Definitions of childhood obesity: current practice

Abstract

Objective: To review current practice in the definition of childhood obesity based on body mass index (BMI), in order to understand why the recommendations of the International Task Force on Obesity (IOTF) have not been fully adopted. Design: Literature search using MEDLINE to identify papers on childhood obesity published in the first 4 months of 2005, and a cited reference search on the recommendations of the IOTF. Results: Citations of the IOTF definition have increased since publication, but less than half of papers on childhood obesity published in the period used the definition. Most used the 95th centile of a national distribution to define obesity. Conclusions: Reasons for using centiles of a national distribution included the need for z-scores or centiles to define underweight or extreme obesity, not available in conjunction with the IOTF definition, inclusion of children under 2 years, and concerns about over-or underestimation of the prevalence of obesity. None of these preclude use of the principle underlying the IOTF definition, that of continuity with the adult definition of BMI of 30 kg/m 2 or more. Adoption of either the IOTF definition or corresponding centiles of a national distribution would enable comparisons of prevalence between countries and over time.

Prevalence of overweight and obesity in Irish school children, using four different definitions

“…Also, as mentioned earlier the IOTF method is further limited in the fact that it has no criteria for identifying children who are underweight and may be at risk of growth faltering (Chinn, 2006). Additionally, other researchers reported a lower sensitivity and specificity when using the IOTF cutoffs for defining overweight and obesity.…”

“…The reference curves for the UK 1990 also used in this study were Prevalence of overweight and obesity in Irish school children JL O'Neill et al derived from 11 UK surveys for which data were collected between 1978 and 1990 (Cole et al, 1995). It is important to note that when using centile cutoffs, independent of age and sex, the prevalence of overweight and obesity is always going to be related to the percentage of the reference population situated above the chosen centile (Power et al, 1997;Bellizzi and Dietz, 1999;Chinn, 2006). Therefore, it is recommended that growth reference charts be fixed at a certain time point to enable changes in prevalence over time be monitored (Power et al, 1997;Bellizzi and Dietz, 1999).…”

“…Recently, Chinn (2006) reviewed reasons why the IOTF recommendations have not been universally enforced. A literature search revealed that the main reasons were use of BMI z-scores, no cutoffs for identifying underweight, IOTF cutoffs under/overestimating the prevalence of obesity and no cutoffs available for under 2 years.…”

“…The problem of comparing with previous studies that had not used the IOTF cutoffs was also highlighted. The author recommended that if using national definitions the IOTF cutoffs should also be applied to enable comparisons be made (Chinn, 2006).…”

Methodological limitations in measuring childhood and adolescent obesity and overweight in epidemiological studies: does overweight fare better than obesity?

“…A number of distinct definitions based on centiles of BMI have been used. Accepted definitions are desirable so that trends over time can be monitored, and comparisons can be made between populations 6 . Moreover, established standards would be of use in clinical practice as well as in epidemiological settings.…”