Association between maternal and paternal employment and their children’s weight status and unhealthy behaviours: does it matter who the working parent is?

An escalating global epidemic of overweight and obesity—“globesity”- is taking over many parts of the world [1], representing serious challenges for public health [2] due to its association with several chronic diseases [3]. Obesity during childhood is a major health concern, given its growing prevalence and long-term health consequences [4]. In 2016, over 340 million children and adolescents aged 5–19 were affected by overweight or obesity worldwide [5]. In Europe, although the prevalence may have stabilized [6], the differences between countries are especially striking, with a prevalence that ranges from more than 40% in southern Europe to less than 10% in northern Europe [7]. In Spain, despite the traditional Mediterranean diet, 28.6% and 10.3% of children between 2 and 17 years old are categorised as overweight and obese, respectively [8]. The prevalence of obesity among children shows a marked social gradient in Spain; 15.4% of children in the lowest social class are affected by obesity compared to 5.4% in the highest social class [9].

The direct cause of obesity is an energy imbalance between calories consumed and calories expended. This simple explanation is related to deep and complex social and technological changes which indirectly affect both diet and physical activity (at school, at work, at home and in leisure time) [10‐14]. One of the social changes that is commonly referred to as a potential indirect cause of the increased overweight and obesity in children is the higher proportion of working women [15]. This implies major changes in family decisions about the division of time between paid work, domestic work and leisure, where “domestic work” includes both the time devoted to preparing meals and the time spent on the education and care of the children, among other tasks.

However, there are competing theoretical arguments about the direction of the relationship between parental employment and childhood overweight or obesity. Theoretically, parents who have a job would have less time to prepare meals with fresh food and supervise the quality and frequency of their children’s food intake. Skipping breakfast, eating fast food, and a quicker eating pace are all positively associated with childhood obesity [16]. Also, less parental supervision may discourage their children’s physical activity in early ages, leading to higher body mass index (BMI) [17]. However, having a job is also associated with higher income and educational levels, which are in turn linked to lower BMI.

In fact, the existing literature has found conflicting results in this area. There is certainly a body of studies that have identified a positive and significant relationship between maternal employment and the probability of children to be affected by overweight or obesity [17‐24]. However, in other studies this result is tempered by the characteristics of the job and the number of hours worked, showing a lower effect or even a non-statistically significant relationship when these factors are controlled for [15, 25‐29]. In addition, the increased involvement of fathers in raising children has been linked with a decreased likelihood that their children would become obese [30], which might offset any potential detrimental effect of the increase in maternal paid working hours. In fact, other studies have found that socioeconomic characteristics are more relevant determinants of children’s weight status [28, 31‐37]. Moreover, the school environment has also been found to play a significant role in tackling obesity and promoting healthy lifestyles among children and adolescents [38‐41]. Schools might create environments in which children eat healthfully and engage regularly in physical activity, leading to physical, emotional, and social benefits [39, 42]. They can also raise awareness and understanding of health risks (e.g. bullying and smoking), increase self-esteem and resistance to social pressure and promote healthy social relationships [43]. Disadvantaged and low academic performing schools are doubly burdened with additional risks, such as higher obesity risk [38]. There has been an emphasis on “whole school” interventions on health, which go beyond individual-focused education, and involve changes to schools’ overall organization, teaching, discipline, school health services, policies, culture, and extra-curricular activities [44, 45].

Our objective was to contribute to the understanding of the complex relationship between children’s excess weight and parental employment. We investigated whether overweight/obesity among children and adolescents in Spain is connected with parental employment, considering separately the situations in which a) only one parent (mother or father) has a job, b) both parents have a job, c) neither of the parents has a job. This approach allows us to explore more deeply the relationships between socioeconomic circumstances and parental employment, and to account for the roles of both fathers and mothers as carers of their children. Our analyses also aimed to determine the extent to which the school’s environment is associated with a higher weight. We also examined whether several unhealthy behaviours among children may be associated with parental employment status, taking into account the socioeconomic context of the family.

Our results indicate that family’s socioeconomic characteristics are determining factors of obesity and unhealthy habits among children. In most cases, these factors overweighted the detrimental effect that the potential lower parental supervision expected in households with working parents might have on children’s weight and obesogenic behaviours. However, in some cases, the labour participation of mothers was found to be linked to higher children’s obesity and unhealthy habits when mothers were the only providers of the households, suggesting that the availability of a non-working father did not offset the impact of maternal employment. To our knowledge, this is the first study that examines the underlying factors that affect children’s weight status, differentiating by parental work status and including the potential effect of the school as an additional level of analysis.

Our results confirm that, in line with previous studies, family’s educational level [31, 33, 37] and incomes are highly associated with childhood obesity/overweight [28, 36, 60], as well as the school environment [38‐41]. In Spain, an immigrant background of the parents has also been associated with obesity among children [35].

