Gender equality in domestic work and sickness absence - a population-based study on women and men in Sweden

ABSTRACT

Division of domestic work by gender has been discussed as part of the explanation why women present a higher sickness absence rate than men. This study aimed to examine the association between gender equality in domestic work and sickness absence. Data from 2,609 co-habiting women and men (aged 19–64) collected in a general population in Sweden were used. Associations between different measures of gender equality in domestic work and numbers of sick-leave days were analyzed with logistic regression analysis adjusted for age, children, and paid work. Results show that women reported lower levels of gender equality than men did. Satisfaction with division of domestic work was in the final model associated with lower odds ratio (OR) for sickness absence in men irrespective of number of days. Work-family conflicts were associated with higher ORs for sickness absence in men, 1–7 sick-leave days (OR 1.51 (CI 1.04–2.18)), and in women, 8–30 days (OR 1.51 (1.00–2.33)). More knowledge on the meaning of gender equality in domestic work in relation to sickness absence for women and men are important for future prevention activities.

KEYWORDS:

• Domestic work
• gender equality
• sickness absence
• sweden

Introduction

Gender stratification in domestic work is well documented in many countries (Cunha et al. 2016; Treas and Tai 2016) and has implications on an individual, organizational, and societal level as one of the main causes for gender gaps in labor participation and income (Angelov, Johansson, and Lindahl 2016; Hook 2010). Previous research on domestic work has also shown associations with adverse effects in relation to health and sickness absence (DePasquale et al. 2016; Josephson et al. 2003; Landstedt, Harryson, and Hammarstrom 2016; Staland-Nyman, Alexanderson, and Hensing 2008).

The sharing of paid and unpaid work tasks among women and men in Sweden is promoted by public policy and aim for gender equality (Borrell et al. 2014). The labor participation rate among women is almost equal to men and gender equality is a highly important issue among the public (Hagqvist, Gillander Gådin, and Nordenmark 2017; OECD 2015). However, despite a cultural context of gender equality values at work and in the family, the division of domestic work by gender is still persistent with women investing more hours in domestic work and taking longer parental leaves than men (Björk Eydal et al. 2015; Statistics Sweden 2018).

From an international perspective, sickness absence rates are high in Sweden. Since long there have also been considerable gender differences in sickness absence with women having higher sickness absence compared to men (Armannsdottir et al. 2013; Försäkringskassan [Swedish Social Insurance Agency] 2018). Moreover, women’s larger share of the domestic work and the inequalities in the division of this work between partners has been suggested as an explanation for the gender difference in sickness absence. However, the evidence for a relationship between gender inequalities in the division of domestic work and sickness absence has been inconsistent (Bekker, Rutteb, and Van Rijswijk 2009; Nyman, Spak, and Hensing 2012; Peters et al. 2018).

Domestic work is often conceptualized as a set of unpaid household work tasks aimed at maintaining the home and support different needs in the family or the extended family (Staland-Nyman 2010). According to Hammarstrom et al. (2013) the concept of gender equality can be defined in terms of absence of discrimination between women and men and is characteristically a foundation for notions of gender fairness (Hammarstrom et al. 2013). Building on these concepts, gender equality in domestic work in this study focuses on the degree to which participants report domestic work tasks to be shared equally between partners and on their satisfaction with this actual division.

Research from several countries provides evidence that women perform the major part of the domestic work, also when performing paid labor (Jolly et al. 2014; Treas and Tai 2016). Facing a heavy burden from domestic work or conflicts between domestic work and paid labor have in earlier studies been found to be associated with, for example, sup-optimal self-rated health in both women and men (Janzen and Hellsten 2018; Leineweber et al. 2013; Molarius et al. 2014; Staland-Nyman, Alexanderson, and Hensing 2008), burnout (Blom et al. 2014; Putnik and Houkes 2011) and sickness absence (Antai et al. 2015; Mortensen et al. 2017; Nyman, Spak, and Hensing 2012; Sabbath et al. 2012). Gender equality in domestic work is however not only a question of women and men sharing specific domestic work tasks in an equal way on an hourly basis. Individual perception, of equality and fairness regarding the division of domestic work between partners has also been found to be of importance in relation to physical as well as mental health outcomes, in both women and men (Bohlin et al. 2013; Claffey and Manning 2010). An American study on employed married women showed that perceived fairness in domestic work played a mediating role on the extent to which the division of domestic work influenced personal distress (Claffey and Mickelson 2009). Results from a Swedish longitudinal study showed that the experience of couple relationship as gender unequal was an important influence on the associations between domestic work and psychological distress in both women and men (Harryson, Strandh, and Hammarstrom 2012).

