Mental health symptoms in Australian general practitioners during the COVID-19 pandemic

Keywords: anxiety, burnout, COVID-19, depression, general practitioners, mental health, primary care, survey.

Medical professionals can experience high levels of work-related stress (Riley 2004; Galbraith et al. 2021). Research shows doctors suffer higher rates of psychological distress, anxiety disorder, depression, suicidal ideation and occupational burnout when compared with the general population (Mata et al. 2015; Rotenstein et al. 2016; Ashraf et al. 2019). Intense work environments, fear of making mistakes, and difficulty balancing work and personal responsibilities are some common stressors related to poor mental health. The Coronavirus disease 2019 (COVID-19) pandemic poses an additional threat to mental health for healthcare workers globally. Being a frontline healthcare worker has been found in many studies to be an independent predictor of worse mental health outcomes, including depression, anxiety, post-traumatic stress disorder and insomnia during the pandemic (Krishnamoorthy et al. 2020; Sanghera et al. 2020; Seyed et al. 2020; Cénat et al. 2021; Smallwood et al. 2021a, 2021b, 2021c, 2021d; Thakur and Pathak 2021).

Notably, most studies examining the mental health impacts of the pandemic have focused on healthcare workers in secondary care (Digby et al. 2021; Hammond et al. 2021; Holton et al. 2021; Withiel et al. 2021), with the experiences and mental health of healthcare workers in primary health care (PHC) not to the fore. Yet, general practitioners (GPs) and other frontline healthcare workers in PHC have also faced considerable challenges during the pandemic, including having to manage workplace exposure and the consequent risk of contracting COVID-19 themselves or transmitting the disease to their family and friends (Cabarkapa et al. 2020; Tse et al. 2020; Siddiqui et al. 2021). Unlike hospital medical staff (HMS), who usually work at relatively large health organisations, Australian GPs work in community clinics of variable size, and may be responsible for the oversight and management of all aspects of the daily running of the clinic. Consequently, GPs have to cope with different workforce pressures and/or financial burdens (Scott 2020; The Royal Australian College of General Practitioners 2020). Furthermore, PHC workers have reported challenges and training needs related to adopting new telehealth practices and using personal protective equipment (PPE; Lee et al. 2020; Sotomayor-Castillo et al. 2021).

Having an effective PHC system is critical, especially in times of crisis, as PHC is known to potentially be the most cost-effective and efficient way to enhance patients’ physical and psychosocial health (PricewaterhouseCoopers Consulting Australia 2020; World Health Organization 2021). In countries like Australia and the UK, the PHC sector forms the fundamental basis of the national healthcare system (Kidd 2020b). Poor mental wellbeing in healthcare workers can affect the quality of patient care (Bauchner and Redberg 2020). It is therefore vital to understand the psychological health needs of our primary carers, so that targeted workplace and wellbeing support can be implemented.

The Australian COVID-19 Frontline Healthcare Workers Study investigated the prevalence and severity of workplace, social, and financial disruptions, and their impact on mental health, in Australian healthcare workers during the first year of the COVID-19 pandemic (Smallwood et al. 2021a). This paper reports a subset of findings from GPs and HMS. We aimed to investigate whether there are any differences between GPs and HMS in terms of their workplace challenges and stressors, the prevalence and predictors of mental health issues, and their coping strategies utilised during the COVID-19 pandemic. We hypothesised that the prevalence and predictors of mental illness, as well as the coping strategies adopted, would significantly differ between medical staff working in primary and secondary care.

The full study methodology has been published (Smallwood et al. 2021a). In brief, a nationwide, voluntary, anonymous, online survey was conducted between 27 August and 23 October 2020, concurrent with the second wave of the pandemic in Australia, which predominantly affected the city of Melbourne in the state of Victoria. At the time of this survey, strict and prolonged stay-at-home orders were in place in Melbourne, Victoria, with varying levels of restrictions nationwide. Restrictions on face-to-face consultations were in place during this time, and GPs were required to bulk bill all telehealth services (The Royal Australian College of General Practitioners 2022). Australian healthcare workers from all health roles, who self-identified as frontline health workers in secondary or primary and community care, were invited to participate. Information regarding the survey was widely disseminated across Australia by hospital leaders, professional societies, colleges, universities, associations, government health departments and the media. Completion of the survey took approximately 20 min.

