Aim Child health varies with body mass index (BMI), but it is unknown by what age or how much this attracts additional population health-care costs. We aimed to determine the (1) cross-sectional relationships between BMI and costs across the first decade of life and (2) in longitudinal analyses, whether costs increase with duration of underweight or obesity.
Methods Participants: Baby (n = 4230) and Kindergarten (n = 4543) cohorts in the nationally representative Longitudinal Study of Australian Children. Outcome: Medicare Benefits Scheme (including all general practitioner plus a large proportion of paediatrician visits) plus prescription medication costs to federal government from birth to sixth (Baby cohort) and fourth to tenth (Kindergarten cohort) birthdays. Predictor: biennial BMI measurements over the same period.
Results Among Australian children under 10 years of age, 5-6% were underweight, 11-18% overweight and 5-6% obese. Excess costs with low and high BMI became evident from age 4-5 years, with normal weight accruing the least, obesity the most, and underweight and overweight intermediate costs. Relative to overall between-child variation, these excess costs per child were very modest, with a maximum of 94 per year at age 4-5 years. Nonetheless, this projects to a substantial cost to government of approximately 13 million per annum for all Australian children aged less than 10 years.
Conclusions Substantial excess population costs provide further economic justification for promoting healthy body weight. However, obese children's low individual excess health-care costs mean that effective treatments are likely to increase short-term costs to the public health purse during childhood.

Additional Metadata
Keywords child, health services research, health-care cost, obesity, overweight, thinness
Persistent URL,
Journal Journal of Paediatrics and Child Health
Clifford, S.A, Gold, L, Mensah, F.K.F, Jansen, P.W, Lucas, N, Nicholson, J.M, & Wake, M. (2015). Health-care costs of underweight, overweight and obesity: Australian population-based study. Journal of Paediatrics and Child Health, 51(12), 1199–1206. doi:10.1111/jpc.12932