A small-scale trial was carried out in the Upper Kinabatangan district of Sabah, Malaysia, to determine the effect of using permethrin-impregnated bednets on malaria transmission. A total of 306 nylon bednets with cotton borders, impregnated at a dose estimated to have been 0.062 g permethrin/m2 of nylon netting, were distributed to 139 households in five villages. At the time of distributing bednets, mass drug administration with Fansidar plus primaquine was also administered to the human population to clear all parasitaemias due to Plasmodium falciparum Welch. In another village, for comparison, mass drug administration was the only intervention. After intervention measures in December 1984 and January 1985, the parasite rates in children declined in all villages during the first month, significantly more in the villages with impregnated bednets than in the control, thus proving that the nets had an impact on malaria. However, after about 2 months, parasite rates started to increase again. After 4-6 months, parasite rates in the villages with bednets approached the rate in the control village without nets. The increase in parasite rates was paralleled by a significant deterioration in the quality, physical condition and the degree of non-utilization of bednets. Entomological evaluation proved the efficacy of permethrin-impregnated nets for controlling Anopheles balabacensis Baisas and other anophelines. Bioassays (1 h exposure) of permethrin-impregnated bednets gave 100% mortality initially and 44-61% mortality after 85-106 days. Mosquito collections in treated bednets were significantly reduced for at least 217 days. The project failed to achieve prolonged suppression of malaria transmission for a combination of entomological, sociological and practical reasons which are discussed in relation to the objectives and implementation of future bednet studies.
Holoendemic malaria transmission in two small isolated forest communities and a coastal village was studied by (1) all night human bait collections of Anopheles species from inside and outside houses and (2) buffalo-biting and CDC light-trapping catches during March and November 1984. During the same period thick and thin blood films were collected from the human population, and spleen rates were determined in children from two to nine years of age. Using both the immunoradiometric assay (IRMA) and the dissection technique, more sporozoite-positive infections were detected in An. balabacensis and An. flavirostris in November than in March. IRMA confirmed the presence of Plasmodium falciparum sporozoites. An average of 76.2% of the An. balabacensis population lived long enough to have reached a point where infectivity with P. falciparum was possible in November. Although fewer than five adult females bit humans per night at any time, a resident could theoretically have received more than 160 infective bites in one year. A high frequency of feeding on humans, coupled with increased anopheline life expectancy, contributed to high estimates of falciparum malaria vectorial capacity (number of infections distributed per case per day); for An. balabacensis (1.44-7.44 in March and 9.97-19.7 in November) and for An. flavirostris (0.19-5.14 in March and 6.27-15.8 in November). These high values may explain the increased malaria parasite rates obtained from at least two forest communities. Correlation between actual and calculated rates of gametocytaemia was poorest in Kapitangan due to inadequate sampling of the human population. In Banggi island, malaria is stable and holoendemic, and the population enjoys a high degree of immunity.