(IRS) in the rural area, excluding the costs of project management and monitoring and .... Business trust .... the small change in the discount rate (b) the relatively.
Tropical Medicine and International Health volume 9 no 1 pp 125–132 january 2004
The cost and cost-effectiveness of malaria vector control by residual insecticide house-spraying in southern Mozambique: a rural and urban analysis Lesong Conteh1, Brian L Sharp2, Elisebeth Streat3, Avertino Barreto3 and Sundragasen Konar2 1 Health Economics and Financing Programme & Gates Malaria Partnership, London School of Hygiene and Tropical Medicine, UK 2 Malaria Research Programme, Durban, South Africa 3 National Malaria Control Programme, Ministry of Health, Mozambique
Summary
objectives To compare two separately funded, but operationally similar, residual household-spraying (RHS) initiatives; one rural and one peri-urban in southern Mozambique. methods The rural programme is a regional project involving the participation and co-ordination of organizations across three countries in southern Africa and is focussed on control in an area in Mozambique of 7552 km2. The second programme focusses on spraying a peri-urban community within a 10-km radius around MOZAL, an aluminium smelter plant of area 410 km2. An ingredients approach was used to derive unit costs for both the rural and peri-urban spraying programmes using detail retrospective cost data and effectiveness indicators. results The economic cost per person covered per year using Carbamates for indoor residual spraying (IRS) in the rural area, excluding the costs of project management and monitoring and surveillance was $3.48 and in the peri-urban area, $2.16. The financial costs per person covered in the rural area and peri-urban area per year were $3.86 and $2.41, respectively. The economic costs per person covered were respectively increased by 39% and 31% when project management and monitoring and surveillance were included. The main driving forces behind the costs of delivering RHS are twofold: the population covered and insecticide used. Computed economic and financial costs are presented for all four insecticide families available for use in RHS. conclusions The results from both these initiatives, especially the rural area, should be interpreted as conservative cost estimates as they exclude the additional health gains that the newly introduced programmes have had on malaria rates in the neighbouring areas of South Africa and Swaziland. Both these initiatives show that introducing an IRS programme can deliver a reduction in malaria-related suffering providing financial support, political will, collaborative management and training and community involvement are in place. keywords malaria, cost analysis, residual household-spraying, southern Africa
Introduction A lot of research has been published on the effectiveness of indoor residual spraying (IRS) (Kouznetsov 1977; Sharp & le Sueur 1996; Curtis & Mnzava 2000; Sharp et al. 2000), but studies on the cost effectiveness of IRS are relatively scarce (Guinness 1997; Goodman et al. 2001; Guyatt et al. 2002b). This analysis aims to help fill this important knowledge gap. Of the few cost effectiveness studies that exist, this paper is the first to compare two separately funded, but operationally similar, RHS initiatives; one rural and one peri-urban. The rural area (referred to as Zone 1), is a regional programme involving the participation and co-ordination
ª 2004 Blackwell Publishing Ltd
of three countries in Southern Africa: Mozambique, Swaziland and South Africa, within a rural project called The Lubombo Spatial Development Initiative (LSDI). There is a growing acceptance that the results of any malaria control programme exclusively in north-eastern KwaZulu-Natal, eastern Swaziland or southern Mozambique, however successful, will only reduce malaria to a certain point, beyond which a regional approach is necessary to achieve further gains. Therefore the three nations involved in the LSDI are channelling their efforts in a regional capacity and thereby benefitting not only the communities in the two districts in southern Mozambique, but also reaping significant externalities by reducing malaria in South Africa and Swaziland (Sharp and Streat, 125
Tropical Medicine and International Health
volume 9 no 1 pp 125–132 january 2004
L. Conteh et al. Cost and cost-effectiveness analysis of malaria vector control
Table 1 Summary of economic costs of malaria in Zone 1A (Mozal project), (USD 2000) Item Indirect costs Lost productivity Direct costs Malaria tests Evacuation costs Drug costs Physician/nurse time Hospital costs Vector control and educational costs*
Cost
% of total
7 81 930 91 4 29 98 3 26 1 84 8 07 27 20
564 846 516 551 267 693 367
29 3 16 4 12 7 30 100
* Community based vector control costs presented here were shared between Zone 1 and 1A. This was reflected in the cost calculations. Tables adapted from Tren and Bate (2001). (R6.50 ¼ 1USD).
Figure 1 Map of study area showing Zones 1 and 1A in Mozambique.
unpublished data), as shown in Figure 1, a map of the study area showing Zones 1 and 1A in Mozambique. The second project, the Mozal project (Zone 1A), focusses on spraying a peri-urban community within a 10-km radius around the development. This initiative was born from an ambitious investment scheme, the construction and commissioning of a highly sophisticated aluminium smelter located in Matola, close to the capital of Maputo and within the larger LSDI area. Attracted by a number of financial incentives, such as tax breaks and competitive labour costs, the Mozal project was seen as a potentially lucrative option by a number of international firms. The developers of Mozal had not forecasted the significant cost imposed by malaria1. Preliminary estimates, which include all malaria cases from the inception of the programme until 10 June 2000, show that the direct and indirect costs of malaria to the Mozal construction 1 During the 1999 malaria season, the company recorded approximately 3500 malaria cases in their workforce of roughly 9000, with a minimum 3-day sick leave on average per episode (Malaria Control Annual Report, 2001).
126
team at just under R18m (approximately US$2.73m 2000). Table 1 presents a breakdown of these costs in more detail (Tren & Bate 2001). The loss of productivity accounts for a large proportion of costs, both for Mozambicans and expatriate staff, unable to carry out their duties during each malaria episode. The cost of evacuating serious cases of malaria to medical centres across South Africa and the intensive vector control programme and education campaign also account for a significant amount of the total costs.
Method Study area Since independence in 1975, malaria control measures in Mozambique have focussed on the major urban centres and consisted of indoor spraying and treatment for those who report ill to health facilities. Prior to these two initiatives, the rural areas in southern Mozambique under analysis had not been involved in any systematic spraying programme. Moreover, bed net use was rare and the purchase of antimalarials over-the-counter minimal because of the rural, limited cash economy (B. Sharp, personal communication). This is markedly different from the neighbouring areas of KwaZulu-Natal and provinces in South Africa, where the South African government spends approximately R50 m per annum (personal communication Department of Health) on malaria control. The severity of malaria in the uncontrolled areas of Mozambique, as determined by the baseline surveys completed in December 1999 and prior to IRS, showed
ª 2004 Blackwell Publishing Ltd
Tropical Medicine and International Health
volume 9 no 1 pp 125–132 january 2004
L. Conteh et al. Cost and cost-effectiveness analysis of malaria vector control
Table 2 Key variables within both the Zone 1 and Zone 1A programmes Zone 1
Zone 1A 1 80 871 Matola
Area No. of health facilities No. of structures
71 047 Namaacha and Matutuine 7552 km2 11 46 906
410 km2 12 85 000
Insecticide Amount of insecticide used Applied per m2
Ficam, Propoxur 1033.5 kg 412 mg
Ficam 2270 kg 412 mg
No. of spray personnel Average no. of structures sprayed per day per person
48 25
48 25
4 52 106 Business trust
2 31 100 MOZAL
Population in 2000 District
Equipment Vehicles Spray pumps Uniforms Funders
infection in children aged 2–
