Healthcare is among the most demanding of all occupations. Physicians, nurses and allied health staff work amid high patient loads, time pressure, exposure to suffering and death, emotional labour and the constant possibility of error. The concept of stress as a physiological and psychological response to demands that tax adaptive capacity was first articulated decades ago [1,2], and later occupational models gave it a workplace form. The job demand–control model links high demands and low decision latitude to mental strain [3], while the effort–reward imbalance model attributes ill-health to sustained high effort that is poorly rewarded [4]. Work stress in physicians and nurses has measurable physiological correlates during routine clinical work [5], and prolonged exposure may culminate in burnout [6]. Nearly half of the physicians surveyed in a large national study in the United States reported at least one symptom of burnout [7], and a meta-analysis found that roughly one in four resident physicians screens positive for depression or depressive symptoms [8]. Sleep is a basic biological need and a sensitive barometer of wellbeing. Consensus statements recommend that healthy adults obtain at least seven hours of sleep per night on a regular basis [9,10], and short sleep duration is associated with higher all-cause mortality [11]. Healthcare professionals are structurally exposed to sleep loss because hospitals run round the clock. Night duty, rotating rosters, extended shifts and on-call work force wakefulness against the circadian drive for sleep, shorten and fragment the sleep that follows and cause chronic sleep debt [12–14]. Studies among nurses have shown that rotating shifts are associated with sleep-related accidents [15], poorer sleep quality and higher job stress [16], and that poor sleep is common across shifts and linked to the quality of the work environment and of care [17,18]. Sleep disturbance is not only a consequence of shift schedules; it is also a manifestation of psychosocial stress. Prospective data show that work stress predicts later insomnia [19], and psychosocial stress is a well-recognised cause of impaired sleep [20]. Several mechanisms have been proposed. Chronic insomnia is accompanied by activation of the hypothalamic–pituitary–adrenal axis [21]; individuals differ in their sleep reactivity, that is, their vulnerability to stress-related sleep disturbance and hyperarousal [22]; and predisposing, precipitating and perpetuating factors interact in the course of insomnia [23], with worry and selective attention to sleep-related threat maintaining it [24]. In this sense, a doctor or nurse who begins to lie awake after a difficult shift may be displaying the first observable sign of occupational strain, well before formal anxiety or depressive disorders emerge. Poor sleep in turn is a robust risk marker for psychological morbidity. Persistent insomnia was associated with a markedly higher risk of new-onset major depression in an early large epidemiological study [25], insomnia in young men predicted subsequent depression [26], and a meta-analysis of longitudinal studies estimated that people with insomnia have roughly twice the odds of developing depression [27]. Chronic insomnia also predicts anxiety [28], and insomnia, depression and anxiety frequently co-occur and often precede one another [29,30]. Among doctors in training, the transition to rotating shifts has been shown to precipitate depression and anxiety in those with high sleep reactivity [31], and long working hours and stressful events predicted depressive symptoms during internship [32]. The consequences extend beyond the individual. Interns working extended shifts made significantly more serious medical errors than those on a schedule that limited work hours [33], had a markedly higher risk of motor vehicle crashes after extended shifts [34] and more percutaneous injuries [35]. A synthesis of the evidence concluded that provider work hours and sleep deprivation compromise safety and performance [36]. Sleep disturbance also predicts occupational injury [37] and carries substantial costs in lost productivity [38]. The Indian context adds particular urgency. India's health workforce is stretched, and long duty hours, high patient loads and limited rest facilities are routine in many public and teaching hospitals. The National Mental Health Survey of India documented a considerable burden of mental disorders in the adult population and a very wide treatment gap [39]. Indian clinical practice guidelines have emphasised that sleep disorders remain under-recognised and under-treated in routine practice [40], and community-based work from India has shown that sleep-disordered breathing is common and often undiagnosed [41–43]. Indian studies among medical students have reported frequent sleep problems and substantial academic stress [44,45], and surveys of young adults using standardised scales have found a considerable burden of depression, anxiety and stress [46]. However, relatively few Indian studies have examined occupational stress, sleep disturbance and psychological morbidity together across the different categories of healthcare professionals, and fewer still have treated sleep disturbance as an indicator of occupational stress. The present study was therefore undertaken with the following aim and objectives. Aim: To study sleep disturbances as a manifestation of occupational stress and their association with psychological morbidity among healthcare professionals. Objectives: To assess the prevalence and pattern of sleep disturbances among doctors, nurses and paramedical staff. To determine the level of occupational stress and its relationship with sleep disturbances. To estimate the prevalence of anxiety and depression (psychological morbidity) and their association with sleep disturbances. To identify occupational and sociodemographic factors independently associated with poor sleep and psychological morbidity.

