Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Gender, time use and overweight and obesity in adults: Results of the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil)

  • Karina Araujo Pinto ,

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Validation, Visualization, Writing – original draft, Writing – review & editing

    karinapinto@ufba.br

    Affiliation School of Nursing, Federal University of Bahia, Salvador, Bahia, Brazil

  • Rosane Harter Griep,

    Roles Validation, Writing – review & editing

    Affiliation Laboratory of Health and Environmental Education, Oswaldo Cruz Institute (Fiocruz), Rio de Janeiro, Rio de Janeiro, Brazil

  • Lucia Rotenberg,

    Roles Validation, Writing – review & editing

    Affiliation Laboratory of Health and Environmental Education, Oswaldo Cruz Institute (Fiocruz), Rio de Janeiro, Rio de Janeiro, Brazil

  • Maria da Conceição Chagas Almeida,

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Validation, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Gonçalo Moniz Institute, Fiocruz, Salvador, Bahia, Brazil

  • Rosane Sousa Barreto,

    Roles Visualization, Writing – review & editing

    Affiliation Climério de Oliveira Maternity Hospital, Federal University of Bahia, Salvador, Bahia, Brazil

  • Estela M. L. Aquino

    Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Institute of Collective Health, Federal University of Bahia, Salvador, Bahia, Brazil

Gender, time use and overweight and obesity in adults: Results of the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil)

  • Karina Araujo Pinto, 
  • Rosane Harter Griep, 
  • Lucia Rotenberg, 
  • Maria da Conceição Chagas Almeida, 
  • Rosane Sousa Barreto, 
  • Estela M. L. Aquino
PLOS
x

Correction

8 Aug 2018: Pinto KA, Griep RH, Rotenberg L, Almeida MdCC, Barreto RS, et al. (2018) Correction: Gender, time use and overweight and obesity in adults: Results of the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil). PLOS ONE 13(8): e0198380. https://doi.org/10.1371/journal.pone.0198380 View correction

Abstract

Perceived time constraints have been highlighted in sociological studies as representing a core issue in determining quality of life. The objective of this study was to test the hypothesis that gender inequalities regarding insufficiency of time play a role in the development of overweight and obesity in adults. The study used baseline data (2008–2010) from the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil), which monitors a cohort of 15,105 civil servants of 35 to 74 years of age. Insufficient time for personal care and leisure due to professional and domestic duties, as detailed in a structured questionnaire, comprised the main exposure variable. The outcome variable was overweight/obesity measured according to body mass index. Prevalence ratios were calculated using multinomial logistic regression. A greater proportion of women compared to men reported insufficient time for personal care and leisure (34.5% versus 23.8%, respectively). The prevalence of overweight was greater in men, while obesity was more common in women. Insufficient time for personal care and leisure was associated with overweight (PR = 1.29; 95% CI: 1.04–1.61) and obesity (PR = 1.65; 95% CI: 1.28–2.12) only in women working over 40 hours/week. No significant association was found for males. These results suggest that the length of the working week influences factors underlying weight gain, possibly issues linked to behavior and/or stress mechanisms. The fact that such an association was restricted to women suggests that the results originate from gender inequalities involving relationships between time and health. The findings of this study provide data on which to base public policies aimed at encouraging the redistribution of domestic responsibilities in the direction of gender equity, as well as macrosocial policies such as providing public schools for workers’ children.

Introduction

Perceived time constraints have been highlighted in sociological studies as a central issue for quality of life [14]. Several characteristics of the modern world account for this perception. Firstly, the increasing contingency of employees working antisocial hours in a society that demands round-the-clock production and availability of services [5]; secondly, the intrusion of work into individuals’ personal life as a consequence of information technology providing access to employees at any time of the day or night [6]; and, finally, the greater participation of women in the job market, with slower changes in the domestic sphere, creating challenges for women with respect to the demands made on them in their personal and professional life [7].

This context also encompasses the debate on the work-family conflict that arises when the roles played in these two spheres are incompatible [8], since the demands of one exhaust all the individual’s time and physical and mental energy, leaving insufficient resources for the activities of the other [9]. The majority of studies dealing with this approach have analyzed the effect of job demands on family life [10], while others have evaluated the implications in the opposite direction [11].

