Pediatric Original Article | Published:

Preventing and treating childhood obesity: time to target fathers

International Journal of Obesity volume 36, pages 1215 (2012) | Download Citation

Abstract

Objective:

To examine the long-term effects of having one overweight or obese parent on child weight status and determine whether these effects vary according to parent sex.

Design:

Prospective study: Longitudinal Study of Australian Children (LSAC).

Subjects:

Two-parent families (N=3285) from the LSAC were included if height and weight data were available for both parents and their child at the 2004 and 2008 time points.

Measurements:

Child weight status category (healthy, overweight, obese) in 2008 when the child was aged 8–9 years. Regression modelling was used to investigate how self-reported parent weight status in 2004 influenced measured child weight status 4 years later.

Results:

Parent body mass index (BMI) was significantly correlated with child BMI, but there was no evidence of sex-specific associations between parent and child BMI correlations. The results from the regression analysis showed that having an overweight or obese father, but a healthy weight mother, significantly increased the odds of child obesity (odds ratio: 4.18, 95% confidence interval (CI): 1.01–17.33 and odds ratio: 14.88, 95% CI: 2.61–84.77, respectively), but the reverse scenario (overweight or obese mother with a healthy weight father) was not a significant predictor of child overweight or obesity (odds ratio: 2.52, 95% CI: 0.38–16.71 and odds ratio: 2.56, 95% CI: 0.31–21.26, respectively).

Conclusions:

Children with overweight or obese fathers are at a higher risk of becoming obese. This suggests that interventions are urgently required to test the efficacy of treating overweight fathers as a key strategy for childhood obesity prevention and/or treatment.

Introduction

Internationally, childhood overweight and obesity have been identified as major public health concerns and have been consistently associated with poor physical and psychological health outcomes.1 In Australia, the prevalence of overweight and obesity in children aged 2–18 years is 21–25%.2 It has recently been argued that in order to curb childhood obesity it is ‘time to shrink a parent’3 thus highlighting the potential value of targeting weight management in overweight parents as a strategy to improve dietary and physical activity behaviours in their children.4, 5, 6, 7

A number of studies have demonstrated that having two overweight parents is associated with higher child weight status compared with having one overweight parent or two healthy weight parents.4, 5, 6 It has also been suggested that father weight status has a stronger relationship with son weight status and similarly mother weight status with daughter weight status.7 However, this effect was not supported in a recent study utilising a large (N=4654) nationally representative sample of UK children aged 7.5 years.8 In addition, it remains unclear whether it matters which parent is targeted for childhood obesity prevention.9

Therefore, in order to refine child obesity prevention and treatment, it is necessary to establish whether, in families where only one parent is overweight or obese, it makes a difference to a child's weight status if it is their mother or their father who is the overweight or obese parent. The aim of this study was to examine within the Australian Longitudinal Study of Children (LSAC), the prospective long-term impact of maternal and paternal weight status on child weight status, when only one member of the parenting couple is overweight or obese.

Materials and methods

We used data from LSAC (http://www.aifs.gov.au/growingup/) to investigate prospective associations between parent weight status as measured in 2004 and childhood weight status as measured in a follow-up assessment conducted in 2008. LSAC is a nationally representative sample of Australian children that is funded by the Australian Government Department of Families, Community Services and Indigenous Affairs.10 LSAC data are collected using a range of methods including face-to-face interviews with the primary parent (usually the study child's mother), self-report questionnaires from both parents, interviewer observation, direct child assessment and teacher/childcare worker questionnaires.11 The study generates extensive information about the children, their families and environments, and includes information on both children and parents' weight status over a 4-year period.10 Families were included in this analysis if height and weight information were available for the study child and their two biological parents at both time points (N=3285).

