When owls struggle at work: chronotype, burnout, and work ability in shift and non-shift employees

Abstract

Evening chronotype has been linked to poorer occupational functioning, yet the mechanisms through which circadian preference relates to work ability remain insufficiently understood. This study examined whether job burnout functions as a psychological pathway connecting chronotype to work ability and whether this mechanism operates similarly in day workers and employees working shifts including night work. A cross-sectional Computer-Assisted Web Interviewing (CAWI) survey was conducted among 1,293 employed adults in Poland. Confirmatory factor analyses (CFA) and multi-group CFAs confirmed the factorial validity and full measurement invariance of all measures across both groups. Mediation models controlling for age and gender revealed a consistent mediation pattern in day and shift workers, with no significant between-group differences in path coefficients or indirect effects. Specifically, morningness was associated with lower burnout and higher work ability, burnout was strongly and negatively related to work ability, and the indirect effect of chronotype on work ability via burnout was significant, indicating partial mediation alongside a remaining direct effect. These findings demonstrate a general psychological mechanism linking chronotype to work ability that operates independently of formal shift work. The structural equivalence across groups suggests that chronotype-related vulnerability is not limited to shift work but may emerge under any rigid work schedule misaligned with individual circadian preferences. Chronotype-sensitive work arrangements and burnout-prevention strategies may therefore support sustainable work ability, particularly among employees with evening chronotypes.

Introduction

Work ability constitutes a key aspect of employee functioning [1], with significant implications for both health and job performance [2–4]. It is defined as a subjective evaluation of an employee’s physical capacity, psychomotor, and sensory abilities that enable the effective performance of work tasks without compromising the safety and health of oneself or others [1, 5]. Empirical evidence indicates that low work ability is associated with a higher risk of absenteeism [3], reduced productivity [4, 6], and an increased likelihood of disability pension and early retirement [3, 7]. Research on the determinants and mechanisms underlying work ability has therefore gained particular importance, especially in the context of increasing labor market demands and population aging. A deeper understanding of these processes can inform the design of effective interventions aimed at extending occupational activity, enhancing employee well-being, and reducing the social and economic costs associated with work disability.

Both systematic reviews [8] and meta-analyses [9] have confirmed the significant role of work environment factors (e.g., high physical and mental demands, low job resources), type of work (e.g., manual vs. cognitive), individual characteristics (e.g., older age, obesity, general health status), and lifestyle factors (e.g., low physical activity, health prevention behaviors) in shaping work ability. In recent years, however, an increasing number of studies have highlighted the role of individual differences related to circadian rhythms in the development of work ability [10–12]. The findings consistently demonstrate that employees with an evening chronotype exhibit lower work ability compared to those with morning or intermediate chronotypes.

However, little is known about the mechanisms underlying the negative effect of the evening chronotype on lower work ability. It may be assumed that individuals with an evening chronotype, although they tend to reach their peak cognitive and physical performance later in the day [13], are often required to adhere to routines that conflict with their natural circadian tendencies (e.g., early morning shifts or strict task deadlines). Such misalignment generates tension between the social clock (external and collective) and the biological clock (internal and individual), leading to insufficient sleep, reduced sleep quality, negative mood at work, and increased fatigue [14–16]. When this circadian misalignment persists over time, it may contribute to more enduring mental health difficulties, including job burnout [15, 17, 18]. Job burnout is a state of chronic physical, emotional, and psychological exhaustion that arises as a result of prolonged stress caused by excessive job demands and low level of resources [19–20]. The aim of the present study was to test both the direct effect of chronotype on work ability and the mediating effect of job burnout.

The direct link between chronotype and work ability

Work ability may be understood as a dynamic construct reflecting the balance between the demands of work and the worker’s resources - including health, skills, motivation, and the work environment - that together determine one’s capability to continue working effectively [8]. This conceptualization of work ability is consistent with the model developed by K. Tuomi and J. Ilmarinen and their team at the Finnish Institute of Occupational Health (FIOH), in which work ability is understood as a state of balance between an employee’s resources (e.g., health status, functional capacities, and skills) and the demands of the job [1, 5].