Participants completed the survey either directly via the online survey link or via the study website (https://covid-19-frontline.com.au/). Data were collected and managed using REDCap electronic data capture tools (Harris et al. 2009). Online consent was obtained from the participants before commencement of the survey. Each participant completed the survey once.

Data collected included demographics, professional background and work arrangements, the impact of the pandemic on employment and finances, organisational leadership and workplace change, exposure to COVID-19, and health and recreational habits. Five validated psychological measurement tools were included: the Generalised Anxiety Disorder scale, Patient Health Questionnaire, abbreviated Impact of Event Scale, abbreviated Maslach Burnout Inventory and abbreviated CD-RISC-2 scale to measure resilience (Smallwood et al. 2021d).

Only data from HMS and GPs were included in this analysis. All analyses were performed using SPSS version 26.0 (IBM Corp., Armonk, NY, USA) or SAS version 9.4 (SAS Institute, Cary, NC, USA). Demographic and socioeconomic characteristics were reported descriptively. For each mental health scale, outcomes were merged into dichotomous categories (none-to-mild or moderate-to-severe symptoms). The associations between medical staff role (GP or HMS) and predictor variables, including workplace changes, training and confidence, impacts on relationships, and mental health outcomes, were assessed using Chi-squared tests with results presented as counts and percentages. Likert scale measures of confidence and resilience were reported as means and standard deviations, and were compared using the independent samples t-test.

Predictors of mental illness symptoms and coping strategies were identified through logistic regression models. Covariates examined in univariable analyses included: age; sex; state; lives alone; lives with children; lives with elderly people; frontline area; practice location; works with COVID-19 patients; anticipates working with COVID-19 patients; received PPE training; concern of transmitting COVID-19 to family; worried about being blamed by colleagues, close friends or relatives infected with COVID-19; changed relationships with partner or friends or family or colleagues; changed household income; household income concerns; and pre-existing mental illness. Multivariable models were developed using stepwise selection and backward elimination procedures before undergoing a final assessment for clinical and biological plausibility. Variables with a P < 0.10 on univariable analyses or those deemed to be clinically relevant were considered for inclusion in the multivariable models. Results from the regression models are presented as odds ratios (OR) with 95% confidence intervals (95% CI) and forest plots. A two-sided P-value <0.05 was chosen to indicate statistical significance.

Online consent was obtained from the participants before commencement of the survey. The Royal Melbourne Hospital Human Research Ethics Committee provided ethics approval for the study (HREC/67074/MH-2020).

Of 9518 responses to the survey, 389 (4%) GPs and 1966 (21%) HMS provided complete responses (Table 1). There were significant differences in age (P < 0.001), sex (P < 0.001), physical health (P = 0.003) and home caring responsibilities (P = 0.005) between GP and HMS participants. GPs worked predominantly in primary care (86%) with some also having roles in hospital departments or COVID-19 testing. HMS most commonly worked in medical specialty areas (53%), emergency departments (17%), anaesthetics and surgery (17%), and intensive care (10%). Both GPs and HMS experienced increased working hours; however, more GPs (35%) than HMS (29%) reported increased unpaid hours (P < 0.001; Table 2). Significantly more GPs (60%) experienced a reduction in household income (P < 0.001) or had income concerns (P < 0.001) compared with HMS.

Table 1.  Participant demographics and relationships.

Table 2.  Workplace environment and exposure to COVID-19.

Only 31% of the GPs received any training to care for COVID-19 patients or use PPE, in which they lacked confidence (Table 2). GPs were significantly less likely than HMS to feel that their mental health was supported by their workplace (P < 0.001), or report that they received useful and timely communication from their workplace (P < 0.001). Significantly more GPs compared with HMS were worried about transmitting the disease to family (P < 0.001) or being blamed by colleagues if they contracted COVID-19 (P = 0.019).

Anxiety and burnout were the most frequently self-reported psychological problems, with burnout reported by more GPs than HMS (P = 0.008; Table 3). From the objective symptom scale data, significantly more GPs reported symptoms consistent with moderate-to-severe emotional exhaustion, compared with HMS (P = 0.002). However, HMS were more likely than GPs to have low levels of personal accomplishment (P = 0.011).

Table 3.  Mental health.

