Abstract
Objectives. To investigate associations between COVID-19-related factors and depressive symptoms among primary care workers (PCWs) in São Paulo, Brazil, and to compare the prevalence of probable depression among PCWs before and during the pandemic.
Methods. In a random sample of primary care clinics, we examined 6 pandemic-related factors among 828 PCWs. We used multivariate Poisson regression with robust variance to estimate prevalence ratios for probable depression. We assessed the prevalence of probable depression in PCWs before and during the pandemic in 2 comparable studies.
Results. Adjusted prevalence ratios were substantial for insufficient personal protective equipment; experiences of discrimination, violence, or harassment; and lack of family support. Comparisons between PCWs before and during the pandemic showed that the prevalence of probable depression among physicians, nurses, and nursing assistants was higher during the pandemic and that the prevalence among community health workers was higher before the pandemic.
Conclusions. Our findings indicate domains that may be crucial to mitigating depression among PCWs but that, with the exception of personal protective equipment, have not previously been examined in this population. It is crucial that governments and communities address discriminatory behaviors against PCWs, promote their well-being at work, and foster family support. (Am J Public Health. 2022;112(5):786–794. https://doi.org/10.2105/AJPH.2022.306723)
Brazil is among the most affected countries of the COVID-19 pandemic, and within Brazil the city of São Paulo has had high numbers of cases and deaths. Brazil’s government has been especially negligent during the pandemic,1,2 with the president encouraging people to ignore physical distancing and mask wearing, promoting the use of ineffective treatments such as chloroquine,1 and providing minimal support for vaccination.3
The COVID-19 outbreak has overstretched the Brazilian health care system, overwhelmed health care workers, and jeopardized the mental health of these workers. Health care workers in general, and primary care workers (PCWs) in particular,4 have been largely neglected, as exhibited by limited government support2 and unfavorable workplace conditions. We sought to identify factors strongly related to depression among PCWs in this context as a means of informing policies and interventions designed to reduce the mental health toll on these workers. We could not identify any previous studies focusing on samples drawn to represent PCWs during the pandemic. We therefore selected factors that have been related to depression among health workers in other countries during the COVID-19 pandemic (e.g., lack of personal protective equipment [PPE]5) or have been related to mental disorders in previous pandemics (e.g., discrimination toward health workers6). We also sought to compare the prevalence of depression among PCWs before and during the pandemic.
Accordingly, we first used data from a sample of PCWs (n = 828) who took part in the COVID-19 Health Care Workers Study (HEROES)–São Paulo (HEROES-SP) to examine the relationship between COVID-19-related factors and depression among PCWs in the city of São Paulo. The study was conducted as part of the larger HEROES global initiative, which examined the mental health effects of the pandemic on health workers in 25 countries.7 Second, we analyzed data from 2 directly comparable subsamples of PCWs, one from HEROES-SP (n = 376) and the other from Panorama of Primary Health Care Workers in São Paulo, Brazil: Depression, Organizational Justice, Violence at Work, and Burnout Assessments (PANDORA-SP; n = 574); these subsamples were restricted to the same region within São Paulo and to the same types of workers (i.e., physicians, nurses, nursing assistants, and community health workers [CHWs]).
The PANDORA-SP subsample was studied before the initial pandemic outbreak, and the HEROES-SP subsample was studied shortly after the initial outbreak.8 We compared these 2 subsamples to examine whether and to what degree overall rates of depression were higher after the pandemic onset. (Note that HEROES-SP is an ongoing longitudinal, cohort study, and only baseline data are reported here.)
METHODS
To examine pandemic-related factors and their association with depression, we conducted an online survey in October and November 2020 of PCWs who took part in HEROES-SP (Figure A, available as a supplement to the online version of this article at http://www.ajph.org). The sample was obtained by randomly selecting 26 primary care clinics in Region 1 of São Paulo and including all of the PCWs in the selected clinics. Of 2106 potential participants, 828 (39.3%) completed the survey. Data on age were missing for 5.2% of the sample; there were no missing data for other variables.
