Discussion

This study evaluated the psychological impact of COVID-19 in a Mexican sample of FHCWs attending COVID-19 patients. Our findings showed that the prevalence of depression, generalized anxiety, post-traumatic stress, and insomnia symptoms were 36%, 21%, 23%, and 24%, respectively. In multivariate analysis, women, residents/fellows and nurses, and being married were significant risk factors for depression. Although not significant, residents/fellows were more likely to experience any symptom compared with nurses.

A month after the first COVID-19 cases were reported in Wuhan (January 29 to February 3, 2020), Lai et al. performed a survey-based study evaluating the mental health outcomes among Chinese frontline- and non - FHCWs exposed to COVID-19 patients. ^{Reference Lai, Ma and Wang4} Compared to our study, the authors reported a lower prevalence of depression (18% vs. 36%), anxiety (16% vs. 21%), and insomnia (12% vs. 24%) among FHCWs. However, the prevalence of post-traumatic stress was higher (42% vs. 23%). ^{Reference Lai, Ma and Wang4} The timing of participant recruitment (early vs. second wave) might explain these results. Several repeated cross-sectional studies report that the prevalence of depression consistently increased from the first to the second wave of the pandemic. In contrast, other mental health symptoms, including anxiety, insomnia, and post-traumatic stress, showed variable changes between each wave of the COVID-19 pandemic. ^{Reference Magnavita, Soave and Antonelli24–Reference Obeidat, Dodin and Hawari26} The type of occupational adversities faced by FHCWs varied during the pandemic course. The uncertainty associated with the management of COVID-19 patients, the lack of readiness and safety measures, ^{Reference Magnavita, Soave and Antonelli24} fear of infection and stigma, ^{Reference Lai, Ma and Wang4} and separation from families were prevalent during the early stages of the pandemic. In contrast, the death of relatives, colleagues, and patients and the prolonged workload and isolation were everyday stressors during later stages. ^{Reference Jáuregui Renaud, Cooper-Bribiesca and Martínez-Pichardo25}

Robles et al. performed a multicenter cross-sectional study to evaluate mental health problems among Mexican FHCWs providing care to patients in COVID-19 and non-COVID-19 centers. ^{Reference Robles, Rodriguez and Vega-Ramirez6} This study was conducted during the first phase (April 17 to May 7, 2020) of the pandemic in Mexico, defined by the WHO as the time interval when a country experiences cases related to well-defined clusters associated with everyday exposure, geographical location, and time. ^{Reference Robles, Morales-Chaine and Bosch5} A second study was conducted by the same group by the second phase (May 8), described by the Mexican Undersecretary as the community-spread stage in Mexico, and August 18, 2020, when according to the Undersecretary, the situation in the country was ‘dreadful.’ ^{Reference Robles, Morales-Chaine and Bosch5} Between the first and second study, the prevalence of depression significantly increased (38% vs. 43%), whereas the rates of anxiety (16% vs. 15%) and post-traumatic stress (38% vs. 40%) did not significantly change. ^{Reference Robles, Morales-Chaine and Bosch5,Reference Robles, Rodriguez and Vega-Ramirez6} After analyzing the mental health outcomes among FHCWs recruited after the previous studies (August and November, 2020), we report a somewhat lower rate of depression (36%), a higher rate of anxiety (21%), but a significantly lower rate of post-traumatic stress (23%). The fact that different scales and cut-off values were used to measure mental health symptoms between both studies and that participants in our study were all providing care at a converted-to-COVID-19-hospital with all the safety measures required, might have led to lower rates of depression and post-traumatic stress.