In GPs, independent predictors for anxiety symptoms included having concerns about household income (P < 0.001), prior mental illness (P = 0.004) and worse relationships with family (P < 0.001; Fig. 1). GPs were less likely to have anxiety symptoms if they had better relationships with their partner (P = 0.024) or lived in regional areas (P = 0.012). Independent predictors for depressive symptoms included prior mental illness (P < 0.001), worse relationships with their partner (P = 0.001) or colleagues (P = 0.024) and worry about being blamed by colleagues for contracting COVID-19 (P = 0.045). Independent predictors for post-traumatic stress disorder included worse relationships with family (P < 0.001), having income concerns (P = 0.013), working with COVID-19 patients (P = 0.037) and prior mental illness (P = 0.019). Independent predictors for emotional exhaustion included worse relationships with friends (P = 0.001) and worry about being blamed by colleagues for contracting COVID-19 (P = 0.013). Having better relationships with family (P = 0.038) and having received PPE training (P = 0.048) were both associated with a sense of higher personal accomplishment.

Fig. 1.  Personal and professional predictors of adverse mental health outcomes among GPs.

Similar coping strategies were adopted by both GPs and HMS, with some notable differences. Significantly fewer GPs used digital applications to track their physical health (P = 0.018; Table 4) or increased alcohol consumption (P < 0.001) during the pandemic compared with HMS. GPs were significantly more likely than HMS to see a doctor or psychologist for mental support (P = 0.027). Overall, <25% of both groups sought professional help. Independent predictors of drinking more alcohol to cope during the pandemic included working with COVID-19 patients (P = 0.007) and worse relationships with family (P = 0.013; Fig. 2). Independent predictors of seeking help from a doctor or psychologist for mental health included having prior mental illness (P < 0.001) and worse relationship with their partner (P = 0.001).

Table 4.  Coping strategies and help-seeking behaviour.

Fig. 2.  Personal and professional predictors of coping strategy utilisation and help-seeking behaviour among GPs.

To our knowledge, this is the first Australian study to specifically examine the psychosocial impacts of the COVID-19 pandemic on GPs. We identified unique workplace challenges and stressors, more burnout issues, a significantly higher prevalence of moderate-to-severe emotional exhaustion, and different coping strategies adopted by GPs compared with HMS. Predictors for both mental health symptoms and coping strategy use by GPs have been described. These findings are important for developing strategies to prevent psychosocial workplace harms and guiding the creation of targeted workplace wellbeing supports in PHC.

The relatively large sample size in our study enabled comparison of the workplace challenges and psychological impacts of COVID-19 between GPs and HMS, and also examination of independent predictors of poor mental health and use of coping strategy among our GP respondents. The response rate in the current survey could not be calculated due to the widespread distribution of the survey. There was potential risk of selection bias and response bias, which may over- or under-represent the exposure to COVID-19, changes in working conditions and impacts on mental health. The proportion of responses received from Victoria, where the majority of COVID-19 cases and restrictions had occurred at the time of the survey, is a likely indication of selection bias. The participants in this survey were skewed towards women respondents. The current study was a single time-point measure to minimise burden on healthcare workers; however, longitudinal data would provide insight into changes in mental health and working conditions throughout the ongoing pandemic.

Compared with HMS, GP respondents felt less supported by their workplace, and that the communication received from their workplace was perceived to be suboptimal. This issue is important, as previous research demonstrated that PHC workers who were satisfied with their organisational support experience significantly lower stress levels than those who were less satisfied (Lee et al. 2020). GPs who worked in smaller practices, in particular those in regional or remote areas (27% in our study), may have had insufficient workplace resources to keep up with and implement changing protocols. The burden related to a plethora of information from multiple different guidelines during the COVID-19 pandemic has been shown in many studies to increase stress levels among PHC workers (Gray and Sanders 2020; Prescott et al. 2020; Wright et al. 2020; Crowley et al. 2021; Dutour et al. 2021; Lum et al. 2021; Sotomayor-Castillo et al. 2021).

