kg/m2, median (IQR) 26.9 (23.5-29.7) 27.2 (23.8-30.1) 0.8
Number of acute symptoms, median (IQR) 11 (7-14) 12 (9-15) 0.01
Obesity 70 (24.1) 32 (26.2) 0.4
Obese (BMI = 30-34) 12 (4.1) 9 (7.4)
Overweight (BMI = 35-39) 57 (19.7) 22 (18.0)
Extremely obese (BMI >40) 1 (0.3) 1 (0.3)
Number of comorbidities 0.05
0 189 (65.0) 75 (61.0)
1 75 (25.9) 26 (21.3)
>1 26 (9.0) 21 (17.2)
Length of follow-up – months, median (IQR) 3.30 (2.00-5.10) 3.75 (2.12-5.55) 0.07
Required hospitalization or ICU 44 (15.2) 21 (17.2) 0.27
BMI – body mass index, ICU – intensive care unit, IQR – interquartile range
*Pearson χ2 test or Wilcoxon rank sum test.

The risk of long COVID increased with age and was higher among individuals aged >50 years (odds ra-
tio (OR) = 2.44; 95% CI = 1.29-4.66) compared to individuals aged 15-30. Individuals with more than five
symptoms in the acute phase (OR = 3.15; 95% CI = 1.37-8.55) and females (OR = 1.55; 95% CI = 0.99-2.44)
were more likely to develop long COVID (Table 3). After adjusting for the number of comorbidities, body
mass index (BMI), respiratory allergy, and smoking in the sensitivity analysis, the results for the primary
variables did not change significantly (Table S4 in the Online Supplementary Document).
We also observed a co-occurrence of long COVID symptoms. At least 50% of individuals who reported fa-
tigue also complained of gustatory dysfunction, anorexia, dysphonia, chest pain, headache, breathlessness,
or myalgia, while 80% of participants with gustatory dysfunction also reported olfactory dysfunction (Fig-
ure S7 in the Online Supplementary Document).