A meta-analysis including Asian studies performed by Pappa et al. reported that women and nurses providing care to COVID-19 patients exhibited higher rates of depression, anxiety, and insomnia than men and medical staff, respectively. ^{Reference Pappa, Ntella and Giannakas27} These results are supported by a systematic review of the psychological impact of the COVID-19 pandemic on Western FHCWs. ^{Reference Danet Danet28} Nevertheless, most studies did not perform a sub-analysis comparing the prevalence of symptoms between residents and attending (or primary) physicians. ^{Reference Pappa, Ntella and Giannakas27,Reference Danet Danet28} In the United States, Schecter et al. also reported a higher prevalence of mental health symptoms among nurses compared to attending physicians, residents, and fellows. ^{Reference Shechter, Diaz and Moise29} Contrariwise, both studies performed in Mexican HCWs by Robles et al., report lower psychiatric symptoms among nurses and attending physicians compared with medical residents and general practitioners. ^{Reference Robles, Morales-Chaine and Bosch5,Reference Robles, Rodriguez and Vega-Ramirez6}

In this study, we found that nurses and residents/fellows experienced significantly more depression and insomnia compared to attending physicians. The prevalence of depression among nurses was 46%. This is similar to the 44% prevalence of depression reported by Hu et al. among nurses providing care to COVID-19 patients in China. ^{Reference Hu, Kong and Li13} Studies suggest that the intense nature of the nurse-patient relationship might result in the high prevalence of burnout and mental health issues among nurses. ^{Reference Chou, Li and Hu30} Gray-Toft and Anderson propose 7 stressors among nurses, including excessive workload, conflict with other nurses, and lack of social support, as well as uncertainty concerning treatment administered, inadequate preparation, conflict with physicians, and death and dying of patients. ^{Reference Gray-Toft and Anderson31} Lou et al. reported that nurses experienced significantly more distress, burnout, and maladaptive coping strategies than physicians during the COVID-19 pandemic, including substance use, denial, behavioral disengagement, and self-blame. ^{Reference Lou, Montreuil and Feldman32} An integrative review by Sastrawan et al. reported that nurses experience less mental health issues by establishing a culture of fostering mutual support among colleagues, promoting teamwork and empowerment, and using coping strategies to preserve integrity, including self-preservation, accepting limitations, and compensation. ^{Reference Sastrawan, Newton and Malik33} Thus, providing psychological and social support services endorsing adaptive coping strategies is necessary to prepare nurses for future public health crises. ^{Reference Lou, Montreuil and Feldman32,Reference Dobson, Malpas and Burrell34}

Even before the COVID-19 pandemic, multiple studies correlate medical residency with the development of psychiatric comorbidities among residents and fellows. ^{Reference Mendonca, Steil and Gois14} Pre-pandemic studies on Mexican residents report a lower prevalence of depressive and anxiety symptoms ranging from 25 - 80% and 39 - 70%, respectively. ^{Reference Dávila-Valero, Alfaro-Mejía and Padilla-Ochoa35,Reference Jimenez-Lopez, Arenas-Osuna and Depression36} Such a finding contrasts with a 2015 meta-analysis by Mata et al., reporting a lower pooled prevalence of depressive symptoms of 29% among resident physicians. ^{Reference Mata, Ramos and Bansal37} In the present study, we did not find any significant risk factors associated with endorsing mental health symptoms among residents and fellows. We propose several hypotheses as to why residents from Mexico experience more psychiatric symptoms both before and during the COVID-19 pandemic. First, mistreatment among medical residents in Mexico is highly prevalent. A study performed in Mexico revealed that 84% of medical residents experienced some form of abuse during their residency, mainly from residents of higher hierarchy and attending physicians. ^{Reference Derive, Martínez and Vera38} Second, a significant number of residents (58%) in our study reported working daily shifts of > 12 hours. Toral-Villanueva et al. reported that working > 12 hours per day tripled the risk of burnout in a study performed on Mexican residents. ^{Reference Toral-Villanueva, Aguilar-Madrid and Juarez-Perez39} Moreover, as reported by Derive et al., 64% of Mexican residents in their study report work overload as a form of punishment. ^{Reference Derive, Martínez and Vera38} Also, residents in Mexico earn 4 times less than US residents, which may eventually lead to emotional distress, mainly depression. ^{Reference Collier, McCue, Markus and Smith40} Finally, as of September 2020, Mexico topped the worldwide rank of COVID-19-related deaths among HCWs. At least 1320 Mexican HCWs died due to COVID-19, representing 1.8% of the total death toll reported at that time. ^{Reference Dominguez-Varela12} As it is obvious, the increased risk of contracting COVID-19 while at work, fear of transmitting the disease to family, and the social stigma among HCWs increased stress levels among medical personnel. ^{Reference Chu, Lee and Stotts41}