The findings that maternal employment is not associated with children’s weight status or unhealthy behaviours when both parents are employed but, for at least some subgroups, working mothers are associated with a higher prevalence of child’s obesity and unhealthy habits when they are the only providers of the household is striking. This may reinforce the conclusions of other studies claiming that mothers, even when they are working outside the home, assume greater responsibility for child care and home management than fathers do [21, 61]. In addition, a closer look at the characteristics of biparental households with only a working mother indicates that these families are likely to be categorized as having limited financial resources, suggesting that unobserved socioeconomic factors might also play a role in the observed association.

The detrimental association of a non-working father on children’s weight and lifestyles, while observing that having a non-working mother is not associated with poorer outcomes also deserves further discussion. This finding might suggest that having a mother as main caregiver may have a greater positive influence on children’s weight and lifestyles than having a father as main caregiver. However, our results should be interpreted in the light of the socioeconomic context at the time the data was collected. Both in 2010 and in 2014 the difference in the level of unemployment between men and women was the smallest in the history of Spain. This was due to the more severe impact of the economic crisis on sectors that employ more men than women. As a result, the unemployment rate among men increased from 6.4% in 2007 to 25.6% in 2013 [62]. This suggests that non-working fathers in our data might in some cases represent newly unemployed men affected by the past recession. In fact, in our data children reported that 59% of non-working fathers were looking for a job, while this percentage was 29% among non-working mothers (however, around 10% of children did not know the reason why their father/mother did not have a job). The potential benefits of the expected increased involvement of fathers in their children’s raising might be fewer in such circumstances.

Our study has certain limitations that should be mentioned. First, given the nature of the data used, it is difficult to establish a causal relationship between parents’ work status and children’s BMI, as parental labour force is likely to be correlated with a number of factors that are also related to children’s weight (e.g. educational background, parent’s weight status, etc.). Second, our analyses are based on the information contained in the HBSC cross-sectional surveys, which albeit of being the richest source of data on children’s health in Spain, the information provided is limited (i.e. there was no information on the number of working hours) and self-reported. Therefore, our estimates might be affected by measurement errors and by omitted variables bias, making causal inferences problematic. Regarding measurement errors, children and adolescents completed the questionnaires on their own using an online platform at their school centres with support from school teachers. However, the self-reported nature of the data might lead to some discrepancies with reality regarding height, weight, family structure and parents’ work status and educational level, as well as regarding the elements needed to estimate family affluence, dietary behaviour, leisure-time habits, etc., especially among younger children. A previous study analysed the validity of self-reported height and weight data within the framework of the HBSC study conducted in Estonia and found a small and age-decreasing underestimation of overweight/obesity prevalences when compared with measured anthropometric data [63]. In Spain, obesity and overweight prevalences estimated by the Spanish National Health Survey in 2011/12 showed very similar values for the age groups included in the HBSC [64]. With respect to the role of omitted variables, the factors included as control variables in our models might only be a subset of all possible factors related to employment and child’s outcomes. For instance, parents’ own health and weight status might affect their ability to have a job and be correlated with their children’s obesity outcomes. Therefore, we might expect that some degree of endogeneity is still affecting our estimates.

Finally, the reasons why fathers do not have a job seem to be different from the reasons for non-working mothers. This, together with our finding indicating that employed mothers might be linked to higher children’s obesity and unhealthy habits when they are the only working parent in the household, emphasises A more balanced sharing of child’s care between men and women might weaken the documented link between maternal employment and children’s unhealthy behaviours and weight. Our results suggest the the need for a new social contract in which the burden of caregiving is equally shared, so that the participation of either parent in the employment market does not involve a risk for their children’s health. Conditions of employment must also favour the conciliation between family and professional life. According to the European Job Quality Index, which is “a measure that encompasses a broad range of work and employment characteristics, including wages, non-wage aspects of employment and work organisation, and collective interest representation” [65], Spain has the third worst position in this ranking. The socioeconomic and educational gradient which we observed in children’s obesity and unhealthy behaviours, as well as the effect of the school environment, also point towards the need to act at the root of the observed inequalities in children’s weight status. On this matter, the evidence shows that complex obesity prevention interventions acting on multiple targets, settings, and risk factors appear both to be more effective and to lower health inequalities than individual actions [66, 67].

In conclusion, our findings indicate that once we control for family wealth and education, the observed association between a working mother and her children’s risk of obesity disappears, except for younger girls living in households where the mother had a job and the father did not. Unhealthy behaviours were also observed in households where the mother was the only employed parent. This study highlights the need for more complete research into the way in which obesity develops in young people, and into the roles that working and non-working parents have in their children’s development.

Further studies should be undertaken, including the use of time, to analyse the relationship between family members, and also to discern whether the results may change over time or working regimes. It would also be desirable to explore a larger number of potential causes of obesity, to assess the importance of reverse causality between education, income, and obesity, and to determine the nature of childcare among biparental families. Further research is also needed to widen the analysis to include families other than biparental ones, although the formers are more heterogeneous, and so more information about additional caregiving would be needed. As we get collectively closer to understanding the development of children’s excess weight, we may be better able to craft appropriate public health interventions to help reduce its current prevalence.