The dynamics of professional and family life are strongly affected by asymmetric gender relationships, involving differentiated demands for men and women [1,3,12,13]. Indeed, the pattern of inequality in the division of household chores between men and women persists and, even in developed countries, within the domestic sphere the burden remains greater for women, particularly if there are children and/or a spouse or a person who requires care [3,11]. The need to conciliate different demands within a daily process of negotiation under time constraints affects the hierarchy of women and men’s priorities, encouraging them to reduce the use of time for themselves, with a possible adverse effect on their health and on their ability to maintain healthy lifestyle habits [3,10,12,13].

Despite the broad scope of sociological literature on time use during the 24 hours of the day, even from the perspective of gender, few studies have focused on the relationship between time use and health. One of the issues that have been investigated is obesity, which has been associated both with time use [2] and with long working hours [14]. The analyses of nutritional patterns and obesity are based on the concept that, although the multi-causality of obesity has been well documented [1517], the interaction between socio-environmental and biological factors in the occurrence of obesity remains to be fully clarified. Empirical results have shown that mechanisms involved in the increasing magnitude of the accumulation of body weight appear to develop as a function of the underlying lifestyle characteristics of individuals in current society [1820].

From a quantitative viewpoint, time is an organizing element for the dynamics of daily life. Thus, excessive time spent on any given activity reduces the time available for other activities, including those such as physical exercise that influence health [21]. Gender differences in time distribution and their relationships with health have seldom been investigated. The hypothesis of the present study is that gender inequalities in the time available for one's own care and leisure play a role in the development of overweight and obesity in adults.

Materials and methods

Ethical approval and participant consent

The ELSA-Brasil study was approved by the internal review boards of the six institutions involved (Hospital de Clínicas de Porto Alegre, Universidade Federal do Rio Grande do Sul, Universidade Federal do Espírito Santo, Universidade Federal de Minas Gerais, Universidade Federal da Bahia, Universidade de São Paulo and the Fundação Osvaldo Cruz) and by the National Committee for Ethics in Research (CEP/CONEP system) under reference #13065. All the study procedures were conducted in compliance with Resolution 196/96 of the National Health Council in effect at that time. All participants provided written, informed consent for their clinical records to be used in this study prior to enrolment.

Data

This study was conducted using baseline data from the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil), a prospective multicenter study currently being carried out in this country with civil servants working at five federal universities and one research center, and whose detailed description has been published elsewhere [22]. In brief, the ELSA-Brasil cohort consists of 15,105 active and retired volunteers of both sexes, ranging in age from 35 to 74 years at baseline, with recruitment having taken place between 2008 and 2010. This study combined participants from three different segments of civil service: faculty members, administrative and technical staff, and support workers. To achieve better distribution, recruitment goals were defined by sex, age and occupational category. At baseline, the participants were submitted to clinical exams and face-to-face interviews. For the present study, data were analyzed from 12,066 women and men still in active employment at that time and for whom the data of interest were available.

Exposure variable

The principal exposure variable was “insufficient time for personal care and leisure because of job- and family-related demands”, obtained from responses to the statement: “Family or professional demands (requirements or requests) prevent you from using the amount of time you would like to have to take care of yourself or for leisure”. The participants were asked to what extent they agreed with this statement, with response options being: “never or almost never”, “rarely”, “sometimes”, “often” or “very often”, rendering a score that ranged from 0 to 4 in that order. This variable was dichotomized, with scores of 3 and 4 points grouped together into a category defined as “yes” and the other scores grouped together into the reference category “no”. The item was found to have temporal stability when its psychometric proprieties were evaluated using the kappa statistic, resulting in a weighted kappa index of 0.70 (0.63–0.77) [23].

Outcome variable

The outcome variable was overweight/obesity, based on the calculation of body mass index (BMI). The anthropometric measurements included weight and height. All the participants’ measurements were taken under baseline conditions consisting of 8–12 hours fasting.

Body weight was measured using Toledo® scales with a capacity of up to 200 kg. Height was measured using Seca® stadiometers appropriate for taking anthropometric measurements for research purposes. The equipment and their respective installation and calibration procedures were standardized for all the research centers involved in the ELSA-Brasil study [24].

Participants were classified as overweight or obese based on their BMI, calculated using the formula BMI = weight(kg)/height(m)2. This is the indirect measurement most commonly used and that recommended by the World Health Organization for the evaluation of overweight/obesity in adults [25]. Overweight was defined as a BMI of 25 to 29 kg/m2 and obesity as BMI ≥ 30 kg/m2. Three categories were thus defined for the outcome variable: normal weight/underweight (1), overweight (2) and obesity (3).

