Where did the nets go? Tracing long-lasting insecticidal net migration on Bioko Island

BACKGROUND: Long-lasting insecticidal nets (LLINs) are commonly distributed through mass distribution campaigns (MDCs). MDCs are often followed by substantial loss of LLINs, affecting coverage. Here, data from an MDC and a malaria indicator survey (MIS) on Bioko Island were used to characterize LLIN mobility. METHODS: Between October 2014 and July 2015, an MDC reached most of the households of the island. All nets distributed were marked with a unique code corresponding to the community where they were distributed. An MIS conducted in August and September 2015 allowed to measure LLIN loss and mobility by verifying the number of existing nets and the community code. Multivariate models were used to identify factors associated with households with LLINs with mismatched codes. A source-sink analysis measuring risk gradient between the source community and the sink household was used to assess the overall effect of migrating nets. RESULTS: The MIS revealed that of the 11,599 LLINs distributed to surveyed households, 7274 (62.7%) were observed during the MIS, indicating that 37.3% were no longer present. Among the 7,274 surviving nets, community codes were registered only for 3,413 (46.9%), and 551 of these (16.1%) revealed a mismatched code. Having open eaves was significantly associated with the presence of mismatched LLINs in households (aOR: 1.61; 95% CI 1.35-1.93, p < 0.001). The source sink analysis showed that while the majority of LLINs migrated from lower to higher risk areas (62.3%), almost a third migrated in the opposite direction. CONCLUSION: While redistribution had a positive impact on some communities by serving populations at higher risk of transmission, many households in high-risk areas lost their nets to areas with lower risk. This, coupled with the significant LLIN loss post-MDC, pointed to critical inefficiencies and prompted a shift in LLIN distribution strategy from MDC to demand-driven fixed distribution points.