Evid Based Nurs 13:110-111 doi:10.1136/ebn1097
  • Prognosis
  • Prospective cohort study

Activity levels inversely associated with weight gain in women only when BMI is less than 25

  1. Manfred Stommel
  1. Michigan State University, East Lansing, Michigan, USA
  1. Correspondence to Manfred Stommel
    College of Nursing, W-149 Owen Graduate Center, Michigan State University, East Lansing, MI 48824, USA; manfred.stommel{at}

Commentary on: [CrossRef][Medline][Web of Science]Google Scholar

Recent revisions in the 2008 (USA) Physical Activity Guidelines were prompted by evidence that the health benefits (both in terms of morbidity and mortality) of moderate physical activity (at least 150 min per week) are substantial1 2 The authors of this paper do not directly address the problem of health and survival benefits associated with physical activity, but rather take an indirect route: they examine to what extent moderate amounts of exercise are enough to prevent weight gain over time, taking the adverse impact of weight gain on morbidity and mortality for granted.

The study follows 34 079 middle-aged women, whose self-reported physical activity, weight and height measures were obtained at baseline and five follow-up interviews spaced 2–3 years apart. The main outcome variables are individual weight changes occurring in any of the five 2- or 3-year intervals, and the odds of gaining at least 2.3 kg in weight during the same intervals. The main predictor variable is a time-dependent, three-category weekly energy expenditure measure, equivalent to <150, 150<420, ≥420 min of moderate-intensity physical activity, which was constructed from responses to questions about a variety of physical activities. The analysis relies mainly on pooled time series regressions, including both linear and logistic models, which control for both time-dependent and time-independent confounders like age, race and time intervals between assessments.

The results show an overall average weight gain of 2.6 kg over the 13-year observation period. However, weight gain during the five intervals was greater among women with the lowest (0.12 kg per interval) and medium physical activity levels (0.11 kg per interval) compared to women with the highest activity level. Furthermore, such differences in weight gain by physical activity levels were observed only for women under the age of 65 and among normal weight women (body mass index (BMI) <25). Similar results obtained with the alternative outcome variable: the odds of gaining at least 2.3 kg of weight over any 2–3-year interval.

The main conclusions are as follows: (1) Once a person is overweight or obese, greater physical activity does not retard weight gain; (2) a high level of physical activity (≥420 min of moderate-intensity physical activity per week) is required to prevent weight gain among normal weight (BMI<25) women; (3) the 2008 Physical Activity Guidelines, with their recommendation of 150 min per week of moderate physical activity, are ‘insufficient for weight gain prevention absent caloric restriction’.

The evidence presented in this study calls into question all three conclusions: (1) Table 2 in the article shows estimates for weight gain among physically inactive obese women that are larger (0.16 kg) than for all physically inactive women (0.12 kg), but while the latter are statistically significant, the former are not. A power analysis for the subgroup of obese women is not provided, yet there are likely to be very few women in the reference group of highly active and obese women. (2) Figure 2 in the study shows that the magnitude of weight gains is about the same for all three physical activity groups as measured at baseline; the regression coefficients in table 2 only show the differences in average weight gains among the three physical activity groups. With an average weight gain of 2.6 kg across all five time intervals (=13 years), the weight gain per interval is about 0.52. Thus, even the women in the most active group (≥420 min of moderate-intensity physical activity) gained weight over the 13-year period. (3) The same point applies a fortiori to the overall claim that the 2008 Physical Activity Guidelines, with their recommendation of 150 min per week of moderate physical activity, are ‘insufficient for weight gain prevention absent caloric restriction’. Based on the data shown, this statement is true, but it neglects to emphasise that even among very active women, physical activity alone is insufficient for weight gain prevention absent caloric restriction!

The study confirms that physical activity alone does not suffice to maintain weight, in particular among women in the age group of 55 to 65. This is not a new finding. Based on the evidence in this study, the recommendation to female patients should be that, in addition to physical activity, any weight containment among middle-aged women needs to address a reduction in calorie intake as well.

The authors of the study take it for granted that moderate weight gain (2.6 kg over a 13-year span) during the fifth or sixth decade of a woman's life leads to greater health hazards. Yet, studies of the morbidity and mortality hazards of obesity show that, holding age and other relevant variables constant, the risks of adverse health outcomes increase with higher BMI. By contrast, evidence on how individual weight gain over time affects morbidity and mortality remains scarce. In particular, the current paper shows the greatest weight gain among normal weight (BMI<25) inactive women, but we don't have evidence that changes in BMI from 20 to 22, for example, are hazardous to health. In addition, we lack evidence on whether it matters at what point in a person's life cycle weight gain occurs. Is gaining a few pounds between the ages of 45 and 65 as damaging as between the ages of 15 and 35? In the absence of such evidence, the clinically prudent thing is to emphasise what we know: moderate exercise reduces many health risks, regardless of whether or not it alone is sufficient to maintain weight.


  • Competing interests None.


Free Sample

This recent issue is free to all users to allow everyone the opportunity to see the full scope and typical content of EBN.
View free sample issue >>

EBN Journal Chat

The EBN Journal Chat offers readers the opportunity to participate in discussion about research articles and commentaries from Evidence Based Nursing (EBN).

How to participate >>

Don't forget to sign up for content alerts so you keep up to date with all the articles as they are published.

Navigate This Article