Study site

This study was conducted with healthcare workers and other hospital-based staff based in the psychiatry departments of three public tertiary hospitals (Charlotte Maxeke Johannesburg Academic Hospital, Chris Hani Baragwanath Academic Hospital, and Helen Joseph Hospital) and two specialized psychiatric hospitals (Sterkfontein Psychiatric Hospital, and Tara Psychiatric Hospital) located in southern Gauteng.

Recruitment & participant selection

Participants were recruited based on the following criteria. Inclusion criteria included: 18 years and older, worked in one of five psychiatric facilities during the COVID-19 pandemic, and were able to provide informed consent. Exclusion criteria included: individuals who were not able to consent due to perceptual or cognitive disabilities. To reduce infection risk, participants were recruited through online means (e.g. email) and through word-of-mouth, using convenience and snowball sampling methods. Additionally, a minimum of five individuals for each of the following hospital divisions were recruited to gather perspectives from each institution: consultants (i.e. head psychiatrists), staff psychiatrists, allied health workers (e.g. psychologists, occupational therapists, social workers), nurses, administrative staff. Data were collected until data saturation for each of the major research themes was obtained (e.g. stress, coping, perceptions of COVID-19, etc.). A minimum of 24 interviews was needed to obtain meaning saturation [34]. Finally, participants who wished to take part in data collection were not turned away as many hospital staff members requested the opportunity to share their experiences under the lockdown and reflect on the challenges of providing healthcare during the pandemic. A major fire in one of the hospitals required redeploying patients and healthcare staff across the public mental healthcare system in the region, which severely limited the availability of participants and prompted our team to end data collection.

Data collection

Interviews included a short survey of socio-demographic characteristics, after which participants were asked open-ended questions that assessed experiences with service delivery during the COVID-19 pandemic, sources of stress and coping, and changes in clinical care across the course of the pandemic. Most interviews were conducted telephonically and ranged between 45 minutes to three hours, with a mean of about 1.25 hours. Research assistants administered the interview using the preferred languages of the participant, which included English, isiXhosa, isiZulu. Interviews were conducted between July 2020 and March 2021. The study protocol was approved by the Human Research Ethics Committee at the University of the Witwatersrand, Johannesburg (Clearance number: M190545).

Data analysis

Interviews were recorded and extensively summarized into field notes which captured key ideas and experiences, and observable behaviors, cues, or interactions. Researchers transcribed and reviewed field notes concurrent to the data collection phase. This process was reflective and systematic, which provided a broader understanding of the data. Researchers then used field notes to develop and discuss key ideas and patterns with qualified qualitative research assistants who were closely connected with the data. Any identified interpretive discrepancies were discussed and resolved at this level. Constantly evaluating and re-evaluating the already grouped data helped the research team to come up with new research questions and observations that informed the next course of data collection to a point where no new themes emerged.

Data from the field notes were coded and extracted using a constructivist research paradigm paired with content analysis [35] and Dedoose 9.0.17 software. Coders both had doctoral degrees in psychology and had focused their training on psychological resilience. Methodological integrity was assured through constant reflexivity and engagement with the literature, as well as through regular discussions about coding patterns, discrepancies, and possible questions. An initial codebook was developed a priori and deductively, based on prior theory and previous publications [12]. Additional codes were added inductively, after an initial read of the data. Coded data were then extracted and thematically analyzed to identify coping strategies.

Intrapersonal coping

Intrapersonal factors in coping include any psychological, mental, or emotional factors within an individual which foster or reinforce the ability of the participant to maintain well-being or continue functioning in daily life. We found five themes of intrapersonal coping: self-reliance, mindset and reappraisal, acceptance, manifestations of agency, and refocused attention.

Material coping

Participants used material resources to protect or support themselves physically, mentally, and emotionally, at home and at work, though some everyday concerns worsened during the pandemic.

Participants attempted to keep healthy by using vitamins, minerals, herbal remedies such as umhlonyane (or African wormwood), and traditional healing practices including ginger, garlic, eucalyptus, and steaming. A few participants who fell ill attempted to relieve symptoms by using flu or allergy medicine, painkillers, or antibiotics. Combinations of practices were also mentioned: “I was drinking umhlonyane, taking vitamin C, and Allergex” (30-year-old participant).

