The American Journal of Tropical Medicine and Hygiene - Volume 68, Issue 4_suppl, April 2003

Randomized controlled trials in sub-Saharan Africa have shown that permethrin-treated bed nets and curtains reduce all-cause child mortality by 15–33% in areas with low or high but seasonal malaria transmission. This report describes the study site for a community-based, group-randomized, controlled trial in an area of high and year-round malaria transmission in western Kenya. We outline the development of the human and physical infrastructure required to conduct this trial and discuss some of the difficulties encountered and lessons learned in conducting it.

This paper describes the study design and methods used in a large community-based, group-randomized, controlled trial of permethrin-treated bed nets (ITNs) in an area with intense, perennial malaria transmission in western Kenya conducted between 1996 and 1999. A multi-disciplinary framework was used to explore the efficacy of ITNs in the reduction of all-cause mortality in children less than five years old, the clinical, entomologic, immunologic, and economic impact of ITNs, the social and behavioral determinants of ITN use, and the use of a geographic information system to allow for spatial analyses of these outcomes. Methodologic difficulties encountered in such large-scale field trials are discussed.

The effect of permethrin-treated bed nets (ITNs) on malaria vectors was studied as part of a large-scale, randomized, controlled trial in western Kenya. Indoor resting densities of fed Anopheles gambiae s.l. and An. funestus in intervention houses were 58.5% (P = 0.010) and 94.5% (P = 0.001) lower, respectively, compared with control houses. The sporozoite infection rate in An. gambiae s.l. was 0.8% in intervention areas compared with 3.4% (P = 0.026) in control areas, while the sporozoite infection rates in An. funestus were not significantly different between the two areas. We estimated the overall transmission of Plasmodium falciparum in intervention areas to be 90% lower than in control areas. Permethrin resistance was not detected during the study period. As measured by densities of An. gambiae s.l., the efficacy of bed nets decreased if one or more residents did not sleep under a net or if bed nets had not been re-treated within six months. These results indicate that ITNs are optimally effective if used every night and if permethrin is reapplied at least biannually.

A group-randomized controlled trial of insecticide (permethrin)-treated bed nets (ITNs) was conducted in an area of high perennial malaria transmission in western Kenya to test the effect of ITNs on all-cause mortality in children 1–59 months of age. Child deaths were monitored over a two-year period by biannual household census in Asembo (1997–1998) and in Gem (1998–1999). Overall, 1,722 deaths occurred in children 1–59 months followed for 35,932 child-years. Crude mortality rates/1,000 child-years were 51.9 versus 43.9 in control and ITN villages in children 1–59 months old. The protective efficacy (PE) (95% confidence interval) adjusted for age, study year, study site, and season was 16% (6–25%). Corresponding figures in 1–11- and 12–59-month-old children in control and ITN villages were 133.3 versus 102.3, PE = 23% (11–34%) and 31.1 versus 28.7, PE = 7% (−6–19%). The numbers of lives saved/1,000 child-years were 8, 31, and 2 for the groups 1–59, 1–11, and 12–59 months old, respectively. Stratified analysis by time to insecticide re-treatment showed that the PE of ITNs re-treated per study protocol (every six months) was 20% (10–29%), overall and 26% (12–37%) and 14% (−1–26%) in 1–11- and 12–59-month-old children, respectively. ITNs prevent approximately one in four infant deaths in areas of intense perennial malaria transmission, but their efficacy is compromised if re-treatment is delayed beyond six months.