Chronotype refers to an individual’s biologically driven tendency to sleep and wake at particular times of the day [21]. It is generally categorized into morning, intermediate, and evening types [22]. The body’s circadian rhythms are governed by an internal biological clock that naturally maintains an approximately 24-hour cycle, while also adapting to external environmental cues called zeitgebers. An individual’s chronotype represents personal variation in the timing at which this synchronization occurs [23]. In contemporary society, social obligations frequently clash with one’s natural sleep - wake tendencies, causing a disruption of the body’s internal circadian timing system - a condition known as circadian misalignment or social jet lag [22, 24, 25]. Research indicates that such disturbances in circadian rhythm are linked to hyperactivation of the hypothalamic–pituitary–adrenal (HPA) axis and elevated cortisol levels, both of which negatively influence mood regulation [26].

Social jet lag is often associated with excessive daytime sleepiness [25], heightened depressive symptoms, anxiety, and irritability [26, 27]. It has also been linked to diminished reward responsiveness and decreased motivation to expend effort in pursuit of rewards [28]. Within the workplace, these outcomes manifest as poorer physical and psychological well-being among employees [2], reduced engagement at work [29], lower productivity [6], and also diminished work ability [10–12]. For example, a large-scale population study of more than 13,000 Finnish employees aged 18–64 years found that those with an evening chronotype - commonly referred to as “owls” - showed lower work ability compared to individuals with morning or intermediate chronotypes. Comparable results were reported in two additional large cohort studies involving several thousand participants [11, 12].

Chronotype and job burnout

Job burnout can be conceptualized as a state of prolonged physical and psychological exhaustion [19], which is caused by job-related stress [20], resulting from an imbalance between job demands and job resources [30]. It may occur across all types of occupations characterized by chronic work-related stress [20]. Although workplace-related factors are currently considered the primary sources of job burnout, an increasing number of studies highlight the significant role of employees’ individual biological characteristics—for instance, chronotype [15, 17, 31–33].

Recent studies conducted in recent years confirm that an evening chronotype is linked to elevated levels of job burnout, including among shift-working healthcare professionals [17], nurses [31, 33], physical therapists [32], and nursing students [15]. This association is commonly explained by the fact that individuals with evening preferences, despite reaching their peak cognitive and physical performance later in the day [13], are frequently required to operate within schedules that are misaligned with their biological predispositions, such as early work shifts or rigidly defined task deadlines. Such misalignment gives rise to a conflict between the social clock (external and shared) and the biological clock (internal and individual). When sustained over time, this discrepancy results in sleep curtailment, reduced sleep quality, and heightened fatigue [14–16], which may, in turn, increase vulnerability to job burnout [15–18]. Inadequate working hours for employees with an evening chronotype also impede recovery after work [34], which may further exacerbate irritability and increase sensitivity to job demands, ultimately contributing to burnout [35].

Job burnout and work ability

Burnout and work ability are two key constructs within occupational health psychology, both referring to the amount of energy and resources an employee can mobilize to cope with job demands and invest in their work at a particular point in time [36]. The notion of work ability primarily highlights the physical dimension of functioning at work, including working conditions, physical stamina, and the competencies and functional capacities necessary to meet occupational demands [8]. In contrast, the burnout framework places particular emphasis on emotional and affective resources [19]. Nevertheless, in both cases, maintaining an appropriate balance between job demands and available resources is considered essential for employee well-being and effective functioning. Burnout can impair an employee’s work ability by depleting their resources. It not only affects their motivation and engagement, but can also negatively influence their attitude toward work and even disrupt cognitive functions, making it harder to concentrate, make decisions, and maintain productivity over time [37]. Thus, research consistently shows that burnout’s negative effect on work ability is most pronounced when it occurs alongside poor job resources. The presence of adequate resources can serve a protective role [38, 39].

Several studies across different occupational groups have shown a negative relationship between burnout and work ability, including research involving teachers [40], human services workers [41], nurses [42], flight attendants [43], nurses [44] and healthcare staff [45]. Evidence from a meta-analysis further indicates that increases in job burnout are associated with decreases in work ability [46]. Based on the cited literature, we formulated two general research hypotheses:

• H1: Evening chronotype is directly associated with lower work ability.

