Police officers work in stressful, dangerous, and demanding conditions where they are expected to respond effectively, responsibly, and emotionally neutral (Załuski and Makara-Studzińska 2022). Policing is one of the most stressful careers in our society (Decker et al. 2022). Police officers are estimated to be exposed to around 900 highly stressful and potentially traumatic events during their careers. These events may be dealing with domestic violence, motor vehicle accidents, injured children, fatalities, and having to use their service weapon to protect others and prevent harm (Hoeve et al. 2021).

Police officers in Australia are required to meet a set of responsibilities; for those employed by the Queensland police force as general duties officers, they are expected to serve their community by protecting life and property, preserving peace and safety, preventing crime, and appropriately upholding the law (Queensland Police Service 2014). Staff positions cannot remain vacant to meet these responsibilities effectively. Currently, every Australian state is facing a crisis, as stated by Ian Leavers from Police Federation Australia: “It is a real challenge, not only recruiting but the retention rates of police across Australia. We are putting the police officers who are left at risk. And we are putting the community at risk” (Coulter and Ackew 2023). Furthermore, Queensland Police Commissioner Katarina Carroll disclosed that the attrition rate for the police force, which was previously approximately 2.7%, has more than doubled to 5.7%. This increase is attributed to heightened workloads and increasing demands placed on officers (Sato 2023). All Australian states currently have vacancies within their police force; Seek lists around 2000 full-time and 1000 part-time jobs for police officers (https://www.seek.com.au/Police-Officer-jobs/in-All-Australia; accessed 2 August 2024).

The difficulties in recruiting and retaining police officers may be attributed to the increased risks associated with this career, for example, cardiovascular disease, metabolic syndromes, burnout, chronic stress, and sleep disturbances (Decker et al. 2022; Garbarino et al. 2019). These adverse consequences can be explained by organizational stressors such as job content and context stressors (Evans and Coman 1993). Content stressors include work hours, shift work, overload, excessive task demands, and role responsibilities. Context stressors instead consist of an unsafe working environment, inadequate career opportunities, and ineffective communication. Employees who work nightshifts report higher work-related stressors than those who work during the day (Ma et al. 2015).

Shift work often requires employees to work night, weekend, and rotating fixed and irregular work hours, typically between 7 pm and 6 am (Caruso 2014; Peterson et al. 2019). Apart from being a source of stress, those working the afternoon or nightshift experience more significant physical and physiological threats (Knowles and Bull 2003; Ma et al. 2015). Past literature has confirmed shift work to be associated with a range of negative consequences, such as poor physical activity, poor nutrition, and reduced recovery time (Załuski and Makara-Studzińska 2022). Working nightshifts increases an individual’s risk of sleep disturbances, stress, and burnout (Athar et al. 2020; Fekedulegn et al. 2016; Ma et al. 2019; Peterson et al. 2019; Shields 2002). The quantity of sleep is also impacted, typically losing 1 to 4 h of sleep per night (Baron and Reid 2014). Failure to obtain sufficient sleep quantity and quality may lead to increased stress and burnout due to an inability to recover from the occupational demands of a police officer (Baek et al. 2022; Peterson et al. 2019).

These stressors are explained within personnel psychology, which describes burnout as a response to organizational stressors (Edú-Valsania et al. 2022). Herbert Freudenberger (1974) was the first to use the term “burnout” and described the state of being burned out as “becoming exhausted by making excessive demands on energy, strength, or resources” (p. 159). This accumulated stress over time leads to exhaustion and demotivation, symptoms that Maslach and Jackson (1981) used to develop the first of many iterations of the Maslach Burnout Inventory (MBI), which measures three dimensions: emotional exhaustion, depersonalization, and reduced personal accomplishment.

The phase model by Golembiewski et al. (1983) explains the cause of burnout as being due to organizational and work stressors that travel through phases. Created around MBI, an individual is categorized within one of eight categories according to their burnout level; individuals within the eighth category adopt one of two coping strategies: active or passive, with the majority choosing a passive coping strategy (Golembiewski 1999). Active coping would observe individuals confronting and contesting their environment, while a passive style sees individuals adopting a submissive posture towards their environment (Filipova 2016).