Study design, setting and duration This was a hospital-based, observational, cross-sectional study conducted in a tertiary care teaching hospital in India over a period of six months. The hospital provides round-the-clock inpatient, emergency and critical care services, which makes its workforce representative of those exposed to shift work and on-call duties. Study population and eligibility The study population comprised doctors (consultants, residents and interns), nursing staff and paramedical staff (laboratory, radiology, physiotherapy and pharmacy personnel). Participants were included if they were aged 21 years or above, had at least six months of continuous service in the hospital and provided written informed consent. Staff were excluded if they had a previously diagnosed sleep disorder (such as obstructive sleep apnoea or narcolepsy), a diagnosed psychiatric illness on treatment, were on regular hypnotic or psychotropic medication, were pregnant or in the postpartum period, or were on leave during the data collection period. Sample size and sampling technique The sample size was calculated using the formula n = Z²pq/d² [47], assuming a prevalence of poor sleep of 60%, an absolute precision of 6% and a 95% confidence level, which gave a minimum of 256 participants. Allowing for about 15% non-response and incomplete questionnaires, the sample was rounded up to 300. Participants were selected by stratified random sampling. The hospital staff roster was divided into three strata (doctors, nurses and paramedical staff), and participants were drawn from each stratum by computer-generated random numbers in proportion to stratum size, giving 90 doctors, 150 nurses and 60 paramedical staff. Study tools Semi-structured proforma. This recorded age, gender, marital status, professional category, department, years of experience, working hours per week, duty pattern (fixed day or rotating shifts) and the number of night duties per month. Occupational Stress Index (OSI). Occupational stress was measured with the OSI developed by Srivastava and Singh [48], an Indian instrument widely used in occupational research. It consists of 46 items rated on a five-point scale and covers twelve job-related dimensions, including role overload, role ambiguity, role conflict, responsibility for persons, powerlessness, poor peer relations, low status and strenuous working conditions. Total scores range from 46 to 230, with higher scores indicating greater stress. For this study, scores of 100 or less were categorised as low, 101 to 140 as moderate and above 140 as high occupational stress. Pittsburgh Sleep Quality Index (PSQI). Sleep quality over the previous month was assessed with the PSQI [49], a 19-item self-rated questionnaire that generates seven component scores (subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleeping medication and daytime dysfunction), each scored from 0 to 3. The global score ranges from 0 to 21, and a score above 5 identifies a poor sleeper with a reported sensitivity of 89.6% and specificity of 86.5% [49]. Epworth Sleepiness Scale (ESS). Daytime sleepiness was assessed with the ESS [50], which asks respondents to rate their likelihood of dozing in eight everyday situations from 0 to 3 (total 0–24). A score above 10 was taken to indicate excessive daytime sleepiness. Hospital Anxiety and Depression Scale (HADS). Psychological morbidity was screened with the HADS [51], which has seven items each for anxiety (HADS-A) and depression (HADS-D), scored 0 to 21 per subscale. A cut-off score of 8 or more on either subscale was used to define probable anxiety or depression, a threshold that balances sensitivity and specificity at approximately 0.80 in a large review of validation studies [52]. Psychological morbidity was defined as a score of 8 or more on HADS-A and/or HADS-D. Data collection procedure After explaining the purpose of the study, written informed consent was obtained. Participants completed the self-administered, anonymous questionnaires in a quiet room during off-duty hours, which took about 20 to 25 minutes. Questionnaires were checked for completeness on the spot, and respondents were assured that individual responses would remain confidential. Participants with high HADS scores or severe sleep complaints were counselled and offered referral to the psychiatry and sleep clinic. Ethical considerations The study was approved by the Institutional Ethics Committee. Participation was voluntary, anonymity was maintained and participants could withdraw at any stage without any consequence. Statistical analysis Data were entered in Microsoft Excel and analysed with SPSS software (version 22.0). Categorical variables were expressed as frequencies and percentages and continuous variables as mean ± standard deviation. The chi-square test was used to compare proportions, one-way analysis of variance to compare means across the three professional categories, and Pearson's correlation coefficient to assess the relationship between continuous scores. Odds ratios (OR) with 95% confidence intervals (CI) were calculated for associations with psychological morbidity. Variables with p < 0.10 on bivariate analysis, together with age and gender, were entered into multivariable logistic regression models to obtain adjusted odds ratios (AOR) for poor sleep and for psychological morbidity. Model fit was assessed with the Hosmer–Lemeshow test and Nagelkerke R². A p-value below 0.05 was considered statistically significant.

