Bed net ownership in Kenya: the impact of 3.4 million free bed nets

Abstract

Background: In July and September 2006, 3.4 million long-lasting insecticide-treated bed nets (LLINs) were distributed free in a campaign targeting children 0-59 months old (CU5s) in the 46 districts with malaria in Kenya. A survey was conducted one month after the distribution to evaluate who received campaign LLINs, who owned insecticide-treated bed nets and other bed nets received through other channels, and how these nets were being used. The feasibility of a distribution strategy aimed at a high-risk target group to meet bed net ownership and usage targets is evaluated.

Methods: A stratified, two-stage cluster survey sampled districts and enumeration areas with probability proportional to size. Handheld computers (PDAs) with attached global positioning systems (GPS) were used to develop the sampling frame, guide interviewers back to chosen households, and collect survey data.

Results: In targeted areas, 67.5% (95% CI: 64.6, 70.3%) of all households with CU5s received campaign LLINs. Including previously owned nets, 74.4% (95% CI: 71.8, 77.0%) of all households with CU5s had an ITN. Over half of CU5s (51.7%, 95% CI: 48.8, 54.7%) slept under an ITN during the previous evening. Nearly forty percent (39.1%) of all households received a campaign net, elevating overall household ownership of ITNs to 50.7% (95% CI: 48.4, 52.9%).

Conclusions: The campaign was successful in reaching the target population, families with CU5s, the risk group most vulnerable to malaria. Targeted distribution strategies will help Kenya approach indicator targets, but will need to be combined with other strategies to achieve desired population coverage levels.

Figure 1

Districts in the bed net distributions and clusters in the evaluation survey.

Figure 2

Distribution of bed net brands owned before and after the campaign*. * Pre-campaign: Number of bed nets = 2,105, Post-Campaign: Number of bed nets = 3,141.

Figure 3

Equity ratios for selected indicators*. *An equity ratio of more than 1 means that the poorest quintile was more likely to receive the intervention, less than 1 means that the wealthiest quintile was more likely to receive the intervention. Sample sizes: Number of households: 2059; Number of Households with CU5s: 1113, Number of Households with Women of Reproductive Age (WORA): 1613; Number of CU5s: 1727, Number or WORAs: 2053, Number of Pregnant Women: 182.