As a response to experiencing a specific strain or tension (Vuorensyrjä and Mälkiä, 2011), chronic stress reduces cognitive performance, motor performance, empathy, friendlessness, and effectiveness in completing work duties (Kohyama 2021). It has previously been indicated that stress experienced by officers is the direct consequence of content and context stressors (Evans and Coman 1993). The risk of burnout is increased for police officers due to stressful working conditions, exposure to traumatic stressors, fatalities, and violence (Kaplan et al. 2017). In addition to the operational challenges faced by law enforcement, rising anti-police sentiment has intensified retention issues across the policing sector. This has in the US been fueled by the Black Lives Matter demonstrations following the deaths of two African Americans in 2014 and the George Floyd protests in 2020, events that have led to increased scrutiny and criticism of policing practices, contributing to a widespread retention crisis. The negative public perception affects officer morale and discourages recruits from joining the force (Adams et al. 2023; Christián et al. 2022; Drew et al. 2024; Mourtgos et al. 2022).

As research continues to develop, a new approach to burnout has been highlighted, with exhaustion and disengagement becoming the focus (Ogińska-Bulik and Juczyński 2021). Initially, burnout was viewed to only occur within people-work, such as social workers and teachers (Gomes et al. 2022; Lubbadeh 2020). While it is now recognized to affect anyone, symptoms of burnout are known to increase in response to extended exposure to stressful environments (Kaplan et al. 2017). As such, job burnout is a syndrome that develops in response to long-term chronic interpersonal and occupational stressors (Ogińska-Bulik and Juczyński 2021). In 2019, the World Health Organization (WHO) recognized burnout as an occupational hazard, with individuals experiencing chronic fatigue, muscle pain, headaches, reduced motivation, irritability, absenteeism, and impaired job performance and relationships (Gomes et al. 2022).

Our research aimed to examine the impact of nightshift work on police officers within the Australian Police Force compared to their dayshift counterparts. We hypothesized that nightshift work would mediate the relationships among sleep quality, stress, and burnout.

Methods

Participants

The study involved 642 Australian police officers who work on a rotating shift roster that includes early day, afternoon, night, and weekend shifts. At the time of the survey, 379 officers (59%) were working nightshifts (6 pm to 6 am), while 263 officers (41%) were on dayshifts (6 am to 6 pm). Informed consent was provided before the survey was answered. Ethics approval was obtained from the human research ethics committee at Southern Cross University (number 2023/096). Participants were recruited through Facebook by posting a survey link on pages and groups commonly visited by police officers. The survey began with a screening question to confirm participants’ occupation, allowing only those who identified as police officers to continue. The recruitment strategy emphasized the significance of the study, its relevance to law enforcement professionals, the anonymity of responses, and the benefits of participation. Additionally, efforts were made to engage with comments and inquiries to encourage participation and build trust among potential respondents.

Measurements

Participants completed the anonymous online survey on their devices by accessing the study link. The relationship between the variables, such as shifts, sleep quality, stress, and burnout, was examined using a cross-sectional design. Participants initially completed demographic questions, including whether they worked day- or nightshift. Subsequently, they answered Likert-scale questions to assess various aspects of sleep quality, stress levels, and burnout.

Athens Insomnia Scale (AIS) was used to measure sleep quality using a 4-point Likert scale (Shahid et al. 2011). This original 8-item scale assesses the participant’s sleep difficulty over the previous month. We, however, only used item 5, which explicitly measures sleep quality; participants rated their overall sleep quality (no matter how long they slept) from 0 (satisfactory) to 3 (insufficient or did not sleep at all). The scale has good internal consistency (Cronbach’s α = 0.87; Enomoto et al. 2018).

Depression Anxiety Stress Scale-21 (DASS-21) was measured using the stress subscale, which consists of 7 items (Lovibond and Lovibond 1995). Scale items were presented as statements, for example, “I found it hard to wind down,” “I found myself getting agitated,” or “I was intolerant of anything that kept me from getting on with what I was doing.” Participants used a 4-point Likert scale from 0 (never) to 3 (almost always). The stress subscale was summed and multiplied by two (ranging from 0 to 42) per scoring instructions (Di Nota et al. 2020). Scores ranged from normal (0–14), mild (15–18), moderate (19–25), severe (26–33), and extremely severe (34 +). The DASS-21 displayed good psychometric properties within adult populations (Shaw et al. 2017). Convergent validity for the stress scale (0.65; Gracielle Pereira et al. 2022) and good internal consistency were demonstrated (Cronbach’s α, 0.82–0.97; Osman et al. 2012). Reliability for each construct was > 0.86 to 0.90 (Hoeve et al. 2021).

