Social Costs of Chronic Heart and Lung Disease

Twenty percent of all people in the United States suffer from some form of chronic heart or circulatory disease. Eight percent suffer from either chronic bronchitis, emphysema, or asthma. In numerous ways, these diseases impose costs on the persons who have them and on society. Costs include the pain and discomfort associated with chronic illness, the resources used to provide medical treatment for the disease, and time lost because of illness. The most obvious form of lost time is years lost due to premature death. However, lost time also takes the form of lost workdays and lost leisure time while a person is alive. An additional loss occurs if a person becomes less productive as a result of disease.

Society has limited resources to devote to combating chronic heart and lung disease. To help decide how to get the best value for money, those whose job it is to apportion these resources must have some idea of the monetary costs of the diseases. Suppose the government can fund a program that will discourage smoking and, as a result, cut the number of new cases of emphysema each year by 10,000. Alternatively, the same funds could be used to educate people about dietary fat which, let us suppose, would reduce by 15,000 the number of heart attacks each year. To decide which of the two investments would provide better value for money—all other things being equal—one must determine the costs of each disease and then hold these costs against the benefits of making the investment and cutting back the number of cases that occur. In monetary terms, the benefits of each prevention program equal the number of cases of disease prevented, times the cost per case.

Researchers at Resources for the Future recently set out to measure some of the social costs of chronic heart and lung disease—specifically the medical and labor-market costs. These cost estimates are based on two national surveys—the National Medical Care Expenditure Survey and the Social Security Survey of Disabled and Non-Disabled Adults. Medical costs include the costs of medication, doctors' visits, and hospitalization. Labor-market costs include the lost earnings of people who stop working because of their disease and the reduced earnings of people who continue to work but cut back their hours or switch jobs.

The major finding in the study, not surprisingly, is that the costs of chronic illness vary greatly from one disease to another. Emphysema and ischemic heart disease (heart attack) have the largest combined medical and labor-market costs. Hypertension, on the other hand, has no labor-market effects and average annual medical expenses of only $200 per case (in 1977 dollars). If efficiency is used as the determining factor for the allocation of funds across disease prevention programs, this information could have important implications. It should be remembered, however, that social costs such as pain and discomfort were not part of the RFF study and that these and other factors carry considerable weight in attitudes toward disease prevention and investment decisions.

Medical costs

It would seem a simple matter to compute the medical costs associated with a disease. All that one need do is to locate people with the disease and record their medical expenses over some period. In reality, it is not at all simple. Until recently, surveys providing such data on specific diseases such as emphysema, heart disease, and hypertension either were too specialized, covered too few people to be generalized to the entire U.S. population, or were unavailable.

Even where such information is at hand, there are many complications. For instance, a person may have more than one disease, making it difficult to attribute a medical expense incurred from a doctor's visit to a specific disease. Also, if people have medical insurance, they do not pay for much of their medical expenses. Thus, asking them how much they themselves have paid will not reveal the lion's share of the expenses.

Fortunately, the 1977-78 National Medical Care Expenditure Survey addressed many of these concerns. It issued a calendar diary to 14,000 households (40,320 persons) selected randomly from the U.S. population. The diary was intended as a tool to help participants accurately report their health care utilization and expenditures for approximately one year. By carefully eliciting a list of all of the diseases each person had and which of these diseases were involved in any episode of illness, the survey facilitated matching specific medical costs to specific diseases. And, by building in questions on the source of payment for each medical expense, the issue of who pays—individual families, insurance companies, or the government—could be addressed. Knowing who pays as well as how much is paid may be useful information in deciding on disease prevention investments.

The RFF study found that of the five diseases examined—hypertension, ischemic heart disease, other (nonspecific) heart disease, emphysema, and chronic bronchitis—medical expenses incurred from heart diseases dwarf those for some of the other diseases (see table 1). For instance, expenses resulting from heart attacks (at $1,256 per person per annum in 1977 dollars) are almost thirteen times higher than those for chronic bronchitis (at $97), but only twice as large as those for emphysema (at $633).

However, most people suffering from any of these five chronic diseases have very low medical expenses. Median expenses (where half the sample has higher expenses and half has lower) for heart attack, for instance, are only $74 per year, while those for chronic bronchitis are $23. In select cases, however, costs can be prohibitive: almost 4 percent of those who had a heart attack in 1977-78 incurred medical expenses exceeding $10,000.

What are the costs of services being used to combat these diseases? It comes as no surprise that hospitalization, if required, is the most expensive service when compared with doctors' visits and drugs during a typical episode. It is perhaps more surprising that average hospitalization expenses exceed expenses for doctors and drugs even when the large number of people having no hospital expenses is counted.

Who pays for these expenses? In general, families pay only a small share of total expenses out of their own pockets--23 percent of medical costs for hypertension and 34 percent for chronic bronchitis, the least costly diseases. They pay only 14 percent of the costs for emphysema and 10 percent for heart diseases.

