Explaining the gender gap in sickness absence

Explaining the gender gap in sickness absence Abstract Background In many western countries, women have a much higher rate of sickness absence than men. To what degree the gender differences in sickness absence are caused by gender differences in health is largely unknown. Aims To assess to what degree the gender gap in sickness absence can be explained by health factors and work- and family-related stressors. Methods Norwegian parents participating in the Tracking Opportunities and Problems (TOPP) study were asked about sickness absence and a range of factors possibly contributing to gender differences in sickness absence, including somatic and mental health, sleep problems, job control/demands, work–home conflicts, parent–child conflicts and stressful life events. Using a cross-sectional design, we did linear regression analyses, to assess the relative contribution from health and stressors. Results There were 557 study participants. Adjusting for health factors reduced the gender difference in sickness absence by 24%, while adjusting for stressors in the family and at work reduced the difference by 22%. A simultaneous adjustment for health factors and stressors reduced the difference in sickness absence by about 28%. Conclusions Despite adjusting for a large number of factors, including both previously well-studied factors (e.g. health, job control/demands) and lesser-studied factors (parent–child conflict and sexual assault), this study found that most of the gender gap in sickness absence remains unexplained. Gender differences in health and stressors account for only part of the differences in sickness absence. Other factors must, therefore, exist outside the domains of health, work and family stressors. Family stressors, gender gap, health, sickness absence, work factors Introduction In many western countries, there is a large and unexplained gender gap in the rates of sickness absence [1]. In many countries, including Norway, women’s sickness absence is about 50% higher than men’s. These gender differences in sickness absence may lead to negative long-term outcomes for women such as reduced income and career opportunities, stigmatization and long-term labour market exclusion [2–4]. There are many hypotheses for why women have higher sickness absence than men, but no single hypothesis has yet been able to explain the considerable gender differences. Socially based hypotheses have suggested that women may be subject to higher levels of stress, possibly aggravated by the effect of the so-called ‘double-burden hypothesis’, the idea that working women also carry more of the burden at home [5]. For example, women may do more of the childcare or household chores, or they may be more involved when relatives become ill. There is some evidence that women experience a higher degree of conflict between demands at work and at home, but this has been found to explain little of the gender differences in sickness absence [5]. One possible explanation could be that women have physically and/or mentally more demanding jobs, or that female-dominated workplaces are at increased risk due to the nature of the work. Some previous studies have supported this view [6], while a more recent study found that gender differences in sickness absence increased when adjusting for type of occupation [7]. Health-related factors may also contribute to gender differences in sickness absence. For instance, women have a higher prevalence of anxiety, depression and musculoskeletal pain conditions [8,9], which are important causes of sickness absence [10,11]. Also, sickness absence in pregnancy has increased considerably in Norway over the last two decades, and now accounts for about 25% of the gender difference in sickness absence [12]. Reasons for the sharp increase in sickness absence in pregnancy are largely unknown and are not related to delayed first pregnancies [12]. While some of these hypotheses may explain the gender gap in sickness absence, a large and unexplained difference remains in the rates of sickness absence between men and women. Most studies have examined possible explanatory factors separately, and in the few studies that do include explanatory factors from several domains, the conclusions vary [13–15]. A recent review article concluded that work–family conflict may contribute to the gender gap, but that there are few studies on this topic [5]. This study aimed to examine the gender gap in sickness absence using comprehensive questionnaire data from the Tracking Opportunities and Problems (TOPP) study [16]. The study included questions about sickness absence and a wide range of factors, including somatic and mental health problems, sleep problems, job control/demands, work–family conflict, parent–child conflict and stressful life events. TOPP is conducted on a sample of Norwegian parents, which is an advantage for studying family stressors. Our aim was to examine gender differences in sickness absence among employed Norwegian parents in a cross-sectional design, while adjusting for a wide variety of factors, to see how much of the gender gap can be explained, and to establish the relative contribution from work- and family-related stressors and from health differences. Methods TOPP is an on-going Norwegian community-based study of parents that were recruited when they attended for vaccination of their child at a municipal health centre [16,17]. In 2011, when their child was 18–20 years old, the parents were sent a questionnaire about sickness absence, health and various stressors. The current study is based on data from this questionnaire. The collection and use of data was approved by the regional committees for medical and health research ethics. Sickness absence was assessed by the following question: ‘Have you been away from work over the past six months due to own illness?’, and the respondents were asked to give the total number of days, weeks and months they had been away from work. The total sickness absence for each participant was then calculated. As a measure of physical health, the participants were asked to report if they, over the past 12 months, had had any of the following disorders: ‘cardiac/vascular disorder’, ‘metabolic disorder’, ‘asthma’, ‘Hay fever/allergy’, ‘skin disease/eczema’, ‘musculoskeletal disorder’, ‘kidney/urinary tract disorder’, ‘fibromyalgia’, ‘gynaecological/genital disorder’, ‘migraine’, ‘diabetes’, ‘abdominal/intestinal disorder’ and ‘cancer’. Responses were coded as 13 separate dichotomous variables. As a measure of mental health, participants were asked about 24 items from the Hopkins Symptom checklist (HSCL-25) [18], which measures the presence of symptoms of anxiety and depression in a non-psychiatric setting. For each item in the checklist the participants rate the degree to which they have felt affected by a given symptom on a four-point scale ranging from 1 (‘not at all’) to 4 (‘extremely affected’). One item from the HSCL-25, ‘Loss of sexual interest or pleasure’, was not included in the questionnaire because some participants perceived this item as inappropriate in a pilot study. The average of each participant’s HSCL-25 score was treated as a continuous variable in the analyses. Sleeping difficulties were assessed by the question ‘Have you, over the past 12 months, experienced sleeping difficulties that affected your work ability?’, and the response alternatives were ‘yes’ and ‘no’. Participants were assessed for the demands and control domains of Karasek and Theorell’s model of work strain by questions from the Job Content Questionnaire [19], including five questions on the psychological demands latitude (demands) and four questions on the decision latitude (control). Responses were coded from 1 to 5, and for each participant, two mean values were calculated, one demands score and one control score. Finally, we defined a continuous variable as a measure of job strain by subtracting the demand score from the control score. This model is supported by previous research [20]. Using 15 items from the Oldenburg Burnout Inventory (OLBI) [21], we measured exhaustion and disengagement from work. For each item, a statement was rated from 1 (‘strongly agree’) to 4 (‘strongly disagree’). Exhaustion was measured with seven items (example: ‘After my work, I regularly feel worn out and weary’), while disengagement was measured with eight items (example: ‘I often talk about my work in a negative way’). Due to a misprint, one exhaustion item from OLBI (‘After my work, I usually feel worn out and weary’) was not included in the questionnaire. For each participant, the average scores for each of the two dimensions (exhaustion and disengagement) were calculated. As a measure of exposure to family stressors, participants were asked to rate the burden from 10 different stressors experienced over the past 12 months. Responses were coded from 1 (‘No burden’) to 4 (‘Very heavy burden’), and for each participant, the mean value for all stressors was estimated. The 10 stressors were: ‘Teenager’s health problems’, ‘Other children’s health problems’, ‘Problems related to sick parents or parents-in-law’, ‘Problems related to structuring the children’s/teenager’s daily life’, ‘Problems with combining work life and child care’, ‘Stressors related to mood swings in children/teenagers’, ‘Worries about what the children/teenager exposes themselves to, or what they may be exposed to, in their spare time’, ‘Problems related to children’s/teenager’s schooling’, ‘Worries about the children’s future’ and ‘Problems related to setting limits for the children/teenagers’. These questions have previously been used in research as measures of family stressors [22]. Conflict with teenagers was assessed using the conflict subscale of the Child Parent Relationship Scale (CPRS, Pianta et al.). Each of six statements is rated from 1 (‘does not fit’) to 5 (‘always fits’), and for each participant, the