However, as mentioned, the relationship between the gender equality in the division of domestic work and its significance for sickness absence has not been fully clarified. Staland-Nyman and colleagues found that not only the amount of household and caring activities but also lack of domestic work equity and marital satisfaction were related to higher risk for sickness absence among professional women in Sweden (Staland-Nyman, Alexanderson, and Hensing 2008). Furthermore, gender equality in the division of domestic work has been discussed to play a role in sickness absence duration. Perceiving unequal division of responsibility for domestic work was, for example, associated with increased odds for insufficient time for recovery activities in a study on co-habiting women in Sweden (Eek and Axmon 2015). In relation to sickness absence such situation may contribute to inadequate rehabilitation with a prolonged sick-leave period as a result (Staland-Nyman et al. 2009).

Earlier research on domestic work exposures has foremost focused on the situation of women, and hence, less knowledge is available about men. Differences may exist in the meaning of measures and predictors to evaluate domestic work tasks and their relationship to health outcomes (Claffey and Manning 2010; Lachance-Grzela and Bouchard 2010; Staland-Nyman, Alexanderson, and Hensing 2008). Moreover, the experiences of gender equality in domestic work in women and men may be related in different ways to sickness absence. The main objective of this study was therefore to examine the relationship between different indicators of gender equality in domestic work among co-habiting women and men and sickness absence.

Materials and methods

Research design and study population

Data were drawn from a large population sample (N = 7,737) in Western Sweden including urban and rural areas inhabiting approximately 1.6 million people (17% of the Swedish population). The current study was based on a random selected population sample (RPS), including individuals 19–64 years of age. The population sample was selected by Statistics Sweden (responsible national authority for official and government statistics) in 2008. Data were collected through a comprehensive questionnaire that covered areas such as socio-demographic factors, paid and unpaid work, family situation, life-events, leisure time, different health aspects and sick-leave. More details have been described elsewhere (Armannsdottir et al. 2013). The response rate in the RPS was 57% (n = 2,234) in women and 44% (n = 1,793) in men (total N = 4027).

Since focus in the present study was on indicators of gender equality in domestic work only participants who reported a marital or co-habiting relationship were included, 1449 women and 1160 men (total n = 2,609).

A drop-out analysis in the total study population showed a higher drop-out rate for women and men in the youngest age group (19–30 years), in the lowest income level (≤149 000 SEK) and among those born outside the Scandinavian countries. Furthermore, the dropout rate was higher among those living alone, compared to married or cohabiting persons. The dropout pattern for the separate samples was overall similar to the pattern in the total study population.

The study complied with the World Medical Association’s Declaration of Helsinki. Participation in the study was based on informed consent. An information letter was enclosed with the mailed questionnaire where information was given on the study aim, the procedure for data collection, the right to abstain from participation or to withdraw consent to participate at any time and institutional affiliation of the researcher together with contact details. Data are shown on an aggregated level and no single individual can be identified. All parts of the study, both questionnaire- and register-based, were approved by the Regional Ethics Committee for Human Research at the Gothenburg University, Sweden, registration number 039–08.

Data analysis

Explanatory variables

In the present study, five measures, earlier found to be associated with health (Harryson, Novo, and Hammarstrom 2012; Lidwall, Marklund, and Voss 2010; Staland-Nyman, Alexanderson, and Hensing 2008), were used as indicators of gender equality in domestic work; equal partner relationship, planning of domestic work, performing of domestic work, satisfaction with division of domestic work and experience of conflicts between work and family.

Equal partner relationship was assessed by the following question: ‘Compared to other couples of your acquaintance, to what degree do you think your partner relationship is on an equal footing?’ Responses were given on a 5-point ordinal scale: (5) To a very high degree; (4) To a high degree; (3) Partly; (2) To a slight degree and (1) To a very slight degree. In the analyses, the response categories were dichotomized into equal relationship (including the categories 4 and 5) and less equal relationship (including the categories 1 to 3).