Weight status

Children's height and weight were measured in both 2004 and 2008 when they were aged 4–5 and 8–9 years old, respectively. Of the 4983 families in the full 2004 LSAC sample, only 37 (0.01%) mothers refused to allow their child's weight to be measured. However, in 15 of these 37 families, mothers still self-reported their own weight, suggesting that the impact of non-response bias is likely to be limited. Measurements were assessed by trained researchers using standard digital scales (Code 79985, Salter; Springvale, VIC, Australia) and a portable rigid stadiometer (Code IPO955, Invicta, Leicester, UK). The International Obesity Task Force age- and sex-specific criteria for body mass index (BMI (kg/m2)) were used to classify children as being healthy weight, overweight or obese.12

Parent self-reports of their height and weight were collected in both 2004 and 2008. Instead of collecting height and weight information during the face-to-face interview, each parent self-reported these data as part of a leave-behind questionnaire. This method of data collection has been shown to help improve the validity of self-reported height and weight measurements.13 Maternal and paternal BMI were calculated and each parent was assigned to one of the three weight status categories (healthy weight (BMI<25), overweight (BMI 25–<30) and obese (BMI30)). Although fewer fathers than mothers reported their height and weight information, there was no evidence of bias in the sample as there was a similar level of non-response for parents of boys and girls across each child weight category.

Statistical analysis

Correlations between maternal, paternal and child BMI were conducted to determine whether there was evidence of sex-specific associations between parent and child weight status across the 2004 and 2008 time points. A weighted multinomial regression model (weighted by 2008 population weights as specified in the LSAC data set) was fitted to the data using 2008 child weight status as the outcome variable and 2004 maternal and paternal weight status as the predictor. We controlled for a number of parent and child covariates, as used in previous LSAC analyses14 including child sex (male or female), English as main language spoken (yes or no), number of siblings at 4-year follow-up (none, 1, 2, or 3 or more) and child weight status as measured in 2004 (healthy, overweight, obese). Parent covariates included their education level at follow-up (less than 12 years, 12 years (that is, completed high school), or more than 12 years) and their weight status in 2008 (healthy, overweight, obese). The Socio-Economic Indexes for Areas (SEIFA) disadvantage score was also included. Socio-Economic Indexes for Areas scores give an indication of the social and economic conditions at the postcode of residence level in Australia.15 Scores are standardised to have a mean of 1000 (s.d.=100), with high scores indicating greater advantage.

Results

Table 1 reports the demographic characteristics of parents and their children. The prevalence of overweight and obesity in our sample closely matched the national prevalence data for both adults (2004)16 and children (2008).17 The majority of families (85%) spoke English as their main language. There was a similar distribution of boys and girls (51.4% boys). Most children had one (46%) or two (33%) siblings, few had three or more siblings (16%), and a minority had no siblings (5%). The mean Socio-Economic Indexes for Areas score in this sample was 1017 (s.d.=61).

Table 1: Characteristics of parents as measured at baseline (2004) and children as measured 4 years later (2008)

Due to the conflicting evidence as to whether or not there are sex-specific associations between parent–child weight status, we first examined the correlations between average 2004 and 2008 parent and child BMI. All correlations were significant (P<0.001). There was no evidence of sex-specific associations in our sample. Both mothers' and fathers' BMI correlated slightly more strongly with daughter BMI (r=0.31 and r=0.29, respectively) than with son BMI (r=0.29 and r=0.25, respectively). As there was no evidence of sex-specific associations, we included child sex as a covariate in our further analysis but did not run separate models for girls and boys.

The results from the regression analysis are reported in Table 2. When compared with the reference condition (neither parent overweight), having two overweight parents was predictive of the development of child overweight. Furthermore, having two obese parents was a stronger predictor of child obesity than two overweight parents. Having an obese father and an overweight mother was predictive of child obesity, but the reverse scenario, an obese mother and an overweight father, was not.