average score was calculated. Sexual assaults is a stressor with considerable gender imbalance [23,24], and since these experiences may have long-lasting effects on the victims, they may potentially contribute to the gender gap in sickness absence. To measure previous occurrences of sexual assault, the participants were asked if they had ever experienced being pushed or forced into sexual acts. In the current study, the responses ‘Yes, I was forced’ and ‘Yes, I was raped’ were defined as occurrences of sexual assault which was used as a dichotomous variable. Information about participants’ level of education was retrieved from a previous data collection in TOPP that was conducted on the same sample in 2008, and the responses were coded from 1 (‘up to 9 years of schooling’) to 5 (‘4 or more years in university/college’). For 53 participants, we did not have information about education, and for these we imputed the missing data using the multiple imputation function in IBM SPSS Statistics version 23 for Windows with choice of method set to automatic and with all variables used in the analyses as predictor variables. The unadjusted gender differences in both sickness absence and in the potentially associated factors were estimated in the whole study population by running independent sample t-tests. For the main analyses, we conducted linear regression analyses with sickness absence as the dependent variable and gender as the independent variable, adjusted for each of the possible confounding factors. First, we ran the analyses by adjusting for each factor separately, then again with all the factors entered into the model one by one, looking at the cumulative effect on the adjusted gender difference in sickness absence. All analyses were carried out in IBM SPSS Statistics 23 for Windows. Results In total, 627 parents participated in the data collection in 2011. Participants who reported occupational status as homemaker, student or unemployed, as well as those reporting pregnancy during the previous year were excluded. After exclusions, the study population consisted of 240 men and 317 women—a total of 557 participants. Gender differences in the examined variables are presented in Table 1. The unadjusted analyses showed significant gender differences in sickness absence, as women had on average 12.6 days of absence over the past 6 months while men only had 5.5 days of absence (i.e. 7.09 excess sick days for women). There were also gender differences in most of the adjusting factors, with significant differences found for some somatic illnesses (i.e. thyroid disease, asthma, gynaecological/genito-urinary disease, migraine, musculoskeletal disorder, fibromyalgia), for mental health, sleep problems, family stressors, sexual victimization and for work factors (job control and exhaustion). All these factors were more frequently reported by women than men. Table 1. Gender differences in the study sample   Men (n = 240)  Women (n = 317)  Significance  Sickness absence (days)  5.45  12.55  **  Age  51.15  48.53  **  Education (range: 1–5)  3.91  3.76    Somatic illness   Cardiac/vascular disorder  6%  6%     Thyroid disorder  3%  13%  **   Diabetes  3%  3%     Cancer  3%  1%     Asthma  5%  9%  *   Hay fever/allergies  16%  18%     Dermatitis/eczema  11%  11%     Kidney/urinary tract disorder  3%  5%     Gynaecological/genital disorder  2%  7%  **   Migraine  8%  14%  *   Abdominal/intestinal disorders  4%  6%     Musculoskeletal symptoms  19%  31%  **   Fibromyalgia  2%  6%  *  Mental illness   Mental distress (Hopkins Symptom Checklist) (range: 0–3)  1.24  1.32  **   Sleep problems  15%  25%  **  Job stressors   Exhaustion (Olbi, range: 1–4)  1.92  2.00  *   Disengagement (Olbi, range: 1–4)  2.00  2.05     Job demands (Karasek, range: 1–5)  3.22  3.29     Job control (Karasek, range: 1–5)  3.70  3.44  **   Control minus demands (range −4 to 4)  0.48  0.15  **  Family stressors   Family stressors (burden) (range: 1–4)  1.33  1.42  **   Teenager conflict (range: 1–5) (Pianta)  1.74  1.81    Sexual victimization (forced/raped)  1%  5%  **    Men (n = 240)  Women (n = 317)  Significance  Sickness absence (days)  5.45  12.55  **  Age  51.15  48.53  **  Education (range: 1–5)  3.91  3.76    Somatic illness   Cardiac/vascular disorder  6%  6%     Thyroid disorder  3%  13%  **   Diabetes  3%  3%     Cancer  3%  1%     Asthma  5%  9%  *   Hay fever/allergies  16%  18%     Dermatitis/eczema  11%  11%     Kidney/urinary tract disorder  3%  5%     Gynaecological/genital disorder  2%  7%  **   Migraine  8%  14%  *   Abdominal/intestinal disorders  4%  6%     Musculoskeletal symptoms  19%  31%  **   Fibromyalgia  2%  6%  *  Mental illness   Mental distress (Hopkins Symptom Checklist) (range: 0–3)  1.24  1.32  **   Sleep problems  15%  25%  **  Job stressors   Exhaustion (Olbi, range: 1–4)  1.92  2.00  *   Disengagement (Olbi, range: 1–4)  2.00  2.05     Job demands (Karasek, range: 1–5)  3.22  3.29     Job control (Karasek, range: 1–5)  3.70  3.44  **   Control minus demands (range −4 to 4)  0.48  0.15  **  Family stressors   Family stressors (burden) (range: 1–4)  1.33  1.42  **   Teenager conflict (range: 1–5) (Pianta)  1.74  1.81    Sexual victimization (forced/raped)  1%  5%  **  Variables shown are either percentages or mean values. *P < 0.05, **P < 0.01. View Large Table 1. Gender differences in the study sample   Men (n = 240)  Women (n = 317)  Significance  Sickness absence (days)  5.45  12.55  **  Age  51.15  48.53  **  Education (range: 1–5)  3.91  3.76    Somatic illness   Cardiac/vascular disorder  6%  6%     Thyroid disorder  3%  13%  **   Diabetes  3%  3%     Cancer  3%  1%     Asthma  5%  9%  *   Hay fever/allergies  16%  18%     Dermatitis/eczema  11%  11%     Kidney/urinary tract disorder  3%  5%     Gynaecological/genital disorder  2%  7%  **   Migraine  8%  14%  *   Abdominal/intestinal disorders  4%  6%     Musculoskeletal symptoms  19%  31%  **   Fibromyalgia  2%  6%  *  Mental illness   Mental distress (Hopkins Symptom Checklist) (range: 0–3)  1.24  1.32  **   Sleep problems  15%  25%  **  Job stressors   Exhaustion (Olbi, range: 1–4)  1.92  2.00  *   Disengagement (Olbi, range: 1–4)  2.00  2.05     Job demands (Karasek, range: 1–5)  3.22  3.29     Job control (Karasek, range: 1–5)  3.70  3.44  **   Control minus demands (range −4 to 4)  0.48  0.15  **  Family stressors   Family stressors (burden) (range: 1–4)  1.33  1.42  **   Teenager conflict (range: 1–5) (Pianta)  1.74  1.81    Sexual victimization (forced/raped)  1%  5%  **    Men (n = 240)  Women (n = 317)  Significance  Sickness absence (days)  5.45  12.55  **  Age  51.15  48.53  **  Education (range: 1–5)  3.91  3.76    Somatic illness   Cardiac/vascular disorder  6%  6%     Thyroid disorder  3%  13%  **   Diabetes  3%  3%     Cancer  3%  1%     Asthma  5%  9%  *   Hay fever/allergies  16%  18%     Dermatitis/eczema  11%  11%     Kidney/urinary tract disorder  3%  5%     Gynaecological/genital disorder  2%  7%  **   Migraine  8%  14%  *   Abdominal/intestinal disorders  4%  6%     Musculoskeletal symptoms  19%  31%  **   Fibromyalgia  2%  6%  *  Mental illness   Mental distress (Hopkins Symptom Checklist) (range: 0–3)  1.24  1.32  **   Sleep problems  15%  25%  **  Job stressors   Exhaustion (Olbi, range: 1–4)  1.92  2.00  *   Disengagement (Olbi, range: 1–4)  2.00  2.05     Job demands (Karasek, range: 1–5)  3.22  3.29     Job control (Karasek, range: 1–5)  3.70  3.44  **   Control minus demands (range −4 to 4)  0.48  0.15  **  Family stressors   Family stressors (burden) (range: 1–4)  1.33  1.42  **   Teenager conflict (range: 1–5) (Pianta)  1.74  1.81    Sexual victimization (forced/raped)  1%  5%  **  Variables shown are either percentages or mean values. *P < 0.05, **P < 0.01. View Large Table 2 lists the excess number of sick days for women (as compared to men) before and after adjustment for each single factor. Most of these adjustments reduced the gender difference in sickness absence, and, ranked by the resulting reduction in beta, the most important explanatory factors were sleep problems, musculoskeletal disease, mental distress, the relationship between demands and control at work, the exhaustion domain of the OLBI, family stressors) and conflict with teenagers. Table 2. Gender differences in sickness absence—adjusted for explanatory factors   Excess sick days for women  95% CI  Effect of adjustment (change in sick days)  Rank of contribution  Not adjusted  7.09  2.27–11.91  –  –  Education  7.07  2.24–11.91  −0.02  15  Health factors   Somatic health    Cardiac/vascular disorder  7.09  2.26–11.91  0  17    Thyroid disorder  6.79  1.90–11.68  −0.30  8    Diabetes  7.06  2.23–11.88  −0.03  13    Cancer  7.35  2.52–12.17  +0.26  20    Asthma  7.03  2.19–11.87  −0.06  12    Hay fever/allergies  7.07  2.24–11.90  −0.02  14    Dermatitis/eczema  7.09  2.27–11.91  0  16    Kidney/urinary tract disorder  6.96  2.13–11.79  −0.13  11    Gynaecological/genital disorder  7.53  2.67–12.39  +0.44  21    Migraine  7.09  2.24–11.94  0  18    Abdominal/intestinal disorder  7.07  2.24–11.90  −0.02  15    Musculoskeletal symptoms  5.72  0.94–10.51  −1.37  2    Fibromyalgia  6.86  2.01–11.71  −0.23  10   Mental health    Mental distress (Hopkins Symptom Checklist)  5.94  1.09–10.79  −1.15  3    Sleep problems  5.72  0.91–10.53  −1.37  1  Stressors   Job stressors    Exhaustion (Olbi)  6.53  1.70–11.36  −0.56  5    Disengagement (Olbi)  6.85  2.03–11.67  −0.24  9    Karasek substitution (control–demand)  6.43  1.52–11.35  −0.66  4     Job demands (Karasek)  7.29  2.43–12.15  +0.20       Job control (Karasek)  6.17  1.26–11.07  −0.92     Family stressors    Family stressors (burden)  6.59  1.72–11.46  −0.50  6    Teenager conflict (Pianta)  6.63  1.85–11.40  −0.46  7  Sexual victimization (forced/raped)  7.19  2.33–12.05  +0.10  19  All health factors  5.42  0.41–10.43  −1.67    All stressors  5.56  0.65–10.47  −1.53    Education + health + stressors  5.10  0.04–10.15  −1.99      