Responsibility for planning of domestic work was assessed by: ‘Who plans most of the housework and other work at home?’ Responsibility for the performing of domestic work was assessed by: ‘How is the housework and other work at home shared?’ Responses on the two questions were given on a six-point scale: (1) I plan/do the most; (2) My partner plans/does the most; (3) Another adult in the household plans/does the most; (4) My partner and I share the planning/performing roughly equally; (5) Another adult in the household and I share the planning/performing roughly equally; (6) Other distribution of the planning/performing. In the analyses, the answers were leveled into, me (I plan/do the most of the work), my partner (My partner plans/does the most of the work) and 50/50 (My partner and I share the planning/performing roughly equally). Other response categories were excluded in the analyses due to small numbers.

Satisfaction with division of the domestic work was assessed by: ‘To what degree are you satisfied with the division of the housework and other work at home? Conflicts between work and family was assessed by; ‘To what degree do you experience a conflict between your paid work and housework or other work at home?’ The response categories in questions on satisfaction and conflicts were: (5) To a very high degree; (4) To a high degree; (3) Partly; (2) To a slight degree and (1) To a very slight degree. In the analyses, categories were dichotomized into high degree (including the categories 4 and 5) and low degree (including the categories 1 to 3).

A brief description of domestic work (including care and supervision activities related to children, aged or sick relatives) was given in the questionnaire to ensure that the participants related their answers not only to household work but to a more comprehensive understanding of the content of domestic work.

Adjustment variables

Three factors, age, having children below 12 years of age living at home and paid work were adjusted for in the analyses since earlier studies have shown associations with both explanatory variables and sickness absence (Lidwall, Marklund, and Voss 2010). Age was verified through census data. In the questionnaire, participants were asked to report if they had any children (biological, adoptive-step- or foster children) currently living at home at least half of the time, and when applicable age of the children was asked for. Paid work was adjusted for by using information from questionnaire data on the number of hours (<15 hours/week, ≥15 hours/week, ≥35 hours/week) the individuals devoted to paid work. In Sweden, full-time work involves ≥35 hours/week. In the analyses, categories were dichotomized into full-time work (≥35 hours/week) and part-time work (<35 hours/week).

Outcome variable

The outcome was self-reported numbers of sick-leave days during the past 12 months. All residents of working age in Sweden are covered by the social sickness insurance. The sickness benefit amounts to approximately 80% of lost income and can be granted if an individual´s ability to work is reduced due to medical reasons. For employees, the first 14 days in a sick-leave spell are paid by the employer (sick-pay period), and any further days are paid by the social insurance. The first 7 days can be self-certified thereafter a medical certificate is required. The outcome variable in the present study was assessed by the following question “Have you been on sick leave on one or more occasions during the past 12 months?” Participants that had been on sick-leave were then asked to state the actual numbers of sick-leave days during the past 12 months. Response categories were; 1–7 days, 8–30 days, 2–3 months, or 4–12 months. In the analysis, the response categories were trichotomized into; 1–7 days, 8–30 days, and > 30 days.

Statistical methods

Data were analyzed using IBM SPSS 21.0. Pearson chi-square test was performed for differences in independent variables between women and men. Spearman correlations were performed for analyses of interrelationships among the indicators of gender equality used in this study. Associations between indicators of equality in domestic work and sickness absence were analyzed with multinomial logistic regression in three models yielding odds ratios (OR) with 95% confidence intervals (CI). Associations were adjusted for age, having children below 12 years of age living at home and the number of hours devoted to paid work (less than 35 hours per week or 35 hours or more per week). The cut off ≥35 hours/week was set on the basis of national and European Union standards for categorization of full-time work. All analyses were performed separately for women and men.

Results

The focus in the present study was on the association of different indicators of gender equality in domestic work on sickness absence. Consequently, only participants who reported a marital or co-habiting relationship and not on sick leave at the time when answering the questionnaire were included. The individuals in the study population were well established at the labor market. The vast majority reported that they were employed, 76% of the women and 78% of the men, and among them 80% or more reported that they had a permanent position. A higher proportion of the men compared to the women reported a full-time paid work, 93 and 64%, respectively. About 50% of the population reported that they had younger children (<12 years of age) living at home. Details on sociodemographic and work-related characteristics of the study population are presented in Table 1.