COVID-specific protections, including masks, sanitizer, and household cleaners, were also important. Some participants regularly cleaned groceries after purchase or bathed before entering their homes from work. Vaccine development was a source of comfort for some participants, “We won’t get the COVID if you take the vaccine and then, you know, you won’t infect other people as well, around you, in your household, you know” (an older participant, age unknown). This comfort, though, was tempered with a concern for the safety and effectiveness of the vaccines in development. COVID-19 testing and PPE were a source of comfort in the instances when it was available. For a minority of participants, the lockdown protocols were comforting, though most participants found them challenging.

Many participants found masks and mask requirements an important method for reducing transmission; in fact, masks were mentioned more frequently than any other COVID protection measure: “You must always wear a mask and to keep social distance” (44-year-old participant). Community members who were not following the rules became a source of stress. Social distancing and restricting movement, particularly in the hospital, was comforting to some participants, who then began to rely on the outdoors as both a method to social distance and a source of emotional and mental support. Finally, use of glass or plastic barriers and thermometers increased sense of safety for some.

Technology became a key resource, and social media groups flourished among healthcare workers both as socioemotional support systems and for communication purposes for healthcare workers: “We had a spiritual wellness group that I was coordinating at work. We used to send the weekly recording to the group, uplift them and then at home over the week also, we would connect and stream online services” (45-year-old nurse). Participants used these groups to keep in touch with family and friends, to show support for colleagues who were ill, or as a means of expressing solidarity. Social media groups provided a mode of communication at work, becoming one method for giving or receiving advice or updating guidelines or protocols. Hospitals also used informational emails and online meeting platforms to connect healthcare workers and management. Online meeting platforms were also used to attend religious services or write exams. A few participants also sought teletherapy for psychological support or counseling.

Lastly, some common or everyday resources changed in valence. Though many participants worried about finances, a minority expressed gratitude for receiving financial assistance, data, or airtime, or for having savings or investments to use during the pandemic: “The university, they gave us free data so we can study from home” (37-year-old male participant). Groceries could be a source of stress when they were tied to finances or when healthcare workers became ill with COVID-19, had to isolate, and were unable to shop. The provision of food became a support, though, when healthcare workers were gifted food hampers at work or received groceries from friends or family while they were ill and could not shop.

Common coping mechanisms for healthcare workers

Related to previous findings of cognitive reappraisal, attitude, positive attitude, or positive framing [15, 16, 18, 26, 30], healthcare workers in this study used mindset and reappraisal as emotion-focused coping strategies [26] to promote optimism and hope. The use of positive mindset and reappraisal within our sample is particularly important, as both may influence selection of other coping strategies and therefore encourage stress optimization as opposed to maladaptive coping techniques [17, 23], which could be leveraged in the development of pandemic-focused interventions [24]. Additionally, making meaning is associated with lower levels of PTSD [22] and so can directly impact mental health in adverse situations. Acceptance [13, 27, 28], self-efficacy [13, 30] and refocusing attention to activities such as exercise or self-distraction [27, 33] have also been used as coping mechanisms by healthcare workers during this and previous pandemics. Refocused attention may be an effective form of emotional regulation [26] and self-efficacy is associated with greater problem-focused coping and job satisfactions [41], whereas acceptance–when used mindfully–may promote both problem-focused engagement and cognitive reappraisal [42, 43]. Interestingly, “deliberate rumination” reminiscent of the introspection mentioned by two participants was found to be associated with post-traumatic growth in one study [16]. It is possible that, to the extent that introspection as a form of self-observation and awareness is related to the process of mindfulness [40, 42], it may encourage problem-focused coping [42].

Social support was one of the most endorsed coping strategies [15], which is notable in the role that relationships may play in decreasing the risk of mortality [44] and mental health issues [45, 46]. Family support specifically has been linked with increased active coping [47] and decreased emotional exhaustion [48], perhaps providing an explanation for its importance in our sample. Support from families operated through multiple pathways and served different purposes, and the flexibility and adaptability displayed in seeking and maintaining relationships may be indicative of the importance of social support systems in times of stress [47]. Additionally, comradery fostered between co-workers provided both emotional support and functional adaptations such as sharing or swapping shifts, which are associated with improved wellbeing through decreased burnout [49] and mental health issues [50, 51], and increased job retention [19, 52]. Communication between colleagues provided interpersonal connection and knowledge, which is associated with increased confidence [7, 13–15, 19, 28, 30, 31]. Good communication and collaboration with management, described as being transparent, clear, and informative were also a source of support, possibly providing a buffer between traumatic stress at work and burnout or intent to resign [13, 51].

