TABLE 2.
Measures of sleep disturbance and mental health in the included studies (n = 59).
| References | Measures of sleep disturbance | Adjustment factors in the model | Measures of mental health | Statistical model |
| Prospective study | ||||
| Hillhouse et al. (27) | Sleep hours | Gender, specialty and citizenship (US versus non-US) | Perceived Stress Scale (PSS) | Hierarchical regression |
| Sørengaard et al. (21) | The Bergen Insomnia Scale (BIS)>2 | / | Hospital Anxiety and Depression Scale (HADS) | Structural equation modelling |
| Wang et al. (28) | Pittsburgh Sleep Quality Index(PSQI) >7 | Age, marital status, offspring status, current work tenure, professional status and depression status. |
The Patient Health Questionnaire-9 (PHQ-9) (item No.9)>1 | Multiple logistic regression models |
| Fang et al. (29) | 24 h total sleep time (TST) | Age, gender, bedtime median and Wake time median |
PHQ-9 | Multivariable linear regression models |
| Cross-sectional studies | ||||
| Ruggiero et al. (30) | PSQI | / | Standard Shiftwork Index Chronic Fatigue Scale (SSICFS) | Simultaneous Multiple Regression |
| Lu et al. (31) | Designed by researchers: choose sleep disturbance or not | / | Maslach Burnout Inventory Human Services Survey (MBI-HSS) | Multiple linear regression |
| Peterson et al. (32) | Designed by researchers:three items | / | The Hospital Anxiety and Depression Scale (HAD) | Pearson correlation |
| Rutledge et al. (33) | PSQI | / | Diary of Ambulatory Behavioral States | Multiple linear regression |
| Stucky et al. (34) | Sleep quality scale>0 and sleep hours per day | Age, gender, familiarity with patients, average patientload and number of admissions in 24 hours | Study-developed instrument containing 10 Likert scale questions | Multiple linear regression |
| Sun et al. (35) | Health status (grade): Do you have difficulty in sleeping (no/slight/serious) | Age | Zung Self-Rating Anxiety Scale (SAS) | General linear regression |
| Aldrees et al. (36) | Sleep hours <6 h per day | / | Maslach Burnout Inventory (MBI) | Multiple logistic regression analysis |
| Yost et al. (22) | Sleep hours per day | / | MBI-HSS | Spearman correlation |
| Chin et al. (37) | Sleep hours <6 h per day | / | Modified Chinese version of the Copenhagen Burnout Inventory (C-CBI) | Multiple logistical regression |
| Qiao et al. (38) | Symptom Checklist 90 (SCL-90): sleep part>2 | / | Maslach Burnout Inventory-General Survey (MBI-GS) | Multiple logistic regression |
| Koyama et al. (39) | Six items related to early insomnia from the Structured Interview Guide for the Hamilton Depression Rating Scale (SIGH-D)>2 | / | Depression (six items) from the brief job stress questionnaire>6 | Multiple logistic regression |
| Vilchez-Cornejo et al. (40) | Designed by researchers: sleep disturbance or not | Number of jobs, sex | MBI-HSS | Multivariate paired logistic regression |
| Cai et al. (41) | Sleep hours per day | / | PHQ-9>10 | Poisson regression |
| Vidotti et al. (42) | Sleep hours ≤ 6 h per day | Age,Marital status, Occupation, Health related variable Exercise, Regular diet, Health status, Psychiatric symptom, Work related variable Hospital rank, Turnover intention, Physician–patient relationship | The 11-item Chalder fatigue scale (CFS) for measuring burnout >4. | Multiple logistical regression |
| Wang et al. (43) | Designed by researchers. Continuous variable from 0 (very poor) to10 (perfect) | / | Professional Quality of Life Scale (ProQOL-CN) | Multiple linear regression |
| Ibrahim et al. (44) | Sleep hours <6 h per day | Age, nationality, exercise, service type, number of meals per day and number of breakfasts | the 21-item Depression Anxiety Stress Scale (DASS-21)>2 | Binary and multinomial logistic regression |
| Dai et al. (45) | PSQI>5 | / | HADS>7 | Multiple logistic regression |
| Ghasemi et al. (46) | PSQI | / | Swedish Occupational Fatigue Inventory (SOFI) | Structural equation modelling |
| Cross-sectional studies | ||||
| Aydin Sayilan et al. (47) | ISI>2 | / | DASS-21 | Multivariable logistic regression |
| Youssef et al. (48) | PSQI | Gender | Post-traumatic Stress Disorder Checklist for DSM-5 (PCL-5) >33 | hierarchical multiple regression |
| Yin et al. (49) | sleep disorder screening | / | MBI-HSS | Not mentioned |
| Weaver et al. (4) | Sleep hours <6 h per day | Age, study site, alcohol drinking, smoking, exercise, hypnotic medication use, and night shift work |
Core symptoms of depression>2 | Multivariable logistic regression |
| Ji et al. (50) | PSQI>7 | SAS>50 | Pearson linear correlation | |
