Intervention Costs and Cost-Effectiveness
Widespread implementation of effective interventions depends on cost and cost-effectiveness considerations.
Organization of STI Control Activities in Poor Countries
In poor countries, patients can typically obtain treatment for an STI in a public sector health care facility. Many countries have publicly funded, stand-alone STI clinics, but the typical pattern is for health care personnel to provide care for STIs as part of their regular practice in general outpatient clinics. Despite the availability of such publicly funded care, or perhaps because of concerns about anonymity, many STI patients in poor countries avoid public facilities in favor of traditional healers and private pharmacies. A recent study of the cost-effectiveness of delivering STI treatment through trained pharmacists in Peruvian cities included reports of the popularity of self-treatment in Brazil, Cameroon, Ghana, Nepal, South Africa, Thailand, Vietnam, and Zambia (Adams and others 2003). After reviewing this literature, Adams and others selected the point estimate of 0.4 as their best guess for the proportion of STI patients seeking treatment from a pharmacy in Lima.
Determinants of the Costs of Interventions
We adopt a government perspective in analyzing the costs and cost-effectiveness of interventions. Thus, we define the costs of an activity as the total budgetary expenditure attributable to that activity—that is, the total budget for buildings, equipment, personnel, and supplies, with adjustments made when buildings are used for multiple purposes.
We define the unit costs of an activity as the total budgetary expenditure during a stated time period divided by the number of units of output during that same period. Because the same activity can have several outputs, this definition necessarily entails some ambiguity. For example, an intermediate output of the delivery of STI treatment services is the patient treated, whereas a more final output is the patient cured. An even more complete measure of output would include the secondary infections averted as a result of the cure.
One of the reasons that many economists prefer cost-benefit analysis to cost-effectiveness analysis is that the former attaches a dollar value to each of the outputs of an activity and then aggregates across the outputs to construct a summary measure of the total benefit of the activity. However, this simple result hides many arbitrary assumptions that are required to value the separate outputs. One of the most arbitrary of these assumptions is the assignment of a dollar value to a healthy life year. So instead we present costs and cost-effectiveness denominated in the outputs for which we have data. We then go as far as we can toward aggregation by adopting the conventions of the healthy life year and the disability-adjusted healthy life year.
Kumaranayake and others' (2004) background study for this chapter reviews the literature on the unit costs of STI treatment. They identify 35 studies on this topic that provide a total of 77 unit cost estimates. These are grouped in table 17.4 by the disease or syndrome being treated and by the output that was costed. Of the 46 estimates of the unit cost of treatment, only 33 could be interpreted as, or converted to, 2001 U.S. dollars, and the same applied to 9 of the 10 estimates of the unit cost of a cure. Table 17.4 summarizes the results of these studies by the disease or syndrome that occasioned the treatment.
[Table .]
The most notable thing about the summary statistics is their variability: the cost per unit of output can vary by a factor of 100 or more. Even though the mean dollar cost per cure could plausibly be almost four times that of treatment alone, the standard errors are so large as to make the difference statistically insignificant.
The same point—that is, that unit costs vary enormously from one site to another—is made by the preliminary results of a study by Dandona and others (2005). The cost per case treated in the study varies by a factor of 10 across the 14 sites. Furthermore, the two sites that treat the fewest and the most cases per year also display the highest costs, a finding that suggests the existence of both economies and diseconomies of scale.
The variables that determine the costs—and therefore the cost-effectiveness—of STI treatment include the following:
-
delivery by the public or private sector
-
economies of scale
-
economies of scope
-
prevalence and incidence
-
epidemic phase
-
transmission efficiency
-
health system characteristics
-
population composition and concentration
-
resource combinations and input prices
-
incentives to providers for high quality and quantity of service delivery
-
willingness to pay for treatment as a function of price, income, and distance
-
stigmatization
-
disutility of condom use.