Excess sick days for women  95% CI  Effect of adjustment (change in sick days)  Rank of contribution  Not adjusted  7.09  2.27–11.91  –  –  Education  7.07  2.24–11.91  −0.02  15  Health factors   Somatic health    Cardiac/vascular disorder  7.09  2.26–11.91  0  17    Thyroid disorder  6.79  1.90–11.68  −0.30  8    Diabetes  7.06  2.23–11.88  −0.03  13    Cancer  7.35  2.52–12.17  +0.26  20    Asthma  7.03  2.19–11.87  −0.06  12    Hay fever/allergies  7.07  2.24–11.90  −0.02  14    Dermatitis/eczema  7.09  2.27–11.91  0  16    Kidney/urinary tract disorder  6.96  2.13–11.79  −0.13  11    Gynaecological/genital disorder  7.53  2.67–12.39  +0.44  21    Migraine  7.09  2.24–11.94  0  18    Abdominal/intestinal disorder  7.07  2.24–11.90  −0.02  15    Musculoskeletal symptoms  5.72  0.94–10.51  −1.37  2    Fibromyalgia  6.86  2.01–11.71  −0.23  10   Mental health    Mental distress (Hopkins Symptom Checklist)  5.94  1.09–10.79  −1.15  3    Sleep problems  5.72  0.91–10.53  −1.37  1  Stressors   Job stressors    Exhaustion (Olbi)  6.53  1.70–11.36  −0.56  5    Disengagement (Olbi)  6.85  2.03–11.67  −0.24  9    Karasek substitution (control–demand)  6.43  1.52–11.35  −0.66  4     Job demands (Karasek)  7.29  2.43–12.15  +0.20       Job control (Karasek)  6.17  1.26–11.07  −0.92     Family stressors    Family stressors (burden)  6.59  1.72–11.46  −0.50  6    Teenager conflict (Pianta)  6.63  1.85–11.40  −0.46  7  Sexual victimization (forced/raped)  7.19  2.33–12.05  +0.10  19  All health factors  5.42  0.41–10.43  −1.67    All stressors  5.56  0.65–10.47  −1.53    Education + health + stressors  5.10  0.04–10.15  −1.99    View Large Table 2. Gender differences in sickness absence—adjusted for explanatory factors   Excess sick days for women  95% CI  Effect of adjustment (change in sick days)  Rank of contribution  Not adjusted  7.09  2.27–11.91  –  –  Education  7.07  2.24–11.91  −0.02  15  Health factors   Somatic health    Cardiac/vascular disorder  7.09  2.26–11.91  0  17    Thyroid disorder  6.79  1.90–11.68  −0.30  8    Diabetes  7.06  2.23–11.88  −0.03  13    Cancer  7.35  2.52–12.17  +0.26  20    Asthma  7.03  2.19–11.87  −0.06  12    Hay fever/allergies  7.07  2.24–11.90  −0.02  14    Dermatitis/eczema  7.09  2.27–11.91  0  16    Kidney/urinary tract disorder  6.96  2.13–11.79  −0.13  11    Gynaecological/genital disorder  7.53  2.67–12.39  +0.44  21    Migraine  7.09  2.24–11.94  0  18    Abdominal/intestinal disorder  7.07  2.24–11.90  −0.02  15    Musculoskeletal symptoms  5.72  0.94–10.51  −1.37  2    Fibromyalgia  6.86  2.01–11.71  −0.23  10   Mental health    Mental distress (Hopkins Symptom Checklist)  5.94  1.09–10.79  −1.15  3    Sleep problems  5.72  0.91–10.53  −1.37  1  Stressors   Job stressors    Exhaustion (Olbi)  6.53  1.70–11.36  −0.56  5    Disengagement (Olbi)  6.85  2.03–11.67  −0.24  9    Karasek substitution (control–demand)  6.43  1.52–11.35  −0.66  4     Job demands (Karasek)  7.29  2.43–12.15  +0.20       Job control (Karasek)  6.17  1.26–11.07  −0.92     Family stressors    Family stressors (burden)  6.59  1.72–11.46  −0.50  6    Teenager conflict (Pianta)  6.63  1.85–11.40  −0.46  7  Sexual victimization (forced/raped)  7.19  2.33–12.05  +0.10  19  All health factors  5.42  0.41–10.43  −1.67    All stressors  5.56  0.65–10.47  −1.53    Education + health + stressors  5.10  0.04–10.15  −1.99      Excess sick days for women  95% CI  Effect of adjustment (change in sick days)  Rank of contribution  Not adjusted  7.09  2.27–11.91  –  –  Education  7.07  2.24–11.91  −0.02  15  Health factors   Somatic health    Cardiac/vascular disorder  7.09  2.26–11.91  0  17    Thyroid disorder  6.79  1.90–11.68  −0.30  8    Diabetes  7.06  2.23–11.88  −0.03  13    Cancer  7.35  2.52–12.17  +0.26  20    Asthma  7.03  2.19–11.87  −0.06  12    Hay fever/allergies  7.07  2.24–11.90  −0.02  14    Dermatitis/eczema  7.09  2.27–11.91  0  16    Kidney/urinary tract disorder  6.96  2.13–11.79  −0.13  11    Gynaecological/genital disorder  7.53  2.67–12.39  +0.44  21    Migraine  7.09  2.24–11.94  0  18    Abdominal/intestinal disorder  7.07  2.24–11.90  −0.02  15    Musculoskeletal symptoms  5.72  0.94–10.51  −1.37  2    Fibromyalgia  6.86  2.01–11.71  −0.23  10   Mental health    Mental distress (Hopkins Symptom Checklist)  5.94  1.09–10.79  −1.15  3    Sleep problems  5.72  0.91–10.53  −1.37  1  Stressors   Job stressors    Exhaustion (Olbi)  6.53  1.70–11.36  −0.56  5    Disengagement (Olbi)  6.85  2.03–11.67  −0.24  9    Karasek substitution (control–demand)  6.43  1.52–11.35  −0.66  4     Job demands (Karasek)  7.29  2.43–12.15  +0.20       Job control (Karasek)  6.17  1.26–11.07  −0.92     Family stressors    Family stressors (burden)  6.59  1.72–11.46  −0.50  6    Teenager conflict (Pianta)  6.63  1.85–11.40  −0.46  7  Sexual victimization (forced/raped)  7.19  2.33–12.05  +0.10  19  All health factors  5.42  0.41–10.43  −1.67    All stressors  5.56  0.65–10.47  −1.53    Education + health + stressors  5.10  0.04–10.15  −1.99    View Large Table 2 includes the cumulative effect of adjusting for all somatic and mental health factors, all family and work stressors, and finally the effect of adjusting for all variables included in the study (health, stressors and education). The unadjusted difference in sickness absence was 7.09 excess sick days over 6 months for women. Adjusting for somatic and mental health problems reduced the difference to 5.42 excess sick days. Adjusting for family- and work-related stressors reduced the difference to 5.56. Adjusting for all factors simultaneously, including education, health and family stressors, reduced the gender difference in sickness absence to 5.10 days. These results are visualized in Figure 1. Figure 1. View largeDownload slide Gender differences in sickness absence (excess sick days for women compared to men). Presented unadjusted, adjusted for health factors, adjusted for stressors and fully adjusted. Figure 1. View largeDownload slide Gender differences in sickness absence (excess sick days for women compared to men). Presented unadjusted, adjusted for health factors, adjusted for stressors and fully adjusted. Discussion This study was not able to entirely explain the gender gap in sickness absence. Adjusting for a range of health factors accounted for only about 22–24% of the observed gender difference in sickness absence. After further adjusting for various stressors, about 72% of the gender difference remained unexplained. About a quarter of the gender difference in sickness absence could be attributed to differences in health. Work and family stressors contributed to a similar extent, but adjusting for these added little beyond the adjustments for health. Previous studies of the popular ‘double-burden hypothesis’ has largely failed to explain why women’s sickness absence rates are so much higher than men’s [5]. There is also strong evidence against the hypothesis that industries or professions employing mostly women cause more sickness absence than those employing mostly men [7]. The gender difference in sickness absence, therefore, remains largely unexplained. The five most important factors were sleep problems, musculoskeletal symptoms, mental distress, the control dimension of Karasek and Theorell’s model of job stress and the exhaustion dimension of burnout. It is noteworthy that these factors are all subjective phenomena vulnerable to reporting bias, and they may all be related to mental health and/or responses to stress. In our study, sleep problems had a strong effect on gender differences in sickness absence, surpassing all other factors. This may be explained by how sleep problems were measured. Participants were asked if they had experienced sleeping difficulties ‘that affected their work ability’. Since this question was phrased in a way that is directly linked to the outcome, it is not surprising that this factor apparently had a large effect. Had the question been phrased differently, the effect of adjusting for sleep problems may have been smaller. Some of women’s excess absence is related to pregnancy and childbirth [25,26]. In the current study, we did not examine sick leave related to birth and pregnancy, as we excluded those who reported recent pregnancies, and instead focused on parents with children in late adolescence/young adulthood. Our study had several limitations. Firstly, all the factors were self-reported and subject to various biases. Possible gender-related bias in self-reporting may be a limitation. For example, if men systematically under-report sexual assaults or family stressors, we may over-estimate how these factors contribute to gender differences in sickness absence. For most of our explanatory factors (Table 1), there were apparent gender differences, but there is no way for us to determine whether these differences reflect true gender differences, or gender differences in reporting. In the latter case, it is possible that we over-estimated how much can be explained by gender differences in health and stressors. Secondly, the cross-sectional design is a limitation which precludes causal interpretations. However, if we had applied a longitudinal design, it is more likely that important health problems or stressors would remain undetected, as they may have occurred after the baseline measurement. This may have led to an under-estimation of the factors’ contribution. Thirdly, the outcome variable in our study was self-reported sickness absence, which may be less accurate than registry-based measures, and possibly subject to recall bias. However, any such bias is likely unrelated to gender. There are also advantages to using self-reported sickness absence: episodes of short absence may be handled between employer and employee without being recorded in official registries, and self-employed people may not obtain sick-notes for short absence. The gender difference in sickness absence in our study (7.1 sick days in excess for women) is