We used subsamples from HEROES-SP and PANDORA-SP to compare the prevalence of depression among PCWs before and after the initial COVID-19 pandemic onset in São Paulo. These subsamples comprised 2 different random samples of all primary care clinics in Region 1 of Sao Paulo. For comparability, we restricted both subsamples to physicians, nurses, nursing assistants, and CHWs. PANDORA-SP was a cross-sectional study conducted in São Paulo in 2012 to examine depression and burnout among PCWs. For that study, we randomly selected 66 primary health clinics in São Paulo. All workers involved in the Family Health Program were eligible to participate. Details on PANDORA-SP have been provided elsewhere.8 The PANDORA-SP response rate was 93%.
Although PANDORA-SP was conducted nearly a decade ago, the context in which PCWs work has remained essentially the same in subsequent years. For instance, (1) the São Paulo primary model, adopted by the secretary of health, has continued to be based on the national family health strategy primary care model adopted in 1994; (2) each Family Health Program team maintains the same staff composition (e.g., physicians, nurses, nursing assistants, and CHWs), is located in the same region, and covers the same vulnerable populations; (3) primary care centers in São Paulo are still coordinated by private institutions in partnership with the local government; and (4) the private institution that managed PCWs in Region 1 at the time of PANDORA-SP is the same institution that has managed these workers during the pandemic.
Table 1 describes the sampling procedures and response rates for PANDORA-SP and HEROES-SP. Table 2 provides data on the characteristics of participants from the PANDORA-SP and HEROES-SP subsamples.
TABLE 1—
PANDORA-SP and HEROES-SP Sampling Procedures and Response Rates: São Paulo, Brazil, 2012 and 2020
| PANDORA-SP | HEROES-SP | |
| No. of primary care centers randomly selected | 66 | 26 |
| No. of primary care workers invited (all primary care workers from the primary care centers selected) | 3141 | 2016 |
| São Paulo regions included in the study | 1–5 | 1 |
| No. of primary care workers in Region 1 who participated in the study | 574 | 376 |
| Response rate, % | 93.0 | 39.3 |
Note. HEROES-SP = COVID-19 Health Care Workers Study–São Paulo; PANDORA-SP = Panorama of Primary Health Care Workers in São Paulo, Brazil: Depression, Organizational Justice, Violence at Work, and Burnout Assessments.
TABLE 2—
Comparable PANDORA-SP and HEROES-SP Subsamples Before and During the COVID-19 Pandemic, by Job Type, Gender, Age, and Skin Color: São Paulo, Brazil, 2012 and 2020
| PANDORA-SP (n = 574), No. (%) | HEROES-SP (n = 376), No. (%) | P | |
| Job type | |||
| Physician | 42 (7.3) | 42 (11.7) | .04 |
| Nurse | 65 (11.3) | 56 (14.9) | .1 |
| Nursing assistant | 136 (23.7) | 93 (24.7) | .71 |
| Community health worker | 331 (57.7) | 185 (49.2) | .01 |
| Gender | |||
| Male | 35 (6.1) | 42 (11.1) | .005 |
| Female | 539 (93.9) | 334 (88.9) | .005 |
| Age group, y | |||
| 18–29 | 147 (25.6) | 77 (20.2) | .07 |
| 30–39 | 239 (41.6) | 175 (46.8) | .13 |
| 40–49 | 129 (22.5) | 97 (25.8) | .23 |
| ≥ 50 | 59 (10.3) | 27 (7.2) | .1 |
| Skin color | |||
| White | 231 (40.3) | 149 (39.6) | .85 |
| Brown | 232 (40.4) | 158 (42.0) | .62 |
| Black | 95 (16.5) | 59 (15.7) | .72 |
| Other | 16 (2.8) | 10 (2.7) | .9 |
Note. HEROES-SP = COVID-19 Health Care Workers Study–São Paulo; PANDORA-SP = Panorama of Primary Health Care Workers in São Paulo, Brazil: Depression, Organizational Justice, Violence at Work, and Burnout Assessments.
