Examining the Relationships between Excess Body Weight, Health and Disability


U.S. Department of Health and Human Services

Examining the Relationships between Excess Body Weight, Health and Disability

Executive Summary

Timothy Waidmann, The Urban Institute
Vicki Freedman, University of Medicine and Dentistry of New Jersey
Christine Himes, Syracuse University
Saad Ahmad, The Urban Institute

May 2008

This report was prepared under contract #HHS-100-03-0011 between the U.S. Department of Health and Human Services (HHS), Office of Disability, Aging and Long-Term Care Policy (DALTCP) and the Urban Institute. For additional information about this subject, you can visit the DALTCP home page at http://aspe.hhs.gov/_/office_specific/daltcp.cfm or contact the ASPE Project Officer, Hakan Aykan, at HHS/ASPE/DALTCP, Room 424E, H.H. Humphrey Building, 200 Independence Avenue, S.W., Washington, D.C. 20201. His e-mail addresses are: Hakan.Aykan@hhs.gov.

Submitted to the Department of Health and Human Services’ Office of the Assistant Secretary for Planning and Evaluation in fulfillment of Task 6 of task order no. 11 under contract HHS-100-03-0011. The authors would like to thank the members of the Technical Advisory Group and Brenda Spillman for valuable contributions to the direction and content of this report. Any errors or omissions are the authors’ alone. The views reflected in this report are those of the authors and do not reflect those of their institutions or funding agency.

The recent rise in the prevalence of overweight and obesity in the United States has raised alarm over present and future impacts on public programs. Of most immediate concern is the effect on increased use of medical services to treat diseases related to excess body weight and fat. However, the apparent link between disability and obesity raises longer-run concerns about the use of long-term care services.

Given the rising rates of obesity at middle ages, it is possible that the declines we have observed in rates of late-life disability will not continue, and long-term care costs could increase. How much obesity will influence late-life disability rates, and the modifiable mechanisms through which obesity influences disability, are key questions. Yet, past projections of the effects of obesity trends on future disability trends have been based on a cross-sectional association between obesity and disability, and the causal pathways between these markers are not well understood. We approach these concerns by attempting to answer the following three research questions.

  • What is the nature of the cross-sectional relationship between obesity and disability, as well as other health outcomes among the elderly?

  • What is the nature of the dynamic relationship among obesity, disability, and other health outcomes in the elderly?

  • If it is found that obesity is a significant independent determinant of health and disability status, what are the intervening mechanisms?

The most alarming findings in the literature focus on the relationship between obesity and work disability rather than on the types of disability (ADL and IADL) that have been declining in recent years and that are most relevant for aging and long-term care policy. In addition, most studies are based on cross-sectional analyses. The purpose of this study is to focus on measures of disability more relevant for elderly individuals and to examine whether cross-sectional associations arise because of actual increased risk of disability onset for those with overweight and obesity, or whether they are simply correlated outcomes without a causal link between them.

The data for this study come from the 1998-2004 waves of the HRS. To answer Question 1, we conducted cross-sectional analyses using pooled data from all of these waves. To address Question 2, we did several analyses using weight and health status in 1998 as predictors of changes in functional status between 1998 and 2004. We also examined changes in one set of measures between 1998 and 2000 as potential predictors of changes in other measures between 2000 and 2004. Finally, to explore potential clinical pathways, we conducted several analyses of the effects of excess weight interacted with specific profiles of chronic disease to determine which disease pathways were most relevant for the development of functional limitation and disability.

Relevant to the first research question, the cross-sectional analyses produced results consistent with other findings in the literature, namely of a strong relationship between obesity and functional limitations as well as ADL disability. The relationship between excess weight and lower body limitations is especially striking with significant limitation risks beginning at values of body mass index (BMI) below the obesity threshold. The correlation with IADL disability was much weaker, and shows that moderate excess weight is actually associated with lower disability prevalence. This finding suggests that as the cohorts with high rates of obesity age, we might expect an increase in the physical limitations that contribute to disability, but that the increase in the level of dependency requiring long-term care will be less dramatic. Further, since the overall trend in disability is most directly a result of downward trends in IADL disability, and IADL disability only appears to increase at extreme levels of obesity, the effect of trends in overweight and obesity on overall disability trends is likely to be modest.

The analyses framed by the second research question were less conclusive. Longitudinal analyses that use baseline weight as a predictor of disability onset generally confirm that among persons who are not already disabled, those who are obese are at greater risk of developing disabilities. Among those reporting disabilities at baseline, the probability of recovery of functioning is lower for those who are obese.

More stringent tests based on longitudinal analyses are less clear, however. We find no relationship between a change in weight and subsequent disability onset or recovery. While this may imply that weight loss interventions among the elderly and near-elderly would be ineffective in reducing future disability, a longer observation period should be studied before such a conclusion is reached. Second, the analyses we conducted to understand the relative timing of disability and excess weight did not produce a clear answer to the question of whether excess weight causes disability or vice versa.

Similarly, in analyzing the third research question, we are unable to find a significantly different pattern of disability onset associated with a particular disease pathway. While clinical evidence suggests differences in the effects of overweight by disease, observational studies like ours may require a longer follow-up period, or more reliable methods for measuring disease than are available in survey data to identify these differences.

Preliminary 2006 data were recently released, and next year the 2008 data from the HRS will be available, providing a ten year follow-up period. The longer time period will provide both a more meaningful estimate of longer-run implications of past obesity trends, but also allow a more complicated set of disability onset models using up to five repeated observations on which hazard models might be more reliably estimated. Any future improvements in the measurement of obesity (such as using actual measurement of height and weight instead of reliance on respondents' self-reports, and using measures of obesity other than BMI) can also lead to improvements in examining the effects of obesity on disability as well as other health outcomes.

The Full Report is also available from the DALTCP website (http://aspe.hhs.gov/_/office_specific/daltcp.cfm) or directly at http://aspe.hhs.gov/daltcp/reports/2008/weight.htm.