Estimates of mortality in children less than five years old using government civil registration statistics (passive surveillance) were compared against statistics generated by active demographic surveillance during a randomized controlled trial of permethrin-treated bed nets (ITNs) in western Kenya. Mortality rates were two-fold lower when estimated through civil registration compared with active prospective surveillance (rate ratio [RR] = 0.51, 95% confidence interval [CI] = 0.44–0.59). While civil registration underestimated deaths, particularly in the neonatal period, the age distribution of deaths in children 1–59 months of age was the same as with active surveillance. Seasonal mortality trends were also similar. There was no agreement between cause of death recorded by active and passive surveillance. Verbal autopsy estimated that half of all deaths were associated with malaria and pneumonia, but civil registration markedly under-reported these illnesses; incidence RR (95% CI) = 0.18 (0.14–0.24), and 0.05 (0.03–0.08), respectively, while over-reporting deaths due to measles (RR = 15.5 [95% CI = 7.3–33.2]). Government statistics under-represent mortality, particularly neonatal mortality, in children less than five years of age in rural areas of Kenya. They can provide accurate information on the age-distribution of deaths among children 1–59 months old, and on seasonal trends, but not on disease-specific mortality.

During a randomized controlled trial of insecticide (permethrin)-treated bed nets (ITNs) in an area with intense malaria transmission in western Kenya, we monitored 20,915 sick child visits (SCVs) by children less than five years of age visiting seven peripheral health facilities. The SCVs were monitored over a four-year period both before (1995–1996) and during the intervention (1997–1998). Results are used to estimate the effect of ITNs on the burden of malaria in this community and to evaluate the potential role of these facilities in assessment of the impact of large-scale public health interventions. Compared with baseline, a 27% greater reduction in the incidence of SCVs was seen in ITN villages than in control villages (37% versus 10%; P = 0.049). A similar reduction was observed in SCVs diagnosed as malaria (35% reduction in ITN villages versus 5% reduction in controls; P = 0.04). Two-hundred sixteen SCVs per 1,000 child-years were prevented; three-fourths of these were in children less than 24 months old. As a consequence of lack of laboratory facilities, severe anemia was rarely (< 2%) diagnosed, regardless of intervention status. No effect of ITNs on the incidence of respiratory tract infections, diarrhea, and other commonly diagnosed childhood illnesses was observed. The ITNs reduced the number of SCVs due to malaria, but had no effect on other illnesses. Routine statistics from these facilities provided useful information on trends in malaria incidence, but underestimated the burden of severe anemia.

Health facility ledgers of 11 rural health facilities in western Kenya were reviewed to evaluate diagnostic and prescribing practices. Clinics lacked laboratory facilities. Of 14,267 sick child visits (SCVs), 76% were diagnosed with malaria and/or upper respiratory infections. Other diagnoses were recorded in less than 5% of SCVs. Although two-thirds of malaria cases were diagnosed with co-infections, less than 3% were concomitantly diagnosed with anemia. Chloroquine and penicillin constituted 94% of prescriptions. Half of children given a sole diagnosis of measles or pneumonia were prescribed chloroquine, and 22% of children with a sole diagnosis of malaria were given penicillin. Antimalarials other than chloroquine were rarely prescribed. Only 12% of children diagnosed with anemia were prescribed iron supplementation, while 53% received folic acid. This study highlights limited diagnostic and prescribing practices and a lack of adherence to national treatment guidelines in rural western Kenya.

The impact of insecticide (permethrin)-treated bed nets (ITNs) on malaria in pregnancy was studied in a rural area in western Kenya with intense perennial malaria transmission. All households in 40 of 79 villages were randomized to receive ITNs by January 1997. The ITNs were distributed in control villages two years later. Complete data on birth outcome were available on 2,754 (89.6%) of 3,072 deliveries. Women (n = 780) were followed monthly throughout pregnancy in 19 of 79 villages. Among gravidae 1–4, ITNs were associated with reductions of 38% (95% confidence interval [CI] = 17–54%) in the incidence of malaria parasitemia and 47% (95% CI = 6–71%) in the incidence of severe malarial anemia (hemoglobin level < 8 g/dL with parasitemia) during pregnancy. At the time of delivery, mean hemoglobin levels were 0.6 g/dL (95% CI = 0.01–1.2 g/dL) higher, the prevalence of placental or maternal malaria was reduced by 35% (95% CI = 20–47%), and the prevalence of low birth weight was reduced by 28% (95% CI = 2–47%) in gravidae 1–4 from ITN villages. No beneficial impact was observed in gravidae five or higher. In areas of intense perennial malaria transmission, permethrin-treated bed nets reduce the adverse effect of malaria during the first four pregnancies.