Table 2: Adjusted odds ratios for the association between mother and father overweight and obesity as measured in 2004 and the development of child overweight and obesity as measured 4 years later

Large differences were found in cases where only one member of the parenting couple was overweight or obese. Having a healthy weight mother and an overweight father increased the odds of a child being obese 4 years later by 318% (95% confidence interval (CI): 1–1633%) and in cases where the father was obese these odds increased to 1388% (95% CI: 161–8377%). Having a healthy weight father and an overweight or obese mother was not a significant predictor of child obesity.

Discussion

Having an overweight or obese father, but a healthy weight mother, increased the odds of a child becoming obese, but the reverse scenario, an overweight or obese mother with a healthy weight father, was not a significant predictor of child obesity. These results suggest that fathers are a key influence in shaping the family environment that leads to the development of child obesity. Further, it suggests that interventions to test the efficacy of treating overweight fathers as an important and novel strategy to impact on weight status in childhood obesity should be a priority as fathers may be a strategic target for prevention and treatment of childhood obesity.

The results from this study provide evidence of the important role that fathers have in the development of children's weight status and this has consequences for their long-term health outcomes. When only one member of the parenting couple was in a higher weight status category, it was the father's and not the mother's weight status that was a significant predictor of later child overweight and obesity. This is particularly important in countries like Australia where substantially more men are overweight than women.17

Our study has a number of strengths. We analysed a large nationally representative sample using a longitudinal study design, which allows for strong conclusions to be drawn relating to the long-term relationships between mother and father weight status on the development of childhood overweight and obesity. Also, the relatively recent data collection (2004–2008) means that these results are current. Additionally, child measures of height and weight were measured objectively by trained researchers, eliminating the chance of parents under or overestimating their child's measurements. A limitation of our study, and numerous similar studies,5, 8, 18 is the use of self-reported parent height and weight measurements. However, it has been suggested that for this age range there is only a modest discrepancy between self-reported and actual weight.19 Also, limiting our sample to two-biological-parent families means that our findings may not generalise to single- or step-parent families. Additionally, this study was unable to take into account parent and child dietary intake and physical activity due to a lack of comprehensive measures of these factors being available in the LSAC data set.

Future research

The associations between father and child obesity found in this study should be confirmed in other populations. Intervention studies, including randomised controlled trials such as the Healthy Dads Healthy Kids programme20 that focused on weight loss in overweight fathers in order to improve both their own and their child's health behaviours are needed to test whether targeting father overweight and obesity is effective in preventing or treating childhood obesity. Furthermore, studies are needed to determine the mediating influences of these relationships. For example, it has previously been found that parents influence children's weight through modelling and control of diet and physical activity behaviours.21 However, we were unable to test these potential mechanisms in relation to fathers' influence using the existing data. Future investigations into the relative contributions of unhealthy parental lifestyle behaviours, using comprehensive measures of dietary and physical activity behaviours, should be conducted in order to examine the links between mother and father weight and lifestyle behaviours and children's weight gain.

Conclusion

The seriousness of the health consequences of childhood overweight and obesity make the accurate targeting of health interventions to reduce future obesity essential. Our findings suggest that interventions are urgently required to test the efficacy of treating overweight fathers as a key strategy for childhood obesity prevention and/or treatment.

References

  1. 1.

    , , . Obesity in children and young people: a crisis in public health. Obes Rev 2004; 5 (Suppl 1): 4–85.

  2. 2.

    , , , . Trends in the prevalence of childhood overweight and obesity in Australia between 1985 and 2008. Int J Obes 2010; 34: 57–66.

  3. 3.

    . Childhood obesity: time to shrink a parent (editorial). Int J Obes 2010; 34: 1–3.

  4. 4.

    , , , , , et al. Predictors of body mass index and associations with cardiovascular risk factors in Australian children: A prospective cohort study. Int J Obes 2005; 29: 15–23.

  5. 5.

    , , , , . Development of overweight in children in relation to parental weight and socioeconomic status. Obesity 2009; 17: 814–820.

  6. 6.

    , , , , . Predicting obesity in young adulthood from childhood and parental obesity. New Engl J Med 1997; 337: 869–873.

  7. 7.