www.jogh.org • doi: 10.7189/jogh.13.06015 5 2023 • Vol. 13 • 06015

 Cazé et al.
RESEARCH THEME 2: COVID-19
Table 3. Multivariable analysis of odds ratio (lower and upper Finally, we evaluated the impact of long COVID on the quality
confidence intervals) for presenting long COVID symptoms of life by administering the EQ-5D-3L questionnaire to 102 out
Characteristics OR (95% CI) P-value of 122 patients with long COVID. Fifty-seven patients (55.8%)
Age group in years, 15-30 Reference reported alteration in at least one dimension of QoL. The dimen-
31-40 1.71 (0.95-3.12) 0.078 sions of pain and depression were the most impacted, with 42
41-50 2.04 (1.02-4.10) 0.045 patients (41.1%) reporting any degree of alteration (Table S5 in
>50 2.44 (1.29-4.66) 0.006 the Online Supplementary Document).
Female sex 1.55 (0.99-2.44) 0.055
Number of symptoms
3.15 (1.37-8.55) 0.012
(>5 acute symptoms)
DISCUSSION
OR – odds ratio, CI – confidence interval
We confirmed that the persistence of symptoms for more than
one month following the acute phase was specifically associated
with and frequently detected after SARS-CoV-2 infection. Risk factors for long COVID included female gen-
der, being aged >50, and exhibiting more than five symptoms during the acute phase. Together, these find-
ings demonstrate high frequency of long COVID, with well-defined demographic and clinical risk factors.
The proportion of patients with COVID-19 who developed long COVID varied from 7.5% to 89% [1-3,20].
Most studies only included patients who tested positive for SARS-CoV-2. One strength of our study is the
inclusion of a comparator group of symptomatic patients, recruited at the same time/location, who tested
negative by SARS-CoV-2 RT-PCR. Additionally, we evaluated most symptoms from the follow-up question-
naire at the time of recruitment in a blinded manner, prior to the SARS-CoV-2 test result. Including a con-
trol group is important, as many residual symptoms attributed to long COVID (such as fatigue and head-
ache) are non-specific and could be triggered or aggravated by stress and psychosocial problems connected
to the global health crisis [24-26]. We observed that non-confirmed COVID-19 cases reported similar re-
sidual symptoms after acute illness, but at a much lower frequency. Few studies have included a control
group of symptomatic patients that tested negative for SARS-CoV-2 [4,27]. In a cohort of healthcare work-
ers, olfactory disorders and hair loss were related to long COVID, while positive and negative cases both
had similar frequencies of exhaustion/burnout and fatigue [27]. In a population-based digital retrospective
cohort including outpatients and well-matched controls, 62 symptoms were associated with long COVID
[4]. The inclusion of a control group is even more valuable for hospitalised cases, considering the increased
risk of sequelae associated with iatrogenic procedures for treating severe disease [28]. Together, these find-
ings highlight the importance of including a control group when determining the frequency of sequelae to
avoid overestimating long COVID cases.
We observed a similar proportion of long COVID in groups with mild and moderate/severe disease [29,30].
We enrolled a small number of severe acute COVID-19 cases and could not conclude an association between
disease severity and sequelae. Severe COVID-19 increases the risk of long COVID from 1.2 to eight times
among hospitalised compared to non-hospitalised cases [31]. Nonetheless, the finding that long COVID
occurs in a significant proportion of individuals after mild disease highlights its potential impact on the
healthcare system, as COVID-19 variants and sub-variants continue to spread worldwide.
Fatigue was the most frequent sequelae and occurred isolated or associated with other clinical manifesta-
tions. As previously reported [32], we also observed a vast impact of long COVID on the quality of life. Al-
though the mechanism is unknown, an exacerbation of the immunological response and multisystemic in-
volvement [33] seems to be involved. Studies investigating the pathogenesis of long COVID are essential to
improve treatment and attenuate the impact.
In line with previous studies [4,5,17,30,34-36], we found that long COVID was more prevalent in women
and could be attributed to their greater propensity for going to a clinic [35]. Females represented 55% of
the COVID group and 64% of the non-COVID group in our study. Sex hormone differences in immune re-
sponse [37] and autoimmunity triggered by SARS-CoV-2 have been suggested to be involved in the disease
pathogenesis of long COVID [36]. Other studies have found that individuals with more symptoms during
the acute phase were at a higher risk for developing long COVID [17,27]. There are conflicting results regard-
ing the role of age in the risk for developing long COVID. In a cohort of healthcare workers in Switzerland,
most of whom had mild disease, young age was associated with a higher risk for developing long COVID
[27], in contrast to our findings. Additional prospective studies including multicentre cohorts with larger
sample sizes and older patients should be performed to evaluate if the risk for developing long COVID dif-
fers among age groups, racial groups, and ethnicities. Prospective studies to develop predictive models and
algorithms should be performed for the early detection of long COVID cases.

2023 • Vol. 13 • 06015 6 www.jogh.org • doi: 10.7189/jogh.13.06015

 Prevalence and risk factors for long COVID

RESEARCH THEME 2: COVID-19

Various organisations define long COVID differently. According to the CDC’s definition used in this study, it
is the persistence of symptoms after four weeks post-onset [15]. Recently, a WHO-coordinated Delphi con-
sensus set a period of more than three months after COVID-19 onset for the case definition of long COVID
[38]. In our study, only 445 of the calls were performed three months after disease onset, which would mean
that we detected long COVID in 68/216 (31.5%) of our patients. We also observed residual symptoms in
17/229 (7.4%) of the non-COVID-19 patients. As observed in our study, the prevalence of long COVID may
vary between studies in the absence of a universal definition [6]. This impacts the estimation of the preva-
lence of long COVID cases globally.
This study has some limitations. Around 36% of patients were lost to follow-up, which may have influenced
the estimates of residual symptoms. These patients had similar demographic and clinical characteristics to
the included ones. Additionally, most participants were young with mild disease, so we cannot generalise
our findings to the elderly population and patients with severe disease. Moreover, the participants were
aware of their diagnosis of COVID-19 when questionnaire 3 was applied, which may have led to an over-
estimation of persistent symptoms. Finally, the short time to follow-up did not allow for an analysis of the
duration of long COVID.

CONCLUSIONS
We detected long COVID in 29.6% of patients with mild COVID-19 disease, with older age, female sex, and
polysymptomatic acute disease as the main risk factors for persistent symptoms. Estimating the prevalence
of long COVID is important for preparing healthcare systems to assist and guide these patients. Further
studies should evaluate the effect of different variants of SARS-CoV-2 on long COVID and the potential im-
pact of vaccines in reducing residual symptoms.