As part of a community-based group-randomized trial on the impact of permethrin-treated bed nets (ITNs) on malaria in pregnancy in a holoendemic area of western Kenya, we assessed their effects on antibody responses to Plasmodium falciparum pre-erythrocytic antigens (recombinant circumsporozoite protein [CSP] and peptides complimentary to the repeat region of the liver stage antigen-1 [LSA-1]) and blood stage antigen (recombinant C-terminal domain of the merozoite surface protein-1 [MSP-119 kD]) in paired maternal/cord plasma samples obtained from 296 deliveries (157 from ITN villages and 139 control villages). Levels of total IgG and IgG subclasses 1–3 to LSA-1 and total IgG and IgG3 to MSP-1 were lower, whereas those of total IgG to CSP were significantly higher in women from ITN villages than those from control villages. In cord plasma, levels of total IgG and IgG2 to LSA-1 and IgG3 to MSP-1 were lower in ITN villages than in control villages, but antibody responses to CSP were similar. Our results suggest that the use of ITNs decreases antibody responses to LSA-1 and MSP-1 antigens in pregnant women with associated reductions in levels of the same antibodies in cord blood. In contrast, ITN use was found to be associated with increased antibody responses to CSP in pregnant women, but had no effect on antibody levels to CSP in cord blood.

As part of a community-based, group-randomized, controlled trial of insecticide-treated bed nets (ITNs) in an area with intense malaria transmission in western Kenya, a birth cohort (n = 833) was followed monthly until the age of 24 months to determine the potential beneficial and adverse effects of reduced malaria exposure during pregnancy and infancy. Malaria transmission and morbidity were comparable pre-intervention. The ITNs reduced malaria attack rates (force of infection) in infancy by 74%, and delayed the median time-to-first parasitemia (4.5 to 10.7 months; P < 0.0001). The incidence of both clinical malaria and moderate-severe anemia (hemoglobin level <7 g/dL) were reduced by 60% (P < 0.001 for both). Protective efficacy was greatest in infants less than three months old and similar in older infants and one-year-old children. Efficacy was lowest in the dry season. Infants from ITN villages experienced better height and weight gain. In areas of intense perennial malaria transmission, ITNs substantially reduce exposure to malaria and subsequent malaria-associated morbidity in children less than 24 months old. Reduced malaria exposure during infancy did not result, with continued ITN use, in increased malaria morbidity in one-year-old children.

Insecticide-treated bed nets (ITNs) have been demonstrated to reduce morbidity and mortality in children less than five years of age. They have also been shown to improve the nutritional status of these children, but little is known about their impact on the nutritional status of school-age children. We evaluated the impact of ITNs on growth, nutritional status, and body composition of primary schoolchildren less than 13 years of age living in an area of intense perennial malaria transmission in western Kenya. The ITNs did not have a significant impact on linear growth or summary measures of protein-energy malnutrition in this age group. This lack of efficacy most likely relates to the reduced burden of malaria in this age group in a setting of stable transmission pressure. Use of ITNs was associated with a change in body composition with an increase in percent lean body mass (1.2%; P = 0.04). This may be consequent to reduced exposure to malaria with subsequent reduced elaboration of pro-inflammatory cytokines known to promote muscle wasting.

The impact of insecticide (permethrin)-treated bed nets (ITNs) on the health of adolescent schoolgirls was investigated during a community-based, randomized, controlled trial of ITNs in western Kenya. Two school-based cross-sectional surveys were conducted to determine the prevalence of malaria and anemia in 644 schoolgirls 12–18 years old in a rural area with intense perennial malaria transmission. In 12- and 13-year-old schoolgirls, ITNs were associated with a reduced prevalence of all cause anemia (hemoglobin level <12 g/dL, 16.9% versus 31.4%, adjusted odds ratio [OR] = 0.38, 95% confidence interval [CI] = 0.21, 0.69%) and a 0.34 g/dL (95% CI = 0.02, 0.66) increase in mean hemoglobin concentrations. No beneficial effect on all-cause anemia (adjusted OR = 0.79, 95% CI = 0.43, 1.45) or hemoglobin concentrations (difference in mean = 0.14 g/dL, 95% CI = −0.24, 0.53) was evident in older girls. In all age groups, no effect was found on malaria parasite prevalence or density, clinical malaria, all-cause morbidity, standard measures of nutritional status and growth, or the use of antimalarials and other medications. ITNs approximately halved the prevalence of mild anemia in young, school-attending, non-pregnant, adolescent girls, but had no impact in older girls or on other malaria-associated morbidity or nutritional status.