Sociodemographic and occupational profile All 300 selected participants returned complete questionnaires (response rate 100%). Their mean age was 33.6 ± 8.4 years, and the largest group (39.3%) was aged 21 to 30 years. Women formed 57.3% of the sample, and 65.3% were married. Half of the participants were nurses (50.0%), 30.0% were doctors and 20.0% were paramedical staff. The mean length of service was 9.2 ± 7.1 years. Rotating shifts were worked by 56.0% of participants, and 51.3% performed four or more night duties per month. The mean working time was 52.8 ± 10.6 hours per week, and 62.0% worked more than 48 hours weekly (Table 1). Table 1: Sociodemographic and occupational characteristics of the study participants (N = 300) Variable Category n (%) Age group (years) 21–30 118 (39.3) 31–40 112 (37.3) 41–50 48 (16.0) >50 22 (7.3) Gender Male 128 (42.7) Female 172 (57.3) Marital status Married 196 (65.3) Unmarried 104 (34.7) Professional category Doctors 90 (30.0) Nurses 150 (50.0) Paramedical staff 60 (20.0) Department Medicine and allied 88 (29.3) Surgery and allied 72 (24.0) ICU / emergency 68 (22.7) Obstetrics-gynaecology / paediatrics 42 (14.0) Laboratory / radiology / others 30 (10.0) Experience (years) < 5 96 (32.0) 5–10 94 (31.3) >10 110 (36.7) Duty pattern Rotating shifts 168 (56.0) Fixed day duty 132 (44.0) Night duties per month ≥ 4 154 (51.3) < 4 146 (48.7) Working hours per week >48 186 (62.0) ≤ 48 114 (38.0) Mean age 33.6 ± 8.4 years; mean experience 9.2 ± 7.1 years; mean working hours 52.8 ± 10.6 per week. Occupational stress, sleep and psychological morbidity The mean OSI score was 128.4 ± 24.6. Occupational stress was low in 16.0%, moderate in 44.0% and high in 40.0% of participants. The mean OSI score was highest among doctors (134.0 ± 22.8), followed by nurses (127.9 ± 24.1) and paramedical staff (121.3 ± 26.0), a difference that was statistically significant (F = 4.90, p = 0.008). The mean global PSQI score was 7.0 ± 3.5, and 66.0% of participants were poor sleepers (PSQI > 5). The average self-reported sleep duration was 5.8 ± 1.3 hours, and 46.0% slept for less than six hours. Excessive daytime sleepiness (ESS > 10) was found in 28.0%. On the HADS, probable anxiety was present in 32.0% and probable depression in 28.0%; overall, 42.0% had psychological morbidity and 18.0% had both anxiety and depression (Table 2). Table 2: Distribution of occupational stress, sleep parameters and psychological morbidity (N = 300) Measure Mean ± SD Category / cut-off n (%) Occupational stress (OSI) 128.4 ± 24.6 Low (≤ 100) 48 (16.0) Moderate (101–140) 132 (44.0) High (> 140) 120 (40.0) Sleep quality (PSQI) 7.0 ± 3.5 Poor sleep (> 5) 198 (66.0) Sleep duration (hours) 5.8 ± 1.3 < 6 hours per night 138 (46.0) Daytime sleepiness (ESS) 8.6 ± 4.4 Excessive (> 10) 84 (28.0) Anxiety (HADS-A) 7.1 ± 3.9 Score ≥ 8 96 (32.0) Depression (HADS-D) 6.4 ± 3.7 Score ≥ 8 84 (28.0) Psychological morbidity – Anxiety and/or depression 126 (42.0) Co-morbid anxiety and depression – Both scores ≥ 8 54 (18.0) OSI: Occupational Stress Index; PSQI: Pittsburgh Sleep Quality Index; ESS: Epworth Sleepiness Scale; HADS: Hospital Anxiety and Depression Scale. Pattern of sleep disturbance Among the seven PSQI components, daytime dysfunction had the highest mean score (1.38 ± 0.86), followed by sleep duration (1.40 ± 0.92) and subjective sleep quality (1.10 ± 0.78). Short sleep duration was the most frequent clinically significant problem (46.0% scoring 2 or more), followed by daytime dysfunction (37.3%), delayed sleep latency (32.0%) and poor subjective sleep quality (28.0%). Regular use of sleeping medication was uncommon (7.3%) but not negligible in a group that had been screened to exclude those on prescribed hypnotics (Table 3). Table 3: Component-wise analysis of the Pittsburgh Sleep Quality Index (N = 300) PSQI component Mean score ± SD Component score ≥ 2, n (%) Subjective sleep quality 1.10 ± 0.78 84 (28.0) Sleep latency 0.96 ± 0.92 96 (32.0) Sleep duration 1.40 ± 0.92 138 (46.0) Habitual sleep efficiency 0.78 ± 0.98 70 (23.3) Sleep disturbances 1.08 ± 0.58 62 (20.7) Use of sleeping medication 0.30 ± 0.79 22 (7.3) Daytime dysfunction 1.38 ± 0.86 112 (37.3) Global PSQI score (0–21) 7.0 ± 3.5 198 (66.0) with score > 5 Each component is scored 0–3; a component score of 2 or more indicates a clinically meaningful disturbance. Factors associated with poor sleep Poor sleep was most common among doctors (73.3%), followed by nurses (66.7%) and paramedical staff (53.3%) (p = 0.039). Gender was not significantly associated with poor sleep (women 69.8% vs men 60.9%, p = 0.110). Strong associations were observed with duty pattern, night duty frequency and working hours: 79.8% of those on rotating shifts were poor sleepers compared with 48.5% on fixed day duty; 85.7% of those with four or more night duties a month compared with 45.2% of