comparable to that in the general population according to registry data. In the first quarter of 2011, the sickness absence rates for men and women in the general Norwegian population were 6 and 9%, respectively [27]. Over a 6-month period, this would amount to an excess of about 5 sick days for women. This indicates that the self-reported sickness absence in our study was reasonably accurate, and also that our population was reasonably representative. Fourthly, as this study uses a population-based sample, some bias in participation would be expected. In general, participants in voluntary questionnaire studies have higher education attainment, are often in better health and have less sickness absence than the general population [10]. The results are, therefore, not necessarily representative of the general population. Work is largely beneficial for health [28], and sickness absence may be a serious economic and social burden for the individual even in countries with generous welfare policies [3]. The gender gap in sickness absence may, therefore, put women at a disadvantage economically as well as professionally. The popular double-burden hypothesis has repeatedly been tested and evidence to support it is still lacking [5]. The selection hypothesis, which suggests that the gender gap is caused by differences in typical male and female industries and occupations, has also been refuted [7]. Our study found that some of the gender differences in sickness absence may be attributed to gender differences in self-reported symptoms and conditions. However, the majority of the gender difference in sickness absence remains unexplained. In the search for explanations for the gender gap, the current study has covered the domains of health and work/family stressors quite comprehensively. Even though we included a wide range of factors, the simultaneous adjustment for health factors and stressors only reduced the difference in sickness absence by about 28%. The main reasons for the gender gap in sickness absence must be outside these domains. Key points In this study, gender differences in health explained 22% of the gender differences in sickness absence. When also taking work and family stressors into consideration, 28% of the differences were explained. Although sickness absence is a health-related welfare benefit, health plays a minor role in explaining the gender gap in sickness absence. The most significant factors explaining the gender gap must be outside the domains of health, work and family stressors. Funding This work was supported by the Research Council of Norway’s Program for Sickness Absence, Work and Health (SYKEFRAVÆR) (grant numbers 218373, 227097). The funding body did not participate in any part of the process of the current paper, including the design of the study, the data collection, analyses, data interpretation and writing of the manuscript. Competing interest None declared. Acknowledgements This research is based on data from the Tracking Opportunities and Problems (TOPP) belonging to the Norwegian Institute of Public Health. We acknowledge all the participating families and their voluntary effort over 18 years, as well as the health care personnel who contributed with the data collection the first three waves. We acknowledge the founder of the TOPP study, Dr Kristin S. Mathiesen, and all other researchers in the TOPP group who contributed with the data collection over the 18 years. We also thank the Research Council of Norway, which has been the main contributor for funding the data collections. References 1. Mastekaasa A. The gender gap in sickness absence: long-term trends in eight European countries. Eur J Public Health  2014; 24: 656– 662. Google Scholar CrossRef Search ADS PubMed  2. Bryngelson A. Long-term sickness absence and social exclusion. Scand J Public Health  2009; 37: 839– 845. Google Scholar CrossRef Search ADS PubMed  3. Markussen S. The individual cost of sick leave. J Popul Econ  2011; 25: 1287– 1306. Google Scholar CrossRef Search ADS   4. Vingård E, Alexanderson K, Norlund A. Swedish Council on Technology Assessment in Health Care (SBU). Chapter 9. Consequences of being on sick leave. Scand J Public Health Suppl  2004; 63: 207– 215. Google Scholar CrossRef Search ADS PubMed  5. Nilsen W, Skipstein A, Østby KA, Mykletun A. Examination of the double burden hypothesis—a systematic review of work-family conflict and sickness absence. Eur J Public Health  2017; 27: 465– 471. Google Scholar CrossRef Search ADS PubMed  6. Laaksonen M, Mastekaasa A, Martikainen P, Rahkonen O, Piha K, Lahelma E. Gender differences in sickness absence—the contribution of occupation and workplace. Scand J Work Environ Health  2010; 36: 394– 403. Google Scholar CrossRef Search ADS PubMed  7. Mastekaasa A, Melsom A. Occupational segregation and gender differences in sickness absence: evidence from 17 European countries. Eur Sociol Rev  2014; 30: 582– 594. Google Scholar CrossRef Search ADS   8. Kringlen E, Torgersen S, Cramer V. A Norwegian psychiatric epidemiological study. Am J Psychiatry  2001; 158: 1091– 1098. Google Scholar CrossRef Search ADS PubMed  9. Woolf AD, Pfleger B. Burden of major musculoskeletal conditions. Bull World Health Organ  2003; 81: 646– 656. Google Scholar PubMed  10. Knudsen AK, Hotopf M, Skogen JC, Overland S, Mykletun A. The health status of nonparticipants in a population-based health study: the Hordaland Health Study. Am J Epidemiol  2010; 172: 1306– 1314. Google Scholar CrossRef Search ADS PubMed  11. Norwegian Labour and Welfare Administration. Sick Leave Statistics 2006–2015 , 2015. https://www.nav.no/no/NAV+og+samfunn/Statistikk/Sykefravar+-+statistikk/Tabeller/legemeldte-sykefrav%C3%A6rstilfeller-4.kv-2006-2015.diagnose-og-kj%C3%B8nn.antall (1 January 2018, date last accessed). 12. Ariansen AM, Mykletun A. Does postponement of first pregnancy increase gender differences in sickness absence? A register based analysis of Norwegian employees in 1993–2007. PLoS One  2014; 9: e93006. Google Scholar CrossRef Search ADS PubMed  13. Casini A, Godin I, Clays E, Kittel F. Gender difference in sickness absence from work: a multiple mediation analysis of psychosocial factors. Eur J Public Health  2013; 23: 635– 642. Google Scholar CrossRef Search ADS PubMed  14. Laaksonen M, Martikainen P, Rahkonen O, Lahelma E. Explanations for gender differences in sickness absence: evidence from middle-aged municipal employees from Finland. Occup Environ Med  2008; 65: 325– 330. Google Scholar CrossRef Search ADS PubMed  15. Smeby L, Bruusgaard D, Claussen B. Sickness absence: could gender divide be explained by occupation, income, mental distress and health? Scand J Public Health  2009; 37: 674– 681. Google Scholar CrossRef Search ADS PubMed  16. Nilsen W, Kjeldsen A, Karevold EB, et al.   Cohort profile: the Tracking Opportunities and Problems Study (TOPP)-study of Norwegian children and their parents followed from infancy to early adulthood. Int J Epidemiol  2017; 46: 1399– 1399. Google Scholar CrossRef Search ADS PubMed  17. Mathiesen KS, Tambs K, Dalgard OS. The influence of social class, strain and social support on symptoms of anxiety and depression in mothers of toddlers. Soc Psychiatry Psychiatr Epidemiol  1999; 34: 61– 72. Google Scholar CrossRef Search ADS PubMed  18. Deane FP, Leathern J, Spicer J. Clinical norms, reliability and validity for the Hopkins Symptom Checklist-21. Aust J Psychol  1992; 44: 21– 25. Google Scholar CrossRef Search ADS   19. Karasek R, Brisson C, Kawakami N, Houtman I, Bongers P, Amick B. The Job Content Questionnaire (JCQ): an instrument for internationally comparative assessments of psychosocial job characteristics. J Occup Health Psychol  1998; 3: 322– 355. Google Scholar CrossRef Search ADS PubMed  20. Courvoisier DS, Perneger TV. Validation of alternative formulations of job strain. J Occup Health  2010; 52: 5– 13. Google Scholar CrossRef Search ADS PubMed  21. Peterson U, Bergström G, Demerouti E, Gustavsson P, Asberg M, Nygren A. Burnout levels and self-rated health prospectively predict future long-term sickness absence: a study among female health professionals. J Occup Environ Med  2011; 53: 788– 793. Google Scholar CrossRef Search ADS PubMed  22. Skipstein A, Janson H, Kjeldsen A, Nilsen W, Mathiesen KS. Trajectories of maternal symptoms of depression and anxiety over 13 years: the influence of stress, social support, and maternal temperament. BMC Public Health  2012; 12: 1120. Google Scholar CrossRef Search ADS PubMed  23. Black MCBKC, Breiding MJ, Smith SG, et al.   National Intimate Partner and Sexual Violence Survey: 2010. Summary Report: The Centers for Disease Control and Prevention, National Center for Injury Prevention and Control, 2011. 24. Finkelhor D, Hotaling G, Lewis IA, Smith C. Sexual abuse in a national survey of adult men and women: prevalence, characteristics, and risk factors. Child Abuse Negl  1990; 14: 19– 28. Google Scholar CrossRef Search ADS PubMed  25. Alexanderson K, Sydsjö A, Hensing G, Sydsjö G, Carstensen J. Impact of pregnancy on gender differences in sickness absence. Scand J Soc Med  1996; 24: 169– 176. Google Scholar CrossRef Search ADS PubMed  26. Sydsjö A, Sydsjö G, Alexanderson K. Influence of pregnancy-related diagnoses on sick-leave data in women aged 16–44. J Womens Health Gend Based Med  2001; 10: 707– 714. Google Scholar CrossRef Search ADS PubMed  27. Statistics Norway. Statistikkbanken 2011 . https://www.ssb.no/statistikkbanken/ (1 January 2018, date last accessed). PubMed PubMed  28. Modini M, Joyce S, Mykletun A, et al.   The mental health benefits of employment: results of a systematic meta-review. Australas Psychiatry  2016; 24: 331– 336. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the Society of Occupational Medicine. All rights reserved. For Permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Occupational Medicine Oxford University Press

Explaining the gender gap in sickness absence

Loading next page...
