The cost-effectiveness of Intermittent Preventive Treatment for malaria in Gambian multigravidae, including examination of indirect costs

Abstract

The aim of this thesis is to estimate the cost-effectiveness of intermittent preventive treatment of pregnant women (IPTp) with sulphadoxine-pyrimethamine (SP) to prevent Low Birth Weight (LBW) and anaemia due to malaria in pregnancy in Gambian multigravidae, including an examination of indirect costs. The study was piggy-backed on a Randomised Controlled Trial (RCT) of the effectiveness of IPTp conducted in the rural area of the country from July 2002 to February 2004. The specific objectives are to: 1. Examine the cost-effectiveness of introducing SP as IPTp for malaria into normal antenatal care for multigravidae women in rural Gambia; ii. Explore various methods of valuing indirect costs and assess the extent to which they affect the cost-effectiveness ratio; and iii. Make policy recommendations as to whether to introduce SP as IPTp on. cost-effectiveness grounds. The rationale for the study is that both national and international policy makers need precise information to determine which intervention strategies are best for prevention and control of malaria in pregnancy, and which strategies represent good investment. They also need to know how the cost-effectiveness of such interventions compared to other public health interventions. Cost-effectiveness analysis (CEA) is a recognised tool for advising policy makers on the value of an intervention. However, in practice, there are few CEA studies of malaria in general and. malaria in pregnancy in particular in developing countries. Previous CEA studies of IPTp in developing countries have been limited to primigravidae and secundagravidae. Moreover, most of them were conducted from the perspective of the provider without incorporating indirect or productivity costs. This could partly be due to lack of consensus on the issue of whether or not to include indirect costs. The data required for the study was collected through several methods. For IPTp intervention costs to the health-care provider, antenatal clinic users and their families, a sample of 884 multigraviade were randomly selected from the main IPTp effectiveness trial sample of 2,688 recruited multigravidae. In addition, several sub-studies such as health facility studies for the IPTp intervention, a hospital study for treatment of LBW and anaemia, a time-use study and collection of secondary data were used to estimate costs. Time-use and employment surveys were used to measure the time and alternative values for unpaid work. The costs data were combined with the effectiveness data from the trial to estimate incremental costs and consequences for Base case I (trial sample) and Base case II (non-users of bednets in trial sample). The study results showed that the net costs of IPTp with SP for multigravidae with and without indirect costs were D1,221,771 and D1,887,607 respectively for Base case I. The inclusion of indirect costs led to a 68% increase in net costs in Base case I. The corresponding figures for Base case II were D315,933 and D453,620. The indirect costs in Base case II constituted a 44% increase in net costs. In terms of effectiveness, the DALY s averted in Base cases I and II were -125.8 and -0.13 respectively so the intervention created resource losses. Sensitivity analysis conducted by varying key costs and effectiveness parameters showed the introduction of a haemo lobin test led to over 400% increase in net costs for both cases and reduction in number of doses led to around a 40% reduction in net costs. Except for the use of the opportunity cost method of valuing indirect costs in Base case I that led to a 12% decrease in net costs, the use of all other human capital wage rates led to a less than 10% change in net costs for both base cases. Sensitivity analysis of giving IPTp to all pregnant women (primigravidae, secundigravidae and multigravidae) found that IPTp dominated the control for the two base cases. The domination remained regardless of whether indirect costs were included or not. However, the magnitude depended on the wage rate used in estimating indirect costs. The inclusion of indirect costs led to resource savings in the range of 9%-20% for Base case I and 10%-23% for Base case II. The general conclusion is that giving IPTp with SP to multigravidae alone is not cost-effective. However, some improvements in health outcomes were observed for those women who do not sleep under bednets. If IPTp were to be given to all pregnant women without regard to gravidae, IPTp was dominant with and without indirect costs. The resource savings varied according to the wage rate used. Howeve there may be a policy dilemma associated with giving IPTp to all women knowing that the trial showed it confer no benefits, and even might carry some risks.