A cluster-randomized vector control trial in Bihar, India, and neighboring Nepal, will test the efficacy of long-lasting impregnated bednets (LLINs, Permanets) for reducing visceral leishmaniasis incidence. The intervention unit is the village (400-1000 people). The study is designed to detect a 50% reduction in Leishmania donovani incidence in intervention compared to control clusters over 2 years. 24 clusters (selected as high incidence during previous years) will be randomly allocated to intervention or control. Following health education, and with informed consent, all households in intervention villages will receive free Permanets (from September 2006). Net usage will be monitored and new nets provided if required. Control villages will not be given untreated nets, as - although commonly used in this region - their effectiveness against sandflies has not been proven. Pre-intervention infection status of villagers (>2 yrs) will be evaluated serologically from finger-prick blood (and past/current disease status noted). Incident infections will be recorded by 3-monthly active search for clinical cases, and by annual serological diagnoses to detect subclinical infections. All villagers (>2yrs) will be leishmanin skin tested at the end of the trial for further subclinical infection detection, and sera from a sub-sample will be tested for antibodies to sandfly saliva antigens (a measure of sandfly exposure). All clinical cases will be given free treatment. Free Permanets will be provided to control villages after the trial. Complementary studies involve entomological surveillance by light traps in a sample of houses and social/economic questionnaire surveys. The entomological surveys will test whether community-wide use of LLINs provides any mass effect, which could protect those in the community who fail to use LLIN for any reason.