GSBSE Faculty publishes on the “Jolly Fat Hypothesis”
Previous cross-sectional research with older adults in the 1970’s led to the conclusion that being overweight as measured by body mass index (BMI) was related to lower levels of depressive symptoms. This phenomenon came to be known as the “jolly fat hypothesis.” The hypothesis was that higher BMI might be protective against symptoms of depression, possibly by reducing risk of frailty.
However, the studies supporting the jolly fat hypotheses have been cross sectional (performed at a single point in time) and thus it has been impossible to answer several important questions: Does BMI predict symptoms of depression; do symptoms of depression predict BMI; or are both pathways predictive over time?
Members of the Maine-Syracuse Longitudinal Study research team, led by psychology graduate student Peter Dearborn, employed the MSLS database and a longitudinal design to answer the questions regarding directionality of the jolly fat phenomenon. Some studies since the 1970s have shown that obesity may be related to higher, not lower, symptoms of depression. However, these results were obtained particularly among younger adults and the relationship between BMI and symptoms of depression among older adults remained less clear. The MSLS is uniquely situated to address this issue. This study of community-dwelling older (n = 638, ages ≥ 50y) adults in the Central New York area was conducted as part of the larger 35-year Maine-Syracuse Longitudinal Study.
Peter J Dearborn, Michael A Robbins, and Merrill F (Pete) Elias examined the bidirectional relationships between symptoms of depression and BMI over a period of five years using objective measures of BMI and physical functioning. They examined both pathways simultaneously because there is research indicating that pathways in both directions may be predictive. In the study, we adjusted for baseline BMI and depressive symptoms as well as demographic variables (age, sex, education, marital status), social variables (social activity, social isolation, recent life events), and physical health variables (chronic health conditions, ability to complete 3 timed performance tasks).
The MSLS investigators found that, although participants underestimated their weight and overestimated their height (producing very different obesity rates 29.2% vs. 38.8%), self-reported BMI and objectively measured BMI related quite similarly to symptoms of depression. Specifically, higher BMI was related to increases in depressive symptoms over a 5-year period. Interestingly, we did not find the reverse relationship of high levels of depressive symptoms predicting increases in BMI.
As in many previous studies, women reported higher levels of depressive symptoms compared with men, but the risks associated with BMI functioned similarly among both sexes.
The results of our study indicate that, as is true for younger adults, older adults with higher BMI are at a risk for developing depressive symptoms over time. Being overweight does not protect against symptoms of depression in elder persons. It is possible that increased weight-stigmatization over the last 20 years has led to older adults “catching up” with younger adults with respect to BMI-associated risk for depressive symptoms. This may explain why our study found the reverse of predictions based on the jolly fat hypothesis. Interestingly additional findings from the analysis indicate that increasing physical function through maintenance of physical activity may be a protective factor in the development of depressive symptoms and that this is true independent of any relationship with weight loss or weigh status.
Link to full article: http://digitalcommons.library.umaine.edu/cgi/viewcontent.cgi?article=1015&context=longitudinal_papers
Peter Dearborn is a senior graduate student in the Department of Psychology, Merrill F. (Pete) Elias, and Michael A. Robbins are faculty members in the Department and cooperating faculty in the Graduate School of Biomedical Science and Engineering.