In addition, to understand differences between the HEROES-SP sample and the overall PCW population in Region 1, we compared the characteristics of HEROES-SP participants and HEROES-SP “minimal responders.” Minimal responders are PCWs who began filling out the HEROES-SP questionnaire but did not finish it and did not complete the Patient Health Questionnaire (PHQ-9; Table A, available as a supplement to the online version of this article at http://www.ajph.org).
Measures
The HEROES questionnaire was originally created in both Spanish and English versions. HEROES-SP employed a Portuguese version. The Spanish version of the HEROES questionnaire was translated to Portuguese and back-translated according to the World Health Organization’s standard procedures.9
Outcome
Depression was assessed with the Brazilian version of the widely used PHQ-9.10 As in previous studies, some from Brazil,11,12 we used a score of 10 or higher to indicate at least moderate symptoms of depression. Participants were asked whether they had experienced any of 9 depressive symptoms (assessed via the Diagnostic and Statistical Manual of Mental Disorders13) in the preceding 2 weeks. Possible responses for each symptom included not at all (0), several days (1), more than half of the days (2), and nearly every day (3).
Exposures in HEROES-SP
We assessed the following factors hypothesized to be related to depression:
-
1.
Level of access to PPE (sufficient, a little insufficient, much/very much insufficient).
-
2.
Experiences of discrimination (“I have felt discriminated due to being a health worker during the pandemic” [yes or no]), violence (“I have experienced violence due to being a health worker during the pandemic” [yes or no]), and harassment (“I have been harassed by family members of patients with COVID-19” [yes or no]); response options ranged from 0 to 2 or 3.
-
3.
Job type (physician, nurse, nursing assistant, CHW, other clinical staff, administrative staff).
-
4.
Financial strain (“Have you felt constantly under financial strain?” [yes or no]).
-
5.
Family support (“I have loved ones who support me when needed?” [yes or no]).
-
6.
Isolation due to COVID-19 (“How many days have you been in isolation for being a suspected or confirmed case of COVID-19?” [0 = no isolation, 1 = 1 day or more in isolation]).
Confounders and Government Mistrust
We considered the following sociodemographic variables as potential confounders: age (18–30, 30–40, 40–50, ≥ 50 years), gender (female, male, other), and self-reported skin color (White, Black, Brown, and other).
To document government mistrust, we assessed participants’ responses to an item regarding their level of trust in the government (none, a little, a lot).
Statistical Analysis
Stata version 14.0 (StataCorp LP, College Station, TX) was used in conducting all of our analyses. We used Poisson regression analyses with robust variance estimates to obtain prevalence ratios (PRs) for relationships between the aforementioned exposures and depression among HEROES-SP participants. This type of regression was selected to minimize overestimation of the associations given that the outcome was frequent in our sample (25%).14 We included all of the aforementioned confounders and pandemic-related factors in the model. The analysis was restricted to the 828 PCWs who completed the survey. For the missing data on age, we used the mean age of the sample. We used the variance inflation factor to investigate multicollinearity. Variance inflation factor values for explanatory variables ranged from 1.00 to 1.09, with a mean of 1.04. No significant correlation was found between a given explanatory variable and any other explanatory variable in the model.
We compared log-pseudolikelihood values between models and selected the model with the lowest value (−453.809) as the final model. In addition, we chose the model with the highest pseudo R2 (0.0839).
We examined the prevalence of depression among PCWs before (i.e., in PANDORA-SP) and after (i.e., in HEROES-SP) the initial COVID-19 pandemic onset overall and by job type. We used the Pearson χ2 test to estimate P values.
Ethical Considerations
Privacy and confidentiality of data were guaranteed in HEROES, HEROES-SP, and PANDORA-SP. Individuals were asked to sign a consent form before participation.
For HEROES, a Web-based platform is being used to collect data across countries. This platform is akin to REDCap in terms of protection and data management. As a means of guaranteeing confidentiality, each participant is issued an identification number created by a code system. Access to the system is restricted to personnel with credentials defined by the study’s steering committee.