Covariates

The major association was adjusted for other covariates indicated in the literature as being associated with overweight/obesity [12,16,17] and time scarcity [21]. Covariates related to the work and family domains were included in the analyses. Regarding the work domain, the responses to a summarized and adapted version of the Swedish Demand-Control-Support Questionnaire [26] were used, with the following covariates being extracted and dichotomized (0 and 1 in that order): job demands (low and high); level of control at work (high and low); and social support at work (high and low). A covariate was also included on the quantification of the working week (≤ 40 hours or > 40 hours/week).

The following covariates related to the family were selected and dichotomized (0 and 1 in that order): being in a stable marital union (“no” for those who were single or not living with a partner and “yes” for those who were married or living with a partner); presence of children (no/yes); presence of a full-time maid (yes/no) and having to care for a sick or handicapped person (no/yes).

Data analysis

All the analyses were performed separately for women and men. Pearson’s chi-square test was used in the descriptive analysis of the exposure and outcome variables, and for all covariates, with p-values ≤ 0.05 being considered statistically significant.

Tests for ordered logistic model (using omodel command) showed that the proportional odds assumption was violated and then we opted to use a multinomial model. All covariates associated with the outcome (p≤0.05) were included in a multinomial logistic regression analysis to calculate prevalence ratios (PR) and 95% confidence intervals (95%CI), taking as a reference the classification normal weight/underweight. In order to identify a potential effect modifier, the magnitude of crude and adjusted associations was analyzed, together with the 95% confidence intervals. Covariates associated with the outcome (but not the exposure) and with the exposure (but not the outcome), and where in addition there was no causal link between exposure and outcome, were included in the final models for women and men. Prevalence ratios (PR) were calculated using the mlogit command of the Stata® software program. The Wald test was used to verify the independence of the variables in the final models [27]. All analyses were conducted using the Stata® statistical software package, version 10.

Results

Of the 12,066 individuals analyzed in this study, 52.2% were women. The proportion of individuals reporting insufficient time for personal care and leisure was greater in the group of women compared to the men (34.5% and 23.8%, respectively) (Table 1 and S1 Table). Both in men and women, insufficient time for personal care and leisure was reported by a significantly higher proportion of workers in the 35-49-year age group, by those with higher education levels, and those self-reporting their skin color as white. All job domain variables were associated with insufficient time for personal care and leisure, both in the group of men and in the group of women.

thumbnail
Table 1. Proportion of individuals reporting insufficient time for personal care and leisure, according to selected variables.

https://doi.org/10.1371/journal.pone.0194190.t001

In the family domain, the greatest proportions of individuals reporting insufficient time for personal care and leisure were in the group of women—but not men—in a stable union, among those who had children and those who cared for a dependent individual with special needs. In the group of men, the only variable related to the family that was found to be significantly associated with having insufficient time for personal care and leisure was the presence of a full-time maid to do the housework (35.5%) (Table 1).

The distribution analysis of overweight/obesity showed a high prevalence of both overweight and obesity in both sexes, although overweight was more common in men (45.0%), while obesity was more prevalent in women (23.9%) (Table 2 and S2 Table).

thumbnail
Table 2. Prevalence of overweight and obesity, according to sex and selected variables.

https://doi.org/10.1371/journal.pone.0194190.t002

Women with less schooling were more likely to be obese (30.1%). Among men, the prevalence of overweight was greater in those with a university education (45.4%). Individuals of 50 years of age or more were more likely to be overweight or obese irrespective of sex. Among women, those who declared themselves black or brown-skinned were more likely to be overweight (36.7%) or obese (27.4%). This trend was not confirmed for the men (Table 2).

Among women, an association was also found between being overweight or obese and all the variables in the family domain. It is noteworthy that the greatest prevalence of overweight (36.3%) was observed in those who had children, while the greatest prevalence of obesity (30.5%) was in the group of women who cared for a sick or handicapped person. Women who were able to count on the support of a full-time maid were less likely to be obese (18.1%), while the lowest prevalence of overweight was found in the group of women with no children (30.0%) (Table 2).