Oldenburg Burnout Inventory (OLBI) measured emotional exhaustion and disengagement with 16 items (Looti 2023). Two example items measuring emotional exhaustion are “After work, I tend to need more time than in the past in order to relax and feel better” and “There are days when I feel tired before I arrive at work.” Two example items measuring disengagement: “It happens more and more often that I talk about my work in a negative way” and “Lately, I tend to think less at work and do my job almost mechanically.” All responses were recorded on a 4-point Likert scale from 1 (strongly agree) to 4 (strongly disagree). Total scores range between 16 and 64. Initial scale development demonstrated adequate internal consistency (Cronbach’s α, 0.70–0.75; Ogunsuji et al. 2022).

Analyses

Before running the analyses, a sensitivity analysis was performed using listwise deletion. Bivariate correlation was then calculated between all variables. A moderation analysis was run using SPSS to examine the effect of moderation on sleep quality. Bootstrapping 5000 was used for the moderation analyses, with data centering occurring before running the analyses. All assumptions were checked before conducting the analyses. This included visual inspection of a standardized residuals histogram, P-plot, and scatterplot, ensuring no normality, linearity, or homoscedasticity breach. Multivariate outliers were checked and found not to be violated (Cook’s distance < 1). Multicollinearity was not a concern, indicating that the regression estimates were stable and reliable; the Durbin-Watson statistic showed that the independence of errors was not violated.

Results

Bivariate Correlations

The relationship between sleep quality and stress was weak but statistically significant (see Table 1). A small positive correlation was also observed between sleep quality and burnout and between stress and burnout.

Table 1 Means, standard deviations, and Pearson’s (r) correlations between variables (N = 642)
Full size table

Stress and Burnout

The statistical analysis revealed that 9.2% of study participants scored within the normal category for stress, 21.3% within the mild stress category, 60.2% within the moderate category, 8.1% in the severe category, and 1.2% in the extremely severe category (34 +). Participants’ burnout scores ranged between 22 and 57, with 58% scoring 40 and above, placing over half of participants within the medium and above burnout category. Cut-off scores — low, medium, high — for the OLBI were suggested by Glowacz et al. (2022). One participant scored within the low range on the OLBI, 613 within the medium range, and 28 scored high. Participants’ scores in stress, burnout, and sleep quality between participants who worked nightshift and those who worked dayshift are shown in Table 2.

Table 2 Participants working nightshift vs dayshift (N = 642)
Full size table

Moderation Analysis of Shift Work and Stress

PROCESS for SPSS was used to test whether the effect of night- and dayshift (predictor) on stress (dependent variable) would be moderated by sleep quality. All assumptions were met. The model was statistically significant, F(8.59, 0.406), p = 0.001, R2 = 0.058, with the full model accounting for 5.8% of the variance in stress. The interaction between shifts and sleep quality was not statistically significant, F(0.406), p = 0.524, R2 = 0.001.

Moderation Analysis of Shift Work and Burnout

PROCESS for SPSS was also used to test whether the effect of night- and dayshift (predictor) on burnout (dependent variable) would be moderated by sleep quality. This model was not statistically significant, F(1.96, 0.103) p = 0.118, R2 = 0.0097, with the full model accounting for 0.97% of the variance in burnout. The interaction between night- and dayshift and sleep quality was also not statistically significant, F(0.103), p = 0.749, R2 = 0.000.

Discussion

Our study aimed to determine whether Australian police officers working nightshifts scored higher on stress and burnout and reported lower levels of sleep quality than their peers’ working dayshifts. Surprisingly, those working nightshifts did not score differently from those working dayshifts. All police officers did, however, present with moderate to high stress and burnout scores, which may be explained by their exposure to highly stressful and even traumatic events during their careers (Decker et al. 2022; Hoeve et al. 2021). In studies in North America, police officers rated their stress levels higher when their shifts were irregular, longer (> 11 h), or mandatory rather than voluntary (Peterson et al. 2019). Similarly, police officers who worked afternoon or nightshifts were exposed to significantly more stressful events than those who worked the dayshifts (Ma et al. 2015).

In contrast to the North American studies, our findings suggest that shift work may not be the most important factor, especially when they rotate through different shift types. Police officers seem to struggle to cope with mounting stress and high levels of burnout. Whether elevated stress and burnout scores can fully explain the difficulties in attracting new employees occurring within all states and territories in Australia is merely speculative but well worth exploring (cf. Coulter and Ackew 2023). Consequently, turnover is a risk for all police officers with elevated stress and burnout levels, regardless of whether they work nightshifts or not.