• H2: Job burnout mediates the effect of chronotype on work ability, such that evening chronotype is associated with higher job burnout, and higher job burnout is associated with lower work ability.

Shift work

Contemporary research, including systematic reviews and meta-analyses, consistently indicates that shift work is associated with an increased risk of mental health problems, such as depression, anxiety, sleep disorders, and burnout [47, 48], as well as with reduced work ability [11]. Particular attention has been paid to the role of individual circadian preferences, which determine tolerance for working at different times of the day and night [13]. Some studies suggest that employees with an evening chronotype are more vulnerable to the adverse effects of shift work, especially when their work schedules are misaligned with their biological predispositions [11, 13]. Conversely, individuals with a morning chronotype may experience greater difficulty tolerating night shifts, resulting in increased fatigue, sleep disturbances, and reduced performance [49]. In light of the above, further research is warranted to comprehensively examine the relationship between chronotype and work ability, taking into account the mediating role of burnout and the specific characteristics of shift work, compared with non-shift workers. Such research would not only enhance understanding of the mechanisms underlying reduced work ability and burnout but also support the development of effective prevention and intervention strategies, including individualized work schedules, psychological support, and programs aimed at strengthening stress resilience.

Method

Participants and procedures

The study included a diverse group of 1,293 working individuals. Men made up 66.4% of the total sample (N = 858), while women constituted the remaining 33.6% (N = 435). The participants’ ages ranged from 18 to 70 years (M = 40.33; SD = 10.84; Me = 39.00). Job tenure varied considerably-from less than one year up to 50 years (M = 14.87; SD = 9.88; Me = 13.00).

All respondents were employed within formal organizational structures and reported having a direct supervisor. The sample represented numerous branches of the economy (identified via a closed-ended question) and a wide array of professions (declared in an open-ended format). The largest group worked in industry (52.9%; N = 684; e.g., machine and production line operators, assemblers, welders), followed by employees in the transportation sector (24.6%; N = 318; e.g., truck and bus drivers, tram operators, signal controllers), construction (18.9%; N = 244; e.g., bricklayers, roofers, welders, civil engineers), and agriculture (3.6%; N = 47; e.g., warehouse staff, machine operators, gardeners, animal breeders). Most participants were employed in the private sector (87.9%; N = 1,137), whereas 12.1% (N = 156) worked in public institutions. Approximately one in five respondents (20.3%; N = 262) held managerial responsibilities, including positions supervising small teams. 671 participants (51.9%) reported working in a shift system that included night shifts, whereas 622 (48.1%) did not. The difference in group sizes was not statistically significant (χ² = 3.841; p = 0.173).

Table 1 summarizes the participants’ educational background and the size of their place of residence.

Table 1.

Characteristics of the participants, Poland, March–April 2025

	n	%
Gender
Male	858	66.4
Female	435	33.6
Age
18–24 years	38	2.9
25–34 years	391	30.2
35–44 years	418	32.3
45–54 years	308	23.8
55–70 years	138	10.7
Place of residence
Rural area	297	23.0
Town ≤ 50,000 inhabitants	275	21.3
Town ≤ 100,000 inhabitants	237	18.3
City ≤ 250,000 inhabitants	223	17.2
City > 250,000 inhabitants	261	20.2
Education
Primary	14	1.1
Vocational	142	11.0
Technical secondary	331	25.6
General secondary	171	13.2
Post-secondary	86	6.7
Currently studying at higher level	9	0.7
Higher education	540	41.8

Participants came from all 16 administrative regions of Poland. The largest proportion resided in the Masovian Voivodeship (18.1%; N = 490). Men accounted for the majority of participants (66.4%), which corresponds to the real gender distribution observed in the occupational groups included in the study. National labour market statistics for Poland indicate that women make up 34.5% of employees in industry, 10.7% in construction, 23.4% in transportation, and 47.3% in agriculture. Consequently, the male predominance within the present sample accurately mirrors the actual demographic composition of these sectors.

