The COVID-19 outbreak was declared a Public Health Emergency of International Concern (PHEIC) on 30 January 2020 by the World Health Organisation, retaining its status for three years till the declaration was lifted on 5 May 2023. In the initial weeks after COVID-19 was first detected in countries around the globe, citizens were engulfed in panic and trepidation as news of spikes in the number of new cases, hospitalisation rates and death tolls flooded the media. Hospitals were overburdened and medical resources ran low. In order to mitigate transmission of the virus and prevent medical systems from reaching breaking point, stringent physical distancing policies such as school and workplace closures, travel restrictions, quarantine, gathering size limits and at the forefront of these, lockdown was imposed in many countries.

These containment measures would have severe economic, health and social impacts. Governments were faced with difficult trade-offs among these factors (1). As regulations were adjusted with the evolving situation, civilians and institutions altered their routines and operational procedures to comply with them. In addition to the need to acclimatise to the new regulations and the dynamics it brought, individuals battled with the constant negative presence of COVID-19 on media, prolonged social deprivation and other stressors that had no foreseeable end in sight, taxing them physically and mentally, causing burnout over time (2, 3).

The concept of burnout was defined by Maslach and Jackson as a psychological syndrome characterised by emotional exhaustion, feelings of cynicism or depersonalisation and reduced personal accomplishment (4). Majority of the research that examined pandemic burnout focused on frontline healthcare workers who faced the harshest impact of this crisis attributed to long working hours without adequate rest, torment from having to make life-and-death decisions hastily, and pain of losing their patients and colleagues (5). Where burnout was originally conceptualised in the workplace context, it has expanded to other chronically stressful situations (6).

Within homes, burnout was recognised to be heightened among young parents who were struggling with full-time parenting and home-schooling while simultaneously working from home (7). Those who were not parents also reported struggles with work–family conflict, in the form of increased interruptions and distractions due to the presence of family and additional household responsibilities. Adding to that burden, there were expectations—whether real, perceived, or self-imposed—regarding one’s availability to respond quickly during remote working as individuals were assumed to be confined to the home. Combined with the impression that saved commuting time could be used to extend work hours, workers found themselves under increased pressure (8).

High levels of burnout have been shown to be significantly associated with depression, anxiety, and insomnia (9). It has also been associated with weakened immune functioning (10). While some individuals suffer the impact of the pandemic more acutely, others appear to cope better. According to the Transactional Model of Stress, individuals respond to the same stressor differently due to individual characteristics and contextual factors (11). Resilience is an individual response to adverse situations that has been found to mitigate the detrimental effects of stress on burnout (12). Resilience refers to the “ability to bounce back quickly” from highly distressing events (2). Smith et al. proposed that resilience involves confronting (rather than avoiding or denying) the stressful event, orienting oneself towards a future positive outcome of the event and actively engaging in efforts to cope with it (13). As an example, a study among emergency response workers during the second COVID-19 wave in Italy showed that problem-focused coping, being able to get past negative feelings and a strong sense of purpose appeared to offer workers protection against burnout (14).

A contextual factor that has been found to buffer the negative effects of stress from adverse life events on mental and physical health is social support (15). Nitsche et al. reported that greater social connectedness during the lockdown period was associated with lower levels of perceived stress as well as general and COVID-19 specific worries (16). According to the stress-buffering hypothesis, the more social support or resources a person has or perceives to have available, the more likely the individual is to feel in control of the stressful situation. Perceived social support has been theorised to prevent a situation from being appraised as highly stressful whereas received support has been theorised to cushion the impacts of stress by assisting with coping (17). Several studies have in turn shown that social support can be a key mechanism in bolstering resilience (18, 19). Conversely, individuals who lacked social interaction were hypervigilant to threats and had higher negative appraisal of threats resulting in increased overall stress (17).

Singapore is a small metropolitan city-state measuring 728.6 km² with a population of 5.64 million (20). It is regarded as one of the medical hubs in the Asia Pacific region (21) and billed itself as an efficient business city with one of the busiest shipping ports in the world (22). While Singapore’s healthcare system remained resilient during the pandemic and its nationwide vaccination campaign was a success with 80% of the population being fully vaccinated by the end of August 2021, a survey across six Asian countries stated that it was the only country to report burnout as the leading factor affecting mental health of residents during the pandemic (3). This contrasted with Japan, South Korea, and Hong Kong whose residents were most affected by public measures to keep the pandemic under control such as mask-wearing and travel restrictions, while Malaysia and Indonesia were most affected by financial burden due to income loss (3). The uniqueness of this finding to Singapore could be a reflection of the competitive and efficiency-driven culture which was embodied even amidst the pandemic.