Employees experiencing burnout are in a vicious cycle of absenteeism, presenteeism, and leaving their careers (Linos et al. 2021). This leaves the remaining employees to fill the job demands, leading to fatigue, mistakes, and themselves experiencing the onset or more severe levels of burnout, ultimately forcing those employees to find a way to cope, potentially entering the same vicious cycle. As a concrete consequence, Asher (2023) presented data from 15 US law enforcement agencies, revealing that officer shortages have increased response times for 911 calls and decreased clearance rates, indicating reduced police efficiency. The growing retention crisis is at least partly due to negative perceptions of police work, which adversely affects officer health, leading to burnout, psychological distress, and higher turnover intentions (cf. Drew et al. 2024). These challenges emphasize the need for strategies to improve recruitment, retention, and the overall wellbeing of officers to ensure effective law enforcement and community safety.

While most of our participants were experiencing burnout, some may have already left their jobs, particularly if they have reached the later stages of burnout (Golembiewski et al. 1983; Gomes et al. 2022). Officers in our study who sit within the high range of burnout may also influence and potentially increase the severity of burnout levels among their colleagues, as described by Maslach and Leiter (2016, p. 106): “burnout can be contagious and perpetuates itself through social interaction on the job.” Burnout can be communicated consciously and unconsciously from one person to another (Bakker et al. 2005). This is particularly pertinent when interpersonal relationships are strong and frequent (Meredith et al. 2020). Queensland Police Service (2014) lists several key accountabilities, such as “Develops positive relationships with team members; actively participates in teamwork and activities to achieve objectives.” Teamwork is generally regarded as a protective factor against burnout. Still, close and frequent interpersonal relationships may also increase the risk of burnout as it may spread through the ranks via emotional contagion (Jun and Costa 2020).

Our study further uncovered that sleep quality had no moderation effect between shifts and stress levels; an individual’s sleep quality did not strengthen or weaken the stress levels experienced by police officers who worked night- or dayshifts. This contrasts with past literature, which discovered that police officers who worked nightshifts reported lower levels of sleep quality than those who worked dayshifts (Almheiri and Rashid 2021). Gerber et al. (2013) also described that officers who worked nightshifts reported significantly more sleep complaints and were less satisfied with their sleep than officers who worked dayshifts. Our results were not unexpected due to the previous findings that night- vs dayshift work did not affect stress or burnout levels.

Some limitations were present, such as using a cross-sectional correlational design. Cause and effect could not be determined; a longitudinal design would be beneficial. The study also used self-report scales. While this allowed for anonymity, it opened the survey to recall bias. This was minimized by asking participants to remember information from a short period; DASS-21 and the Oldenburg Burnout Inventory had a recall period of 1 week, while the Athens Insomnia Scale required a recall period of 1 month. However, all participants adhered to a rotating shift schedule, which implies that prior nightshifts may influence their subsequent dayshifts and vice versa. If participants could be randomly assigned to either nightshifts or dayshifts for an extended duration, it would allow for a more precise assessment of the impact of shift work on their stress levels, burnout, and related outcomes. Another potential limitation is asking the participants to disclose their stress and burnout levels, which may be feared to reach their employers and negatively affect future careers (cf. Marshall et al. 2021). We did, however, ensure anonymity and limit demographic questions to occupation — to ensure they were police officers — and whether they worked nightshift or dayshift. We avoided asking about their age, years in the job, shift length, and job position within the force, even though that information would be beneficial.

Assessing sleep quality, burnout, and stress levels in a sample of more than 600 Australian police officers allowed a greater understanding of the population. It is noteworthy that over 60% of all participants report moderate to high stress levels and close to 60% score higher than 40 on the Oldenburg Burnout Inventory. If the Australian police force continues its current trajectory, an increasing number of officers will leave to pursue other careers or retire, resulting in even lower retention rates.

Conclusions

Our study revealed that nightshift work did not directly affect sleep quality, stress, or burnout levels among Australian police officers. These findings provide valuable insight into organizational theory, specifically the phase model of burnout and the stressors that may contribute to the development of burnout. Police officers’ workload, lack of autonomy at work, and lack of perceived social support need urgent attention to reverse the trend towards accelerating turnover.