Incorporating Cost Criteria in Health Policymaking:
The Case of Vaccination

By Mary R. Anderlik
Health Law & Policy Institute

Cost has always mattered in health care. The uninsured have received less care. Providers have migrated to high-dollar specialties and services with generous reimbursement rates. The buck has stopped outside the door of expert panels developing guidelines for clinicians. Now that may be changing.

In February this year, the Advisory Committee on Immunization Practices (ACIP) of the Centers for Disease Control and Prevention scaled back an earlier recommendation concerning the new vaccine Prevenar. Prevenar targets the bacterium streptococcus pneumoniae, the leading cause of pneumonia, meningitis, and ear infections. Initially, ACIP recommended that all children under 5 years old be vaccinated; this was revised to all children under age 2 plus high-risk children in the 2 to 5 age group. A four-dose series of the vaccine costs $232. According to the New York Times, this was only the second time ACIP debated potential economic effects in arriving at a recommendation.

Several forms of economic analysis are employed in health care. Cost-benefit analysis is especially controversial because it involves the monetarization of costs and benefits. In cost-effectiveness analysis, two or more paths to the same benefit are compared. The comparison takes place in monetary terms, but the benefit itself does not need to be assigned a monetary value. Some of the complexities and perplexities of economic analysis are highlighted in a study of the pneumococcal conjugate vaccine published in the March 15 issue of JAMA.

In constructing their model, the researchers had to make numerous judgment calls, e.g., whether to account for the possibility that the vaccine would prove ineffective against some variants of the disease-causing bacterium (they did). In general, the researchers favored conservative assumptions about the benefits of vaccination, reflected in estimates of cost-savings. Significantly, their calculations of cost per life-year saved do not assign any value to the reductions in morbidity (physical distress short of death) or psychological distress that would accompany projected reductions in cases of pneumonia, meningitis, and ear infections following vaccination. Another critical aspect of any economic analysis is perspective. The researchers in this case conducted their analysis from the societal perspective and from the perspective of a health plan. The two perspectives can be quite different. For example, some major costs of childhood illness such as work lost because parents stay home do not fall on health plans.

Working within the constraints of their model, the researchers found that vaccination of healthy infants would result in net savings for society if the vaccine cost less than $46 per dose. Health plans would not realize net savings until the vaccine cost fell below $18 per dose. The manufacturer’s list price is currently $58 per dose. At that price, infant vaccination would cost society $80,000 per life-year saved and a health plan $176,000 per life-year saved (net of cost-savings). The cost-effectiveness of an additional program to administer one "catch-up" dose of vaccine to children aged 2 to 4.9 years would vary depending on a child’s age and level of risk. For example, from the societal perspective, the break-even vaccine cost would be $32 for a 4- to 4.9-year-old not in day care. For a 2- to 2.9-year-old in day care, cost-savings would be realized if the vaccine cost less than $135 per dose.

What are policy makers to do with this kind of information? In evaluating the results of cost-effectiveness or related forms of economic analysis, it is exceedingly important to understand what has been included and what has been left out. One also needs to know what methodologies, if any, have been used to assign a value to non-monetary costs and benefits. In a good study, sensitivity analyses will be conducted to assess the effects of changes in key assumptions.

Given the limits of such studies, ACIP charted what appears to be a wise middle course. ACIP recommended vaccination of infants even though such a policy was not predicted to result in cost-savings at the societal level. In doing so, ACIP recognized the importance of non-economic factors. On the other hand, ACIP responded to the extraordinary variability in cost-effectiveness for older children with a more nuanced recommendation.

Economic analysis may be valuable to policy makers in a number ways. Results might be used in negotiations with vaccine manufacturers over price. And, in light of the divergence between societal and health plan perspectives, it might make sense for policy makers to take steps to bring the two into closer alignment. These steps could include subsidies to health plans for specific services as well as structures that force health plans to internalize social costs so they care about social benefits.