Among the men who cared for a sick or disabled person and those who had a full-time maid, no statistically significant differences were found in the distribution of body weight. Those who lived with a partner and those who had children were the groups in which the prevalence rates of overweight (45.4% and 45.6%, respectively) and obesity (21.3% and 21.5%, respectively) were highest (Table 2). The highest prevalence rates of overweight (36.6%) and obesity (25.8%) were found in the group of women who reported high levels of control at work.

When selecting the covariates for inclusion in the models, the length of the working week was identified as an effect modifier in the women; therefore, stratification was made as a function of this variable. The covariates education, age, ethnicity, control at work, children, caring for a sick or handicapped person and presence of a full-time maid were identified as confounding factors and were included for simultaneous adjustment in the final model for the group of women. Wald’s test confirmed the independence of covariates and the good fit of the model. Having insufficient time for personal care and leisure was associated with overweight (PR = 1.29) and obesity (PR = 1.65) only in the group of women whose working week consisted of more than 40 hours, even following adjustment for confounding variables (Table 3).

thumbnail
Table 3. Insufficient time for personal care and leisure and overweight/obesity in women, according to working hours.

https://doi.org/10.1371/journal.pone.0194190.t003

In the group of men, no modifier of the principal effect was found and the covariates selected for adjustment were education, age and children (Table 4). Nevertheless, simultaneous analysis showed no evidence of any association between insufficient time for personal care and leisure and overweight/obesity in this group.

thumbnail
Table 4. Insufficient time for personal care and leisure and overweight/obesity in men.

https://doi.org/10.1371/journal.pone.0194190.t004

Discussion

Understanding the relationship between time and health in an epoch marked by the acceleration of social times is a complex challenge [21]. The results of our study indicate that the perception of not having enough time available for personal care and leisure because of household chores and family demands is associated with a greater prevalence of overweight and obesity in women whose working week consists of more than 40 hours. These findings are supported in the literature on time use in the light of gender theories, which emphasize the disadvantages for women resulting from the unequal distribution of household chores [3,12]. This study is particularly relevant since it takes into consideration the competing demands for an individual’s time from both the public and private domains, with the findings revealing the implications of time constraints resulting from excessive demands. This approach, seldom used in research studies, is similar to the “gender-sensitive” indicators mentioned in the literature [1] as relevant in highlighting the asymmetric experience of women and men in relation to time use and revealing possible social determinants of quality of life.

Studies based on time use in which participants keep a record of their activities over a certain period of time also suggest an association between temporal aspects of feeding and obesity. For instance, in a nationally representative sample of the North American population, time spent in food preparation was inversely related to BMI in women [28]. One possible mechanism involved in weight gain is that greater number of working hours (including paid and unpaid work) may be related to higher levels of stress (with higher blood cortisol rates) and, consequently, a greater likelihood of increasing food intake and/or of choosing food with a high fat and sugar content [18,20,29]. In fact, the perception of a scarcity of time may be related to preparing fewer healthy meals at home and consuming more fast food and convenience food, particularly if there is no full-time maid [29]. Experts in obesity, diet and physical activity consider time pressure as an important social trend underlying the increasing obesity rates [30].

The finding that of the women working more than 40 hours a week, those who did not have a full-time maid were more likely to be obese, should be used with caution when comparing these results directly with data in the literature. Having a full-time maid is a practice that occurs in few countries, especially in that with high gender inequality and high internal rural-urban migration and international migration. In Brazil, having women of low socioeconomic status that perform domestic chores in middle- and upper-class homes on a full-time basis is common practice [31,32]. A similar practice is observed in Malaysia [33]. Unlike the pattern of domestic service in Brazil, in some countries female immigrants are employed as full-time maids [34]. Despite the difficulty in drawing comparisons with other countries, the present findings can be considered in line with the positive association found in the Whitehall II Study between long working hours and obesity [19]. This view remits to the conception of time as a resource for good health [35], since people need time in which to practice physical activities, to visit a doctor, or to recover from work. Furthermore, it is possible that, when the individual chiefly responsible for cooking is the housemaid, then women from wealthier families who employ maids will be less involved with everyday cooking, possibly resulting in less exposure to food intake and more weight-watching.