The research was conducted using a quantitative, cross-sectional approach based on the Computer-Assisted Web Interviewing (CAWI) technique. Respondents were selected at random from an existing online panel through a probability-based sampling procedure. Eligibility criteria included current employment within an organizational structure and affiliation with one of the economic sectors targeted in the project. Data were collected between March and April 2025. All procedures adhered to the ethical principles of the Declaration of Helsinki. Informed consent was obtained from all individual participants included in the study. Participation was voluntary, and the anonymity and confidentiality of all responses were fully ensured. The study was approved by the Bioethics Committee of the Institute of Rural Health in Lublin (Resolution No. 22/2023).

Measures

Chronotype

Chronotype was assessed using the Polish version of the Composite Scale of Morningness (CSM). The CSM is a self-report questionnaire designed to measure individual circadian preference, ranging from morningness to eveningness [50]. The Polish adaptation was validated by Jankowski [51]. The questionnaire consists of 13 items referring to preferred times of going to bed and waking up, as well as subjective alertness and sleepiness at different times of day. Responses are given on 4- or 5-point Likert scales, and a total score is calculated. Higher scores indicate a stronger preference for morningness.

Job burnout

Burnout was assessed using the burnout scale from the Copenhagen Psychosocial Questionnaire (COPSOQ) [52], in Polish version [53]. The conceptual background of this scale is closely related to the Copenhagen Burnout Inventory (CBI), which operationalizes burnout as work-related exhaustion rather than a multidimensional syndrome [19]. Exhaustion represents the core aspect of job burnout [20] and shows the strongest and most consistent association with work ability [44]. Moreover, using a single-dimension measure reduces participant burden and facilitates data interpretation, which is particularly advantageous in studies with large or heterogeneous occupational samples. Although multidimensional instruments capture additional facets of burnout, such as cynicism and reduced professional efficacy, these components are less directly relevant to the primary outcomes examined in the present study. The burnout scale consists of four items assessing the frequency and intensity of exhaustion related to work. Responses are provided on a five-point Likert scale. Following the COPSOQ scoring procedure, item scores are transformed to a 0–100 scale, with higher values indicating a higher level of burnout.

Work ability

Work ability was assessed using the Polish version of the Work Ability Index (WAI) [5, 54, 55]. The WAI is a standardized self-report measure composed of seven domains assessing different aspects of an employee’s ability to work. These include: (1) current work ability compared with lifetime best, (2) work ability in relation to physical and mental job demands, (3) self-estimated work impairment due to disease, (4) personal prognosis of work ability two years ahead, (5) mental resources, (6) number of physician-diagnosed diseases, and (7) number of sickness absence days during the previous 12 months. A global WAI score was calculated and used in the analyses.

Statistical analyses

To verify the factorial structure of the instruments, confirmatory factor analyses (CFA) were conducted for each questionnaire, followed by multi-group CFA (MGCFA) to test measurement invariance between day and shift workers. Model fit was evaluated using CFI, TLI, RMSEA, and SRMR, with conventional cut-off criteria applied (CFI/TLI ≥ 0.90; RMSEA/SRMR ≤ 0.08). Measurement invariance was established when changes in fit indices did not exceed ΔCFI = 0.010 and ΔRMSEA = 0.015 [56]. The potential influence of common method bias was additionally examined using a latent common method factor approach.

After confirming the psychometric adequacy and invariance of the measures, the final hypothesis-testing model was simplified by using observed variables (sum-score) for each construct. Given the satisfactory model fit and full measurement invariance across groups, as well as adequate reliability coefficients, sum scores were treated as sufficiently reliable observed indicators of the constructs for structural path modeling. This decision was made to maintain model parsimony and avoid unnecessary complexity in the multi-group mediation analyses. The hypothesized mediation model was estimated for the full sample and separately for day and shift workers. Age and gender were included as control variables, given previous evidence that age tends to correlate with chronotype (morningness increases with age; [57]), job burnout [58], and work ability [59], and that gender differences have also been observed in studied constructs [60]. Mediation was conceptualized as a significant indirect effect of chronotype on work ability through burnout, after controlling for age and gender. The significance of the indirect path was tested using bias-corrected bootstrap confidence intervals based on 5,000 resamples, which provide a robust estimate of mediation effects without assuming normality of the indirect distribution [61]. All models were estimated using full information maximum likelihood (FIML) to handle missing data [62]. Statistical significance was set at p < 0.05 (two-tailed). To formally test whether structural paths differed between groups, Wald χ² tests [63] were applied to equality-constrained parameters across day and shift workers. All analyses were conducted in R (version 4.3.3) using the lavaan package [64] for structural equation modeling.

