Impact of weight loss on lifetime medical spending among obese Medicare beneficiaries

Obesity and obesity-related conditions, such as diabetes and cardiovascular disease, are major drivers of U.S. health care spending in general, and Medicare spending specifically. In the absence of obesity, the total annual Medicare budget ($408 billion in 2009) could be reduced by an estimated 8.5%. We examined the impact of hypothetical one year weight loss of at least 15%, 10% or 5% on Medicare savings among newly-enrolled (age 65) beneficiaries with BMI ≥35 (Class II/III Obesity). Our model (developed by Yang and Hall, 2008) incorporates the dynamic relationships among weight, chronic disease, acute medical events, functional status, death, health services use, and associated health care costs. Model parameters were obtained from the 1992-2001 Medicare Current Beneficiary Survey (MCBS) and 2001 Medicare claims data; 2001 reimbursement rates were adjusted to 2009 dollars. Population-based and per-capita lifetime savings, respectively, were defined as total savings conferred by weight loss in the first year of Medicare enrollment among all targeted beneficiaries over their remaining lifetimes and for one patient over that beneficiary's remaining lifetime.

A 15%, 10% or 5% weight loss in the first year of Medicare enrollment (age 65) among all beneficiaries with BMI ≥35 would confer an estimated lifetime Medicare savings of $60 billion, $26 billion, or $4 billion, respectively. A 15%, 10%, or 5% weight loss in the first year of Medicare enrollment (age 65) for one beneficiary with BMI ≥35 would confer per capita lifetime Medicare savings of $32,543, $14,293, or $2,137, respectively.

Because model parameters were obtained from 1999-2001 survey and claims data, savings are likely underestimated. For example, whereas only 5% of MCBS participants reported BMI levels ≥35, current estimated prevalence of Class II/III Obesity among similarly-aged adults exceeds 12.5%. In addition, because our model is predominantly based on data derived from a cohort born between 1920 and 1930, findings may not extend to the current cohort of Medicare enrollees ("Baby Boomers"). Moreover, health services costs do not include Part D or newer (and more expensive) treatments. Finally, our model assumes that chronic conditions are enduring and that associated costs for these conditions correspondingly persist.

Despite the potential for underestimation, this model highlights the substantial savings that Medicare may realize as a result of weight loss among obese, newly-enrolled beneficiaries. Findings provide support for increased access to safe and effective weight loss treatments.