Table 1. Sociodemographic and work-related characteristics of the study population (N = 2,609)

	Study population^1
Characteristics	Women n (% within sex)	Men n (% within sex)	Chi-2 (df)
Partner			2469.2***(1)
Married/cohabiting with a man	1436 (99.1)	22 (1.9)
Married/cohabiting with a woman	13 (0.9)	1138 (98.1)
Age groups			20.81*** (3)
19–30	239 (16.5)	130 (11.2)
31–40	359 (24.2)	259 (22.3)
41–50	358 (24.7)	315 (27.2)
51–64	493 (34.0)	456 (39.3)
Educational level^2			22.89*** (2)
Primary education	223 (15.5)	217 (18.8)
Secondary education	564 (39.1)	522 (45.1)
University or college	654 (45.4)	418 (36.1)
Occupational class^2			83.36*** (3)
Manual±	470 (33.4)	398 (35.2)
Non-manual±±	641 (45.5)	345 (30.5)
High level	285 (20.2)	384 (33.9)
Students	12 (0.8)	5 (0.4)
Labor market status^2			136.74***(1)
Employed	1071 (76.4)	876 (77.9)
Other ±±±	331 (23.6)	249 (22.1)
Employment status^2			83.89***(4)
Permanent position	1067 (83.8)	858 (80.3)
Temporary position	131(9.2)	45 (4.2)
Other	91 (7.0)	166 (15.5)
Managerial position^2
Yes	213 (16.3)	337 (34.8)	108.03***(1)
No	1090 (83.7)	707 (65.2)
Hours/week in paid work^2			289.28***(1)
Full-time (≥35 hours/week)	824 (63.8)	1007 (93.2)
Part-time (<35 hours/week)	467 (36.2)	73 (6.8)
Children <12 years of age living at home^2			1.35 (1)
Yes	688 (48.7)	525 (46.4)
No	725 (51.3)	607 (53.6)
Number of sick leave days past 12 months			14.89** (3)
0 days	855 (59.0)	749 (64.6)
1–7 days	382 (26.4)	296 (25.5)
8–30 days	155 (10.7)	83 (7.2)
> 30 days	57 (3.9)	32 (2.8)

¹Participants on sick leave or disability pension > 75% at the time of the study were omitted

2 Dispersed numbers of participants owing to internal dropouts

±Unskilled/Skilled manual, ±±Low/Medium Non manual

±±± respondents reporting their labor market status as self-employed, student, disability pensioner (part-time), on parental leave, job seeker, unpaid work and other

* p < .05, ** p < .01, *** p < .001.

Differences between men and women in indicators of gender equality in domestic work

The self-report on indicators of gender equality in domestic work differed between women and men (Table 2). A higher proportion of women than men (19.1% vs. 14.5%) reported a low degree of equality in their relationship. Concerning division of domestic work tasks, a higher proportion of women than men reported themselves as main responsible for planning (61.0% vs. 5.1%) and performing (41.3% vs. 4.2%) of domestic work. Almost twice as many women as men (36.2% vs. 19.0%) reported a low degree of satisfaction with the division of domestic work. More women than men (18.8% vs. 14.8%) also reported that they perceived a high degree of conflicts between paid work and domestic work (i.e., family and household work tasks). Significantly more men than women however reported an equal division of the planning and the performing of domestic work tasks in the household. A higher proportion of men than women (81% vs. 63.8) also reported a high degree of satisfaction with how the domestic work tasks were divided in the household (Table 2).

Table 2. Indicators of gender equality in domestic work and differences between women and men (Pearson chi-2 test) (N = 2609)

		Women		Men
Indicators of gender equality in domestic work		n	%	n	%	Chi-2 (df)
Equality in relationship	Low degree High degree	259 1096	19.1 80.9	152 899	14.5 85.5	9.04** (1)
Planning domestic work	Myself Equal division Partner	879 516 22	62.0 36.4 1.6	57 539 523	5.1 48.2 46.7	1163.99** (2)
Performing domestic work	Myself Equal division Partner	576 784 36	41.3 56.2 2.6	47 694 377	4.2 62.1 33.7	714.21** (2)
Satisfied with division	Low degree High degree	524 924	36.2 63.8	220 936	19.0 81.0	92.72** (1)
Conflicts paid work – family	High degree Low degree	247 1067	18.8 81.2	160 923	14.8 85.2	6.82** (1)