Contextual and subjective nature of coping strategies

Participants disagreed on the value of some potential resources, which reflects the contextual and subjective nature of coping [17, 18]. Environmental demands [18] and access [53, 54] play a large role in determining whether a material resource mediates a coping or stress response [55]. Socioeconomic status affected coping by influencing whether participants felt stress, anxiety, and worry about financing groceries and transportation. Access similarly influenced perspectives on PPE, technology, and communication–when available and adequate they were supportive, but their absence resulted in stress.

The role of families were simultaneously supportive and stressful, and previous findings have suggested that families can be both sources of stress [48] and coping [47, 48] in nurses trying to achieve a work-life balance. Though families were generally discussed positively, the concerns felt by the participants for family members’ health and safety were serious and added nuance to the supportive nature of families, which itself was qualified by frustrations that lockdown was preventing in-person interaction. Participants who lived with their families had the comfort of in-person support but were also worried about transmitting COVID-19 from work to their families. Families could be a source of financial stress, as participants took in family members who had lost employment because of the government lockdowns. It appears that, similar to previous findings [48], family support provided more emotional coping during the pandemic than before, but that the pandemic also increased family-related stressors.

Some factors seemed variably supportive, depending upon how they were conceptualized [20, 56] and operationalized [57]. Acceptance has been found to be supportive in healthcare workers during the COVID-19 pandemic [13, 33] but the results of this study suggest that nuance may be warranted, as acceptance was practiced by many participants but with different manifestations. Some participants accepted the newness and consequent likelihood of errors during the pandemic, allowing for mistakes and implying the ability to correct early errors. In fact, some participants described an adaptive acceptance [43], where they worked to understand the situation, assess available resources, and determine best practices for managing outcomes due to the pandemic. Other participants described a passive acceptance [58] that the pandemic was the new reality and therefore expressed a desire to live life as usual, without focusing on the virus. In a minority of interviews, participants fatalistically accepted the inevitability of contracting the virus. This is in line with recent literature which suggests that acceptance may be misused, leading to maladaptive and counterintuitive suppression of emotions and denial of their impact [42]. The nature of passive or fatalistic coping is in opposition to the open reflection required for adaptive acceptance [40] and can be difficult to interpret. Van Breda [58] noted different levels of protective coping afforded through different types of acceptance within the South African context, suggesting that passive acceptance may be associated with helplessness and diminished agency or self-efficacy. Additionally, fatalism could discourage the use of protective factors such as masks, and future research should clarify the different understandings and uses of acceptance in relation to effective coping.

Interconnectedness of coping mechanisms

This study suggests a high degree of interconnectedness between and within systems and levels of coping, while also demonstrating the flexibility of coping strategies through the integrated use of higher order adaptive processes [20]. These systems and processes do not operate in isolation and can function to either strengthen or weaken a global support network [59]. Most notable is the effect of work on participants’ intrapersonal attitudes during the pandemic, with some participants expressing gratitude for employment and others finding personal fulfillment and value in their work, which helped to buffer them against stigma. As mentioned above, these reappraisals and positive mindsets may be of particular interest as a central factor connecting other coping strategies [23], such that a positive mindset may in turn influence healthcare workers sense of agency or interpersonal interactions both at home and at work. Meaning making has also been found to be associated with active acceptance [42, 43, 60] which may be related to distraction [20], illustrating the complexity of coping networks. Additionally, strong teamwork and clear communication decreased stress by allowing individuals to express themselves while also providing planning, structure, and pandemic training, which in turn increased confidence facilitated work between colleagues. Future research should further explore these interactions between coping strategies, their impact on the coping network, and the association between these intra-network interactions and stress outcomes in order to inform the development of interventions to support healthcare workers.