The aims of this study were to identify the extent of burnout in the population, sociodemographic correlates of the population that were most affected by burnout and understand the relationships between COVID-19 related stress, burnout, resilience, and social support. Based on Lazarus and Folkman’s Transactional Model of Stress and earlier studies that demonstrated the associations between stress, burnout and resilience (23–25), we hypothesized that (i) greater perceived COVID-19 stressors predict higher levels of burnout and (ii) the impact of perceived COVID-19 stressors on burnout is mediated by resilience. Additionally, in line with the stress-buffering hypothesis and earlier studies showing that social support acts as an effective mechanism to boost resilience (19, 26, 27), we hypothesized that (iii) greater perceived social support predicts lower levels of burnout, and (iv) social support increases individual resilience.

The sociodemographic profile and classification of the sample by Personal-and Work-related burnout severity are presented in Table 1. Mean scores on the resilience, social support, COVID-19 stress and Personal-and Work-related burnout scales and their respective standard deviation are summarised in Table 2.

Twenty-two percent (22%) of the sample reported elevated levels of work-related burnout which was slightly lower than the proportion found among non-clinical staff of a community mental health service assessed in 2019 prior to the outbreak (25%) (46). The percentage of those who reported elevated personal-related burnout in our study was even lower at 19%. While we were not able to identify other studies conducted in the general population for comparison, our mean burnout scores (Personal: 31; Work-related: 31) were lower than those among samples working non-clinical jobs reported in other countries during the pandemic. To illustrate, a study in Thailand among librarians reported Personal and Work-related burnout scores of 44 and 42 (47) while another among teachers in Ireland reported scores of 65 and 61, respectively (48). A possible explanation offered by See et al. who observed lower burnout rates among physicians in Singapore and Hong Kong (31 and 31%) compared to their counterparts in the US (45–55%) pre-pandemic, was that local work culture and values including collectivism, persistence and guanxi (respecting social orders and protecting others’ reputation) could have blunted overall self-reports of burnout (31). This explanation however conflicts with the survey findings alluding to the high level of burnout in Singapore compared to other countries (3). Likely, burnout symptoms may have alleviated as our study extended till September 2022 when majority of the safety measures were relaxed.

However, as predicted, higher personal-and work-related burnout was observed among those who experienced greater COVID-19 related stress during the second year of the pandemic. Longitudinal studies have shown that excessive and prolonged stress that is not ameliorated leads to feelings of physical and mental exhaustion, cynicism and depersonalisation, and low personal efficacy, which are the hallmarks of burnout (49). Additionally, the impact of COVID-19 stressors on elevated burnout levels was partially accounted for by reduced resilience, corroborating earlier research (23, 25, 50, 51). Our finding also extends the relevance of this mediational relationship that has largely been documented in workplace settings among frontline staff such as doctors, nurses and police officers to those in the home and community settings.

Resilience has been described as mental fortitude to navigate unpleasant challenges or positive adaptation to adversity (51, 52). The Conservation of Resources (COR) theory asserts that individuals strive to acquire, maintain and protect valuable resources (e.g., finances, health, energy) and experience stress when access to essential resources are threatened. Based on this theory, resilience constitutes a personal resource that supports an individual’s ability to bounce back from negative emotional states, flexibly adapt to the changing demands of stressful experiences, allowing the individual to recoup their resources (53). For instance, in the context of the pandemic, resilient individuals may set boundaries to manage their new work-life arrangements or adjust their financial goals while seeking alternate sources of income. Some attributes of resilient individuals include optimism, hope, enthusiasm, and the ability to develop meaning from hardship (54). The resilience trait thus creates a positive feedback loop that sustains a “gain spiral” preserving an individual’s well-being during challenging times (55).

Our findings further supported the hypothesis that those with higher perceived social support will be less burnt out both on the personal and work fronts and that resilience has a role to play in these relationships. Our findings were similar to that of Shang and Yang (26) who demonstrated that higher social support was associated with lower athlete burnout and this relationship was partially mediated by resilience. Our results espouse assertions that resilience can be augmented by supportive social environments. When dealing with a crisis such as the COVID-19 epidemic, effective social support may reduce respondent’s worries in a similar way that illness support groups do. When respondents had support networks to confide in, their concerns and worries can be better understood and soothed. The individual could also obtain strength, confidence and inspiration from others undergoing similar challenges as them, thereby reducing distress and burnout in the long run (56).

The protective effects of resilience in enhancing well-being and preventing psychological morbidity is well-established and resilience training programmes have gained popularity in the recent years. Presently, resiliency training is a loosely defined set of interventions aimed at enhancing resilience through a range of therapeutic approaches such as mindfulness, stress-management or cognitive behavioral techniques and may feature explicit teaching of emotion-regulation, optimism and self-efficacy (57). Although the lack of standardisation of resilience programmes has been criticised, there is no “one size fits all” formula as different aspects of resilience could be differentially valued based on factors such as nature of the stressor, culture, life-stage etc. Nevertheless, in regard to the pandemic, Kauderer et al. recommended that interventions could include effortfully maintaining positive emotions even in the presence of negative ones (e.g., seeking gratitude for what has not been), reframing negative circumstances (e.g., conceptualising quarantine as a chance to pick up a new skill), maintaining social connectedness (e.g., virtually with restrictive measures) and practicing spirituality (e.g., turning to religion for guidance or engaging in meditation) (57). Moving further upstream, efforts to improve community resilience such as continuous investment in public mental health surveillance and programmes, clear and up-to-date accessible communication between government agencies and the public to reduce anxiety and confusion and cultivating a culture of strong community support and cohesiveness can better prepare societies for future outbreaks (58).

