Figure 2 reports the share of people with a hospitalization for any of these conditions. Twenty-two percent of people were admitted to a hospital with some cardiovascular disease. Ischemic heart disease is the most common admission, accounting for 8 percent of admissions. Stroke and heart failure are also common, as are other circulatory diseases. Admissions for peripheral vascular disease and hypertension are much less common.
A person who had a hospital admission for cardiovascular disease and is disabled may or may not have been disabled because of that condition. The NLTCS does not reliably determine the precise condition that leads to each disability (Landrum and Stewart 2006). We make two alternative assumptions about the probability of being disabled by cardiovascular disease. The first assumption, a less restrictive assumption, labels someone as disabled from cardiovascular disease if they were admitted to a hospital with cardiovascular disease in the previous five years. The more restrictive assumption subsets this group to those for whom the most disabling condition was cardiovascular disease3 – where the list of conditions by disability status is reported in Table 1. Thus, a person who had a stroke and hip fracture would be termed disabled because of cardiovascular disease by the first measure, but not by the second measure. Fortunately, our results are very similar regardless of the definition used
Figure 3 shows the probability of being disabled by cardiovascular disease. Using the less restrictive measure, the decline in disability is 1.4 percentage points, or 22 percent of the 6.3 percentage point total reduction in disability. Using the more restrictive measure, the decline is 0.9 percentage points, or 14 percent of the total decline. In each case, cardiovascular disease is a substantial share of the total decline.
As noted above, the conclusion that cardiovascular disease is a substantial share of disability decline contradicts an earlier literature that suggested that ‘marginal survivors’ contribute to an increase in disability. If more people survive strokes, the argument went, the share of the elderly with disabilities would rise. The finding of a reduction in disability suggests, in contrast, two other hypotheses: either fewer people are suffering cardiovascular disease events; or those who have always survived such events are less disabled now than they were formerly, i.e. the incidence of disability among cardiovascular disease patients is falling. These possible effects are demonstrated in equation :