RESULTS
The majority of PCWs in the HEROES-SP sample reported no (18.6%) or little (69.1%) trust in the government. Table 3 presents data on the frequency of pandemic-related factors and confounders. The mean age of the sample was 36.2 years (SD = 9.2), and most participants were women (n = 712; 85.9%). In terms of pandemic-related factors, 51.6% of the participants (n = 427) had insufficient access to PPE, and 38.1% (n = 315) reported at least one experience of discrimination, violence, or harassment at work. Table 3 also provides data on the prevalence of probable depression (i.e., PHQ score above 10). Overall, 25% of the participants had probable depression. Some of the categories included in Table 4 were combined in the Poisson regression to avoid cells with small numbers, reduce the number of categories, and clarify presentation. For example, we combined physicians, nurses, and nursing assistants (in contrast to CHWs, who had limited contact with patients) because they were frontline workers during the data collection period.
TABLE 3—
Exposures and Potential Confounders, by Probable Depression: HEROES-SP, São Paulo, Brazil, 2020
| Total, No. (%) | Probable Depression | P a | ||
| No, No. (%) | Yes, No. (%) | |||
| Exposures | 620 (74.9) | 208 (25.1) | < .001 | |
| Access to personal protective equipment | ||||
| Sufficient | 460 (55.6) | 369 (81.3) | 91 (19.8) | |
| A little insufficient | 252 (30.4) | 182 (72.2) | 70 (27.8) | |
| Much/very much insufficient | 116 (14.0) | 69 (59.5) | 47 (40.5) | |
| Experiences of discrimination, harassment, or violence | < .001 | |||
| 0 | 334 (40.3) | 281 (84.1) | 53 (15.9) | |
| 1 | 315 (38.1) | 237 (75.3) | 78 (24.7) | |
| 2 or 3 | 179 (21.6) | 102 (57.0) | 77 (43.0) | |
| Financial strain | < .001 | |||
| No | 598 (72.2) | 473 (79.1) | 125 (20.9) | |
| Yes | 230 (27.8) | 147 (63.9) | 83 (36.1) | |
| Isolation due to COVID-19 | .02 | |||
| No | 400 (48.3) | 315 (78.8) | 85 (21.2) | |
| Yes | 428 (51.7) | 305 (71.3) | 123 (28.7) | |
| Support from family | < .001 | |||
| Yes | 762 (92.0) | 590 (77.4) | 172 (22.6) | |
| No | 66 (8.0) | 30 (45.5) | 36 (54.5) | |
| Confounder: gender | .66 | |||
| Female | 712 (85.9) | 535 (75.1) | 177 (24.9) | |
| Male | 116 (14.1) | 85 (73.3) | 31 (26.7) | |
| Confounder: age group, y | .06 | |||
| 18–29 | 214 (25.8) | 153 (71.4) | 61 (28.6) | |
| 30–39 | 382 (46.3) | 296 (77.5) | 86 (22.5) | |
| 40–49 | 181 (21.8) | 127 (70.2) | 54 (29.8) | |
| ≥ 50 | 51 (6.1) | 44 (86.3) | 7 (13.7) | |
| Confounder: skin color | .37 | |||
| White | 383 (46.3) | 289 (75.5) | 94 (24.5) | |
| Black | 111 (1.4) | 88 (79.3) | 23 (20.7) | |
| Brown | 316 (38.2) | 228 (72.1) | 88 (27.9) | |
| Other | 18 (2.2) | 15 (83.3) | 3 (16.7) | |
| Job type | .01 | |||
| Physician | 42 (5.1) | 30 (71.4) | 12 (28.6) | |
| Nurse | 56 (6.8) | 40 (71.4) | 16 (28.6) | |
| Nursing assistant | 93 (11.2) | 70 (75.3) | 23 (24.7) | |
| Community health worker | 185 (22.3) | 147 (79.5) | 38 (20.5) | |
| Other clinical staffb | 165 (19.9) | 137 (83.0) | 28 (17.0) | |
| Administrative staff | 287 (34.7) | 196 (68.3) | 91 (31.7) | |
Note. HEROES-SP = COVID-19 Health Care Workers Study–São Paulo. The sample size was 828.