 
/lp/ou_press/explaining-the-gender-gap-in-sickness-absence-5FqTqmY7RT
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the Society of Occupational Medicine. All rights reserved. For Permissions, please email: journals.permissions@oup.com
ISSN
0962-7480
eISSN
1471-8405
D.O.I.
10.1093/occmed/kqy062
Publisher site
See Article on Publisher Site

Abstract

Abstract Background In many western countries, women have a much higher rate of sickness absence than men. To what degree the gender differences in sickness absence are caused by gender differences in health is largely unknown. Aims To assess to what degree the gender gap in sickness absence can be explained by health factors and work- and family-related stressors. Methods Norwegian parents participating in the Tracking Opportunities and Problems (TOPP) study were asked about sickness absence and a range of factors possibly contributing to gender differences in sickness absence, including somatic and mental health, sleep problems, job control/demands, work–home conflicts, parent–child conflicts and stressful life events. Using a cross-sectional design, we did linear regression analyses, to assess the relative contribution from health and stressors. Results There were 557 study participants. Adjusting for health factors reduced the gender difference in sickness absence by 24%, while adjusting for stressors in the family and at work reduced the difference by 22%. A simultaneous adjustment for health factors and stressors reduced the difference in sickness absence by about 28%. Conclusions Despite adjusting for a large number of factors, including both previously well-studied factors (e.g. health, job control/demands) and lesser-studied factors (parent–child conflict and sexual assault), this study found that most of the gender gap in sickness absence remains unexplained. Gender differences in health and stressors account for only part of the differences in sickness absence. Other factors must, therefore, exist outside the domains of health, work and family stressors. Family stressors, gender gap, health, sickness absence, work factors Introduction In many western countries, there is a large and unexplained gender gap in the rates of sickness absence [1]. In many countries, including Norway, women’s sickness absence is about 50% higher than men’s. These gender differences in sickness absence may lead to negative long-term outcomes for women such as reduced income and career opportunities, stigmatization and long-term labour market exclusion [2–4]. There are many hypotheses for why women have higher sickness absence than men, but no single hypothesis has yet been able to explain the considerable gender differences. Socially based hypotheses have suggested that women may be subject to higher levels of stress, possibly aggravated by the effect of the so-called ‘double-burden hypothesis’, the idea that working women also carry more of the burden at home [5]. For example, women may do more of the childcare or household chores, or they may be more involved when relatives become ill. There is some evidence that women experience a higher degree of conflict between demands at work and at home, but this has been found to explain little of the gender differences in sickness absence [5]. One possible explanation could be that women have physically and/or mentally more demanding jobs, or that female-dominated workplaces are at increased risk due to the nature of the work. Some previous studies have supported this view [6], while a more recent study found that gender differences in sickness absence increased when adjusting for type of occupation [7]. Health-related factors may also contribute to gender differences in sickness absence. For instance, women have a higher prevalence of anxiety, depression and musculoskeletal pain conditions [8,9], which are important causes of sickness absence [10,11]. Also, sickness absence in pregnancy has increased considerably in Norway over the last two decades, and now accounts for about 25% of the gender difference in sickness absence [12]. Reasons for the sharp increase in sickness absence in pregnancy are largely unknown and are not related to delayed first pregnancies [12]. While some of these hypotheses may explain the gender gap in sickness absence, a large and unexplained difference remains in the rates of sickness absence between men and women. Most studies have examined possible explanatory factors separately, and in the few studies that do include explanatory factors from several domains, the conclusions vary [13–15]. A recent review article concluded that work–family conflict may contribute to the gender gap, but that there are few studies on this topic [5]. This study aimed to examine the gender gap in sickness absence using comprehensive questionnaire data from the Tracking Opportunities and Problems (TOPP) study [16]. The study included questions about sickness absence and a wide range of factors, including somatic and mental health problems, sleep problems, job control/demands, work–family conflict, parent–child conflict and stressful life events. TOPP is conducted on a sample of Norwegian parents, which is an advantage for studying family stressors. Our aim was to examine gender differences in sickness absence among employed Norwegian parents in a cross-sectional design, while adjusting for a wide variety of factors, to see how much of the gender gap can be explained, and to establish the relative contribution from work- and family-related stressors and from health differences. Methods TOPP is an on-going Norwegian community-based study of parents that were recruited when they attended for vaccination of their child at a municipal health centre [16,17]. In 2011, when their child was 18–20 years old, the parents were sent a questionnaire about sickness absence, health and various stressors. The current study is based on data from this questionnaire. The collection and use of data was approved by the regional committees for medical and health research ethics. Sickness absence was assessed by the following question: ‘Have you been away from work over the past six months due to own illness?’, and the respondents were asked to give the total number of days, weeks and months they had been away from work. The total sickness absence for each participant was then calculated. As a measure of physical health, the participants were asked to report if they, over the past 12 months, had had any of the following disorders: ‘cardiac/vascular disorder’, ‘metabolic disorder’, ‘asthma’, ‘Hay fever/allergy’, ‘skin disease/eczema’, ‘musculoskeletal disorder’, ‘kidney/urinary tract disorder’, ‘fibromyalgia’, ‘gynaecological/genital disorder’, ‘migraine’, ‘diabetes’, ‘abdominal/intestinal disorder’ and ‘cancer’. Responses were coded as 13 separate dichotomous variables. As a measure of mental health, participants were asked about 24 items from the Hopkins Symptom checklist (HSCL-25) [18], which measures the presence of symptoms of anxiety and depression in a non-psychiatric setting. For each item in the checklist the participants rate the degree to which they have felt affected by a given symptom on a four-point scale ranging from 1 (‘not at all’) to 4 (‘extremely affected’). One item from the HSCL-25, ‘Loss of sexual interest or pleasure’, was not included in the questionnaire because some participants perceived this item as inappropriate in a pilot study. The average of each participant’s HSCL-25 score was treated as a continuous variable in the analyses. Sleeping difficulties were assessed by the question ‘Have you, over the past 12 months, experienced sleeping difficulties that affected your work ability?’, and the response alternatives were ‘yes’ and ‘no’. Participants were assessed for the demands and control domains of Karasek and Theorell’s model of work strain by questions from the Job Content Questionnaire [19], including five questions on the psychological demands latitude (demands) and four questions on the decision latitude (control). Responses were coded from 1 to 5, and for each participant, two mean values were calculated, one demands score and one control score. Finally, we defined a continuous variable as a measure of job strain by subtracting the demand score from the control score. This model is supported by previous research [20]. Using 15 items from the Oldenburg Burnout Inventory (OLBI) [21], we measured exhaustion and disengagement from work. For each item, a statement was rated from 1 (‘strongly agree’) to 4 (‘strongly disagree’). Exhaustion was measured with seven items (example: ‘After my work, I regularly feel worn out and weary’), while disengagement was measured with eight items (example: ‘I often talk about my work in a negative way’). Due to a misprint, one exhaustion item from OLBI (‘After my work, I usually feel worn out and weary’) was not included in the questionnaire. For each participant, the average scores for each of the two dimensions (exhaustion and disengagement) were calculated. As a measure of exposure to family stressors, participants were asked to rate the burden from 10 different stressors experienced over the past 12 months. Responses were coded from 1 (‘No burden’) to 4 (‘Very heavy burden’), and for each participant, the mean value for all stressors was estimated. The 10 stressors were: ‘Teenager’s health problems’, ‘Other children’s health problems’, ‘Problems related to sick parents or parents-in-law’, ‘Problems related to structuring the children’s/teenager’s daily life’, ‘Problems with combining work life and child care’, ‘Stressors related to mood swings in children/teenagers’, ‘Worries about what the children/teenager exposes themselves to, or what they may be exposed to, in their spare time’, ‘Problems related to children’s/teenager’s schooling’, ‘Worries about the children’s future’ and ‘Problems related to setting limits for the children/teenagers’. These questions have previously been used in research as measures of family stressors [22]. Conflict with teenagers was