those with fewer; and 80.6% of those working more than 48 hours weekly compared with 42.1% of those working less. A clear dose–response relationship was seen with occupational stress, the proportion of poor sleepers rising from 29.2% in the low-stress group to 62.1% in the moderate-stress group and 85.0% in the high-stress group (p < 0.001) (Table 4). Table 4: Association of poor sleep (PSQI > 5) with occupational and sociodemographic factors (N = 300) Variable Category Total (n) Poor sleep n (%) χ² (df) p-value Professional category Doctors 90 66 (73.3) 6.48 (2) 0.039* Nurses 150 100 (66.7) Paramedical staff 60 32 (53.3) Gender Male 128 78 (60.9) 2.55 (1) 0.110 Female 172 120 (69.8) Duty pattern Rotating shifts 168 134 (79.8) 32.2 (1) < 0.001* Fixed day duty 132 64 (48.5) Night duties / month ≥ 4 154 132 (85.7) 54.8 (1) < 0.001* < 4 146 66 (45.2) Working hours / week > 48 186 150 (80.6) 46.8 (1) < 0.001* ≤ 48 114 48 (42.1) Occupational stress Low 48 14 (29.2) 49.2 (2) < 0.001* Moderate 132 82 (62.1) High 120 102 (85.0) *Statistically significant (chi-square test). Percentages are row percentages. Correlation between stress, sleep and psychological scores The OSI score showed a significant moderate-to-strong positive correlation with the PSQI score (r = 0.58, p < 0.001), and weaker but significant correlations with ESS, HADS-A and HADS-D scores. The PSQI score correlated positively with both HADS-A (r = 0.52) and HADS-D (r = 0.49), and weekly working hours were modestly correlated with poor sleep quality (r = 0.35). All correlations were statistically significant at p < 0.001 (Table 5). Table 5: Pearson correlation between continuous study variables (N = 300) Variables correlated Correlation coefficient (r) p-value OSI score and PSQI score 0.58 < 0.001 OSI score and ESS score 0.37 < 0.001 OSI score and HADS-A score 0.46 < 0.001 OSI score and HADS-D score 0.41 < 0.001 PSQI score and ESS score 0.44 < 0.001 PSQI score and HADS-A score 0.52 < 0.001 PSQI score and HADS-D score 0.49 < 0.001 ESS score and HADS-D score 0.28 < 0.001 Weekly working hours and PSQI score 0.35 < 0.001 OSI: Occupational Stress Index; PSQI: Pittsburgh Sleep Quality Index; ESS: Epworth Sleepiness Scale; HADS-A/D: Hospital Anxiety and Depression Scale – Anxiety/Depression. Sleep disturbance and psychological morbidity Psychological morbidity was far more frequent among poor sleepers than good sleepers. Anxiety was seen in 42.4% of poor sleepers versus 11.8% of good sleepers (OR 5.53; 95% CI 2.84–10.75), depression in 36.4% versus 11.8% (OR 4.29; 95% CI 2.20–8.36) and either condition in 54.5% versus 17.6% (OR 5.60; 95% CI 3.13–10.01). The co-existence of anxiety and depression was nearly five times as likely in poor sleepers (OR 5.12; 95% CI 2.11–12.4) (Table 6). Table 6: Association between sleep quality and psychological morbidity Outcome (HADS ≥ 8) Poor sleepers (n = 198) n (%) Good sleepers (n = 102) n (%) Odds ratio (95% CI) p-value Anxiety 84 (42.4) 12 (11.8) 5.53 (2.84–10.75) < 0.001 Depression 72 (36.4) 12 (11.8) 4.29 (2.20–8.36) < 0.001 Anxiety and/or depression 108 (54.5) 18 (17.6) 5.60 (3.13–10.01) < 0.001 Both anxiety and depression 48 (24.2) 6 (5.9) 5.12 (2.11–12.40) < 0.001 Good sleepers are the reference group. CI: confidence interval. Multivariable analysis In the logistic regression model for poor sleep, high occupational stress (AOR 5.84; 95% CI 2.62–13.01), four or more night duties per month (AOR 4.12; 95% CI 2.31–7.35), more than 48 working hours per week (AOR 2.78; 95% CI 1.58–4.90) and rotating shifts (AOR 2.36; 95% CI 1.31–4.25) were independent predictors. Moderate stress also remained significant (AOR 2.71), as did being a doctor compared with paramedical staff (AOR 2.15). Gender and nursing profession were not independently associated (Table 7). In the model for psychological morbidity, poor sleep was the strongest independent predictor (AOR 4.72; 95% CI 2.51–8.87), followed by high occupational stress (AOR 3.18; 95% CI 1.62–6.24) and excessive daytime sleepiness (AOR 1.92; 95% CI 1.08–3.41), whereas duty pattern variables, gender and age lost significance after adjustment (Table 8). Table 7: Multivariable logistic regression: independent predictors of poor sleep (PSQI > 5) Predictor Reference category Adjusted OR (95% CI) p-value Doctors Paramedical staff 2.15 (1.01–4.58) 0.047* Nurses Paramedical staff 1.64 (0.84–3.20) 0.148 Female gender Male 1.38 (0.78–2.45) 0.270 Rotating shift duty Fixed day duty 2.36 (1.31–4.25) 0.004* Night duties ≥ 4 / month < 4 / month 4.12 (2.31–7.35) < 0.001* Working hours > 48 / week ≤ 48 / week 2.78 (1.58–4.90) < 0.001* Moderate occupational stress Low stress 2.71 (1.28–5.74) 0.009* High occupational stress Low stress 5.84 (2.62–13.01) < 0.001* *Statistically significant. Hosmer–Lemeshow p = 0.42; Nagelkerke R² = 0.46. Age was included as a covariate (not shown). Table 8: Multivariable logistic regression: independent predictors of psychological morbidity (HADS-A and/or HADS-D ≥ 8) Predictor Reference category Adjusted OR (95% CI) p-value Poor sleep (PSQI > 5) Good sleep 4.72 (2.51–8.87) < 0.001* Moderate occupational stress Low stress 1.74 (0.91–3.33) 0.094 High occupational stress Low stress 3.18 (1.62–6.24) 0.001* Excessive daytime sleepiness (ESS > 10) ESS ≤ 10 1.92 (1.08–3.41) 0.026* Female gender Male 1.46 (0.86–2.48) 0.163 Night duties ≥ 4 / month < 4 / month 1.31 (0.76–2.26) 0.331 Working hours > 48 / week ≤ 48 / week 1.52 (0.88–2.63) 0.133 Age > 40 years ≤ 40 years 0.88 (0.45–1.72) 0.707 *Statistically significant. Hosmer–Lemeshow p = 0.58; Nagelkerke R² = 0.38.