Acknowledgements: We wish to thank all patients who were willing to participate in the study. In addition, we would
like to thank Hospital Aeroporto and Municipal Health Secretariats of Irecê and Campo Formoso for logistical support
and to Roberta de Carvalho Freitas for patient-related support.
Funding: This study was supported by MCTI/CNPq/FNDCT/MS/SCTIE/Decit (nº 07/2020). This study was partial-
ly supported by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Finance Code
001 - and by CNPq 409004/2022-7. M.B.N., A.B. and V.S.B. are research fellows from CNPq, the Brazilian National
Research Council.
Authorship contributions: ABC, TCS and VB structured the original draft. VB and CB were responsible for concep-
tualization and methodology. APB, GLS, ACAA, MQAB, NRS, and ABC collected patient data. TCS made the formal
analysis statistics. AMPB and VB participated in the project administration and the funding acquisition. ABC, TCS, JD,
ACB, LPGC, MBN, ABN, CGB substantively revised the manuscript. All authors read and approved the final manuscript.
Disclosure of interest: The authors completed the ICMJE Disclosure of Interest Form (available upon request from the
corresponding author) and disclose no relevant interests.
Additional material
Online Supplementary Document

1 Chen C, Haupert SR, Zimmermann L, Shi X, Fritsche LG, Mukherjee B. Global Prevalence of Post- Coronavirus Disease
(COVID-19) Condition or Long COVID: A Meta-Analysis and Systematic Review. J Infect Dis. 2022;226:1593-607. Med-
REFERENCES

line:35429399 doi:10.1093/infdis/jiac136
2 Nasserie T, Hittle M, Goodman SN. Assessment of the Frequency and Variety of Persistent Symptoms Among Patients
With COVID-19: A Systematic Review. JAMA Netw Open. 2021;4:e2111417. Medline:34037731 doi:10.1001/jamanet-
workopen.2021.11417
3 United States Centers for Disease Control and Prevention. Nearly One in Five American Adults Who Have Had COVID-19
Still Have ‘’Long COVID’’. 2022. Available: https://www.cdc.gov/nchs/pressroom/nchs_press_releases/2022/20220622.
htm. Accessed: 17 September 2022.
4 Subramanian A, Nirantharakumar K, Hughes S, Myles P, Williams T, Gokhale KM, et al. Symptoms and risk factors for
long COVID in non-hospitalized adults. Nat Med. 2022;28:1706-14. Medline:35879616 doi:10.1038/s41591-022-01909-w
5 Amin-Chowdhury Z, Ladhani SN. Causation or confounding: why controls are critical for characterizing long COVID. Nat
Med. 2021;27:1129-30. Medline:34140704 doi:10.1038/s41591-021-01402-w
6 Alkodaymi MS, Omrani OA, Fawzy NA, Shaar BA, Almamlouk R, Riaz M, et al. Prevalence of post-acute COVID-19 syn-
drome symptoms at different follow-up periods: a systematic review and meta-analysis. Clin Microbiol Infect. 2022;28:657-
66. Medline:35124265 doi:10.1016/j.cmi.2022.01.014