Similar to the study of Sotomayor-Castillo et al. (2021), we found a significant proportion of GPs did not receive training on PPE usage or the care of COVID-19 patients. Failure to adequately train GPs is highly concerning, given that GPs are often the first point-of-contact for people with suspected COVID-19, and caring for people with mild COVID-19 infection (Australian Government Department of Health 2020a, 2020b; Kidd 2020a), Respiratory Protection Programs – providing governance, education and protection for workers at risk of respiratory hazards – are established in other industries and gradually adopted in hospitals during the pandemic (Ryan 2001; Victorian Respiratory Protection Program guidelines 2020; Australian NSW Government 2021; Williams et al. 2021). Yet, the majority of Australian GPs have not been offered respiratory fit-testing (The Royal Australian College of General Practitioners 2021). A similar program for primary care clinicians is urgently needed. This would protect not only physical health, but also mental health, as many studies have shown that inadequate PPE training contributes to increased stress levels in primary healthcare workers (Lee et al. 2020; Crowley et al. 2021; Sotomayor-Castillo et al. 2021).

A significant stressor associated with mental health symptoms for GPs in our study was concern regarding finances during the pandemic, with 60% experiencing a decrease in income. In contrast to HMS, who are generally salaried, Australian GPs are paid fee-for-service, often as part of small businesses, of which they may be part owners. Financial pressures arose from fewer patients attending GP clinics due to lockdowns or fears of contracting COVID-19, and restructuring of the GP practice system to telehealth (Kidd 2020a; Wright et al. 2020). GPs could only charge the minimum rebate for telehealth consultations initially, despite having to cover for the implementation costs, along with day-to-day expenditures. Thus, the shift to telehealth and its operating expenses exacerbated the financial challenges they experienced.

The significant concerns from GPs about spreading COVID-19 to their family members, and being blamed by colleagues, were important findings. Perceived stigma associated with being a frontline worker (Juan et al. 2020; Kirk et al. 2021; Tasnim et al. 2021), as well as the worry about inadequate infection protection (Tasnim et al. 2021; Torrente et al. 2021), have been identified as risk factors for anxiety and depression among healthcare workers internationally. From a practical point of view, if a GP becomes infected with COVID-19, the impacts include generating additional work burden for colleagues, and potentially causing colleagues to be furloughed or even temporary closure of their clinic. Furthermore, during this pandemic, shaming of infected healthcare workers by the press may lead to patients seeking alternative GPs (ABC News 2020).

In our study, significantly more GPs compared with HMS reported having burnout symptoms and experienced moderate-to-high levels of emotional exhaustion. Internationally, studies have demonstrated high levels of stress and burnout symptoms amongst GPs during the COVID-19 pandemic (Dutour et al. 2021; Lum et al. 2021; Stafie et al. 2021; Lange et al. 2022). Burnout can be detrimental, at both personal and professional levels, including increased risk of substance abuse and suicide, and impaired quality of patient care (Lebares et al. 2018; Hert 2020; Iacobucci 2021). Moreover, one study showed that burnout had a direct effect on GPs’ turnover intention (Zhang et al. 2021). Similar to the UK, the Australian PHC workforce has been under a significant amount of pressure for some time, with the ever-expanding workload due to GP shortage, (The Guardian 2021) and the growing aging population (Dingwall et al. 2020). Crisis events, such as the pandemic, further exacerbate the problem; for example, the UK is currently facing a potential exodus of GPs due to a ‘burnout workforce’ (The Guardian 2021).

Despite the fact that a considerable proportion of GPs exhibited moderate-to-severe mental health symptoms, <25% sought professional mental health care. Studies showed that GPs preferred self-treatment when dealing with sensitive psychological conditions, because of embarrassment, concerns of confidentiality breaches and also the culture of medicine, which tended to discourage admission of vulnerabilities (Thompson et al. 2001; Davidson and Schattner 2003). Therefore, more acceptable psychological interventions are urgently needed and should be readily available to all healthcare workers (The British Psychological Society 2020).

The unique workplace challenges and stressors faced by GPs during the COVID-19 pandemic, and their associations with poorer mental health symptoms, are undeniable. Workplace challenges related to policy development, communication, training, PPE and financial burden can be overcome with effective organisational and government leadership. New, long-term approaches to provide acceptable psychological support for GPs, who suffer from significant emotional exhaustion, are urgently required. It is essential to protect our GPs’ wellbeing and workforce, which is fundamental to the functioning of the national health system.

Data available upon reasonable request from the corresponding author.

The authors declare that they have no conflicts of interest.

The Royal Melbourne Hospital Foundation and the Lord Mayor’s Charitable Foundation kindly provided financial support for this study.