Women with better education levels, those caring for a sick or handicapped person, and those who reported high job demands and a longer working week were those most likely to perceive time for their personal care and leisure as being insufficient, which could be interpreted as an expression of the excess burden they are subjected to in the two domains of work and family. The total number of hours worked by women, taking into consideration the time spent at work and the time spent in the domestic environment, has been reported to be a relevant factor when evaluating health outcomes in this group [36]. In other studies, researchers have also found significant gender differences with respect to the quantity and quality of free time, with women experiencing more disadvantages and greater restrictions and fragmentation of leisure time [1,7].

The approach used in the present study is innovative in Brazil and, even in the international literature few studies have opted to use body weight as the endpoint in analyses on time-based work-family conflict [37,38]. In the group of women, there was an expressive inverse association between education level and body weight, showing the important effect of schooling in reducing the prevalence of overweight and obesity in this group. In men, higher education played a significant role in decreasing the incidence of obesity; however, the frequency of overweight increased. The results of the present study are in line with previous findings from studies conducted in Brazil, including data from a large survey [39] in which education was found to be directly associated with overweight/obesity in men and inversely in women. These results may be explained in part by the arguments that better-educated women tend to value and pursue the ideal of thinness [16], even if it is a consequence of socio-cultural pressure [40], since education, social class and income are factors that are usually positively associated. Considering this relationship, higher socioeconomic status influences the need to work longer hours [14] but, on the other hand, also increases an individual’s ability to access health-promoting resources such as physical exercise programs [16,40], to live in a safe enough neighborhood to be able to exercise outside, and to have access to a greater variety of healthy foods etc.

Relatively few studies have been conducted on factors related to the family in association with overweight and obesity in adults, except those more closely associated with lifestyle such as marital status and the presence of children [11,41]. Our results confirm the findings of other studies on work-family interplay, describing how factors related to the family are more relevant for women than for men [12,36,41,42]. For the women, factors such as not having the support of a full-time maid, having to care for a sick or disabled person and having children were associated with a greater likelihood of being overweight. Having children was a factor associated with overweight and obesity in men, corroborating findings from international studies [35,43].

Although there have been reports in the international literature of an association between job-related factors (low control at work, greater demands and a greater effort-reward imbalance) and overweight/obesity [18], this was not confirmed in the present study. Only the length of the working week was found to be relevant in modifying the effect of the principal association in women. The composition of the cohort may have affected these results, since in academic life time constraints are a result of having to synchronize different demands, particularly in the group of faculty members [4].

The magnitude of the prevalence of overweight and obesity in men and women in this study merits particular attention, as it highlights the increase in these conditions compared to the findings of previous studies conducted in Brazil, reinforcing the importance of understanding the mechanisms that contribute to this occurrence [39,43,44].

The main association found here reinforces the relevant debate on the use of time and gender equity in modern times; however, due to the inherent limitations of a cross-sectional study, it is impossible to establish the temporal sequence of events with any certainty. It should be emphasized that the hypotheses tested here may be confirmed in future analyses conducted using the longitudinal data from the ELSA-Brazil study and other longitudinal studies. Notwithstanding, the ELSA-Brasil cohort is composed of civil servants who volunteered to participate in the study, which indicates the very selective nature of this study sample and, as such, any broad characterization of these results should, in principle, be interpreted with caution.

Conclusions

These results suggest that the length of the working week influences factors underlying weight gain, possibly issues linked to behavior and/or stress mechanisms. It appears difficult for women not to become involved in domestic duties and, when such demands are combined with long working hours, the likelihood of gaining weight appears to be greater. The fact that this association was restricted to women suggests that the present findings are derived from gender inequalities involving relationships between time and health. The results of this study provide data on which to base public policies aimed at encouraging the redistribution of domestic responsibilities in the direction of gender equity, as well as macrosocial policies such as providing public schools for workers’ children.

Supporting information

S1 Table. Perception of insufficient time for personal care and leisure, according to sex.

https://doi.org/10.1371/journal.pone.0194190.s001

(DOCX)

S2 Table. Distribution of body mass index (kg/m2), according to age groups and sex.

https://doi.org/10.1371/journal.pone.0194190.s002

(DOCX)