This cross-sectional study examined sleep disturbance as an expression of occupational stress and as a correlate of psychological morbidity among healthcare professionals in a tertiary care hospital. Three principal findings emerged: two-thirds of the staff were poor sleepers; poor sleep rose steeply with the level of occupational stress and with shift-related work demands; and poor sleepers carried a four- to five-fold higher burden of anxiety and depression than good sleepers. The proportion of poor sleepers (66.0%) and the average sleep duration (5.8 hours) fall below the recommended sleep time for healthy adults [9,10]. This is considerably higher than the 30–35% prevalence of insomnia symptoms and the roughly 10% prevalence of chronic insomnia disorder reported for general populations [53], which underscores the occupational nature of the problem. Our finding is consistent with studies of nurses in other countries, where a majority of staff were poor sleepers on the PSQI irrespective of shift [18], and with reports that shift work in nursing is associated with sleep and health problems [16,17]. Indian data on sleep disorders are mostly from community surveys or students: sleep problems were common among medical students in western Maharashtra [44], and sleep-disordered breathing was frequent in urban Indian adults [41,43]. The high rate observed here among practising professionals suggests that sleep difficulties that begin during training in India may persist and intensify with clinical responsibility. The prominence of daytime dysfunction and short sleep duration among the PSQI components, rather than medication use, also indicates that the problem is largely behavioural and structural, and thus modifiable, rather than the result of an established, treated sleep disorder; the Indian guidelines also point to widespread under-recognition of such problems [40]. The graded rise in poor sleep from 29.2% (low stress) to 85.0% (high stress), the correlation between OSI and PSQI scores (r = 0.58) and the persistence of stress as the strongest independent predictor after adjusting for shift pattern and working hours (AOR 5.84) support the view that sleep disturbance is more than a by-product of long hours. Prospective work has shown that work stress predicts subsequent insomnia [19], and psychosocial stress is a recognised cause of impaired sleep [20]. Biological hyperarousal, with activation of the stress axis [21], and individual differences in sleep reactivity [22] offer plausible mechanisms. The behavioural cycle of worry, attention to sleeplessness and compensatory habits described in cognitive models of insomnia [23,24] may then perpetuate the disturbance even when the stressor eases. The OSI is built on Indian occupational conditions [48], and the finding that doctors scored highest on this index is in line with the stress profile reported among Indian medical trainees and students [45] and with the high level of burnout reported among physicians elsewhere [7]. The comparatively higher stress and poorer sleep among doctors in our sample probably reflect heavy on-call duties, responsibility for life-and-death decisions and longer working hours. Frequent night duty (AOR 4.12), rotating shifts and working more than 48 hours a week were independently associated with poor sleep. These findings concur with the long-standing evidence that night work and extended shifts disrupt circadian alignment and shorten sleep [12–14], and with studies reporting poorer sleep and greater job stress among rotating-shift nurses [16] and more sleepiness-related accidents in nurses on rotating schedules [15]. The observation that staff with long hours had almost double the prevalence of poor sleep is also consistent with reports of safety consequences of extended work hours in trainee doctors [33–36]. Women had a higher proportion of poor sleep than men, but the difference was not significant after adjustment, which suggests that the occupational exposure rather than gender drives the risk in this setting. Anxiety and depression were present in 32.0% and 28.0% of participants, respectively. The proportion with depression is