* p < .05, ** p < .01

Interrelationships among indicators of gender equality in domestic work

The intercorrelations among the indicators of gender equality in domestic work differed for women and men (Table 3). There was a strong, significant and negative correlation in women between perceived equality of the relationship on the one hand, and planning (−.26**) and performing (−.35**) of domestic work on the other hand. This indicates that women in the study considered their relationship as less equal when they had to plan and perform most of the domestic work in the household. For men, however, this relationship was weakly positive (correlations varied between .02 and .20**) indicating that these factors were less important consider the degree of equality in the relationship. For both men and women, a positive relationship between perceived equality in the relationship and satisfaction with the division of domestic work was found. The correlations varied between .29** for men and .49** for women. Conflicts between paid work and family (i.e. domestic work task in family and household) was negatively related to satisfaction with division of domestic work. Furthermore, women who plan and perform most of the domestic tasks were less satisfied about this division of work (−.42** and −.56**). For men, this relationship was absent (for planning), or weakly positive (for performing) (Table 3).

Table 3. Spearman correlations for indicators of gender equality in domestic work for men/women

Indicators ^a	1	2	3	4	5
1. Equality of relationship	–
2. Planning domestic work	.11**/-.26**	–
3. Performing domestic work	.20**/-.35**	.51**/.54**	–
4. Satisfied with division	.29**/.49**	.02/-.42**	.08**/-.56**	–
5. Conflicts work – family	−.10**/-.12**	.00/.08**	.00/.08**	−.17**/-.13**	–

^aIndicators 1, 4 and 5: range 1 (yes) to 2 (no). Indicators 2 and 3: range 1 (100% myself) to 3 (100% partner).

** p < .01

Associations between indicators of gender equality in domestic work and sickness absence

Two of the examined indicators of gender equality in domestic work were associated with self-reported number of sick-leave days during the past 12 months. After control for age, children living at home and paid work situation, satisfaction with division of domestic work was associated with number of sick-leave days in men and perceiving conflicts between work and family (i.e. domestic work tasks related to family and household) was associated to number of sick-leave days both in women and men.

Men who reported a higher degree of satisfaction with the division of domestic work showed lower ORs for sickness absence (i.e. number of sick-leave days) during the past 12 months compared to men who reported a lower degree of satisfaction with the division of domestic work. This result was found regardless of the number of sick-leave days; 1–7 days, (OR 0.66 (CI 0.47-0-93)), 8–30 days, (OR 0.56 (CI 0.33–0.95)) and >30 days, (OR 0.28 (CI 0.14–0.58)). The figures were only slightly changed after adjustment for age, having children below 12 years of age living at home and finally paid work situation (Table 4).

Table 4. Crude and adjusted Odds ratios (OR) with 95% confidence interval (CI) for associations between indicators of gender equality and number of sickness absence days past 12 months, (N = 2609)

Number of sick-leave days past 12 months
	1–7 sick-leave days
	Model 1^1 OR (95%CI)		Model 2^2 OR (95%CI)		Model 3^3 OR (95%CI)
Indicator of gender equality	Women	Men	Women	Men	Women	Men
Satisfied with division
Low degree ^a	1	1	1	1	1	1
High degree	1.01 (.72–1.30)	.66 (.47-.93)	1.00 (.78–1.28)	.67 (.48-.94)	1.00 (.78–1.31)	.66 (.46-.92)
Conflict work – family
Low degree^b	1	1	1	1	1	1
High degree	1.26 (.92–1.73)	1.67 (1.16–2.04)	1.25 (.91–1.72)	1.55 (1.07–2.34)	1.29 (.93–1.78)	1.51 (1.04–2.18)
	8–30 sick-leave days
	Model 1^1 OR (95%CI)		Model 2^2 OR (95%CI)		Model 3^3 OR (95%CI)
Indicator of gender equality	Women	Men	Women	Men	Women	Men
Satisfied with division
Low degree ^a	1	1	1	1	1	1
High degree	.91 (.64–1.29)	.56 (.33-.95)	.90 (.63–1.29)	.56 (.36-.96)	.95 (.65–1.36)	.56 (.32-.97)
Conflict work – family
Low degree^b	1	1	1	1	1	1
High degree	1.52 (.99–2.33)	.83 (.40–1.72)	1.53 (1.0–2.37)	.75 (.36–1.56)	1.51 (1.0–2.33)	.75 (.36–1.57)
	>30 sick-leave days
	Model 1^1 OR (95%CI)		Model 2^2 OR (95%CI)		Model 3^3 OR (95%CI)
Indicator of gender equality	Women	Men	Women	Men	Women	Men
Satisfied with division
Low degree ^a	1	1	1	1	1	1
High degree	1.04 (.59–1.83)	.28 (.14-.58)	1.08 (.61–1.89)	.27 (.13-.56)	1.32 (.72–2.43)	.23 (.10-.52)
Conflict work – family
Low degree^b	1	1	1	1	1	1
High degree	1.61 (.82–3.17)	1.16 (.39–3.43)	1.74 (.87–3.47)	1.23 (.41-.3.72)	1.89 (.96–3.80)	1.20 (.40–3.63)