assessed using the conflict subscale of the Child Parent Relationship Scale (CPRS, Pianta et al.). Each of six statements is rated from 1 (‘does not fit’) to 5 (‘always fits’), and for each participant, the average score was calculated. Sexual assaults is a stressor with considerable gender imbalance [23,24], and since these experiences may have long-lasting effects on the victims, they may potentially contribute to the gender gap in sickness absence. To measure previous occurrences of sexual assault, the participants were asked if they had ever experienced being pushed or forced into sexual acts. In the current study, the responses ‘Yes, I was forced’ and ‘Yes, I was raped’ were defined as occurrences of sexual assault which was used as a dichotomous variable. Information about participants’ level of education was retrieved from a previous data collection in TOPP that was conducted on the same sample in 2008, and the responses were coded from 1 (‘up to 9 years of schooling’) to 5 (‘4 or more years in university/college’). For 53 participants, we did not have information about education, and for these we imputed the missing data using the multiple imputation function in IBM SPSS Statistics version 23 for Windows with choice of method set to automatic and with all variables used in the analyses as predictor variables. The unadjusted gender differences in both sickness absence and in the potentially associated factors were estimated in the whole study population by running independent sample t-tests. For the main analyses, we conducted linear regression analyses with sickness absence as the dependent variable and gender as the independent variable, adjusted for each of the possible confounding factors. First, we ran the analyses by adjusting for each factor separately, then again with all the factors entered into the model one by one, looking at the cumulative effect on the adjusted gender difference in sickness absence. All analyses were carried out in IBM SPSS Statistics 23 for Windows. Results In total, 627 parents participated in the data collection in 2011. Participants who reported occupational status as homemaker, student or unemployed, as well as those reporting pregnancy during the previous year were excluded. After exclusions, the study population consisted of 240 men and 317 women—a total of 557 participants. Gender differences in the examined variables are presented in Table 1. The unadjusted analyses showed significant gender differences in sickness absence, as women had on average 12.6 days of absence over the past 6 months while men only had 5.5 days of absence (i.e. 7.09 excess sick days for women). There were also gender differences in most of the adjusting factors, with significant differences found for some somatic illnesses (i.e. thyroid disease, asthma, gynaecological/genito-urinary disease, migraine, musculoskeletal disorder, fibromyalgia), for mental health, sleep problems, family stressors, sexual victimization and for work factors (job control and exhaustion). All these factors were more frequently reported by women than men. Table 1. Gender differences in the study sample   Men (n = 240)  Women (n = 317)  Significance  Sickness absence (days)  5.45  12.55  **  Age  51.15  48.53  **  Education (range: 1–5)  3.91  3.76    Somatic illness   Cardiac/vascular disorder  6%  6%     Thyroid disorder  3%  13%  **   Diabetes  3%  3%     Cancer  3%  1%     Asthma  5%  9%  *   Hay fever/allergies  16%  18%     Dermatitis/eczema  11%  11%     Kidney/urinary tract disorder  3%  5%     Gynaecological/genital disorder  2%  7%  **   Migraine  8%  14%  *   Abdominal/intestinal disorders  4%  6%     Musculoskeletal symptoms  19%  31%  **   Fibromyalgia  2%  6%  *  Mental illness   Mental distress (Hopkins Symptom Checklist) (range: 0–3)  1.24  1.32  **   Sleep problems  15%  25%  **  Job stressors   Exhaustion (Olbi, range: 1–4)  1.92  2.00  *   Disengagement (Olbi, range: 1–4)  2.00  2.05     Job demands (Karasek, range: 1–5)  3.22  3.29     Job control (Karasek, range: 1–5)  3.70  3.44  **   Control minus demands (range −4 to 4)  0.48  0.15  **  Family stressors   Family stressors (burden) (range: 1–4)  1.33  1.42  **   Teenager conflict (range: 1–5) (Pianta)  1.74  1.81    Sexual victimization (forced/raped)  1%  5%  **    Men (n = 240)  Women (n = 317)  Significance  Sickness absence (days)  5.45  12.55  **  Age  51.15  48.53  **  Education (range: 1–5)  3.91  3.76    Somatic illness   Cardiac/vascular disorder  6%  6%     Thyroid disorder  3%  13%  **   Diabetes  3%  3%     Cancer  3%  1%     Asthma  5%  9%  *   Hay fever/allergies  16%  18%     Dermatitis/eczema  11%  11%     Kidney/urinary tract disorder  3%  5%     Gynaecological/genital disorder  2%  7%  **   Migraine  8%  14%  *   Abdominal/intestinal disorders  4%  6%     Musculoskeletal symptoms  19%  31%  **   Fibromyalgia  2%  6%  *  Mental illness   Mental distress (Hopkins Symptom Checklist) (range: 0–3)  1.24  1.32  **   Sleep problems  15%  25%  **  Job stressors   Exhaustion (Olbi, range: 1–4)  1.92  2.00  *   Disengagement (Olbi, range: 1–4)  2.00  2.05     Job demands (Karasek, range: 1–5)  3.22  3.29     Job control (Karasek, range: 1–5)  3.70  3.44  **   Control minus demands (range −4 to 4)  0.48  0.15  **  Family stressors   Family stressors (burden) (range: 1–4)  1.33  1.42  **   Teenager conflict (range: 1–5) (Pianta)  1.74  1.81    Sexual victimization (forced/raped)  1%  5%  **  Variables shown are either percentages or mean values. *P < 0.05, **P < 0.01. View Large Table 1. Gender differences in the study sample   Men (n = 240)  Women (n = 317)  Significance  Sickness absence (days)  5.45  12.55  **  Age  51.15  48.53  **  Education (range: 1–5)  3.91  3.76    Somatic illness   Cardiac/vascular disorder  6%  6%     Thyroid disorder  3%  13%  **   Diabetes  3%  3%     Cancer  3%  1%     Asthma  5%  9%  *   Hay fever/allergies  16%  18%     Dermatitis/eczema  11%  11%     Kidney/urinary tract disorder  3%  5%     Gynaecological/genital disorder  2%  7%  **   Migraine  8%  14%  *   Abdominal/intestinal disorders  4%  6%     Musculoskeletal symptoms  19%  31%  **   Fibromyalgia  2%  6%  *  Mental illness   Mental distress (Hopkins Symptom Checklist) (range: 0–3)  1.24  1.32  **   Sleep problems  15%  25%  **  Job stressors   Exhaustion (Olbi, range: 1–4)  1.92  2.00  *   Disengagement (Olbi, range: 1–4)  2.00  2.05     Job demands (Karasek, range: 1–5)  3.22  3.29     Job control (Karasek, range: 1–5)  3.70  3.44  **   Control minus demands (range −4 to 4)  0.48  0.15  **  Family stressors   Family stressors (burden) (range: 1–4)  1.33  1.42  **   Teenager conflict (range: 1–5) (Pianta)  1.74  1.81    Sexual victimization (forced/raped)  1%  5%  **    Men (n = 240)  Women (n = 317)  Significance  Sickness absence (days)  5.45  12.55  **  Age  51.15  48.53  **  Education (range: 1–5)  3.91  3.76    Somatic illness   Cardiac/vascular disorder  6%  6%     Thyroid disorder  3%  13%  **   Diabetes  3%  3%     Cancer  3%  1%     Asthma  5%  9%  *   Hay fever/allergies  16%  18%     Dermatitis/eczema  11%  11%     Kidney/urinary tract disorder  3%  5%     Gynaecological/genital disorder  2%  7%  **   Migraine  8%  14%  *   Abdominal/intestinal disorders  4%  6%     Musculoskeletal symptoms  19%  31%  **   Fibromyalgia  2%  6%  *  Mental illness   Mental distress (Hopkins Symptom Checklist) (range: 0–3)  1.24  1.32  **   Sleep problems  15%  25%  **  Job stressors   Exhaustion (Olbi, range: 1–4)  1.92  2.00  *   Disengagement (Olbi, range: 1–4)  2.00  2.05     Job demands (Karasek, range: 1–5)  3.22  3.29     Job control (Karasek, range: 1–5)  3.70  3.44  **   Control minus demands (range −4 to 4)  0.48  0.15  **  Family stressors   Family stressors (burden) (range: 1–4)  1.33  1.42  **   Teenager conflict (range: 1–5) (Pianta)  1.74  1.81    Sexual victimization (forced/raped)  1%  5%  **  Variables shown are either percentages or mean values. *P < 0.05, **P < 0.01. View Large Table 2 lists the excess number of sick days for women (as compared to men) before and after adjustment for each single factor. Most of these adjustments reduced the gender difference in sickness absence, and, ranked by the resulting reduction in beta, the most important explanatory factors were sleep problems, musculoskeletal disease, mental distress, the relationship between demands and control at work, the exhaustion domain of the OLBI, family stressors) and conflict with teenagers. Table 2. Gender differences in sickness absence—adjusted for explanatory factors   Excess sick days for women  95% CI  Effect of adjustment (change in sick days)  Rank of contribution  Not adjusted  7.09  2.27–11.91  –  –  Education  7.07  2.24–11.91  −0.02  15  Health factors   Somatic health    Cardiac/vascular disorder  7.09  2.26–11.91  0  17    Thyroid disorder  6.79  1.90–11.68  −0.30  8    Diabetes  7.06  2.23–11.88  −0.03  13    Cancer  7.35  2.52–12.17  +0.26  20    Asthma  7.03  2.19–11.87  −0.06  12    Hay fever/allergies  7.07  2.24–11.90  −0.02  14    Dermatitis/eczema  7.09  2.27–11.91  0  16    Kidney/urinary tract disorder  6.96  2.13–11.79  −0.13  11    Gynaecological/genital disorder  7.53  2.67–12.39  +0.44  21    Migraine  7.09  2.24–11.94  0  18    Abdominal/intestinal disorder  7.07  2.24–11.90  −0.02  15    Musculoskeletal symptoms  5.72  0.94–10.51  −1.37  2    Fibromyalgia  6.86  2.01–11.71  −0.23  10   Mental health    Mental distress (Hopkins Symptom Checklist)  5.94  1.09–10.79  −1.15  3    Sleep problems  5.72  0.91–10.53  −1.37  1  Stressors   Job stressors    Exhaustion (Olbi)  6.53  1.70–11.36  −0.56  5    Disengagement (Olbi)  6.85  2.03–11.67  −0.24  9    Karasek substitution (control–demand)  6.43  1.52–11.35  −0.66  4     Job demands (Karasek)  7.29  2.43–12.15  +0.20       Job control (Karasek)  6.17  1.26–11.07  −0.92     Family stressors    Family stressors (burden)  6.59  1.72–11.46  −0.50  6    Teenager conflict (Pianta)  6.63  1.85–11.40  −0.46  7  Sexual victimization (forced/raped)  7.19  2.33–12.05  +0.10  19  All health factors  