From Pearson χ2 test.
Psychologist, dentist, pharmacist, nutritionist, social worker, occupational therapist, physiotherapist, or physical educator.
TABLE 4—
Relationships Between Pandemic-Related Factors and Depression: HEROES-SP, São Paulo, Brazil, 2020
| Crude PR (95% CI) | APR (95% CI) | |
| Access to personal protective equipment | ||
| Sufficient (Ref) | 1 | 1 |
| A little insufficient | 1.40 (1.07, 1.84) | 1.25 (0.96, 1.62) |
| Much/very much insufficient | 2.04 (1.53, 2.72) | 1.53 (1.14, 2.04) |
| Job type | ||
| Physician, nurse, or nursing assistant (Ref) | 1 | 1 |
| Other clinical staffa | 0.63 (0.42, 0.95) | 0.63 (0.42, 0.95) |
| Community health worker | 0.76 (0.53, 1.11) | 0.69 (0.48, 1.00) |
| Administrative staff | 1.18 (0.88, 1.58) | 0.96 (0.71, 1.30) |
| Experiences of discrimination, harassment, or violence | ||
| 0 (Ref) | 1 | 1 |
| 1 | 1.56 (1.14, 2.13) | 1.36 (1.00, 1.86) |
| 2 or 3 | 2.71 (2.00, 3.65) | 2.06 (1.53, 2.78) |
| Financial strain | ||
| No (Ref) | 1 | 1 |
| Yes | 1.72 (1.36, 2.17) | 1.54 (1.22, 1.94) |
| Isolation due to COVID-19 | ||
| No (Ref) | 1 | 1 |
| Yes | 1.18 (1.04, 1.34) | 1.15 (1.01, 1.29) |
| Support from family | ||
| Yes (Ref) | 1 | 1 |
| No | 2.41 (1.86, 3.12) | 2.48 (1.92, 3.21) |
Note. APR = adjusted prevalence ratio; CI = confidence interval; HEROES-SP = COVID-19 Health Care Workers Study–São Paulo; PR = prevalence ratio. The sample size was 828. Data were derived from a Poisson regression model with robust variance adjusted according to gender, age group, skin color, and all of the factors included in the table.
Psychologist, dentist, pharmacist, nutritionist, social worker, occupational therapist, physiotherapist, or physical educator.
Table 4 presents crude and adjusted results with respect to relationships between pandemic-related factors and probable depression. The adjusted results showed that prevalence ratios for probable depression were 1.5 or higher among participants who had insufficient access to PPE (adjusted PR (APR] = 1.53; 95% confidence interval [CI] = 1.14, 2.04); reported 2 or more experiences of discrimination, violence, or harassment (APR = 2.06; 95% CI = 1.53, 2.78); reported financial strain (APR = 1.54; 95% CI = 1.22, 1.94); or did not receive family support (APR = 2.48; 95% CI = 1.93, 3.21). Prevalence ratios were significantly lower for CHWs and other clinical staff (e.g., psychologists) than for physicians, nurses, and nursing assistants.
We used data from the PANDORA-SP (n = 574) and HEROES-SP (n = 376) subsamples to compare the prevalence of probable depression among PCWs before and during the pandemic (Table B, available as a supplement to the online version of this article at http://www.ajph.org). The results showed that the prevalence of probable depression was lower before than during the pandemic among physicians (9.5% vs 28.6%), nurses (16.9% vs 28.6%), and nursing assistants (19.2% vs 24.7%; Table B). These differences were statistically significant for physicians but not for nurses or nursing assistants. When we combined physicians, nurses, and nursing assistants into a single category, the prevalence of probable depression was higher before (16.9%) than during (26.7%) the pandemic (χ2 = 6.18, P = .01). The prevalence of probable depression among CHWs was higher before the pandemic (25.1%) than during the pandemic (20.5%), although the difference was not statistically significant (χ2 = 1.35, P = .24).