Results

Preliminary analyses

Table 2 presents descriptive statistics and correlation coefficients among the variables.

Table 2.

Descriptive statistics and correlations between variables

Variable	Min, Max	M (SD)	Skewness	Kurtosis	ω	Pearson’s r coefficient
						Gender (0 – female, 1 – male)	Age	Chronotype (morningness)	Job Burnout
Chronotype (morningness)	14.0, 55.0	39.8 (5.90)	-0.379	0.646	0.847	-0.088***	0.142***	-
Job Burnout	0, 100	34.9 (19.3)	0.168	-0.505	0.885	-0.152***	0.154***	-0.096***	-
Work Ability	7.00, 37.0	29.2 (5.47)	-0.753	0.231	0.785	-0.009	-0.229***	0.191***	-0.476***

ω – McDonald’s Omega

*** p < 0.001

The results presented in Table 2 showed that the skewness and kurtosis values for all variables ranged between − 1 and 1. Each of the three variables also demonstrated a satisfactory level of internal consistency.

Each of the three questionnaires was subjected to a confirmatory factor analysis (CFA) for the entire sample, as well as a multi-group CFA testing the model fit across employees working night shifts and those working exclusively during the day. The results are presented in Table 3.

Table 3.

Model fit indices and measurement invariance tests for the chronotype, burnout, and work ability scales

Tool	Model	CFI	TLI	RMSEA	SRMR	ΔCFI	ΔRMSEA
Chronotype (CSM)	Full sample	0.936	0.916	0.069	0.056
	Configural	0.932	0.910	0.072	0.058
	Metric	0.930	0.916	0.069	0.062	−0.002	−0.002
	Scalar	0.923	0.915	0.070	0.064	−0.007	+ 0.000
	Strict	0.921	0.921	0.067	0.064	−0.002	−0.002
Burnout (COSPOQ II subscale)	Full sample	0.991	0.974	0.090	0.014
	Configural	0.991	0.974	0.096	0.014
	Metric	0.991	0.984	0.075	0.022	−0.001	−0.021
	Scalar	0.992	0.990	0.060	0.022	+ 0.001	−0.015
	Strict	0.992	0.993	0.051	0.027	−0.000	−0.009
Work Ability Index	Full sample	0.986	0.970	0.056	0.043
	Configural	0.987	0.973	0.054	0.042
	Metric	0.986	0.977	0.049	0.049	−0.0013	−0.0043
	Scalar	0.984	0.980	0.047	0.051	−0.0013	−0.0028
	Strict	0.978	0.976	0.051	0.064	−0.0068	+ 0.0040

“Full sample” refers to the analysis conducted without splitting the sample into day and shift workers. ΔCFI and ΔRMSEA indicate the change in model fit relative to the preceding, less less restrictive invariance model

Overall, the CFA models for all three instruments demonstrated satisfactory fit to the data. The single-factor models of the Chronotype, Burnout, and Work Ability scales showed acceptable values of CFI and TLI (≥ 0.90) and SRMR (≤ 0.08). Although the RMSEA values for some models were slightly above the conventional cut-off of 0.08, this is common in models with few indicators and low degrees of freedom, and does not necessarily indicate poor fit [65]. The SRMR values consistently below 0.08 further support the adequacy of model fit.

Subsequent multi-group CFAs confirmed that each scale exhibited full measurement invariance across day and shift workers. Changes in fit indices between increasingly constrained models (ΔCFI ≤ 0.01 and ΔRMSEA ≤ 0.015) remained well within recommended thresholds [56], indicating that the measurement structure of all three constructs was equivalent across the two groups.

In the next step of analysis, to assess the potential impact of common method variance, we also estimated a model that included all questionnaire items as observed variables, with latent factors representing the subscales and a global common latent method factor with equal loadings on all items (orthogonal to the substantive factors). The method factor collapsed (k ≈ 0, p = 0.94), yielding no improvement in fit over the baseline model (ΔCFI = 0, ΔRMSEA = 0, ΔAIC = + 2; ΔBIC = + 8). Thus, there is no evidence for a pervasive common-method bias in these data.

