Self-Reported Energy Balance Data Cannot Be Trusted

Imagine your clients having to record their food consumption and energy expenditure as part of an important research study. The details might get a little sketchy from day to day, right?

It’s not that we intend to cheat or lie about how much we eat and exercise every day, but it seems to be part of the human condition both to underestimate the amounts and to misremember important details about them, say the authors of a paper published online in the International Journal of Obesity (December 23, 2014; doi: 10.1038/ijo.2014.199). The somewhat disconcerting part is that scientists have relied for years on analyzing such self-reported data to draw conclusions about how much we eat and exercise, as well as
to build public health policy with far-reaching implications. While there have been doubts in the scientific community through the years about the margin of error for self-reported data, the authors of this paper are calling for a serious reconfiguring of energy balance collection methods for research— methods that will infuse more objectivity into the process.

They suggest using technology—for example, digital photography, and chewing and swallowing monitors—to more visually and accurately measure energy consumed. They also discuss using technology to capture energy expenditure data.

The journal’s associate editors John Brameld and David Stensel included their own commentary on the piece: “Over many years, there has been a heavy reliance on self-reported energy intake and physical activity data in human observational and epidemiological studies relating to obesity. Whilst knowledge in the field of energy balance has certainly advanced as a result of these studies it has been clear for some time that self-report data can be highly inaccurate. Energy balance assessment is a crucial methodological issue for the field, since the potential inaccuracies of self-report data make the findings of many studies questionable at best and sometimes incorrect and misleading. This has potentially serious implications for health care policy and clinical decision making. The paper . . . clearly and succinctly examines this issue. [The authors] highlight that ‘bad data lead to bad conclusions.’They make a plea for resources to be directed at the development of objective, precise and accurate measures of energy balance. This call is long overdue and attention to this issue can only enhance the quality of evidence-based recommendations for the future.”