References

  1. 1. Chatzitheochari S, Arber S. Class, gender and time poverty: a time-use analysis of British workers’ free time resources. Br J Sociol. 2012; 63: 451–471. pmid:22950463
  2. 2. Kolodinsky JM, Goldstein AB. Time use and food pattern influences on obesity. Obesity. 2011; 19: 2327–2335. pmid:21617636
  3. 3. Mi-Hye C. Gender, leisure and time constraint: employed men and women’s experience. Development and Society. 2006; 35: 83–105.
  4. 4. Ylijoki O-H, Mäntylä H. Conflicting time perspectives in academic work. Time Society. 2003; 12: 55–78.
  5. 5. Adam B. Timewatch: the social analysis of time. Cambridge: Polity Press; 1995.
  6. 6. Kaufman-Scarborough C. Time use and the impact of technology: examining workspaces in the home. Time Society. 2006; 15: 57–80.
  7. 7. Bittman M, Wajcman J. The rush hour: the character of leisure time and gender equity. Social Forces. 2000; 79: 165–189.
  8. 8. Greenhaus JH, Beutell NJ. Sources of conflict between work and family roles. The Academy of Management Review. 1985; 10: 76–88.
  9. 9. Edwards JR, Rothbard NP. Mechanisms linking work and family: clarifying the relationship between work and family constructs. The Academy of Management Review. 2000; 25: 178–199.
  10. 10. Carlson DS, Kacmar KM, Williams LJ. Construction and initial validation of a multidimensional measure of work–family conflict. Journal of Vocational Behavior. 2000; 56: 249–276.
  11. 11. Bianchi SM, Milkie MA. Work and family research in the first decade of the 21st century. Journal of Marriage and Family. 2010; 72: 705–725.
  12. 12. Berntsson L, Lundberg U, Krantz G. Gender differences in work–home interplay and symptom perception among Swedish white-collar employees. J Epidemiol Community Health. 2006; 60: 1070–1076. pmid:17108304
  13. 13. Matias M, Fontaine AM. [Work and family balance: the psychological mechanism of spillover]. Psic: Teor e Pesq. 2012; 28: 235–244.
  14. 14. Courtemanche C. Longer hours and larger waistlines? the relationship between work hours and obesity. SSRN. 2009. Available: http://ssrn.com/abstract=975099 or http://dx.doi.org/10.2139/ssrn.975099. Accessed 7 Feb 2016.
  15. 15. Wang Y, Beydoun MA. The obesity epidemic in the United States—gender, age, socioeconomic, racial/ethnic, and geographic characteristics: a systematic review and meta-regression analysis. Epidemiol Rev. 2007; 29: 6–28. pmid:17510091
  16. 16. McLaren L. Socioeconomics status and obesity. Epidemiol Rev. 2007; 29: 29–48. pmid:17478442
  17. 17. Sund ER, Jones A, Midthjell K. Individual, family, and area predictors of BMI and BMI change in an adult Norwegian population: findings from the HUNT study. Soc Sci Med. 2010; 70: 1194–1202. pmid:20163904
  18. 18. Kouvonen A, Kivimäki M, Cox SJ, Cox T, Vahtera J. Relationship between work stress and body mass index among
45,810 female and male employees. Psychosom Med. 2005; 67: 577–583. pmid:16046370
  19. 19. Lallukka T, Lahelma E, Rahkonen O, Roos E, Laaksonen E, Martikainen P, et al. Associations of job strain and working overtime with adverse health behaviors and obesity: evidence from the Whitehall II Study, Helsinki Health Study, and the Japanese Civil Servants Study. Soc Sci Med. 2008; 66: 1681–1698. pmid:18261833
  20. 20. Moore CJ, Cunningham SA. Social position, psychological stress, and obesity: a systematic review. J Acad Nutr Diet. 2012; 112: 518–526. pmid:22709702
  21. 21. Strazdins L, Griffin AL, Broom DH, Banwell C, Korda R, Dixon J, Paolucci F. Glover J. Time scarcity: another health inequality? Environment and Planning 2011; 43:545–559.
  22. 22. Aquino EM, Barreto SM, Bensenor IM, Carvalho MS, Chor D, Duncan BB, et al. Brazilian Longitudinal Study of Adult Health (ELSA-Brasil): objectives and design. Am J Epidemiol. 2012; 175: 315–324. pmid:22234482
