Jump to comment:

• Erroneous assumptions invalidate the calculated effects of reducing ultra-processed food consumption on child and adolescent obesity
  Kevin D Hall
  Published on: 30 December 2021

• Published on: 30 December 2021

  Erroneous assumptions invalidate the calculated effects of reducing ultra-processed food consumption on child and adolescent obesity

  • Kevin D Hall, Senior Investigator National Institutes of Health

  The manuscript by Livingston et al. presents a simulation study to estimate changes in prevalence of child and adolescent overweight and obesity in the USA if ultra-processed foods were eliminated. The main assumptions of the study can be traced to two publications from my research group at the National Institutes of Health (1, 2). Unfortunately, the assumptions used by the authors are very likely to be invalid.

  The first assumption regards calculating the changes in energy intake that would result from eliminating ultra-processed food. The data used to estimate energy intake changes derive from our study where we admitted 20 adult volunteers to the metabolic ward at the NIH Clinical Center and randomized them to receive either a diet with ~88% of energy from ultra-processed foods (NOVA category 4) or a diet with 0% ultra-processed foods and ~81% unprocessed foods (NOVA category 1) for 2 weeks followed immediately by the alternate diet for an additional 2 weeks (1). Mean ad libitum energy intake was observed to be ~500 kcal/d greater during the ultra-processed diet period. Livingston et al. used these data and made the following assumptions: a) our results in a small number of adults participating in an inpatient controlled feeding study could be quantitatively extrapolated to a large population of free-living children and adolescents; b) energy intake differences are linearly related to the percentage of energy from ultra-processed foods; and c) the calculated ene...

  Show More

  The manuscript by Livingston et al. presents a simulation study to estimate changes in prevalence of child and adolescent overweight and obesity in the USA if ultra-processed foods were eliminated. The main assumptions of the study can be traced to two publications from my research group at the National Institutes of Health (1, 2). Unfortunately, the assumptions used by the authors are very likely to be invalid.

  The first assumption regards calculating the changes in energy intake that would result from eliminating ultra-processed food. The data used to estimate energy intake changes derive from our study where we admitted 20 adult volunteers to the metabolic ward at the NIH Clinical Center and randomized them to receive either a diet with ~88% of energy from ultra-processed foods (NOVA category 4) or a diet with 0% ultra-processed foods and ~81% unprocessed foods (NOVA category 1) for 2 weeks followed immediately by the alternate diet for an additional 2 weeks (1). Mean ad libitum energy intake was observed to be ~500 kcal/d greater during the ultra-processed diet period. Livingston et al. used these data and made the following assumptions: a) our results in a small number of adults participating in an inpatient controlled feeding study could be quantitatively extrapolated to a large population of free-living children and adolescents; b) energy intake differences are linearly related to the percentage of energy from ultra-processed foods; and c) the calculated energy intake differences would persist indefinitely. None of these assumptions are likely to be true and therefore the estimated change in energy intake that would result from elimination of ultra-processed foods in free-living children and adolescents is likely to be in error.

  The second assumption of Livingston et al. regards their calculation of the expected weight changes resulting from reductions in energy intake. The authors cited a mathematical model developed by my research group to simulate energy balance and body composition dynamics during childhood growth (2). Rather than use the published dynamic mathematical model to simulate expected weight losses for a given reduction in energy intake, Livingston et al. used an equation that represents the approximate excess energy intake at a given age per kg of established excess weight. The authors misrepresented this approximate equation as a “weight reduction model” and thereby miscalculated the expected changes in weight that would occur for a given reduction in energy intake. Thus, the weight changes that would result from the assumed (but likely erroneous) reduction in energy intake are invalid.

  References

  1. Hall KD, Ayuketah A, Brychta R, et al. Ultra-Processed Diets Cause Excess Calorie Intake and Weight Gain: An Inpatient Randomized Controlled Trial of Ad Libitum Food Intake. Cell Metab 2019;30(1):67-77.e3. doi: 10.1016/j.cmet.2019.05.008 [published Online First: 2019/05/21]

  2. Hall KD, Butte NF, Swinburn BA, et al. Dynamics of childhood growth and obesity: development and validation of a quantitative mathematical model. Lancet Diabetes Endocrinol 2013;1:97-105.

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top