5.42  0.41–10.43  −1.67    All stressors  5.56  0.65–10.47  −1.53    Education + health + stressors  5.10  0.04–10.15  −1.99      Excess sick days for women  95% CI  Effect of adjustment (change in sick days)  Rank of contribution  Not adjusted  7.09  2.27–11.91  –  –  Education  7.07  2.24–11.91  −0.02  15  Health factors   Somatic health    Cardiac/vascular disorder  7.09  2.26–11.91  0  17    Thyroid disorder  6.79  1.90–11.68  −0.30  8    Diabetes  7.06  2.23–11.88  −0.03  13    Cancer  7.35  2.52–12.17  +0.26  20    Asthma  7.03  2.19–11.87  −0.06  12    Hay fever/allergies  7.07  2.24–11.90  −0.02  14    Dermatitis/eczema  7.09  2.27–11.91  0  16    Kidney/urinary tract disorder  6.96  2.13–11.79  −0.13  11    Gynaecological/genital disorder  7.53  2.67–12.39  +0.44  21    Migraine  7.09  2.24–11.94  0  18    Abdominal/intestinal disorder  7.07  2.24–11.90  −0.02  15    Musculoskeletal symptoms  5.72  0.94–10.51  −1.37  2    Fibromyalgia  6.86  2.01–11.71  −0.23  10   Mental health    Mental distress (Hopkins Symptom Checklist)  5.94  1.09–10.79  −1.15  3    Sleep problems  5.72  0.91–10.53  −1.37  1  Stressors   Job stressors    Exhaustion (Olbi)  6.53  1.70–11.36  −0.56  5    Disengagement (Olbi)  6.85  2.03–11.67  −0.24  9    Karasek substitution (control–demand)  6.43  1.52–11.35  −0.66  4     Job demands (Karasek)  7.29  2.43–12.15  +0.20       Job control (Karasek)  6.17  1.26–11.07  −0.92     Family stressors    Family stressors (burden)  6.59  1.72–11.46  −0.50  6    Teenager conflict (Pianta)  6.63  1.85–11.40  −0.46  7  Sexual victimization (forced/raped)  7.19  2.33–12.05  +0.10  19  All health factors  5.42  0.41–10.43  −1.67    All stressors  5.56  0.65–10.47  −1.53    Education + health + stressors  5.10  0.04–10.15  −1.99    View Large Table 2. Gender differences in sickness absence—adjusted for explanatory factors   Excess sick days for women  95% CI  Effect of adjustment (change in sick days)  Rank of contribution  Not adjusted  7.09  2.27–11.91  –  –  Education  7.07  2.24–11.91  −0.02  15  Health factors   Somatic health    Cardiac/vascular disorder  7.09  2.26–11.91  0  17    Thyroid disorder  6.79  1.90–11.68  −0.30  8    Diabetes  7.06  2.23–11.88  −0.03  13    Cancer  7.35  2.52–12.17  +0.26  20    Asthma  7.03  2.19–11.87  −0.06  12    Hay fever/allergies  7.07  2.24–11.90  −0.02  14    Dermatitis/eczema  7.09  2.27–11.91  0  16    Kidney/urinary tract disorder  6.96  2.13–11.79  −0.13  11    Gynaecological/genital disorder  7.53  2.67–12.39  +0.44  21    Migraine  7.09  2.24–11.94  0  18    Abdominal/intestinal disorder  7.07  2.24–11.90  −0.02  15    Musculoskeletal symptoms  5.72  0.94–10.51  −1.37  2    Fibromyalgia  6.86  2.01–11.71  −0.23  10   Mental health    Mental distress (Hopkins Symptom Checklist)  5.94  1.09–10.79  −1.15  3    Sleep problems  5.72  0.91–10.53  −1.37  1  Stressors   Job stressors    Exhaustion (Olbi)  6.53  1.70–11.36  −0.56  5    Disengagement (Olbi)  6.85  2.03–11.67  −0.24  9    Karasek substitution (control–demand)  6.43  1.52–11.35  −0.66  4     Job demands (Karasek)  7.29  2.43–12.15  +0.20       Job control (Karasek)  6.17  1.26–11.07  −0.92     Family stressors    Family stressors (burden)  6.59  1.72–11.46  −0.50  6    Teenager conflict (Pianta)  6.63  1.85–11.40  −0.46  7  Sexual victimization (forced/raped)  7.19  2.33–12.05  +0.10  19  All health factors  5.42  0.41–10.43  −1.67    All stressors  5.56  0.65–10.47  −1.53    Education + health + stressors  5.10  0.04–10.15  −1.99      Excess sick days for women  95% CI  Effect of adjustment (change in sick days)  Rank of contribution  Not adjusted  7.09  2.27–11.91  –  –  Education  7.07  2.24–11.91  −0.02  15  Health factors   Somatic health    Cardiac/vascular disorder  7.09  2.26–11.91  0  17    Thyroid disorder  6.79  1.90–11.68  −0.30  8    Diabetes  7.06  2.23–11.88  −0.03  13    Cancer  7.35  2.52–12.17  +0.26  20    Asthma  7.03  2.19–11.87  −0.06  12    Hay fever/allergies  7.07  2.24–11.90  −0.02  14    Dermatitis/eczema  7.09  2.27–11.91  0  16    Kidney/urinary tract disorder  6.96  2.13–11.79  −0.13  11    Gynaecological/genital disorder  7.53  2.67–12.39  +0.44  21    Migraine  7.09  2.24–11.94  0  18    Abdominal/intestinal disorder  7.07  2.24–11.90  −0.02  15    Musculoskeletal symptoms  5.72  0.94–10.51  −1.37  2    Fibromyalgia  6.86  2.01–11.71  −0.23  10   Mental health    Mental distress (Hopkins Symptom Checklist)  5.94  1.09–10.79  −1.15  3    Sleep problems  5.72  0.91–10.53  −1.37  1  Stressors   Job stressors    Exhaustion (Olbi)  6.53  1.70–11.36  −0.56  5    Disengagement (Olbi)  6.85  2.03–11.67  −0.24  9    Karasek substitution (control–demand)  6.43  1.52–11.35  −0.66  4     Job demands (Karasek)  7.29  2.43–12.15  +0.20       Job control (Karasek)  6.17  1.26–11.07  −0.92     Family stressors    Family stressors (burden)  6.59  1.72–11.46  −0.50  6    Teenager conflict (Pianta)  6.63  1.85–11.40  −0.46  7  Sexual victimization (forced/raped)  7.19  2.33–12.05  +0.10  19  All health factors  5.42  0.41–10.43  −1.67    All stressors  5.56  0.65–10.47  −1.53    Education + health + stressors  5.10  0.04–10.15  −1.99    View Large Table 2 includes the cumulative effect of adjusting for all somatic and mental health factors, all family and work stressors, and finally the effect of adjusting for all variables included in the study (health, stressors and education). The unadjusted difference in sickness absence was 7.09 excess sick days over 6 months for women. Adjusting for somatic and mental health problems reduced the difference to 5.42 excess sick days. Adjusting for family- and work-related stressors reduced the difference to 5.56. Adjusting for all factors simultaneously, including education, health and family stressors, reduced the gender difference in sickness absence to 5.10 days. These results are visualized in Figure 1. Figure 1. View largeDownload slide Gender differences in sickness absence (excess sick days for women compared to men). Presented unadjusted, adjusted for health factors, adjusted for stressors and fully adjusted. Figure 1. View largeDownload slide Gender differences in sickness absence (excess sick days for women compared to men). Presented unadjusted, adjusted for health factors, adjusted for stressors and fully adjusted. Discussion This study was not able to entirely explain the gender gap in sickness absence. Adjusting for a range of health factors accounted for only about 22–24% of the observed gender difference in sickness absence. After further adjusting for various stressors, about 72% of the gender difference remained unexplained. About a quarter of the gender difference in sickness absence could be attributed to differences in health. Work and family stressors contributed to a similar extent, but adjusting for these added little beyond the adjustments for health. Previous studies of the popular ‘double-burden hypothesis’ has largely failed to explain why women’s sickness absence rates are so much higher than men’s [5]. There is also strong evidence against the hypothesis that industries or professions employing mostly women cause more sickness absence than those employing mostly men [7]. The gender difference in sickness absence, therefore, remains largely unexplained. The five most important factors were sleep problems, musculoskeletal symptoms, mental distress, the control dimension of Karasek and Theorell’s model of job stress and the exhaustion dimension of burnout. It is noteworthy that these factors are all subjective phenomena vulnerable to reporting bias, and they may all be related to mental health and/or responses to stress. In our study, sleep problems had a strong effect on gender differences in sickness absence, surpassing all other factors. This may be explained by how sleep problems were measured. Participants were asked if they had experienced sleeping difficulties ‘that affected their work ability’. Since this question was phrased in a way that is directly linked to the outcome, it is not surprising that this factor apparently had a large effect. Had the question been phrased differently, the effect of adjusting for sleep problems may have been smaller. Some of women’s excess absence is related to pregnancy and childbirth [25,26]. In the current study, we did not examine sick leave related to birth and pregnancy, as we excluded those who reported recent pregnancies, and instead focused on parents with children in late adolescence/young adulthood. Our study had several limitations. Firstly, all the factors were self-reported and subject to various biases. Possible gender-related bias in self-reporting may be a limitation. For example, if men systematically under-report sexual assaults or family stressors, we may over-estimate how these factors contribute to gender differences in sickness absence. For most of our explanatory factors (Table 1), there were apparent gender differences, but there is no way for us to determine whether these differences reflect true gender differences, or gender differences in reporting. In the latter case, it is possible that we over-estimated how much can be explained by gender differences in health and stressors. Secondly, the cross-sectional design is a limitation which precludes causal interpretations. However, if we had applied a longitudinal design, it is more likely that important health problems or stressors would remain undetected, as they may have occurred after the baseline measurement. This may have led to an under-estimation of the factors’ contribution. Thirdly, the outcome variable in our study was self-reported sickness absence, which may be less accurate than registry-based measures, and possibly subject to recall bias. However, any such bias is likely unrelated to gender. There are also advantages to using self-reported sickness absence: episodes of short absence may be handled between employer and employee without being recorded in official registries, and self-employed people may not obtain sick-notes for short absence. The gender difference in sickness