Moving in the same direction
Madam
We appreciate the opportunity to respond to the above letter regarding our findings on the downward trend in childhood obesity in the populations of Fleurbaix and Laventie in France(Reference Romon, Lommez, Tafflet, Basdevant, Oppert, Bresson, Ducimetière, Charles and Borys1).
Yes, the theoretical model underpinning the intervention study is not described. But the aim of our paper was not to describe an intervention study designed to reduce the prevalence of obesity. The two towns were the setting of three consecutive studies, none of which was designed to prevent childhood obesity. The first evaluated the effect of a school-based nutrition programme on family habits. This programme was designed to improve children’s knowledge of foods and food processing. This programme was elaborated in 1991, when in France childhood obesity was not seen as a challenge, and physical activity was not targeted at all during this period.
The next study was an observational study about the determinants of body fat mass changes and several papers were published on this topic(Reference de Lauzon-Guillain, Romon, Musher-Eizenman, Heude, Basdevant and Charles2–Reference Maillard, Charles, Lafay, Thibult, Vray, Borys, Basdevant, Eschwège and Romon7). As Katan(Reference Katan8) pointed out, it was not a randomised controlled trial based on a treatment theory, it was an observational study.
We had the opportunity to monitor a population exposed to a nutritional information programme for 12 years and during this period, while childhood obesity prevalence was increasing all over the world and also in France, particularly in northern France where the study was located(Reference Romon, Duhamel, Collinet and Weill9), we observed a progressive decline in childhood overweight prevalence from 2000 to 2004.
During these studies, families living in these towns were given regular questionnaires about food habits and physical activity that may have sensitised them and there was an increasing commitment of the community at large, including various stakeholders, about ‘healthy lifestyles’. In 2004, we measured children’s body fat mass both in Fleurbaix and Laventie and in two comparison towns in the same geographical environment which, while not strictly ‘control’ towns, had similar sociodemographic characteristics and population sizes.
The confounding of several intervention components, and the relatively weak study design, do indeed not allow for the evaluation of independent effects or the interaction between intervention components. However we think that this observation needed to be published, if only because it indicates the importance of having all local stakeholders move in the same direction.
