previously (Freedman and Martin, 1999). Educational gains are expected to continue, although at a slower rate. On the other hand, the recent trend of increases in disability at younger ages, although small and starting from a very low level, may have negative implications for the future elderly population (Lakdawalla et al., 2004). Some studies suggest that the gains in mortality from reductions in smoking and better control of blood pressure might be reversed in the coming years by high rates of obesity (Cutler et al., 2007; Olshansky et al., 2005). Another study found that baby boomers on the verge of retirement are in poorer health than pre-retirees 12 years ago (Soldo et al., 2006). Trends in illness and disability will influence the need for health services among the future older adult population, though the direction and magnitude of their effects are not entirely clear. Still, even if disability rates among older adults continue to decline, the size of the future older adult population is so large that, overall, the total need for services can be expected to increase (Johnson et al., 2007).

Many efforts to project the future incidence or prevalence of disease assume that the health status of individuals in a given age-sex category will remain constant, and, therefore, the projections depend only on changes in the age and sex composition of the population (Goldman et al., 2004). This assumption may prove incorrect in the future. Nonetheless, for many health conditions this type of projection offers the best available estimates. Examples of such projections include the following:

• The proportion of older adults with self-reported, doctor-diagnosed arthritis will rise from 34 percent in 2005 to 48 percent in 2050 (Fontaine et al., 2007).

• The prevalence of diabetes among older adults will rise from 5 million in 2005 to 10.6 million in 2025 and to 16.8 million in 2050 (Boyle et al., 2001).

• 7.7 million people will have Alzheimer’s disease in 2030, up from 4.9 million in 2007 (Alzheimer’s Association, 2007).

Assuming no change in current prevalence rates for disability, 26 million of the 75 million older adults alive in 2040 will have limitations in at least one IADL, 16 million will have at least one ADL limitation, and 3 million will be institutionalized (Waidmann and Liu, 2000).

Changes in the health marketplace will likely influence the demand for services as well. A number of medical advances and technologies may