DISCUSSION
This study of PCWs was conducted in São Paulo, Brazil, shortly after the initial phase of the COVID-19 pandemic. The social context included neglect of health workers and denial of the pandemic by the president, reflected here in the 87.7% of participants who reported little or no trust in the government. Overall, 25% of PCWs reported symptoms consistent with at least moderate depression (i.e., a score above 10 on the PHQ).
Three pandemic-related factors were strongly associated with probable depression: insufficient PPE; experiences of discrimination, violence, or harassment; and limited family support. Directly comparable studies of PCWs before and during the pandemic indicated that those who were frontline workers during the pandemic period (physicians, nurses, nursing assistants) had a higher prevalence of depression during that period; in contrast, CHWs, who could not see patients in the community for safety reasons, had a lower prevalence of probable depression during the pandemic. In this section, we offer potential explanations for the 3 associations with depression.
The higher prevalence of probable depression among PCWs who perceived that their PPE was insufficient is consistent with the findings of other studies of health workers during the pandemic.5,15 For example, in the early stages of the COVID-19 outbreak in Wuhan, China, dissatisfaction with PPE was associated with a higher prevalence of depression.16 Limited access to PPE can lead to increased fear of becoming infected and infecting close contacts (especially older relatives), which in turn can affect mental health. Therefore, workplace prevention strategies for PCWs that ensure adequate access to PPE during the pandemic are essential for protection of mental health as well as protection from infection.
Most PCWs reported experiencing discrimination, violence, or harassment because they were health workers. Those who reported 2 or more of these experiences had a markedly higher prevalence of probable depression than those who did not report any such experiences. Studies conducted during the pandemic in other countries, especially low- and middle-income countries, have revealed similar experiences among health workers.17,18 For example, health workers have been attacked with eggs and physically assaulted in Mexico and have been beaten, stoned, and evicted from their homes in India.19 Menon et al.17 noted that the spread of misinformation on COVID-19 has increased fear of health care workers as potential sources of infection. Although reports of harassment of health workers had been increasing in clinics and hospitals worldwide20 before the pandemic, such reports seem to have now risen to a higher level.
PCWs who reported not having support from their family had a higher prevalence of probable depression. We could not identify previous reports on family support and depression among PCWs during the pandemic. In the general population, lack of social support (not only from family members) during the pandemic has been associated with poor mental health,21 including anxiety and depression. Also, studies of health workers before the pandemic have suggested that lack of support from family members is associated with poor mental health outcomes.22
Strengths and Limitations
The strengths of this study include the focus on PCWs, who are an essential component of health care in many countries; the investigation of pandemic-related factors (e.g., discrimination, violence, or harassment and family support) that have not been examined in previous studies of health workers, and the well-defined target population (i.e., a random sample of primary care clinics in a defined area). Other strengths include the response rate of 40%, which is much higher than the rates in other online surveys conducted in meaningful target populations during the pandemic, and the use of directly comparable subsamples before and during the pandemic.
However, there were also limitations. We cannot infer causality from a cross-sectional study, although strong associations do identify the most likely factors that could play a causal role. The PHQ-9 was developed as a screen for depression, and, although commonly used in epidemiological studies as a proxy for depression, it is not equivalent to a clinical diagnosis. In addition, potential participants did not have to reach out, as they were required only to respond to an e-mail. Although this mitigates volunteer bias to some degree, there may still be response bias; for example, PCWs without mental health problems might have been more likely to respond. The converse is also possible (i.e., PCWs with mental health problems might have been more likely to respond). We note that such response bias based on mental health status could have affected our estimate of the prevalence of probable depression.
Another possible limitation is that exposure to violence was underreported owing to recall bias.23 Moreover, our findings might not be generalizable to other locales in Brazil and elsewhere in Latin America where primary services are less developed. However, many primary care teams face similar challenges in large urban settings within low- and middle-income countries. With respect to comparisons between PANDORA-SP and HEROES-SP, it is possible that time effects influenced the outcomes. This possibility is unlikely, however, because work-related contextual conditions remained mostly the same from 2012 (PANDORA-SP) to 2020 (HEROES-SP).