¹Model 1, crude model, 2 Model 2, adjusted for age and having children below 12 years of age ³ Model 3, adjusted for variables in Model 2 and number of hours in paid work.

^aReference category is low degree of satisfaction with division of domestic work ᵇ Reference category is low degree of conflicts between paid work and unpaid domestic work in family.

Conflict between work and family (i.e. domestic work tasks related to family and household) was related to the number of sick-leave days during the past 12 months among both women and men. The analyses showed higher ORs for men to report 1–7 days of sick-leave during past 12 months compare to men not reporting such a conflict (OR 1.67 (CI 1.16–2.04)) while women who perceived conflicts between work and family had higher ORs for reporting 8–30 sick-leave days (OR1.52 (CI .99–2.33)). In the final model after adjustments for age, children, and paid work, the associations were significant with OR 1.51 (CI 1.04–2.18) for men and OR 1.51 (CI 1.00–2.33) for women, (Table 4).

No significant relationships between the other examined indicators: equal relationship, planning of domestic work, and planning of domestic work and number of sick-leave days.

Discussion

In order to emphasize the complexity of domestic work, different indicators were analyzed. Although the vast majority (80% or more) of the women and men rated their relationship as equal, the factual planning and performing of domestic work were only divided equally in approximately half of the population. This is in line with the high positive scores in Sweden on gender equality despite equality being not fully achieved in the reality (Björk Eydal et al. 2015; Hagqvist, Gillander Gådin, and Nordenmark 2017). The women in this study perceived their relationships as more unequal than the men, which was in line with the results of Harryson and colleagues (Harryson, Novo, and Hammarstrom 2012). Also in line with earlier studies (Leineweber et al. 2013; Sabbath et al. 2012), women in this study perceived more conflicts between work and family than the men.

Interrelationships between the examined indicators varied by gender and similar to a Canadian study, women doing more domestic work were more likely to perceive dis-satisfaction with the division (MacDonald, Phipps, and Lethbridge 2005). Even though, satisfaction with division of domestic work was correlated to reporting having an equal relationship in both women and men, the correlation was less strong in men. This result indicates that how domestic work is divided on a daily basis between married/co-habiting partners’ may influence the perception of equality in the partner relationship.

As mentioned, earlier findings regarding gender inequality in the division of domestic work and sickness absence have been inconsistent. In this study, we did not find an association between the division of domestic work (i.e., the actual planning or performing of the work), between partners and sickness absence. For men however, the perceived satisfaction with how the domestic work was divided seemed to decrease the probability for sickness absence. This might suggest that they had less health problems or were able to return to work earlier if they were satisfied with how the domestic work at home was divided between them and their partner. For women, satisfaction with the division of domestic work was not related to sickness absence in this study. In a similar way, Harryson, Strandh, and Hammarstrom (2012) showed gender (in)equality in the domestic sphere to be differently related to psychological distress in women and men. In that study, taking the whole responsibility for household work was associated with psychological distress in women while taking less than half of the responsibility in men. (Harryson, Novo, and Hammarstrom 2012).

Earlier studies have suggested that an unequal division, as well as negotiations within the relationship to achieve a more equal division of tasks, could create stress in women and conflict in the partner relationship (Evertsson 2013; MacDonald, Phipps, and Lethbridge 2005). These results indicate that the achievement of higher equality in domestic work comes with a prize for women where there is a trade-off between getting a more even division of work tasks and effort to get there. Among the women in this study, such factors might have counterbalanced the possible positive effects of satisfaction with the division of domestic work between partners, and thus lead to lack of relationship between satisfaction and sickness absence.

