This article is part of the supplement: Scaling-up health services in low- and middle-income settings
Research
Methods for evaluating delivery systems for scaling-up malaria control intervention
-
* Corresponding author: Jayne Webster Jayne.Webster@lshtm.ac.uk
1 Disease Control and Vector Biology Unit, London School of Hygiene and Tropical Medicine, Keppel Street, London, WC1E 7HT, UK
2 Health Policy Unit, London School of Hygiene and Tropical Medicine, Keppel Street, London, WC1E 7HT, UK
BMC Health Services Research 2010, 10(Suppl 1):S8 doi:10.1186/1472-6963-10-S1-S8
Published: 2 July 2010Abstract
Background
Despite increased resources over the past few years the coverage of malaria control interventions is still inadequate to reach national and international targets and achieve the full potential of the interventions to improve population health. One of the reasons for this inadequate coverage of efficacious interventions is the limited understanding of the optimum delivery systems of the interventions in different contexts. Although there have been debates about how to deliver interventions, the methods for evaluating the effectiveness of different delivery systems have rarely been discussed. Delivery of interventions is relatively complex and a thorough evaluation would need to look holistically at multiple steps in the delivery process and at multiple factors influencing the process. A better understanding of the strength of the evidence on delivery system effectiveness is needed in order to optimise delivery of efficacious interventions.
Methods
A literature review was conducted of methods used to evaluate delivery systems for insecticide treated nets, intermittent preventive treatment in pregnant women, and treatment for malaria in children.
Results
The methodology of delivery system evaluations varied. There were inconsistencies between objectives and methods of the evaluations including inappropriate outcome measures and unnecessary controls. There were few examples where the delivery processes were adequately described, or measured. We propose a cross sectional observational study design with attribution of the outcomes to a specific delivery system as an appropriate method for evaluating delivery systems at scale.
Conclusions
The proposed evaluation framework is adaptable to natural experiments at scale, and can be applied using data from routine surveys such as the Demographic and Health Surveys, modified by the addition of one to two questions for each intervention. This framework has the potential to enable wider application of rigorous evaluations and thereby improve the evidence base on which decisions about delivery systems for malaria control and other public health interventions are taken.