by Krista McLellan, RD
Why this article?
Why I chose it: Many studies on obesity and weight are designed to study the short-term weight loss of obese individuals. However, this study highlights various independent factors and their possible relationship to weight gain in non-obese individuals. In my opinion, obesity can sometimes occur suddenly due to traumatic or stressful life events. Alternatively, many individuals experience weight gain and obesity from early childhood and throughout adulthood. Weight gain is also associated with many chronic diseases. For this reason, it is beneficial to shift research from the obese population to focus on long-term behavior related to weight gain in the general population. With this type of study, the findings can be applied to the non-obese population as an important preventative measure to chronic disease.
Dariush Mozaffarian, M.D. or Dr. P.H., Toa Hao, M.P.H., Eric B. Rimm, Sc.D., Walter C. Willett, M.D. or Dr. P.H. and Frank B. Hu, M.D., Ph.D,. Changes in Diet and Lifestyle and Long-Term Weight Gain in Women in Men. N Engl J Med 2011; 364 2392-2404.
Type of Study: Data analysis on history, lifestyle, and health practices/behaviours from the three separate cohort studies and their independent correlations with weight gain.
Overview: Gaining an understanding of which foods and behaviours related to health are associated with weight gain (WG) is preliminary to understanding how to influence them. This study identifies key factors that aid and impede the maintenance of a healthy energy balance. Therefore, these findings are very beneficial to disease prevention.
Methods: Data analysis from three separate cohort studies: NHS, NHS2, HPFS 120,877 adult men and women health professionals, non-obese and free from chronic disease. Follow ups were from 1986-2006, 1991-2003, and 1986-2006 at 4 year intervals. Cohort and sex specific results were similar and pooled with the use of an inverse-variance weighted meta-analysis.
Key findings: Participants gained an average of 3.35lbs in each 4 year increment. WG was most strongly associated with increased consumption of potato chips (1.69lbs), potatoes (1.28lbs) including French fries, sugar sweetened beverages (1lb), unprocessed red meats (.95lbs), and processed meats (.93lbs). Weight gain was inversely associated with intake of veggies (-.22lb), whole grains (.-.37lbs), fruits (-.49lbs, nuts (-.57lbs) and yogurt (-0.82lbs).
Other lifestyle factors:
Physical activity (-1.76lbs); alcohol use (0.41lb per drink per day); smoking (new quitters: 5.17lbs, former smokers: 0.14lb); sleep (weight gain <6 hrs and >8hrs); and TV watching (0.31hrs per day).
Bottom line: Independent dietary and lifestyle factors have been identified as being associated with long-term weight gain and could also have a substantial aggregate effect. This information could be useful in developing strategies to prevent obesity.
In terms of dietetic practice, this study reinforces what we already know to be sage dietary advice - a plant-based diet containing primarily whole grains, vegetables, fruits, nuts and fermented dairy has been proven to be most protective against weight gain. Also, lifestyle factors such as adequate exercise and sleep are also strongly related to less weight gain and the maintenance of a healthy body weight.
Strengths/limitations: This interpretation of the cohort data could be more meaningful if portion sizes were adjusted for energy. A great example of this was noted in an online comment on the study: one 4-6 oz portion of unprocessed red meat can vary widely in caloric content, information that if available may allow for greater interpretation as to why it had such a high correlation with weight gain over the 4 year increments. The more obvious limitation of the study is what foods were more frequently consumed past the energy balance tipping point. Different lifestyle factors were also measured independently, limiting our ability to better understand the combined effects of different factors.
Access the article by choosing the PDF file below or clicking this link: http://www.nejm.org/doi/full/10.1056/Nejmoa1014296