Cost and quality of operational larviciding using drones and smartphone technology.

BACKGROUND: Larval Source Management (LSM) is an important tool for malaria vector control and is recommended by WHO as a supplementary vector control measure. LSM has contributed in many successful attempts to eliminate the disease across the Globe. However, this approach is typically labour-intensive, largely due to the difficulties in locating and mapping potential malarial mosquito breeding sites. Previous studies have demonstrated the potential for drone imaging technology to map malaria vector breeding sites. However, key questions remain unanswered related to the use and cost of this technology within operational vector control. METHODS: Using Zanzibar (United Republic of Tanzania) as a demonstration site, a protocol was collaboratively designed that employs drones and smartphones for supporting operational LSM, termed the Spatial Intelligence System (SIS). SIS was evaluated over a four-month LSM programme by comparing key mapping accuracy indicators and relative costs (both mapping costs and intervention costs) against conventional ground-based methods. Additionally, malaria case incidence was compared between the SIS and conventional study areas, including an estimation of the incremental cost-effectiveness of switching from conventional to SIS larviciding. RESULTS: The results demonstrate that the SIS approach is significantly more accurate than a conventional approach for mapping potential breeding sites: mean % correct per site: SIS = 60% (95% CI 32-88%, p = 0.02), conventional = 18% (95% CI - 3-39%). Whilst SIS cost more in the start-up phase, overall annualized costs were similar to the conventional approach, with a simulated cost per person protected per year of $3.69 ($0.32 to $15.12) for conventional and $3.94 ($0.342 to $16.27) for SIS larviciding. The main economic benefits were reduced labour costs associated with SIS in the pre-intervention baseline mapping of habitats. There was no difference in malaria case incidence between the three arms. Cost effectiveness analysis showed that SIS is likely to provide similar health benefits at similar costs compared to the conventional arm. CONCLUSIONS: The use of drones and smartphones provides an improved means of mapping breeding sites for use in operational LSM. Furthermore, deploying this technology does not appear to be more costly than a conventional ground-based approach and, as such, may represent an important tool for Malaria Control Programmes that plan to implement LSM.

Conditional trust: Community perceptions of drone use in malaria control in Zanzibar.

Background: The potential of drones to support public health interventions, such as malaria vector control, is beginning to be realised. Although permissions from civil aviation authorities are often needed for drone operations, the communities over which they fly tend to be ignored: How do affected communities perceive drones? Is drone deployment accepted by communities? How should communities be engaged? Methods: An initiative in Zanzibar, United Republic of Tanzania is using drones to map malarial mosqutio breeding sites for targeting larval source management interventions. A community engagement framework was developed, based on participatory research, across three communities where drones will be deployed, to map local perceptions of drone use. Costs associated with this exercise were collated. Results: A total of 778 participants took part in the study spanning a range of community and stakeholder groups. Overall there was a high level of acceptance and trust in drone use for public health research purposes. Despite this level of trust for drone operations this support was conditional: There was a strong desire for pre-deployment information across all stakeholder groups and regular updates of this information to be given about drone activities, as well as consent from community level governance. The cost of the perception study and resulting engagement strategy was US$24,411. Conclusions: Mapping and responding to community perceptions should be a pre-requisite for drone activity in all public health applications and requires funding. The findings made in this study were used to design a community engagement plan providing a simple but effective means of building and maintaining trust and acceptability. We recommend this an essential investment.