close to the pooled estimate for resident physicians [8], and the overall burden of psychological morbidity (42.0%) is well above the rate of mental disorders in the Indian general adult population documented by the National Mental Health Survey [39], though the screening cut-off used here identifies probable cases and not clinical diagnoses. Poor sleepers had about five times the odds of anxiety and four times the odds of depression, and poor sleep remained the strongest predictor of psychological morbidity after adjusting for stress and work pattern (AOR 4.72). This agrees with epidemiological evidence that insomnia is associated with new-onset depression [25–27] and anxiety [28], and that the three conditions frequently co-exist [29,30]. Among doctors in training, sleep reactivity predicted depression and anxiety following the transition to rotating shifts [31], and long working hours were among the factors predicting depressive symptoms during internship [32]. Our results mirror those observations in a mixed group of practising professionals. Importantly, the association persisted even when occupational stress was in the model, which implies that sleep disturbance carries its own risk for psychological morbidity beyond the stress that precipitated it. The significant independent contribution of excessive daytime sleepiness points in the same direction. Given the wide treatment gap for mental disorders in India [39], workplace identification of sleep complaints could serve as an accessible entry point for detecting those in need of help. First, sleep should be treated as an occupational health indicator, and brief screening with tools such as the PSQI and HADS could be incorporated into annual health check-ups for hospital staff. Second, rostering practices need review. Limiting continuous duty hours, capping consecutive night shifts, providing adequate recovery time after night duty and ensuring access to rest rooms are feasible measures, and interventions that reduced intern work hours have been shown to cut serious medical errors [33]. Planned napping during night duty improved alertness and performance in emergency physicians and nurses [54]. Third, institution-level stress reduction is needed through workload redistribution, adequate staffing, fair reward and supportive supervision, in line with the job-strain and effort–reward frameworks [3,4]. Fourth, individual-level programmes are valuable: non-pharmacological interventions, especially cognitive-behavioural strategies, are effective for insomnia [55], and meditation-based programmes can reduce anxiety and depression [56]. Improved sleep may also lessen the injury risk and productivity losses attributed to sleep disturbance [37,38]. Finally, institutions should make confidential counselling and referral for sleep and mental health problems easily accessible in order to narrow the gap highlighted by Indian national data [39,40]. Strengths and limitations The strengths of this study include the use of standardised, validated instruments, including an Indian occupational stress scale, stratified random sampling across professional categories, and adjustment for key confounders in multivariable models. Several limitations should be acknowledged. The cross-sectional design precludes causal inference and the direction of association between stress, sleep and psychological morbidity may be bidirectional. Self-reported measures are subject to recall and social desirability bias and no objective sleep measurement (actigraphy or polysomnography) was used, so that sleep apnoea and other primary sleep disorders could not be fully excluded. The HADS is a screening tool and does not provide clinical diagnoses. The study was carried out in a single institution and the findings may not be generalisable to private, rural or primary care settings. Finally, unmeasured factors such as personal and family stressors, caffeine and screen use, commuting time and baseline chronotype might have influenced the results. Prospective multicentre studies, ideally with objective sleep measures, are needed to confirm these associations and test interventions.

Sleep disturbance is highly prevalent among healthcare professionals and is closely linked to occupational stress, night duty, rotating shifts and long working hours. Poor sleep is, in turn, strongly associated with anxiety and depression, independent of the level of occupational stress. Sleep complaints in doctors, nurses and paramedical staff should therefore be regarded as an early warning signal of occupational strain and an actionable target for prevention. A combined approach of routine sleep and mental health screening, humane duty rosters, organisational stress reduction and accessible support services is recommended to protect the wellbeing of the healthcare workforce and the safety of the patients they serve.