Although previous studies have detected associations between negative professional performance (e.g., malpractice or absenteeism) and depression among health care workers, we did not examine such associations during the pandemic. Also, we did not collect data on satisfaction with work and home life, which could have been another limitation. Finally, we used job type as a proxy for income, which may not have been the best choice of a single socioeconomic status measure in a short online survey. We are expanding measurement of socioeconomic status in the upcoming follow-up questionnaire, adding income alongside other indicators such as assets and household size.
Public Health Implications
Our study has potentially important implications with respect to mitigating the adverse consequences of the pandemic on PCWs’ mental health. First, it underscores that PCWs must have adequate PPE. Second, it suggests that, at least in some contexts, it is vital for governments and communities to dispel misinformation and in other ways ensure that PCWs are appreciated for their hazardous work rather than targeted for harassment. Third, PCWs appear to be vulnerable when family support is limited, and this should be considered in the design of policies and interventions. Although we cannot infer causality from a cross-sectional study and we have noted that our results are not necessarily generalizable to other settings, we believe that these areas need more attention and are likely to be relevant in most settings.
Previous studies have highlighted that depression among health care workers has serious consequences, including absenteeism24 and malpractice,25 and that it can hamper the sustainability of health systems. In the context of an unprecedent global crisis, these consequences may be even worse. The associations we observed between depression and negative professional performance and satisfaction with one’s work environment need to be investigated further.
ACKNOWLEDGMENTS
We acknowledge all of the researchers and health care workers who contributed to the COVID-19 Health Care Workers Study (HEROES).
CONFLICTS OF INTEREST
The authors have no potential conflicts of interest to report.
HUMAN PARTICIPANT PROTECTION
The HEROES-SP COVID-19 Health Care Workers Study (HEROES)–São Paulo was approved by the institutional review board of the city of São Paulo, Brazil, and by the ethics committee of the Associação Santa Marcelina. Participants provided informed consent.
Footnotes
See also Teoh et al., p. 703.
REFERENCES
- 1.Barberia LG, Gómez EJ. Political and institutional perils of Brazil’s COVID-19 crisis. Lancet. 2020;396(10248):367–368. doi: 10.1016/S0140-6736(20)31681-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Castro MC, Kim S, Barberia L, et al. Spatiotemporal pattern of COVID-19 spread in Brazil. Science. 2021;372(6544):821–826. doi: 10.1126/science.abh1558. [DOI] [PubMed] [Google Scholar]
- 3.Barbara V.2021. https://www.nytimes.com/2021/07/05/opinion/brazil-fake-news-bolsonaro.html?referringSource=articleShare
- 4.Li Y, Scherer N, Felix L, Kuper H. Prevalence of depression, anxiety and post-traumatic stress disorder in health care workers during the COVID-19 pandemic: a systematic review and meta-analysis. PLoS One. 2021;16(3):e0246454. doi: 10.1371/journal.pone.0246454. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Zhang SX, Liu J, Afshar Jahanshahi A, et al. At the height of the storm: healthcare staff’s health conditions and job satisfaction and their associated predictors during the epidemic peak of COVID-19. Brain Behav Immun. 2020;87:144–146. doi: 10.1016/j.bbi.2020.05.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Bai Y, Lin C-C, Lin C-Y, Chen J-Y, Chue C-M, Chou P. Survey of stress reactions among health care workers involved with the SARS outbreak. Psychiatr Serv. 2004;55(9):1055–1057. doi: 10.1176/appi.ps.55.9.1055. [DOI] [PubMed] [Google Scholar]
- 7.Susser E, Alvarado R, Mascayano F.2021. https://clinicaltrials.gov/ct2/show/NCT04352634
- 8.da Silva ATC, Lopes C de S, Susser E, Menezes PR. Work-related depression in primary care teams in Brazil. Am J Public Health. 2016;106(11):1990–1997. doi: 10.2105/AJPH.2016.303342. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.World Health Organization. Process of translation and adaptation of instruments guidelines. 2021. http://www.who.int/substance_abuse/research_tools/translation/en
- 10.Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16(9):606–613. doi: 10.1046/j.1525-1497.2001.016009606.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.de Lima Osório F, Mendes AV, Crippa JA, Loureiro SR. Study of the discriminative validity of the PHQ-9 and PHQ-2 in a sample of Brazilian women in the context of primary health care. Perspect Psychiatr Care. 2009;45(3):216–227. doi: 10.1111/j.1744-6163.2009.00224.x. [DOI] [PubMed] [Google Scholar]
- 12.Munhoz TN, Nunes BP, Wehrmeister FC, Santos IS, Matijasevich A. A nationwide population-based study of depression in Brazil. J Affect Disord. 2016;192:226–233. doi: 10.1016/j.jad.2015.12.038. [DOI] [PubMed] [Google Scholar]
- 13. Diagnostic and Statistical Manual of Mental Disorders . 2000.