  23. 23. Pinto KA, Menezes GM de S, Griep RH, Lima KTR dos S, Almeida M da C, Aquino EML. Work-family conflict and time use: psychometric assessment of an instrument in ELSA-Brazil. Rev Saúde Pública. 2016; 50:39.
  24. 24. Schmidt MI, Griep RH, Passos VM, Luft VC, Goulart AC, Menezes GM, et al. Strategies and development of quality assurance and control in the ELSA-Brasil. Rev Saúde Pública. 2013; 47(Suppl. 2): 105–112.
  25. 25. World Health Organization. Obesity and overweight. Fact sheet N. 311. Updated January 2015. Available: http://www.who.int/mediacentre/factsheets/fs311/en/. Accessed 30 Jan 2016.
  26. 26. Alves MGM, Chor D, Eduardo Faerstein, Lopes CS, Werneck GL. Short version of the "job stress scale": a Portuguese-language adaptation. Rev Saúde Pública. 2004; 38: 164–171. pmid:15122370
  27. 27. Kleinbaum DG, Klein M. Logistic regression: a self-learning text. 3rd ed. New York: Springer-Verlag; 2010.
  28. 28. Zick CD, Stevens RB, Bryant WK. Time use choices and healthy body weight: a multivariate analysis of data from the American Time Use Survey. Int J Behav Nutr Phys Act. 2011; 8: 84. pmid:21810246
  29. 29. Jabs J, Devine CM. Time scarcity and food choices: an overview. Appetite. 2006; 47: 196–204. pmid:16698116
  30. 30. Banwell C, Hinde S, Dixon J, Sibthorpe B. Reflections on expert consensus: a case study of the social trends contributing to obesity. Eur J Public Health. 2005; 15: 564–568. pmid:16141305
  31. 31. Fry G, Kemper K. A subnational perspective for comparative research: education and development in northeast Brazil and northeast Thailand. Comparative Education. 1996; 33: 330–360.
  32. 32. Brites J. [Friendship and inequality: gender, generation and class between domestic workers and their employers]. Cad Pagu. 2007; 29: 91–109.
  33. 33. Talib J, Mohamad Z, Mamat M. Effects of parenting style on children development. World Journal of Social Sciences. 2011; 1: 14–35.
  34. 34. Glantz NM. Moving maids: dynamics of domestic service and development. Política y Cultura. 2005; 23: 83–102.
  35. 35. Strazdins L, Welsh J, Korda R, Broom D, Paolucci F. Not all hours are equal: could time be a social determinant of health? Sociology of Health & Illness. 2016; 38: 21–42.
  36. 36. Artazcoz L, Borrell C, Benach J, Cortès I, Rohlfs I. Women, family demands and health: the importance of employment status and socio-economic position. Soc Sci Med. 2004; 59: 263–274. pmid:15110418
  37. 37. Grzywacz JG, Arcury TA, Márin A, Carrillo L, Burke B, Coates ML, et al. Work-family conflict: experiences and health implications among immigrant latinos. J Appl Psychol. 2007; 92: 1119–1130. pmid:17638469
  38. 38. Allen TD, Armstrong J. Further examination of the link between work-family conflict and physical health: the role of health-related behaviors. American Behavioral Scientist. 2006; 49: 1204–1221.
  39. 39. Gigante DP, Barros FC, Post CL, Olinto MT. [Prevalence and risk factors of obesity in adults]. Rev Saúde Pública. 1997; 31: 236–246. pmid:9515260
  40. 40. Filozof C, Gonzalez C, Sereday M, Mazza C, Braguinsky J. Obesity prevalence and trends in Latin-American countries. Obes Rev. 2001; 2: 99–106. pmid:12119667
  41. 41. Sobal J, Rauschenbach B, Frongillo EA. Marital status changes and body weight changes: a US longitudinal analysis. Soc Sci Med. 2003; 56: 1543–1555. pmid:12614704
  42. 42. Griffin JM, Fuhrer R, Stansfeld SA, Marmot M. The importance of low control at work and home on depression and anxiety: do these effects vary by gender and social class? Soc Sci Med. 2002; 54: 783–798. pmid:11999493
  43. 43. Linhares RS, Horta BL, Gigante DP, Dias-da-Costa JS, Olinto MT. [Distribution of general and abdominal obesity in adults in a city in southern Brazil]. Cad Saude Publica. 2012; 28: 438–447. pmid:22415176
  44. 44. Alves JG, Falcão RW, Pinto RA, Correia JB. Obesity patterns women in a slum area in Brazil. J Health Popul Nutr. 2011; 29: 286–289. pmid:21766564