1. Selye H. The Stress of Life. New York: McGraw-Hill; 1956. 2. Lazarus RS, Folkman S. Stress, Appraisal, and Coping. New York: Springer; 1984. 3. Karasek RA. Job demands, job decision latitude, and mental strain: implications for job redesign. Adm Sci Q. 1979;24(2):285-308. 4. Siegrist J. Adverse health effects of high-effort/low-reward conditions. J Occup Health Psychol. 1996;1(1):27-41. 5. Rutledge T, Stucky E, Dollarhide A, Shively M, Jain S, Wolfson T, et al. A real-time assessment of work stress in physicians and nurses. Health Psychol. 2009;28(2):194-200. 6. Maslach C, Jackson SE. The measurement of experienced burnout. J Organ Behav. 1981;2(2):99-113. 7. Shanafelt TD, Boone S, Tan L, Dyrbye LN, Sotile W, Satele D, et al. Burnout and satisfaction with work-life balance among US physicians relative to the general US population. Arch Intern Med. 2012;172(18):1377-85. 8. Mata DA, Ramos MA, Bansal N, Khan R, Guille C, Di Angelantonio E, et al. Prevalence of depression and depressive symptoms among resident physicians: a systematic review and meta-analysis. JAMA. 2015;314(22):2373-83. 9. Watson NF, Badr MS, Belenky G, Bliwise DL, Buxton OM, Buysse D, et al. Recommended amount of sleep for a healthy adult: a joint consensus statement of the American Academy of Sleep Medicine and Sleep Research Society. Sleep. 2015;38(6):843-4. 10. Hirshkowitz M, Whiton K, Albert SM, Alessi C, Bruni O, DonCarlos L, et al. National Sleep Foundation's sleep time duration recommendations: methodology and results summary. Sleep Health. 2015;1(1):40-3. 11. Cappuccio FP, D'Elia L, Strazzullo P, Miller MA. Sleep duration and all-cause mortality: a systematic review and meta-analysis of prospective studies. Sleep. 2010;33(5):585-92. 12. Åkerstedt T. Shift work and disturbed sleep/wakefulness. Occup Med (Lond). 2003;53(2):89-94. 13. Rajaratnam SM, Arendt J. Health in a 24-h society. Lancet. 2001;358(9286):999-1005. 14. Kecklund G, Axelsson J. Health consequences of shift work and insufficient sleep. BMJ. 2016;355:i5210. 15. Gold DR, Rogacz S, Bock N, Tosteson TD, Baum TM, Speizer FE, et al. Rotating shift work, sleep, and accidents related to sleepiness in hospital nurses. Am J Public Health. 1992;82(7):1011-4. 16. Lin SH, Liao WC, Chen MY, Fan JY. The impact of shift work on nurses' job stress, sleep quality and self-perceived health status. J Nurs Manag. 2014;22(5):604-12. 17. Admi H, Tzischinsky O, Epstein R, Herer P, Lavie P. Shift work in nursing: is it really a risk factor for nurses' health and patients' safety? Nurs Econ. 2008;26(4):250-7. 18. Gómez-García T, Ruzafa-Martínez M, Fuentelsaz-Gallego C, Madrid JA, Rol MA, Martínez-Madrid MJ, et al. Nurses' sleep quality, work environment and quality of care in the Spanish National Health System: observational study among different shifts. BMJ Open. 2016;6(8):e012073. 19. Linton SJ. Does work stress predict insomnia? A prospective study. Br J Health Psychol. 2004;9(Pt 2):127-36. 20. Åkerstedt T. Psychosocial stress and impaired sleep. Scand J Work Environ Health. 2006;32(6):493-501. 21. Vgontzas AN, Bixler EO, Lin HM, Prolo P, Mastorakos G, Vela-Bueno A, et al. Chronic insomnia is associated with nyctohemeral activation of the hypothalamic-pituitary-adrenal axis: clinical implications. J Clin Endocrinol Metab. 2001;86(8):3787-94. 22. Drake C, Richardson G, Roehrs T, Scofield H, Roth T. Vulnerability to stress-related sleep disturbance and hyperarousal. Sleep. 2004;27(2):285-91. 23. Spielman AJ, Caruso LS, Glovinsky PB. A behavioral perspective on insomnia treatment. Psychiatr Clin North Am. 1987;10(4):541-53. 24. Harvey AG. A cognitive model of insomnia. Behav Res Ther. 2002;40(8):869-93. 25. Ford DE, Kamerow DB. Epidemiologic study of sleep disturbances and psychiatric disorders: an opportunity for prevention? JAMA. 1989;262(11):1479-84. 26. Chang PP, Ford DE, Mead LA, Cooper-Patrick L, Klag MJ. Insomnia in young men and subsequent depression: the Johns Hopkins Precursors Study. Am J Epidemiol. 1997;146(2):105-14. 27. Baglioni C, Battagliese G, Feige B, Spiegelhalder K, Nissen C, Voderholzer U, et al. Insomnia as a predictor of depression: a meta-analytic evaluation of longitudinal epidemiological studies. J Affect Disord. 2011;135(1-3):10-9. 28. Neckelmann D, Mykletun A, Dahl AA. Chronic insomnia as a risk factor for developing anxiety and depression. Sleep. 2007;30(7):873-80. 29. Taylor DJ, Lichstein KL, Durrence HH, Reidel BW, Bush AJ. Epidemiology of insomnia, depression, and anxiety. Sleep. 2005;28(11):1457-64. 