Letter by Mullie et al Regarding Article, “Estimated Global, Regional, and National Disease Burdens Related to Sugar-Sweetened Beverage Consumption in 2010”
To the Editor:
In their study on mortality, diabetes mellitus, and cancer, Singh et al1 concluded that worldwide >180 000 deaths could be attributed to sugar-sweetened beverage (SSB) consumption. To calculate the attributable deaths, data were derived from prospective cohort studies that examined associations between SSB intake and risks of premature death. Such an approach focusing on a single exposure was successful in the past, when behaviors like smoking had an impact on cancer occurrence and cardiovascular diseases. However, this reductionist approach has strong limitations in nutritional epidemiology. Drinking SSBs is not a behavior that can be isolated like smoking, but rather a behavior that is part of a more general lifestyle. Several studies have related excessive SSB consumption to greater fast food attendance, more frequent smoking, physical inactivity, and lower socioeconomic status,2–4 all factors known to be associated with premature death. Hence, obesogenic and diabetogenic factors are not independent from each other. If excessive SSB consumption is just 1 component of an unhealthy behavior or a marker of deprivation, it will be extremely difficult to disentangle the health effects specific to SSB intake from health effects specific to deprivation and other unhealthy behaviors. In observational studies, in particular, trying to single out the specific contribution of 1 component to the obesity epidemic is elusive because unhealthy behaviors are highly correlated. We thus contend that the clustering of unhealthy behaviors in subgroups of a population has probably more responsibility for the observed epidemic of adiposity and diabetes mellitus than the quality and quantity of the individual dietary components found in that population.
Because unhealthy behaviors are inextricably intertwined, it is impossible to determine the exact relationship between 1 of these unhealthy behaviors and premature death. As a consequence, studies examining 1 particular unhealthy behavior are likely to wrongly attribute an indeterminable number of premature deaths to that unhealthy behavior. In this respect, we believe that the putative causal relationship between SSBs and premature death advanced by Singh et al1 is biased and highly exaggerated.
A reductionist view that all is about SSBs should be replaced by pragmatic concepts embracing the multiplicity of components involved in the occurrence of adiposity and metabolic disorders. The best way to combat adiposity issues is to tackle all unhealthy cluster components at the same time, and, thus, interventions for controlling the adiposity epidemic must be multifactorial and part of comprehensive programs helping people to quit the obesogenic behavior cluster they are trapped in.
Patrick Mullie, PhD
Philippe Autier, MD
Peter Boyle, PhD
International Prevention Research Institute
- © 2016 American Heart Association, Inc.
- Singh GM,
- Micha R,
- Khatibzadeh S,
- Lim S,
- Ezzati M,
- Mozaffarian D