The findings also showed that interference between work and family might be detrimental to health in both women and men. This association was significant significant also after the individuals paid work situation and if they had younger children living at home were considered. Despite a welfare system with public policy and services aiming to ease the interplay between work and family, conflicts between these two areas were related to higher probability for sickness absence in both women and men. Similar results have been seen in earlier studies in relation to women (Clays et al. 2009; Nilsen et al. 2017) and in men in higher socio-economic positions (Lidwall, Marklund, and Voss 2010). In the present study we did not analyze the two directions of interference between work-family versus family-work, meaning that this study had no possibility to distinguish which domain influences the result the most. An earlier study on Swedish data found that work-to-family interference was more common than family-to-work interference in both men and women and furthermore related to long-term sickness absence (>14 days) among men with higher socio-economic status (Lidwall, Marklund, and Voss 2010). However, in the present study the paid work situation was controlled for in terms of the number of hours the individuals devoted to their paid work every week and the association with sickness absence remained significant. Of importance is also that even if the directions and strength of associations were similar in men and women, a significantly higher proportion of women reported that they were exposed to work-family conflicts, which might suggest a potential higher burden in women from this kind of exposure.

The study did not find any other of the examined indicators of gender equality in domestic work to be associated with sickness absence. Part of the explanation could be the multi-factorial and complex nature of decisions about sick leave, which include individual health, sickness insurance legislation, and professional as well as workplace factors involved in that were not investigated in this study. However, it might also be related to the way indicators were operationalized and the self-report nature. In Sweden, gender equity is a strong societal norm that could have influenced the result.

This study addressed a topic of relevance for the individual, for work organizations as well as for the society. Sickness absence is an issue of major concern in many countries and the significance of gender equality in domestic work has been lifted as a potential influencing factor during the recent years. Moreover, gender equality in domestic work has been discussed to promote better work–life balance among partners and in families. The strength of the present study is that the findings represent people in the working age, retrieved from a general population. Although data collection was conducted in 2008, we judge the data still to be highly relevant for the study purpose. The data encompass a comprehensive measure on the question of interest which is not often the case in larger population surveys. The data were collected for both women and men which was an asset, since more comprehensive questions related to domestic work has foremost been collected from women. According to recent statistics, the distribution of domestic work between partners has not changed significantly (Statistics Sweden 2018). Moreover, we have no reason to believe that the association to sickness absence would change over the years. Instead, recent reports put forward gender equality in domestic work as an important factor in promoting work–life balance and reducing stress-related mental illness (Lidwall et al. 2018; Swedish Social Insurance Agency 2020).

A limitation is that indicators of gender equality in domestic work were single-item measures. However, compared to paid labor the concept of domestic work is less operationalized and this study contributes methodologically to the assessment of the concept. The study could analyze several indicators of gender equality in domestic work and is among one of the few that also included the “planning and organization” aspect of domestic work. As mentioned earlier future studies need to develop new instruments and scales tested for use in both women and men, which takes the complexity of domestic work and gender division of this work into account. In relation to this, it is important to note that possible differences in division of domestic work between heterosexual and same-sex couples were not ruled out in the study due to lack of data. Methodological reflections should be made regarding the higher drop-out among women and men in the youngest age group (19–30 years). This could possibly have led to an underrepresentation of a more equal division of domestic work, but such underrepresentation will probably not have affected the association to sickness absence. Furthermore, it has been suggested that age age and having small children having small children. In this study, both age and small children living at home were adjusted for in the analyses as well as the individuals paid work situation. The cross-sectional design does not allow drawing conclusions about causality. However, the findings suggest that perceptions of domestic equality could possibly have different consequences for men and women. Major strengths were the large quantitative sample and analyses shedding light on a topic of great importance for societies with dual-earner couples/families.

Conclusions

Few studies exist on the relation between gender equality in domestic work and sickness absence. Interestingly, only men profited from satisfaction with the division of domestic work, in terms of lower probability for sickness absence, while work and family conflicts increased the probability for sickness absence in women and men alike. The findings expand the knowledge on gender equality in domestic work and sickness absence. However, more research is needed to further elaborate the conceptual differences between different indicators of gender equality in domestic work and sickness absence considering that predictors in different indicators may differ between genders, family structures, and cultural contexts.

Disclosure statement

No potential conflict of interest was reported by the author(s).