We determined the nutritional status of children less than five years of age in an area in rural western Kenya with intense malaria transmission, a high prevalence of severe anemia and human immunodeficiency virus, and high infant and under-five mortality (176/1,000 and 259/1,000). No information is available on the prevalence of malnutrition in this area. Three cross-sectional surveys were conducted between 1996 and 1998 to monitor the effect of insecticide-treated bed nets on child morbidity. Anthropometric indices are presented for 2,103 children collected prior to and during intervention (controls only). The prevalence of stunting (Z-scores for height-for-age [HAZ] <−2), wasting (Z-scores for weight-for-height [WHZ] <−2) and being underweight (Z-scores for weight-for-age [WAZ] <−2) was 30%, 4%, and 20%, respectively. This was severe (Z-score <−3) in 12% (stunting), 1% (wasting), and 5% (underweight) of the children. Few children less than three months of age were malnourished (<2%), but height-for-age and weight-for-age deficits increased rapidly in children 3–18 months of age, and were greatest in children 18–23 months old (44% stunted and 34% underweight). While the mean HAZ and WAZ stabilized from 24 months of age onwards, they still remained substantially below the reference median with no evidence of catch-up growth. Malnutrition is likely to interact with infectious diseases, placing children 3–24 months of age at high risk of premature death in this area.

Information on the impact of insecticide (permethrin)-treated bed nets (ITNs) from randomized controlled trials in areas of intense perennial malaria transmission is limited. As part of a large-scale, community-based, group-randomized controlled trial of the effect of ITNs on childhood mortality in a holoendemic area in western Kenya, we conducted three cross-sectional surveys in 60 villages to assess the impact of ITNs on morbidity in 1,890 children less than three years old. Children in ITN and control villages were comparable pre-intervention, but after the introduction of ITNs, children in intervention villages were less likely to have recently experienced illness requiring treatment (protective efficacy [95% confidence intervals] = 15% [1–26%]), have an enlarged spleen (32% [20–43%]), be parasitemic (19% [11–27%]), have clinical malaria (44% [6–66%]), have moderately severe anemia (hemoglobin level < 7.0 g/dL; 39% [18–54%]), or have a pruritic body rash, presumably from reduced nuisance insect bites (38% [24–50%]). Use of ITNs was also associated with significantly higher mean weight-for-age Z-scores and mid-upper arm circumferences. There was no evidence, however, that ITNs reduced the risk of helminth infections, diarrhea, or upper or lower respiratory tract infections. The ITNs substantially reduced malaria-associated morbidity and improved weight gain in young children in this area of intense perennial malaria transmission.

As part of a large community-based trial on the impact of insecticide (permethrin)-treated bed nets (ITNs) on childhood morbidity and mortality in an area of intense perennial malaria transmission in western Kenya, we assessed the effects of ITNs on malaria-specific humoral responses in young children. The IgG responses to Plasmodium falciparum pre-erythrocytic antigens circumsporozoite protein (CSP) and liver stage antigen-1 (LSA-1) and the blood stage antigen merozoite surface protein-1 (MSP-119 kD) in children less than three years old were investigated during a series of cross-sectional surveys. At 14 and 22 months after the introduction of ITNs, the frequencies and levels of IgG to CSP and LSA-1 were significantly lower in children from ITN villages than in children from control villages (P < 0.001). In contrast, the prevalence of IgG to MSP-1 was significantly higher in children from ITN villages at 14 months (P = 0.0069), but not at 22 months. Our results show that decreased exposure by ITNs reduces IgG responses to pre-erythrocytic antigens, but there was no evidence that two years of ITN use compromises IgG responses to blood stage antigens in these young children in this malaria holoendemic area.

