What Are Appropriate Tasks for Cost-Effectiveness Analysis?
Cost-effectiveness analysis can offer no help for many important policy-making tasks. It essentially provides information about the costs of improving health by means of a particular intervention. As with any investment decision, the price of something is an important, but not the only, consideration. For example, the cost of building a school—like the cost of building a clinic—will vary depending on its size and location and the materials used. Those choices will affect the cost of schooling per student, which may affect the number of children who can attend and perhaps the quality of their learning. However, without information about price, decision makers cannot see the trade-offs involved in addressing other concerns.
"Cost-effectiveness analysis . . . provides information about the costs of improving health by means of a particular intervention."
Thus the question becomes how policy makers, health program administrators, researchers, and others can make the best use of cost-effectiveness analysis. Three types of comparisons become immensely easier with cost-effectiveness analysis:
-
comparisons of different interventions for the same disease
-
comparisons of different interventions for reaching specific segments of a population
-
comparisons of different interventions for different diseases.
Using the cost-effectiveness ratio is most straightforward when comparing interventions that address the same disease or risk factor and differ only in the mode of delivery. In this case, cheaper interventions generally result in greater health gains. For example, addressing vitamin A deficiency by means of capsule distribution has a similar impact on health as fortifying sugar; however, capsule distribution costs about US$6 to US$12 per DALY averted, whereas sugar fortification costs about US$33 to US$35 per DALY averted. Another way of looking at this is to note that for the same cost, capsule distribution could reach three to five times more people than fortified sugar (DCP2, Chapter 28). This is a clear indication that more health gain is possible by spending resources on capsule distribution.
" . . . addressing vitamin A deficiency by . . . capsule distribution has a similar impact on health as fortifying sugar; however, capsule distribution costs about US$6 to US$12 per DALY averted, averted whereas sugar fortification costs about US$33 to US$35 . . ."
However, even in this simple example, decision makers might need to take other factors into account, in particular, that different interventions may reach different people. The cost-effectiveness analysis treats all health gains equally, whereas in public policy, distribution issues are also important. For example, capsule distribution might only reach people who attend health centers, while sugar fortification would only reach people who buy sugar. Depending on the characteristics and behaviors of the population with vitamin A deficiency, fortification might, in practice, be both more effective and more equitable. Fortification would still be costlier per DALY, so decision makers would have to decide whether the additional cost of achieving the more equitable outcome is affordable relative to other uses of the same funds.
Cost-effectiveness analysis is also useful when comparing interventions that address different diseases or risk factors. Scarce resources will generate more health improvements when they are applied to interventions that are more cost-effective. If the cost-effectiveness analysis uses number of deaths averted as its measure of health gain, then allocating resources to more cost-effective interventions will avert the most deaths. For example, spending US$1 million on expanding the traditional vaccination schedule for children to include a second opportunity for measles immunization would avert between 800 and 66,000 deaths, depending largely on the prevalence of measles. In contrast, spending the same amount of money to expand the schedule to include Hib vaccine would avert between 10 and 800 deaths and including yellow fever vaccine would avert between 300 and 900 deaths.
". . . spending US$1 million . . . to include a second opportunity for measles immunization would avert between 800 and 66,000 deaths, . . ."
If instead the analysis uses DALYs as the measure of health gain, then allocating resources to the most cost-effective interventions will maximize years of healthy life. For instance, US$1 million spent on nevirapine and breastfeeding substitutes to prevent HIV-infected mothers from transmitting HIV to their children would yield a gain of 5,000 to 20,000 DALYs, whereas the same amount of money spent to expand immunization coverage with standard children's vaccines would yield a gain of between 50,000 and 500,000 DALYs.
Thus cost-effectiveness should not be the exclusive basis for making health-related public policy decisions and should be complemented with information about distributional consequences. For public policy makers, these two kinds of information establish the trade-offs inherent in allocating funds to different interventions.