Our study also showed that younger age was associated with both personal and work-related burnout with the 21–35 years age group reporting the highest level of burnout. Various studies have reported comparable age trends (59–61). Huang et al. reasoned that age affects how individuals perceive and cope with stress; as one matures, they acquire traits and psychological capital that improve their resilience towards stressors. They may also become more skilful in rallying support that can buffer the impact of stress (61). Furthermore, the 21–35 years age group are a part of the “sandwich generation” that are caring for both young children and older adult parents, who are more vulnerable to the virus, a point earlier identified by other researchers (59, 62). In addition, those in their 30s are likely to be middle managers who have been reported to be most stressed during the pandemic, as a result of increased complexities in their work that could jeopardise their position in the organisation or career aspirations, in contrast to those 40 years and older who have achieved more stability in their careers (3, 50).

Next, we found that individuals who were recently unemployed demonstrated higher levels of work-related burnout compared to those who were employed. Safety measures that persisted in the second year of the pandemic gave rise to and exacerbated stressful work environments (63). Healthcare sectors continued to have high caseloads whereas industries such as aviation and hospitality were forced to make budget cuts that added to workloads, resulting in reduced job satisfaction (64, 65). Additionally, increased virtual meetings due to remote working and screen time led to “Zoom fatigue” and digital exhaustion (66), while the absence of in-person interaction led to isolation and feeling of lack of support. These factors have been associated with burnout and turnover intentions (64, 65, 67). Thus, it is possible that such factors could have caused individuals in our study to resign, explaining the association between recent unemployment and higher work-related burnout. Indeed, another local survey among 1,002 workers in Singapore aged 16–55 years reported that 46% experienced increased stress, 44% perceived heavier workloads, 33% felt more burnout, 20% felt isolated, 49% realised that they do not like their current job and 24% planned to leave their current employer in the next 6 months (68) mirroring the trend observed in the United States dubbed the Great Resignation where monthly resignations in 2021 were the highest in country’s 20-year history (67). Various studies have emphasized the importance of providing employer, peer and job support to reduce burnout and turnover intentions (69).

Finally, it was observed that those with secondary school education (compared to university) and Malay ethnicity were associated with higher personal burnout. Individuals with lower education levels tend to occupy non-PMET (Professionals, Managers, Executives, Technicians) vocations, that are generally lower paying and face higher job insecurity. Similarly, local reports indicate that those of Malay ethnicity are concentrated in lower rung vocations or in sales and service industries that tend to be most affected by the outbreak (70). These groups may face the impact of the pandemic more severely due to financial instability, higher risk of exposure to the virus, a lack of resources and conducive home environment to manage home-based learning, less help with childcare, and means to afford medical care among other stressors (58), highlighting widening social inequalities during this period.

In Singapore, national reserves and past budgetary savings allowed the government to introduce a series of initiatives swiftly to protect jobs, support households and companies. For instance, the Workfare Special Payment, Self-employed Person Relief Scheme and COVID-19 Support Grant were implemented to provide income support and alleviate financial hardship. Efforts from various government agencies were also aimed at creating jobs and re-skilling workers (58). In contrast, mental health interventions and responses were inadequate and slow, placing a spotlight on investments needed in mental health infrastructure, programmes, and research to enhance and accelerate the public’s psychological preparedness for adversity. Recently, the National Mental Health and Well-being Strategy was launched to tackle mental health issues. The strategy comprises not only an expansion of mental health services but the promotion of well-being through whole-of-society efforts that involve an individual’s microsystem such schools, workplaces, healthcare, and social services. The new strategy also involves identifying varying mental health needs in the population using a tiered care model and emphasizes a preventative approach to improving mental health (71). Continued tracking of the nation’s psychological health is needed to better understand the aftermath of the pandemic and assess the impact of the newly launched strategy.

The data supporting the conclusions of this article will be made available, upon request to the corresponding author.

The studies involving humans were approved by National Healthcare Group, Singapore Domain Specific Review Board (Ref: 2021/00566). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

ShS: Conceptualization, Investigation, Project administration, Writing – original draft, Writing – review & editing. ET: Formal analysis, Investigation, Writing – review & editing. SaS: Project administration, Supervision, Writing – review & editing, Methodology. YT: Project administration, Writing – review & editing. SG: Project administration, Writing – review & editing. RT: Project administration, Writing – review & editing. PS: Project administration, Writing – review & editing. YZ: Project administration, Writing – review & editing. PW: Project administration, Writing – review & editing. ST: Project administration, Writing – review & editing. MS: Conceptualization, Funding acquisition, Investigation, Methodology, Supervision, Writing – review & editing.