The effect of insecticide (permethrin)-treated bed nets (ITNs) on the spatial distribution of malaria vectors in neighboring villages lacking ITNs was studied during a randomized controlled trial of ITNs in western Kenya. There was a trend of decreased abundance of Anopheles gambiae with decreasing distance from intervention villages both before (P = 0.027) and after (P = 0.002) introduction of ITNs, but this trend was significantly stronger after ITNs were introduced (P = 0.05). For An. funestus, no pre-intervention trend was observed (P = 0.373), but after the intervention, a trend of decreased abundance with closer proximity to intervention compounds developed (P = 0.027). Reduction in mosquito populations in villages lacking ITNs was most apparent in compounds located within 600 meters of intervention villages. Sporozoite infection rates decreased in control areas following the introduction of ITNs (P < 0.001 for both species), but no spatial association was detected between sporozoite rates and distance to nearest intervention village. We conclude that high coverage of ITNs is associated with a community-wide suppression of mosquito populations that is detectable in neighboring villages lacking ITNs, thereby affording individuals residing in these villages some protection against malaria.

Spatial analyses of the effect of insecticide (permethrin)-treated bed nets (ITNs) on nearby households both with and without ITNs was performed in the context of a large-scale, group-randomized, controlled mortality trial in Asembo, western Kenya. Results illustrate a protective effect of ITNs on compounds lacking ITNs located within 300 meters of compounds with ITNs for child mortality, moderate anemia, high-density parasitemia, and hemoglobin levels. This community effect on nearby compounds without nets is approximately as strong as the effect observed within villages with ITNs. This implies that in areas with intense malaria transmission with high ITN coverage, the primary effect of insecticide-treated nets is via area-wide effects on the mosquito population and not, as commonly supposed, by simple imposition of a physical barrier protecting individuals from biting. The strength of the community effect depended upon the proportion of nearby compounds with treated nets. To maximize their public health impact, high coverage with treated nets is essential.

Prior to implementation of a randomized controlled trial of insecticide (permethrin)-treated bed nets (ITNs) in western Kenya, ethnographic studies were conducted to understand local perceptions of disease, sleeping patterns, and other factors that might affect use of ITNs. Educational activities took place prior to distribution, but immediately after distribution in Asembo only approximately half of the ITNs were in use. A qualitative study was then conducted to identify the community’s perceptions about ITNs and the ITN project. While participants ranked malaria as important and recognized that malaria prevention could be beneficial, they believed ITNs would be only partly effective due to the perception that malaria has multiple causes. Concerns expressed included fear of the insecticide, thought by some to be a toxic family planning aid, the taking of blood during clinical studies, and the mixing up of family ITNs during net re-treatment, which would violate cultural taboos. Attempts were made to allay fears by improved communication on these subjects and modification of the study design.

Adherence with permethrin-treated bed net (ITN) use and their proper deployment was directly observed in 2,178 individuals (784 households) participating in a large-scale trial of ITNs on child mortality in western Kenya. The ITNs were distributed free of charge to ensure high coverage, resulting in a ratio of 1.46 persons per ITN. Approximately 30% of ITNs present were unused. The overall percentage adherence was 72.3%. The probability of adherence by individuals depended strongly on age (relative risk [RR] = 0.86, 95% confidence limit [CL] − 0.78–0.94), in which children less than five years of age were less likely to use ITNs than older individuals, and temperature, in which ITNs were more likely to be used in periods of cooler weather. A marginally significant diminution in adherence during the second year of the project was also observed (RR = 0.83, 95% CL = 0.68–1.01). Mosquito numbers, relative wealth, number of house occupants, and the educational level of the head of the household had no effect on adherence. In unstructured questioning of house residents, excessive heat was often cited as a reason for not deploying the child’s ITN. The most important reason for non-adherence was disruption of sleeping arrangements, indicating that ITNs were not readily redeployed in the face of shifting sleeping patterns due to visitors, funerals, house construction, and other events. Challenges faced by health education programs to maximize adherence with ITN use are discussed.

