How many mosquito nets are needed to achieve universal coverage? Recommendations for the quantification and allocation of long-lasting insecticidal nets for mass campaigns

Abstract

Background: Long-lasting insecticidal nets are an effective tool for malaria prevention, and "universal coverage" with such nets is increasingly the goal of national malaria control programmes. However, national level campaigns in several countries have run out of nets in the course of distribution, indicating a problem in the method used to estimate the quantity needed.

Presentation of hypothesis: A major reason for the shortfall in estimation is the mismatch between the quantification factor used to plan procurement and the allocation algorithm used at community level, in particular the effect of needing to add an additional net to households with an odd number of inhabitants. To solve this problem a revised quantification factor is suggested.

Testing hypothesis: Based on data from a broad range of household surveys across Africa, the effect of odd-numbered households on numbers of nets distributed is estimated via two frequently used allocation methods. The impact of these algorithms on the proportion of households reaching a person to net ratio of 2:1, a frequently used marker of universal coverage is then calculated.

Implications: In order to avoid stock-outs of nets during national coverage campaigns, it is recommended to use a quantification factor of 1.78 people per net, with an additional allocation factor suggested to account for other common problems at the community level resulting in a final recommended ratio of 1.60 people per net. It is also recommend that community level allocation procedures be aligned with procurement estimates to reduce shortages of nets during campaign distributions. These analyses should enable programme managers to make evidence-based decisions and support a more efficient and effective use of LLIN distribution campaign resources.

Figure 1

Correlation of the ratio of persons to net obtained with odd number correction and mean household size (red filled diamonds) compared to that between mean nets per household (same allocation rule) and mean household size (blue open circles)

Figure 2

Correlation between mean number of net users (if net used) and mean number of people per sleeping place. Ideally all values should fall into the intersection of the two boxes.

Figure 3

Concentration curve and index comparing the equity of universal coverage in Northern Bahr el Ghazal, Southern Sudan, between the actual distribution by residents and the distribution had one net been given per sleeping place (N = 502).