A comparison was also conducted between shift workers whose schedules included night shifts and employees working exclusively during the day. No statistically significant differences were observed for any of the analyzed variables: chronotype – day workers M = 40.0 (SD = 5.72) vs. shift workers M = 39.6 (SD = 6.07), t(1291) = 1.159, p = 0.247, d = 0.059; burnout – day workers M = 34.7 (SD = 18.59) vs. shift workers M = 35.2 (SD = 20.01), t_Welch(1280) = -0.498, p = 0.619, d = -0.026; work ability – day workers M = 29.2 (SD = 5.47) vs. shift workers M = 29.2 (SD = 5.47), t(1291) = 0.173, p = 0.863, d = 0.009.

Hypothesis testing

Figure 1 presents the tested mediation model. Path estimates for the full sample, as well as for employees working in a shift system including night shifts and for day workers, are shown in Table 4.

Fig. 1.

Estimated mediation model *** p < 0.001. Standardized coefficients from the model estimated on the full sample are shown. For clarity, control variables are omitted

Table 4.

Path coefficients from the estimated models

Path	Group	B	SE	95CI%LL	95CI%UL	p	β	R 2
a (Chronotype → Burnout)	Full sample	-0.439	0.085	-0.609	-0.274	< 0.001	-0.134	0.063
	Shift workers	-0.391	0.115	-0.618	-0.165	< 0.001	-0.120	0.109
	Day workers	-0.510	0.122	-0.747	-0.266	< 0.001	-0.155	0.032
b (Burnout → WA)	Full sample	-0.125	0.007	-0.138	-0.112	< 0.001	-0.441
	Shift workers	-0.116	0.010	-0.135	-0.098	< 0.001	-0.395
	Day workers	-0.132	0.009	-0.150	-0.113	< 0.001	-0.483
c’ (Direct effect Chronotype → WA)	Full sample	0.157	0.021	0.115	0.199	< 0.001	0.169
	Shift workers	0.148	0.032	0.087	0.213	< 0.001	0.155
	Day workers	0.166	0.028	0.112	0.222	< 0.001	0.184
Indirect effect	Full sample	0.055	0.011	0.033	0.078	< 0.001	0.059
	Shift workers	0.046	0.014	0.019	0.075	0.001	0.048
	Day workers	0.067	0.017	0.035	0.102	< 0.001	0.075
Total effect	Full sample	0.212	0.024	0.165	0.259	< 0.001	0.229	0.285
	Shift workers	0.193	0.035	0.127	0.262	< 0.001	0.202	0.347
	Day workers	0.233	0.033	0.169	0.301	< 0.001	0.258	0.226
Controlled variables
Age → Burnout	Full sample	0.308	0.046	0.220	0.399	< 0.001	0.170
	Shift workers	0.168	0.061	0.049	0.288	0.006	0.097
	Day workers	0.456	0.066	0.326	0.585	< 0.001	0.240
Gender → Burnout	Full sample	-6.235	0.957	-8.103	-4.286	< 0.001	-0.161
	Shift workers	-3.988	1.366	-6.541	-1.381	0.004	-0.107
	Day workers	-8.514	1.419	-11.243	-5.667	< 0.001	-0.213
Age → WA	Full sample	-0.096	0.012	-0.119	-0.072	< 0.001	-0.186
	Shift workers	-0.089	0.017	-0.123	-0.057	< 0.0001	-0.174
	Day workers	-0.102	0.017	-0.134	-0.070	< 0.001	-0.196
Gender → WA	Full sample	-0.699	0.243	-1.171	-0.230	0.004	-0.064
	Shift workers	-0.679	0.357	-1.386	0.024	0.058	-0.062
	Day workers	-0.733	0.338	-1.387	-0.069	0.030	-0.067
Covariances		Cov.	SE	95CI%LL	95CI%UL	p	r
Age ↔ Chronotype	Full sample	8.860	1.695	5.537	12.157	< 0.001	0.141
	Shift workers	8.339	2.367	3.820	13.135	< 0.001	0.136
	Day workers	9.285	2.416	4.402	14.010	< 0.001	0.145
Gender ↔ Chronotype	Full sample	-0.258	0.075	-0.400	-0.112	0.001	-0.088
	Shift workers	-0.052	0.104	-0.249	0.152	0.619	-0.018
	Day workers	-0.471	0.108	-0.680	-0.256	< 0.001	-0.156
Gender ↔ Age	Full sample	-0.071	0.137	-0.342	0.191	0.605	-0.013
	Shift workers	0.007	0.193	-0.376	0.376	0.972	0.001
	Day workers	-0.154	0.195	-0.542	0.224	0.429	-0.029