absence in our study (7.1 sick days in excess for women) is comparable to that in the general population according to registry data. In the first quarter of 2011, the sickness absence rates for men and women in the general Norwegian population were 6 and 9%, respectively [27]. Over a 6-month period, this would amount to an excess of about 5 sick days for women. This indicates that the self-reported sickness absence in our study was reasonably accurate, and also that our population was reasonably representative. Fourthly, as this study uses a population-based sample, some bias in participation would be expected. In general, participants in voluntary questionnaire studies have higher education attainment, are often in better health and have less sickness absence than the general population [10]. The results are, therefore, not necessarily representative of the general population. Work is largely beneficial for health [28], and sickness absence may be a serious economic and social burden for the individual even in countries with generous welfare policies [3]. The gender gap in sickness absence may, therefore, put women at a disadvantage economically as well as professionally. The popular double-burden hypothesis has repeatedly been tested and evidence to support it is still lacking [5]. The selection hypothesis, which suggests that the gender gap is caused by differences in typical male and female industries and occupations, has also been refuted [7]. Our study found that some of the gender differences in sickness absence may be attributed to gender differences in self-reported symptoms and conditions. However, the majority of the gender difference in sickness absence remains unexplained. In the search for explanations for the gender gap, the current study has covered the domains of health and work/family stressors quite comprehensively. Even though we included a wide range of factors, the simultaneous adjustment for health factors and stressors only reduced the difference in sickness absence by about 28%. The main reasons for the gender gap in sickness absence must be outside these domains. Key points In this study, gender differences in health explained 22% of the gender differences in sickness absence. When also taking work and family stressors into consideration, 28% of the differences were explained. Although sickness absence is a health-related welfare benefit, health plays a minor role in explaining the gender gap in sickness absence. The most significant factors explaining the gender gap must be outside the domains of health, work and family stressors. Funding This work was supported by the Research Council of Norway’s Program for Sickness Absence, Work and Health (SYKEFRAVÆR) (grant numbers 218373, 227097). The funding body did not participate in any part of the process of the current paper, including the design of the study, the data collection, analyses, data interpretation and writing of the manuscript. Competing interest None declared. Acknowledgements This research is based on data from the Tracking Opportunities and Problems (TOPP) belonging to the Norwegian Institute of Public Health. We acknowledge all the participating families and their voluntary effort over 18 years, as well as the health care personnel who contributed with the data collection the first three waves. We acknowledge the founder of the TOPP study, Dr Kristin S. Mathiesen, and all other researchers in the TOPP group who contributed with the data collection over the 18 years. We also thank the Research Council of Norway, which has been the main contributor for funding the data collections. References 1. Mastekaasa A. The gender gap in sickness absence: long-term trends in eight European countries. Eur J Public Health  2014; 24: 656– 662. Google Scholar CrossRef Search ADS PubMed  2. Bryngelson A. Long-term sickness absence and social exclusion. Scand J Public Health  2009; 37: 839– 845. Google Scholar CrossRef Search ADS PubMed  3. Markussen S. The individual cost of sick leave. J Popul Econ  2011; 25: 1287– 1306. Google Scholar CrossRef Search ADS   4. Vingård E, Alexanderson K, Norlund A. Swedish Council on Technology Assessment in Health Care (SBU). Chapter 9. Consequences of being on sick leave. Scand J Public Health Suppl  2004; 63: 207– 215. Google Scholar CrossRef Search ADS PubMed  5. Nilsen W, Skipstein A, Østby KA, Mykletun A. Examination of the double burden hypothesis—a systematic review of work-family conflict and sickness absence. Eur J Public Health  2017; 27: 465– 471. Google Scholar CrossRef Search ADS PubMed  6. Laaksonen M, Mastekaasa A, Martikainen P, Rahkonen O, Piha K, Lahelma E. Gender differences in sickness absence—the contribution of occupation and workplace. Scand J Work Environ Health  2010; 36: 394– 403. Google Scholar CrossRef Search ADS PubMed  7. Mastekaasa A, Melsom A. Occupational segregation and gender differences in sickness absence: evidence from 17 European countries. Eur Sociol Rev  2014; 30: 582– 594. Google Scholar CrossRef Search ADS   8. Kringlen E, Torgersen S, Cramer V. A Norwegian psychiatric epidemiological study. Am J Psychiatry  2001; 158: 1091– 1098. Google Scholar CrossRef Search ADS PubMed  9. Woolf AD, Pfleger B. Burden of major musculoskeletal conditions. Bull World Health Organ  2003; 81: 646– 656. Google Scholar PubMed  10. Knudsen AK, Hotopf M, Skogen JC, Overland S, Mykletun A. The health status of nonparticipants in a population-based health study: the Hordaland Health Study. Am J Epidemiol  2010; 172: 1306– 1314. Google Scholar CrossRef Search ADS PubMed  11. Norwegian Labour and Welfare Administration. Sick Leave Statistics 2006–2015 , 2015. https://www.nav.no/no/NAV+og+samfunn/Statistikk/Sykefravar+-+statistikk/Tabeller/legemeldte-sykefrav%C3%A6rstilfeller-4.kv-2006-2015.diagnose-og-kj%C3%B8nn.antall (1 January 2018, date last accessed). 12. Ariansen AM, Mykletun A. Does postponement of first pregnancy increase gender differences in sickness absence? A register based analysis of Norwegian employees in 1993–2007. PLoS One  2014; 9: e93006. Google Scholar CrossRef Search ADS PubMed  13. Casini A, Godin I, Clays E, Kittel F. Gender difference in sickness absence from work: a multiple mediation analysis of psychosocial factors. Eur J Public Health  2013; 23: 635– 642. Google Scholar CrossRef Search ADS PubMed  14. Laaksonen M, Martikainen P, Rahkonen O, Lahelma E. Explanations for gender differences in sickness absence: evidence from middle-aged municipal employees from Finland. Occup Environ Med  2008; 65: 325– 330. Google Scholar CrossRef Search ADS PubMed  15. Smeby L, Bruusgaard D, Claussen B. Sickness absence: could gender divide be explained by occupation, income, mental distress and health? Scand J Public Health  2009; 37: 674– 681. Google Scholar CrossRef Search ADS PubMed  16. Nilsen W, Kjeldsen A, Karevold EB, et al.   Cohort profile: the Tracking Opportunities and Problems Study (TOPP)-study of Norwegian children and their parents followed from infancy to early adulthood. Int J Epidemiol  2017; 46: 1399– 1399. Google Scholar CrossRef Search ADS PubMed  17. Mathiesen KS, Tambs K, Dalgard OS. The influence of social class, strain and social support on symptoms of anxiety and depression in mothers of toddlers. Soc Psychiatry Psychiatr Epidemiol  1999; 34: 61– 72. Google Scholar CrossRef Search ADS PubMed  18. Deane FP, Leathern J, Spicer J. Clinical norms, reliability and validity for the Hopkins Symptom Checklist-21. Aust J Psychol  1992; 44: 21– 25. Google Scholar CrossRef Search ADS   19. Karasek R, Brisson C, Kawakami N, Houtman I, Bongers P, Amick B. The Job Content Questionnaire (JCQ): an instrument for internationally comparative assessments of psychosocial job characteristics. J Occup Health Psychol  1998; 3: 322– 355. Google Scholar CrossRef Search ADS PubMed  20. Courvoisier DS, Perneger TV. Validation of alternative formulations of job strain. J Occup Health  2010; 52: 5– 13. Google Scholar CrossRef Search ADS PubMed  21. Peterson U, Bergström G, Demerouti E, Gustavsson P, Asberg M, Nygren A. Burnout levels and self-rated health prospectively predict future long-term sickness absence: a study among female health professionals. J Occup Environ Med  2011; 53: 788– 793. Google Scholar CrossRef Search ADS PubMed  22. Skipstein A, Janson H, Kjeldsen A, Nilsen W, Mathiesen KS. Trajectories of maternal symptoms of depression and anxiety over 13 years: the influence of stress, social support, and maternal temperament. BMC Public Health  2012; 12: 1120. Google Scholar CrossRef Search ADS PubMed  23. Black MCBKC, Breiding MJ, Smith SG, et al.   National Intimate Partner and Sexual Violence Survey: 2010. Summary Report: The Centers for Disease Control and Prevention, National Center for Injury Prevention and Control, 2011. 24. Finkelhor D, Hotaling G, Lewis IA, Smith C. Sexual abuse in a national survey of adult men and women: prevalence, characteristics, and risk factors. Child Abuse Negl  1990; 14: 19– 28. Google Scholar CrossRef Search ADS PubMed  25. Alexanderson K, Sydsjö A, Hensing G, Sydsjö G, Carstensen J. Impact of pregnancy on gender differences in sickness absence. Scand J Soc Med  1996; 24: 169– 176. Google Scholar CrossRef Search ADS PubMed  26. Sydsjö A, Sydsjö G, Alexanderson K. Influence of pregnancy-related diagnoses on sick-leave data in women aged 16–44. J Womens Health Gend Based Med  2001; 10: 707– 714. Google Scholar CrossRef Search ADS PubMed  27. Statistics Norway. Statistikkbanken 2011 . https://www.ssb.no/statistikkbanken/ (1 January 2018, date last accessed). PubMed PubMed  28. Modini M, Joyce S, Mykletun A, et al.   The mental health benefits of employment: results of a systematic meta-review. Australas Psychiatry  2016; 24: 331– 336. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the Society of Occupational Medicine. All rights reserved. For Permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Occupational MedicineOxford University Press

Published: Apr 17, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off