www.jogh.org • doi: 10.7189/jogh.13.06015 7 2023 • Vol. 13 • 06015

 Cazé et al.
RESEARCH THEME 2: COVID-19
7 Kelly BD, Gulati G. Long COVID: the elephant in the room. QJM: An International Journal of Medicine. 2022;115:5-6.
REFERENCES Medline:34850227 doi:10.1093/qjmed/hcab299
8 Durstenfeld MS, Peluso MJ, Peyser ND, Lin F, Knight SJ, Djibo A, et al. Factors Associated with Long Covid Symptoms in
an Online Cohort Study. Open Forum Infect Dis. 2023;10:ofad047. Medline:36846611 doi:10.1093/ofid/ofad047
9 A l-Aly Z, Bowe B, Xie Y. Long COVID after breakthrough SARS-CoV-2 infection. Nat Med. 2022;28:1461-7. Med-
line:35614233 doi:10.1038/s41591-022-01840-0
10 H an Q, Zheng B, Daines L, Sheikh A. Long-Term Sequelae of COVID-19: A Systematic Review and Meta-Analysis of
One-Year Follow-Up Studies on Post-COVID Symptoms. Pathogens. 2022;11:269. Medline:35215212 doi:10.3390/patho-
gens11020269
11 Nagib W. Genomahcov – Fiocruz [Network Genomics Firocruz]. 2023. Available: http://www.genomahcov.fiocruz.br/.
Accessed: 28 April 2023.
12 I nstituto de Biologia Molecular do Paraná. Consultas – Agência Nacional de Vigilância Sanitária [Consultations – Nation-
al Health Surveillance Agency]. 2023. Available: https://consultas.anvisa.gov.br/#/saude/25351441245202096/. Accessed:
16 June 2023.
13 S eegene. AllplexTM SARS-CoV-2 Assay. 2023. Available: https://consultas.anvisa.gov.br/api/consulta/produ-
tos/25351395442202026/anexo/T15632158/nomeArquivo/IFU_80102512493.pdf?Authorization=Guest. Accessed: 28
April 2023.
14 Cuschieri S. The STROBE guidelines. Saudi J Anaesth. 2019;13 Suppl 1:S31-4. Medline:30930717 doi:10.4103/sja.
SJA_543_18
15 C enters for Disease Control and Prevention. Long COVID. 2022. Available: https://www.cdc.gov/coronavirus/2019-ncov/
long-term-effects/. Accessed: 28 April 2023.
16 Lopez-Leon S, Wegman-Ostrosky T, Perelman C, Sepulveda R, Rebolledo PA, Cuapio A, et al. More than 50 long-term
effects of COVID-19: a systematic review and meta-analysis. Sci Rep. 2021;11:16144. Medline:34373540 doi:10.1038/
s41598-021-95565-8
17 Sudre CH, Murray B, Varsavsky T, Graham MS, Penfold RS, Bowyer RC, et al. Attributes and predictors of long COVID.
Nat Med. 2021;27:626-31. Medline:33692530 doi:10.1038/s41591-021-01292-y
18 E uroQol. EQ-5D User Guides – EQ-5D. 2022. Available: https://euroqol.org/publications/user-guides. Accessed: 17 Sep-
tember 2022.
19 Santos M, Cintra MACT, Monteiro AL, Santos B, Gusmão-filho F, Andrade MV, et al. Brazilian Valuation of EQ-5D-3L
Health States. Med Decis Making. 2016;36:253-63. Medline:26492896 doi:10.1177/0272989X15613521
20 M ichelen M, Manoharan L, Elkheir N, Cheng V, Dagens A, Hastie C, et al. Characterising long COVID: a living systematic
review. BMJ Glob Health. 2021;6:e005427. Medline:34580069 doi:10.1136/bmjgh-2021-005427
21 B atiha GE-S, Al-kuraishy HM, Al-Gareeb AI, Welson NN. Pathophysiology of Post-COVID syndromes: a new perspective.
Virol J. 2022;19:158. Medline:36210445 doi:10.1186/s12985-022-01891-2
22 R aveendran AV, Jayadevan R, Sashidharan S. Long COVID: An overview. Diabetes Metab Syndr. 2021;15:869-75. Med-
line:33892403 doi:10.1016/j.dsx.2021.04.007
23 R Core Team. R: The R Project for Statistical Computing. 2019. Available: https://www.r-project.org/. Accessed: 28 April
2023.
24 Bartoszek A, Walkowiak D, Bartoszek A, Kardas G. Mental Well-Being (Depression, Loneliness, Insomnia, Daily Life Fa-
tigue) during COVID-19 Related Home-Confinement – A Study from Poland. Int J Environ Res Public Health. 2020;17:7417.
Medline:33053787 doi:10.3390/ijerph17207417
25 Usher K, Bhullar N, Jackson D. Life in the pandemic: Social isolation and mental health. J Clin Nurs. 2020;29:2756-7.
Medline:32250493 doi:10.1111/jocn.15290
26 Brooks SK, Webster RK, Smith LE, Woodland L, Wessely S, Greenberg N, et al. The Psychological Impact of Quarantine
and How to Reduce it: Rapid Review of the Evidence. Lancet. 2020;395:912-20. Medline:32112714 doi:10.1016/S0140-
6736(20)30460-8
27 Strahm C, Seneghini M, Güsewell S, Egger T, Leal O, Brucher A, et al. Symptoms compatible with long-COVID in health-
care workers with and without SARS-CoV-2 infection – results of a prospective multicenter cohort. Clin Infect Dis.
2022;75:e1011-9. Medline:35090015 doi:10.1093/cid/ciac054
28 PHOSP-COVID Collaborative Group. Clinical characteristics with inflammation profiling of long COVID and associ-
ation with 1-year recovery following hospitalisation in the UK: a prospective observational study. Lancet Respir Med.
2022;10:761-75. Medline:35472304 doi:10.1016/S2213-2600(22)00127-8
29 Townsend L, Dyer AH, Jones K, Dunne J, Mooney A, Gaffney F, et al. Persistent fatigue following SARS-CoV-2 infection is
common and independent of severity of initial infection. PLoS One. 2020;15:e0240784. Medline:33166287 doi:10.1371/
journal.pone.0240784
30 Bai F, Tomasoni D, Falcinella C, Barbanotti D, Castoldi R, Mulè G, et al. Female gender is associated with “long COVID”
syndrome: a prospective cohort study. Clin Microbiol Infect. 2022;28:611.e9-e16. Medline:34763058 doi:10.1016/j.
cmi.2021.11.002
31 Peghin M, Palese A, Venturini M, De Martino M, Gerussi V, Graziano E, et al. Post-COVID-19 symptoms 6 months
after acute infection among hospitalized and non-hospitalized patients. Clin Microbiol Infect. 2021;27:1507-13. Med-
line:34111579 doi:10.1016/j.cmi.2021.05.033
32 K im Y, Kim S-W, Chang H-H, Kwon KT, Hwang S, Bae S. One Year Follow-Up of COVID-19 Related Symptoms and Patient
Quality of Life: A Prospective Cohort Study. Yonsei Med J. 2022;63:499. Medline:35619573 doi:10.3349/ymj.2022.63.6.499