Gender was coded as: 0 – Female, 1 – Male

B/Cov. – unstandardized estimate,  β/r – standardized coefficient, WA Work ability

Across the full sample, a more morning-oriented chronotype was associated with lower burnout, while more evening-oriented individuals tended to experience higher burnout. Burnout, in turn, showed a strong negative link with work ability, and chronotype also exerted a positive direct effect on work ability. The indirect effect through burnout was significant, indicating partial mediation. In the full sample, the indirect pathway accounted for approximately 26% of the total effect of chronotype on work ability (24% among shift workers and 29% among day workers). Thus, Hypothesis 1 (chronotype positively related to work ability) and Hypothesis 2 (the mediating role of job burnout) were both supported. The same pattern emerged in both subgroups, confirming that the relationships were consistent across shift and day workers.

Although some of the estimated path coefficients appeared slightly stronger in either the day or night worker group, formal Wald χ² tests were conducted to examine whether these differences were statistically significant. In Table 5, the results of equality tests for specific paths are presented. None of the constraints reached significance, suggesting that the relationships among chronotype, burnout, and work ability were comparable for both groups. Thus, the mediation model demonstrated structural stability across groups.

Table 5.

Results of wald tests comparing path equality between day and shift workers

Path tested (equality across groups)	χ2(1)	p
Chronotype → Burnout (a₁ = a₂)	0.549	0.459
Burnout → Work Ability (b₁ = b₂)	1.646	0.199
Chronotype → Work Ability (c′₁ = c′₂)	0.191	0.662
Indirect effect (a₁×b₁ = a₂×b₂)	1.079	0.299

Discussion

The aim of the study was to examine the relationship between chronotype and work ability, as well as the mechanism through which this relationship occurs. The variable selected for testing as a potential mediator was job burnout. We assumed that employees with an evening chronotype, who achieve peak cognitive and physical performance later in the day [13], are often required to operate according to schedules that are misaligned with their natural predispositions, such as early work shifts or rigidly fixed task deadlines. This discrepancy gives rise to a conflict between the social clock (external and shared) and the biological clock (internal and individual). When such misalignment persists over time, it results in sleep deprivation, reduced sleep quality, and increased fatigue [14–16], which may contribute to job burnout [15–18] and, in the longer term, to lower work ability [66, 67].

The study findings confirmed both a direct positive relationship between chronotype and work ability, as well as the mediating role of job burnout. Individuals with an evening chronotype exhibited higher levels of job burnout, which, in turn, was associated with lower work ability. A likely mechanism underlying the relationship between evening chronotype and occupational burnout is the previously described social jet lag. Meeting daily work demands requires substantial physical and psychological effort. When occupational tasks must be performed at times misaligned with an individual’s natural circadian preferences, the demand for mobilizing personal resources necessary to maintain high performance further increases. Such intensified effort is associated with elevated psychophysiological costs [68], which contribute to gradual resource depletion, the development of chronic fatigue [69], and - in the longer term - a reduction in work ability [66, 67].

Work schedule (shift vs. day work) was included to examine whether exposure to circadian misalignment associated with shift work modifies the relationships between chronotype, burnout, and work ability. The results showed no significant differences in path coefficients or indirect effects between shift and day workers, indicating structural equivalence of the mediation model across groups. This suggests that the adverse consequences of chronotype–schedule misalignment are not limited to formal shift work but extend to any rigid or socially constrained schedule. Evening chronotype was associated with higher burnout risk and reduced work ability regardless of work schedule, highlighting chronotype as a broader individual vulnerability factor. Interpreted through the lens of the “social clock,” these findings indicate that misalignment between employees’ natural circadian preferences and imposed schedules can impair well-being and work functioning across occupational contexts, underscoring the importance of flexible scheduling to mitigate burnout and support sustainable work performance.