    , , , , , . Assortative weight gain in mother-daughter and father-son pairs: an emerging source of childhood obesity. Longitudinal study of trios (EarlyBird 43). Int J Obes 2009; 33: 727–735.

  8. 8.

    , , . No evidence of large differences in mother–daughter and father–son body mass index concordance in a large UK birth cohort. Int J Obes 2010; 34: 1191–1192.

  9. 9.

    , , , . Family involvement in weight control, weight maintenance and weight-loss interventions: a systematic review of randomised trials. Int J Obes 2003; 27: 987–1005.

  10. 10.

    , , , , , et al. Introducing the Longitudinal Study of Australian Children. The Australian Institute of Family Studies: Melbourne, Australia, 2002.

  11. 11.

    , , . LSAC Technical Paper No. 1: Sample Design. Australian Institute of Family Studies: Melbourne, Australia, 2005.

  12. 12.

    , , , . Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 2000; 320: 1240–1243.

  13. 13.

    , , . Effects of age on validity of self-reported height, weight, and body mass index: findings from the Third National Health and Nutrition Examination Survey, 1988–1994. J Am Diet Assoc 2001; 101: 28–34.

  14. 14.

    , , , . Preschooler obesity and parenting styles of mothers and fathers: Australian National Population Study. Pediatrics 2007; 120: 1520–1527.

  15. 15.

    Australian Bureau of Statistics (ABS). Socio-Economic Indexes for Areas (SEIFA), vol. 1.1. ABS: Canberra, Australia, 2001.

  16. 16.

    Australian Bureau of Statistics (ABS). National Health Survey: Summary of Results. ABS: Canberra, Australia, 2006.

  17. 17.

    Australian Bureau of Statistics (ABS). National Health Survey: Summary of Results. ABS: Canberra, Australia, 2009.

  18. 18.

    , , , , . Impact of parental BMI on the manifestation of overweight 5–7 year old children. Eur J Nutr 2002; 41: 132–138.

  19. 19.

    , , , . Use of self-report to monitor overweight and obesity in populations: some issues for consideration. Aus NZ J Public Health 2000; 24: 96–99.

  20. 20.

    , , , , , et al. The ‘Healthy Dads, Healthy Kids’ randomized controlled trial: efficacy of a healthy lifestyle program for overweight fathers and their children. Int J Obes 2011; 35: 436–447.

  21. 21.

    . Childhood overweight and the relationship between parent behaviors parenting style, and family functioning. Ann Am Acad Political Soc Sci 2008; 615: 12–37.

Download references

Acknowledgements

This study was supported by a Strategic Initiatives research grant funded by the Faculty of Health, The University of Newcastle. This paper used data from the Longitudinal Study of Australian Children. The study was conducted in partnership between the Department of Families, Housing, Community Services and Indigenous Affairs (FaHCSIA), the Australian Institute of Family Studies (AIFS) and the Australian Bureau of Statistics (ABS). The findings reported in this paper are those of the authors and should not be attributed to FaHCSIA, AIFS or the ABS.

Author information

Affiliations

  1. Family Action Centre, University of Newcastle, Callaghan, New South Wales, Australia

    • E Freeman
    •  & R Fletcher
  2. School of Biomedical Health Sciences and Pharmacy, Faculty of Health, University of Newcastle, Callaghan, New South Wales, Australia

    • C E Collins
    • , T Burrows
    •  & R Callister
  3. School of Education, University of Newcastle, Callaghan, New South Wales, Australia

    • P J Morgan

Authors

  1. Search for E Freeman in:

  2. Search for R Fletcher in:

  3. Search for C E Collins in:

  4. Search for P J Morgan in:

  5. Search for T Burrows in:

  6. Search for R Callister in:

Competing interests

The authors declare no conflict of interest.

Corresponding author

Correspondence to E Freeman.

About this article

Publication history

Received

Revised

Accepted

Published

DOI

https://doi.org/10.1038/ijo.2011.198

Further reading