A study of mothers’ perceptions regarding bed nets and malaria was conducted before and after a randomized controlled trial of insecticide (permethrin)-treated bed nets (ITNs) in western Kenya. Awareness about the trial and the rationale for bed net use increased by the end of the trial. Knowledge that mosquitoes caused malaria also increased; however, a higher proportion of mothers from control, rather than intervention villages, cited this (44.4% versus 27.9%; P < 0.001). Mothers from intervention villages were more knowledgeable about the use and maintenance of bed nets and re-treatment with insecticide. Both groups specified advantages of ITNs. Mothers from intervention villages noted practical advantages such as protection against bedbugs and falling roof debris. Few (< 1%) mothers indicated that ITNs protected children against malaria. Intervention homes used significantly fewer mosquito coils, insect spray, medicines, and burned cow dung less often compared with those in control villages. Mothers were willing to pay approximately U.S. $ 4.5 for a regular bed net, but only U.S. 10.5 cents (intervention) and 0.036 (control) for re-treating a bed net. This study suggests that, despite two years of experience of use, bed nets and insecticides would not be purchased as a household priority in this impoverished rural community.

We measured the two-week household-level economic impact of insecticide (permethrin)-treated bed nets (ITNs) used to prevent malaria among children less than five years of age in Asembo, Kenya. The ITNs induced a two-week reduction of 15 Kenyan shillings (KSH) (U.S. $0.25; P < 0.0001) in health care expenditures, but a statistically insignificant 0.5 day (P = 0.280) reduction in household time lost due to caring for sick children. The equivalent annual threshold cost was estimated at U.S. $6.50 (95% confidence interval = 3.12–9.86). If the actual purchase price and maintenance costs of ITNs were greater than this threshold, then households would pay more than they would save (and vice-versa). Both seasonal effects and number of children per household had larger impacts than ITNs on health care expenditures and time lost from household activities. Health care expenditures by a household without ITNs and one child were only 32 KSH per two weeks (U.S. $0.50; P = 0.002), leaving little opportunity for household-level, ITN-induced direct savings. The widespread adoption of the ITNs will therefore probably require a subsidy.

This study compared the costs and effects of insecticide (permethrin)-treated bed net (ITN) use in children less than five years of age in an area of intense, perennial malaria transmission in western Kenya. The data were derived from a group-randomized controlled trial of ITNs conducted between 1996 and 1999. The annual net cost per life-year gained was U.S. $34 and the net annual cost per all-cause sick child clinic visit averted was U.S. $49. After taking into account a community effect (protection from malaria afforded to non-ITN users who lived within 300 meters from users) these estimates decreased to U.S. $25 and U.S. $38, respectively. This study provides further evidence that ITNs are a highly cost-effective use of scarce health care resources.

The fifth, and probably last, large-scale, group-randomized, controlled trial of insecticide (permethrin)-treated bed nets (ITNs) showed that ITNs are efficacious in reducing all-cause post-neonatal mortality in an area of intense, perennial malaria transmission. The trial helped to define pregnant women and infants as target groups for this intervention in high transmission settings. High population coverage with ITNs in both target and non-target groups may be critical to enhance health and survival in pregnant women and infants. The proportion of households with ITNs (coverage), the proportion of individuals properly deploying ITNs each night (adherence), and the proportion of nets properly treated with insecticide (treatment) are the three key determinants of effectiveness of large-scale ITN programs. These three simple outcomes should serve as the basis for program objectives and monitoring and evaluation efforts. Coverage effects and economic analysis support the proposition that ITNs may be viewed as a public good, worthy of public support. Research should continue to improve the intervention tools (the net, the insecticide, and methods for durable treatment and re-treatment) and their deployment.