30. Ohayon MM, Roth T. Place of chronic insomnia in the course of depressive and anxiety disorders. J Psychiatr Res. 2003;37(1):9-15. 31. Kalmbach DA, Pillai V, Cheng P, Arnedt JT, Drake CL. Shift work disorder, depression, and anxiety in the transition to rotating shifts: the role of sleep reactivity. Sleep Med. 2015;16(12):1532-8. 32. Sen S, Kranzler HR, Krystal JH, Speller H, Chan G, Gelernter J, et al. A prospective cohort study investigating factors associated with depression during medical internship. Arch Gen Psychiatry. 2010;67(6):557-65. 33. Landrigan CP, Rothschild JM, Cronin JW, Kaushal R, Burdick E, Katz JT, et al. Effect of reducing interns' work hours on serious medical errors in intensive care units. N Engl J Med. 2004;351(18):1838-48. 34. Barger LK, Cade BE, Ayas NT, Cronin JW, Rosner B, Speizer FE, et al. Extended work shifts and the risk of motor vehicle crashes among interns. N Engl J Med. 2005;352(2):125-34. 35. Ayas NT, Barger LK, Cade BE, Hashimoto DM, Rosner B, Cronin JW, et al. Extended work duration and the risk of self-reported percutaneous injuries in interns. JAMA. 2006;296(9):1055-62. 36. Lockley SW, Barger LK, Ayas NT, Rothschild JM, Czeisler CA, Landrigan CP. Effects of health care provider work hours and sleep deprivation on safety and performance. Jt Comm J Qual Patient Saf. 2007;33(11 Suppl):7-18. 37. Salminen S, Oksanen T, Vahtera J, Sallinen M, Härmä M, Salo P, et al. Sleep disturbances as a predictor of occupational injuries among public sector workers. J Sleep Res. 2010;19(1 Pt 2):207-13. 38. Rosekind MR, Gregory KB, Mallis MM, Brandt SL, Seal B, Lerner D. The cost of poor sleep: workplace productivity loss and associated costs. J Occup Environ Med. 2010;52(1):91-8. 39. Gururaj G, Varghese M, Benegal V, Rao GN, Pathak K, Singh LK, et al. National Mental Health Survey of India, 2015-16: Summary. Bengaluru: National Institute of Mental Health and Neuro Sciences; 2016. NIMHANS Publication No. 128. 40. Gupta R, Das S, Gujar K, Mishra KK, Gaur N, Majid A. Clinical practice guidelines for sleep disorders. Indian J Psychiatry. 2017;59(Suppl 1):S116-38. 41. Reddy EV, Kadhiravan T, Mishra HK, Sreenivas V, Handa KK, Sinha S, et al. Prevalence and risk factors of obstructive sleep apnea among middle-aged urban Indians: a community-based study. Sleep Med. 2009;10(8):913-8. 42. Sharma SK, Vasudev C, Sinha S, Banga A, Pandey RM, Handa KK. Validation of the modified Berlin questionnaire to identify patients at risk for the obstructive sleep apnoea syndrome. Indian J Med Res. 2006;124(3):281-90. 43. Sharma SK, Ahluwalia G. Epidemiology of adult obstructive sleep apnoea syndrome in India. Indian J Med Res. 2010;131:171-5. 44. Giri PA, Baviskar MP, Phalke DB. Study of sleep habits and sleep problems among medical students of Pravara Institute of Medical Sciences Loni, Western Maharashtra, India. Ann Med Health Sci Res. 2013;3(1):51-4. 45. Supe AN. A study of stress in medical students at Seth G.S. Medical College. J Postgrad Med. 1998;44(1):1-6. 46. Sahoo S, Khess CR. Prevalence of depression, anxiety, and stress among young male adults in India: a dimensional and categorical diagnoses-based study. J Nerv Ment Dis. 2010;198(12):901-4. 47. Cochran WG. Sampling Techniques. 3rd ed. New York: John Wiley & Sons; 1977. 48. Srivastava AK, Singh AP. Construction and standardization of an occupational stress index: a pilot study. Indian J Clin Psychol. 1981;8:133-6. 49. Buysse DJ, Reynolds CF 3rd, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res. 1989;28(2):193-213. 50. Johns MW. A new method for measuring daytime sleepiness: the Epworth Sleepiness Scale. Sleep. 1991;14(6):540-5. 51. Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale. Acta Psychiatr Scand. 1983;67(6):361-70. 52. Bjelland I, Dahl AA, Haug TT, Neckelmann D. The validity of the Hospital Anxiety and Depression Scale: an updated literature review. J Psychosom Res. 2002;52(2):69-77. 53. Ohayon MM. Epidemiology of insomnia: what we know and what we still need to learn. Sleep Med Rev. 2002;6(2):97-111. 54. Smith-Coggins R, Howard SK, Mac DT, Wang C, Kwan S, Rosekind MR, et al. Improving alertness and performance in emergency department physicians and nurses: the use of planned naps. Ann Emerg Med. 2006;48(5):596-604. 55. Morin CM, Culbert JP, Schwartz SM. Nonpharmacological interventions for insomnia: a meta-analysis of treatment efficacy. Am J Psychiatry. 1994;151(8):1172-80. 56. Goyal M, Singh S, Sibinga EM, Gould NF, Rowland-Seymour A, Sharma R, et al. Meditation programs for psychological stress and well-being: a systematic review and meta-analysis. JAMA Intern Med. 2014;174(3):357-68.