In the present sample, women, being single, and of young age (≤ 40 years) were deemed as significant risk factors for developing depression among FHCWs providing care to COVID-19 patients. This is consistent with the studies by Robles and coworkers reporting significantly increased depression, insomnia, and post-traumatic stress disorder in Mexican FHCWs who are single, women, and under 40 years of age. ^{Reference Robles, Morales-Chaine and Bosch5,Reference Robles, Rodriguez and Vega-Ramirez6} Hormonal influences and genetic, environmental, and social factors may contribute to the increased prevalence of mental health disorders in women compared to men. ^{Reference Kuehner42} Aside from their professional role as FHCWs, women are usually the primary caregivers within their families and, thus, experience more significant concerns than men about the health of their relatives. ^{Reference Rabbani, Khan, Piryani, Khan and Abid43} Moreover, evidence suggests women also experienced an increased workload related to attending to childcare needs, which in many countries is conducted by women in response to school closures. ^{Reference Borrescio-Higa and Valenzuela44} Therefore, we believe that sociocultural expectations could at least partially explain the increased prevalence of mental health symptoms in women. For future pandemics, we argue in favor of policies that alleviate financial stress, provide psychological support, and attend to the needs of women to reduce the risk of developing mental health problems. ^{Reference Borrescio-Higa and Valenzuela44}

Marital status also seems to play a role in developing psychological issues among FHCWs. We found a higher prevalence of depression among non-partnered FHCWs. Interestingly, most single FHCWs were women (72%) and under 40 years of age (93%). These findings were also reported in Mexico during the H1N1 pandemic (2009), where there was a higher risk of depression, anxiety, and stress among women and single people. ^{Reference Elizarraras-Rivas, Vargas-Mendoza and Mayoral-Garcia45} Teng et al. reported that marriage was significantly associated with severe depression among men FHCWs, whereas being single was related to severe mental health problems in women. ^{Reference Teng, Su and Chen46} Marital satisfaction enhances self-satisfaction and mental well-being in women and men, leading to less depression, anxiety, and stress. ^{Reference Abbas, Aqeel and Abbas47} However, women with distressing emotions are more likely to experience comfort from their children or the love of their spouses. ^{Reference Teng, Su and Chen46}

The experience of the previous SARS pandemic in 2005 showed that vulnerability and loss of control, the human-to-human transmissibility of the virus, self-health and health of others, and changes in work, as well as being isolated are potential sources of distress among HCWs. ^{Reference Wong, Yau and Chan48} Although the COVID-19 pandemic seems to be over, at least in many countries, mental health disorders developed during the pandemic among FHCWs continue to cause havoc. A large multicenter prospective cohort, the COVID-19 Health care wOrkErs (HEROES) study, which recruited over 34000 HCWs from 26 countries in Latin America, Europe, Africa, and the Middle-East, as well as Asia, aims to describe and track longitudinal trajectories of psychiatric symptoms and disorders among HCWs at different phases of the COVID-19 pandemic. ^{Reference Mascayano, van der Ven and Moro49} The study is ongoing, and countries are currently collecting data. Hopefully, this study will provide valuable information that allows healthcare systems and policymakers to develop evidence-based interventions to improve the mental health of HCWs. ^{Reference Melnyk, Kelly and Stephens50}