The present study makes a few important theoretical and practical contributions. From a theoretical perspective, it extends chronobiological and occupational health research by demonstrating that job burnout functions as a key psychological mechanism linking chronotype to work ability. By empirically confirming the mediating role of burnout, the findings help integrate chronotype research with models of occupational stress and resource depletion, showing that the disadvantage of an evening chronotype in organizational contexts is not direct but operates through prolonged strain and exhaustion. From a practical standpoint, the findings of this study provides valuable insights for organizations and professionals involved in employee health and well-being. Understanding the relationship between chronotype and work ability may inform the design of more flexible and individualized work schedules that accommodate employees’ natural temporal preferences. Furthermore, such knowledge could support burnout prevention programs by aligning workload and work rhythms with employees’ biological predispositions - ultimately promoting healthier, more effective, and sustainable participation in the workforce across different chronotypes.

Limitations and future research

Several limitations of the present study should be acknowledged. First, the cross-sectional design, based on a single measurement occasion, precludes drawing firm conclusions about the directionality and causality of the observed relationships between chronotype, job burnout, and work ability. Although the findings are consistent with theoretical assumptions suggesting that evening chronotype increases the risk of burnout, which in turn impairs work ability, alternative causal pathways cannot be ruled out. The relationships between chronotype, burnout, and work ability are likely not unidirectional, but instead form a process of reciprocal interactions characterized by self-reinforcing amplification over time. However, capturing this process is possible in multi-wave longitudinal studies.

Second, the gender imbalance in the study sample represents another important limitation. Nearly two-thirds of the participants were men. Although there is no strong empirical evidence suggesting that the health-related consequences of an evening chronotype differ substantially between women and men, this disproportion limits the generalizability of the findings. Caution is therefore warranted when extending the results to more gender-balanced or female-dominated occupational groups. Third, all constructs included in the analyses were assessed using self-report measures. While validated instruments were employed, self-report data are inherently susceptible to common method variance, social desirability, and subjective bias.

Future studies would benefit from incorporating more objective indicators, such as actigraphy-based assessments of sleep–wake patterns, physiological markers of stress, or supervisor-rated measures of work performance and functioning. Moreover, the influence of additional variables not included in the present study cannot be excluded and may affect the strength or nature of the observed associations. Finally, the study did not examine potential moderating variables. It is likely that both work-related factors (e.g., job resources, flexibility of working hours, autonomy) and individual characteristics (e.g., self-regulation capacity, resilience, or coping styles) may significantly shape the relationship between chronotype, burnout, and work ability. Including such moderators in future research would allow for a more nuanced understanding of when and for whom evening chronotype poses a risk to occupational functioning. In sum, future studies should address these limitations by employing longitudinal designs, more gender-balanced samples, multimethod assessments, and a broader set of mediating and moderating variables to further elucidate the complex relationship between chronotype and work ability.

Authors' contributions

Conceptualization and study design [LB, LK], Methodology [LB, LK], Data collection [LB, LK, SW, AN, JB], Data analysis and interpretation [LB, LK], Manuscript drafting [LB, LK, SW, AN, JB], Supervision [LB, JB], Critical revision of the manuscript [LB, LK, SW, AN, JB], Project administration [LK].

Funding

This work was supported by the National Centre for Research and Development within the 6th stage of the “Governmental Programme for Improvement of Safety and Working Conditions” (project No. IV.PN.02 entitled “Development of a package of tools to assess psychosocial determinants of safety and health at work,” project manager: Łukasz Kapica).

Data availability

The datasets used and analysed during the current study available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study was approved by the Bioethics Committee of the Institute of Rural Health in Lublin (Resolution No. 22/2023). The study was conducted in accordance with the Declaration of Helsinki and was approved by the relevant institutional ethics committee. All participants provided informed consent to participate in the study.

Consent for publication

Not applicable. No individual-level identifying data are included in this manuscript.

Competing interests

The authors declare no competing interests.