For the more serious diseases, families pay an even lower share of the costs. It falls to insurance companies to pick up most of the tab for diseases such as heart attack (46 percent), reflecting the high proportion of expenses for the hospitalization component and the high degree of coverage afforded this type of expense by health insurance plans. The insurance share for emphysema is large (28 percent) for much the same reason.

A third type of funding source—the government—pays up to 44 percent of the total cost of some diseases. Medicare, available to all persons 65 and over, covers 41 percent of the costs of other (non-specific) heart diseases and between 17 percent and 35 percent of the costs of the other diseases studied. Medicaid, available only to low-income persons, funds between 7 and 17 percent of expenses for the five conditions studied.

Although a minority of total medical costs are paid for directly by patients and their families, family funding is proportionately the largest source of payment for a majority of patients. This is because most patients incur a number of small expenses and families bear a larger percentage of these expenses than of large expenses such as hospitalization. This situation may reflect deductibility clauses in insurance policies, the exclusion of drugs from coverage by some policies, or other factors. But irrespective of who pays initially, the costs of medical care are borne in the long run by society at large.

When chronic disease strikes, a person may stop working altogether or may continue working but work fewer hours or switch to a less demanding job. For each disease examined, the RFF study sought to determine how likely it is that a person with that disease would stop working. Researchers also looked at how much individual earnings were likely to drop if work was continued. To find this information, they used responses to the 1978 Social Security Survey of Disabled and Non-Disabled Adults, because the survey asked respondents to identify their chronic diseases and to state whether they work and, if so, what they earn.

RFF researchers used this survey as a basis for discerning differences between the working habits of healthy people and chronically ill people. They compared the frequency with which each group worked to measure how the likelihood of working is affected by chronic diseases and, therefore, the labor-market costs of these diseases. Of the five chronic heart and five chronic lung diseases studied, only six decrease the probability of working—emphysema, heart attack, and stroke being the most important, followed by chronic bronchitis, arteriosclerosis, and other heart diseases (see table 2).

The age of onset

One important issue addressed by the study is whether the labor-market effects of a disease depend on the age when the disease first began—the age of onset. It is often argued that the effects of a disease at any age are smaller if the age of onset is earlier. According to this argument, a man is more likely to be working at age 40 if he injured his back when he was 25 rather than 35, because he has had longer to adjust to the injury.

RFF findings appeared to contradict this argument, at least for the diseases studied. Researchers found that emphysema, arteriosclerosis, and heart attack reduce the probability that a man will continue to work if his disease begins between 45 and 54, but not if it begins later, between 55 and 65. They also found that having a stroke at any time after age 45 reduces the chance of working, but that the effects are greatest if the stroke occurs between ages 45 and 54.

One reason for the surprising findings may be that persons who first contract a chronic disease between 45 and 54 have had it longer than persons who get the disease after 55; the longer the duration, the more severe the disease may be. This condition is especially true for degenerative diseases. Indeed, when RFF researchers took into consideration the duration of disease, they found that contracting emphysema or arteriosclerosis between ages 45 and 54 affects the chance that someone works only if that person has had the disease at least six years. By contrast, if a heart attack occurs between 45 and 54, the chance that a person works is lowered only for the first five years following the attack.

Among people with chronic heart and lung disease who are still able to work, how big is the drop in earnings likely to be? Using the same 1978 survey, RFF researchers found that only two diseases—emphysema and heart attack—lower earnings for people who continue to work. In the case of emphysema, the fall in earnings does not begin until a person has had the disease at least six years. Once it does fall, however, the drop is large—equal to 62 percent of earnings. Having a heart attack reduces earnings an average of 45 percent for people who continue working after an attack.

The combined effects of chronic heart and lung disease on the probability of working and on earnings if a person continues to work are substantial (table 2). Emphysema, heart attack, and stroke cause the largest annual earnings losses; however, the time pattern of losses is very different for the three diseases. Losses due to a heart attack are largest in the five years following the attack (equal to 55 percent of earnings, on average) and then decline. Losses associated with emphysema do not begin until a person has had the disease at least six years and then, up to age 55, equal 73 percent of earnings. By contrast, asthma, allergies, and hypertension appear to have no effects on earnings.

What are the implications of the RFF findings for programs to reduce the incidence of heart and lung disease (bearing in mind that only a portion of the social costs of these diseases was measured)?

First, the monetary benefits of programs to reduce the incidence of chronic disease depend on the number of cases prevented by the program as well as on the cost per case.

Second, it does not necessarily follow that a program to reduce the incidence of emphysema—for example—will yield higher benefits per dollar spent than a program to reduce hypertension, even though the former disease has far higher social costs per case.

Because the implications of the RFF findings are not straightforward, they do not suggest what decision must be made in choosing one prevention program over another. But making rational decisions about such programs can be aided by better pinpointing the medical costs and labor costs that would be saved if any one program versus another is put into place.