2023 • Vol. 13 • 06015 8 www.jogh.org • doi: 10.7189/jogh.13.06015

 Prevalence and risk factors for long COVID

RESEARCH THEME 2: COVID-19

33 A bboud H, Abboud FZ, Kharbouch H, Arkha Y, El Abbadi N, El Ouahabi A. COVID-19 and SARS-Cov-2 Infection: Patho-
physiology and Clinical Effects on the Nervous System. World Neurosurg. 2020;140:49-53. Medline:32474093 doi:10.1016/j.
REFERENCES wneu.2020.05.193
34 Petersen MS, Kristiansen MF, Hanusson KD, Danielsen ME, Á Steig B, Gaini S, et al. Long COVID in the Faroe Islands: A
Longitudinal Study Among Nonhospitalized Patients. Clin Infect Dis. 2021;73:e4058-63. Medline:33252665 doi:10.1093/
cid/ciaa1792
35 Augustin M, Schommers P, Stecher M, Dewald F, Gieselmann L, Gruell H, et al. Post-COVID syndrome in non-hospi-
talised patients with COVID-19: a longitudinal prospective cohort study. Lancet Reg Health Eur. 2021;6:100122. Med-
line:34027514 doi:10.1016/j.lanepe.2021.100122
36 Seeßle J, Waterboer T, Hippchen T, Simon J, Kirchner M, Lim A, et al. Persistent symptoms in adult patients one year after
COVID-19: a prospective cohort study. Clin Infect Dis. 2022;74:1191-8. Medline:34223884
37 Stewart S, Newson L, Briggs TA, Grammatopoulos D, Young L, Gill P. Long COVID risk - a signal to address sex hor-
mones and women’s health. Lancet Reg Health Eur. 2021;11:100242. Medline:34746909 doi:10.1016/j.lanepe.2021.100242
38 Soriano JB, Murthy S, Marshall JC, Relan P, Diaz JV. A clinical case definition of post-COVID-19 condition by a Delphi
consensus. Lancet Infect Dis. 2022;22:e102-7. Medline:34951953 doi:10.1016/S1473-3099(21)00703-9

www.jogh.org • doi: 10.7189/jogh.13.06015 9 2023 • Vol. 13 • 06015