- 14.Coutinho LMS, Scazufca M, Menezes PR. Methods for estimating prevalence ratios in cross-sectional studies. Rev Saude Publica. 2008;42(6):992–998. doi: 10.1590/S0034-89102008000600003. [DOI] [PubMed] [Google Scholar]
- 15.Khajuria A, Tomaszewski W, Liu Z, et al. Workplace factors associated with mental health of healthcare workers during the COVID-19 pandemic: an international cross-sectional study. BMC Health Serv Res. 2021;21(1):262. doi: 10.1186/s12913-021-06279-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Zhu Z, Xu S, Wang H, et al. COVID-19 in Wuhan: sociodemographic characteristics and hospital support measures associated with the immediate psychological impact on healthcare workers. EClinicalMedicine. 2020;24:100443. doi: 10.1016/j.eclinm.2020.100443. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Menon V, Padhy SK, Pattnaik JI. Stigma and aggression against health care workers in India amidst COVID-19 times: possible drivers and mitigation strategies. Indian J Psychol Med. 2020;42(4):400–401. doi: 10.1177/0253717620929241. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Ñamendys-Silva SA. Health care workers on the frontlines of COVID-19 in Mexico. Heart Lung. 2020;49(6):795. doi: 10.1016/j.hrtlng.2020.09.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.McKay D, Heisler M, Mishori R, Catton H, Kloiber O. Attacks against health-care personnel must stop, especially as the world fights COVID-19. Lancet. 2020;395(10239):1743–1745. doi: 10.1016/S0140-6736(20)31191-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.di Martino V.2021. https://www.who.int/violence_injury_prevention/injury/en/WVsynthesisreport.pdf
- 21.Li F, Luo S, Mu W, et al. Effects of sources of social support and resilience on the mental health of different age groups during the COVID-19 pandemic. BMC Psychiatry. 2021;21(1):16. doi: 10.1186/s12888-020-03012-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Ma H, Qiao H, Qu H, et al. Role stress, social support and occupational burnout among physicians in China: a path analysis approach. Int Health. 2020;12(3):157–163. doi: 10.1093/inthealth/ihz054. [DOI] [PubMed] [Google Scholar]
- 23.Hogarth KM, Beattie J, Morphet J. Nurses’ attitudes towards the reporting of violence in the emergency department. Australas Emerg Nurs J. 2016;19(2):75–81. doi: 10.1016/j.aenj.2015.03.006. [DOI] [PubMed] [Google Scholar]
- 24.Baba VV, Galperin BL, Lituchy TR. Occupational mental health: a study of work-related depression among nurses in the Caribbean. Int J Nurs Stud. 1999;36(2):163–169. doi: 10.1016/S0020-7489(99)00002-4. [DOI] [PubMed] [Google Scholar]
- 25.Shanafelt TD, Balch CM, Bechamps G, et al. Burnout and medical errors among American surgeons. Ann Surg. 2010;251(6):995–1000. doi: 10.1097/SLA.0b013e3181bfdab3. [DOI] [PubMed] [Google Scholar]