| Furihata et al. (51) | Sleep hours per day | Demographic and clinical variables and comorbid psychological symptoms |
Short health anxiety inventory (SHAI) | Hierarchical multivariable regression |
| Cheng et al. (52) | Sleep hours per day | / | CBI>50 | Multivariate linear regression |
| Higgins et al. (53) | PSQI>5 | / | MBI | Multiple regression analysis |
| Ng et al. (54) | The PROMIS Sleep Disturbance Short Form | / | PHQ-9 | Structural Equation Modelling |
| Ding et al. (55) | Sleep Habits Questionnaire (CHAS) | / | Goldberg’s General Health Questionnaire (GHQ-28) | Linear regression |
| Eva et al. (56) | PSQI>5 | / | Generalized Anxiety Disorder 7-item scale (GAD-7) ≥ 4 | Multivariate logistic regression |
| Tu et al. (57) | Design by researchers: sleep disturbance or not | / | MBI-HSS | Multiple logistic regression analysis |
| Magnavita et al. (58) | Sleep Condition Indicator (SCI) | Age, marital status, offspring status, current work tenure, professional status and depression status |
PHQ-9: item No.9 >1 | Multivariate logistic regression |
| Korkmaz et al. (59) | PSQI>5 | / | Beck Anxiety Inventory (BAI) >7 | Pearson correlation |
| Secosan et al. (60) | PSQI | / | Swedish Occupational Fatigue Inventory (SOFI) | Pearson correlation |
| Teo et al. (61) | Designed by researchers: good sleep quality or poor sleep quality. | Age, sex | Generalized Anxiety Disorder 7-item (GAD-7)>5. The Zung Self-Rating Depression Scale (SDS)>50 | Multivariate logistic regression |
| Tasdemir et al. (62) | ISI ≥ 8 | / | General Health Questionnaire-12 (GHQ-12) > 16 | Multiple linear regression |
| Wang et al. (63) | Sleep duration ≤ 6 h per day/PROMIS Sleep Disturbance Short Form | Social support, work-family conflict, and negative behaviors at work | The 10-item version of the Center for Epidemiologic Studies Depression Scale (CES-D)>10 | Multivariate linear regression |
| Mousavi et al. (64) | PSQI | / | SAS | Spearman correlation |
| Simonetti et al. (65) | PSQI ≥ 5 | / | HADS-Anxiety part >10. HADS-depression part >7. | Spearman Correlation |
| Zhang et al. (13) | PSQI>7 | / | The Impact of Event Scale (IES-R) >33 | Multivariate logistic regression |
| Zhang et al. (66) | ISI>7 | / | IES-R>7 | Multivariate logistic regression |
| Yitayih et al. (67) | PSQI | / | CES-D>16 | Multivariate logistic regression |
| Chen et al. (68) | ISI>7 | / | DASS-21: depression part>4, anxiety part>3 or stress part>7. | structural equation modelling |
| Kandemir et al. (69) | ISI | / | The Maslach Burnout Inventory-Short Form (MBI-SF) | Spearman correlation |
| Aydin et al. (70) | ISI>5 | Age, body mass index, sex (if appropriate), physical activity, household income, working time, night shifts, visiting friends constantly, religious or not, marital status, siblings or not, experienced a major life event or not, current year of residency, smoking status, alcohol consumption, coffee intake, and specialty. | PHQ-9>5 | Multivariate logistic regression |
| Cross-sectional studies | ||||
| Chang et al. (71) | PSQI>5 | / | The Link Burnout Questionnaire (LBQ)>0 | Linear regression |
| Mokros et al. (72) | Sleep hours per day | Demographic and clinical variables, and comorbid psychological symptoms. | short health anxiety inventory (SHAI)>27 | Multivariate logistic regression |
| Abdelghani et al. (73) | Sleep quality scale: 0-10 scale (Reverse scoring). | / | GAD-7 for measuring anxiety>10; PHQ-9 for measuring depression>10. | Multivariate logistic regression |
| Abu-Elenin (74) | Designed by researchers for measuring sleep quality: good/general/bad/very bad | / | GAD-7 for measuring anxiety>10; PHQ-9 for measuring depression>10. | Linear regression |
| Pang et al. (18) | Designed by researchers:choose sleep satisfaction or sleep deprivation | / | MBI | Multivariate logistic regression |
| Abbas et al. (77) | PSQI>5 | / | 12-item General Health Questionnaire>2 | Pearson correlation |
| Olagunju et al. (76) | PSQI>5 | / | CES-D>15 | Pearson correlation |
| Hsieh et al. (77) | PSQI>7 | / | GAD-7 ≥ 4 | Pearson correlation |
| Jiang et al. (78) | One item extracted out of PSQI | / | PTSD checklist for DSM-5 (PCL-5) ≥ 33 | Hierarchical regression analysis |
| Geng et al. (79) | PSQI ≥ 6 | / | PHQ-9>5 | Multivariate logistic regression |
| Garcia et al. (80) | How many nightmares did you have that woke you up?—If the answer was greater than one time, the sleep quality was poor. | / | ‘I felt stressed’, rated on a scale of 0 (not at all) to 4 (extremely) | Multilevel models |