Our study innovates by considering the joint characteristics of mothers and fathers to longitudinally examine their potential impact on later childhood adiposity. To our knowledge, there are no previous studies investigating clusters of parental sociobehavioral factors and how these factors are related to later adiposity outcomes in offspring. In this particular sample, we identified three distinct maternal groups based on selected sociobehavioral characteristics measured at study entry, of which two groups differed in terms of offspring adiposity measures. Among fathers, we identified four paternal clusters, but no associations with adiposity in offspring were observed. Our results showed that children from mothers with obesity who are smokers and who have lower educational attainment have higher adiposity measures than children from mothers who are overweight, nonsmokers and who have higher educational attainment. This was true for all adiposity measures of the offspring examined in this study, namely, the BMI z score, android/gynoid fat ratio (a measure of visceral adiposity), fat mass index, and body fat percentage, which strengthens the findings, as they are consistent across outcomes.
Among all three maternal clusters, the characteristics that distinguished them and made them unique were BMI, smoking status and educational attainment. Although PAs differed across clusters, they were not significantly different. Although educational attainment may be complicated to modify, weight loss in mothers was previously associated with lower offspring BMI over time (24). Studies on the heredity of obesity have shown that 50–80% of the variability in body weight measured using BMI could be attributed to genetic factors (37, 38). The risk for obesity in a child increases according to the weight status of the parents (39), and it is even greater when both parents are living with obesity (39). However, across the life course, the risks associated with having parents with obesity may be offset by favorable environments (40). In the QUALITY cohort, in which all participants had at least one biological parent living with obesity, our results suggested that maternal weight status plays a more important role in offspring adiposity than paternal weight status.
Surprisingly, we did not find any associations with childhood adiposity within the father’s clusters. The four paternal clusters were differentiated mainly by age, smoking status, and education attainment, where the older group was composed of fathers who had a higher educational attainment and smoked. Although we did not observe associations between paternal clusters and offspring adiposity measures, a positive association with offspring adiposity has been reported when paternal weight status is examined as a single factor (41), as well as a negative association when paternal education is examined individually (42). A large Chinese study with 4513 children reported that only paternal smoking before conception was associated with childhood adiposity (43), and our data included only paternal smoking status postconception, more specifically, when children were aged 8–10 years. This might be one of the reasons why we did not find associations with our paternal sociobehavioral clusters. A number of other possible reasons could explain this finding in our population. This might be because paternal BMI was comparable and equally elevated among all four clusters or because maternal intrauterine exposure plays a stronger role in offspring adiposity. Alternately, it could be that gender roles in the household at data collection provided mothers with greater responsibilities related to diet and other lifestyle behaviors (e.g., cooking, buying groceries). When investigating paternal direct effects, a significant confounding factor often stems from the strong correlation with maternal exposures, attributable to shared environments and assortative mating tendencies (44).
It is widely shown in the literature that parental education is inversely associated with childhood obesity (13, 16) and that a higher level of parental education is a protective factor against obesity in children (16, 17, 23). Parental education is an indicator of socioeconomic status (SES) (23). Gray et al. showed a significant relationship between children’s weight status and parental income, where children of low-income parents were three times more likely to be at risk of developing or being overweight (13). A study performed in the US showed that both lower income and mothers’ lower education attainment (high school vs college degree or higher) were significant predictors of obesity in third and fifth grades (14). Our results show that the cluster of mothers with higher educational attainment had children with lower adiposity, despite the positive parental history of obesity. However, in our sample, financial stress or parental occupational status did not differ across maternal clusters, suggesting that, in our sample, education superseded financial status with respect to its impact on offspring adiposity.
Parental smoking, more specifically maternal smoking, is a sociobehavioural factor associated with overweight and obesity, and it is associated with an increased risk of a child having overweight or obesity by 40% (45, 46). An Israeli study reported that parental smoking was an independent risk factor for both overweight and obesity (47), and other studies also reported that parental smoking was associated with a higher BMI in offspring (20, 22, 48). We observed that the cluster with nonsmoking mothers was associated with lower adiposity in offspring. The majority (63.3%) of mothers who reported smoking when participants were 8–10 years old were already smoking during pregnancy (data not shown), where there is a growing body of evidence indicating that there is a link between maternal smoking status preconception and prenatal and offspring adiposity (46, 49). It may also be that mothers who are smokers engage in other suboptimal lifestyle behaviors (e.g., unhealthy diet, increased screen time), which could be 'transmitted' to the offspring and result in greater adiposity.
It has been shown that, when considered an independent factor, an increase in general parental physical activity decreases the odds of adolescents having obesity (19, 24). In the present study, mothers from the cluster with the highest adiposity in offspring were more physically active, meeting the current Canadian 24-Hour Movement Guidelines for Adults (50). These conflicting findings likely result from the fact that parental PA information was self-reported in our study, with self-reported PA being less reliable than objectively measured PA (51). Although maternal age (21) and parental alcohol consumption (24) have been reported to be associated with adiposity in offspring, they were not discriminatory variables across maternal clusters in the present study.
The clustering approach is the main strength of our study. Previous studies have reported unique associations between parental characteristics and childhood adiposity, but no studies have examined joint characteristics. We innovated by modeling the effects of combined exposures to several parental risk factors, which is closer to real-life clinical situations where multiple sociobehavioral characteristics cooccur among individuals. Other strengths are the multiple objective and reproducible measures of adiposity in offspring, such as dual energy X-ray absorptiometry (DXA) (52), and the consistency of findings across outcomes. Among the study limitations, our sample was limited to a very specific population of White children with at least one biological parent with obesity, who are also known to have higher SES than the general Quebec population. The overall small sample size resulted in small sample sizes for specific clusters. We also acknowledge that in our sample, at least one parent had obesity, potentially contributing to overestimating the influence of parental BMI. In addition to parental BMI, maternal and paternal characteristics were self-reported, which may have induced information bias (e.g., underestimation of smoking and overestimation of physical activity).