Community point distribution of insecticide-treated bed nets and community health worker hang-up visits in rural Zambia: a decision-focused evaluation

In 2013, the Zambian Ministry of Health through its National Malaria Control Programme distributed over two million insecticide-treated bed nets (ITNs) in four provinces using a door-to-door distribution strategy, and more than 6 million ITNs were allocated to be distributed in 2014. This study was commissioned to measure attendance rates at a community point distribution and to examine the impact of follow-up community health worker (CHW) hang-up visits on short and medium-term ITN retention and usage with a view of informing optimal ITN distribution strategy in Zambia. Households received ITNs at community point distributions conducted in three rural communities in Rufunsa District, Zambia. Households were then randomly allocated into five groups to receive CHW visits to hang any unhung ITNs at different intervals: 1-3, 5-7, 10-12, 15-17 days, and no hang-up visit. Follow-up surveys were conducted among all households at 7-11 weeks after distribution and at 5-6 months after distribution to measure short- and medium-term household retention and usage of ITNs. Of the 560 pre-registered households, 540 (96.4 %) attended the community point distribution. Self-installation of ITNs by households increased over the first 10 days after the community point distribution. Retention levels remained high over time with 90.2 % of distributed ITNs still in the household at 7-11 weeks and 85.7 % at 5-6 months. Retention did not differ between households that received a CHW visit and those that did not. At 7-11 weeks, households had an average of 73.8 % of sleeping spaces covered compared to 80.3 % at 5-6 months. On average, 65.6 % of distributed ITNs were hanging at 7-11 weeks compared to 63.1 % at 5-6 months. While a CHW hang-up visit was associated with increased usage at 7-11 weeks, this difference was no longer apparent at 5-6 months. This evaluation revealed that (1) the community point distributions achieved high attendance rates followed by acceptable rates of short-term and medium-term ITN retention and usage, as compared to reported rates achieved by door-to-door distributions in the recent past, (2) CHW hang-up visits had a modest short-term impact on ITN usage but no medium-term effect, and (3) community point distributions can yield sizeable time savings compared to door-to-door distributions.