Impact of nighttime human behavior on exposure to malaria vectors and effectiveness of using long-lasting insecticidal nets in the Ethiopian lowlands and highlands

Abstract

Background: Ethiopia continues to grapple with a persistent malaria burden, characterized by ongoing transmission and recurrent outbreaks. Human behavior influences both malaria exposure and the effectiveness of vector interventions, complicating malaria control efforts. Implementing tailored strategies that account for the complex interplay between human activities and vector behavior remains a challenge in both high- and low-transmission areas in Ethiopia, particularly for vulnerable highland populations and temporary labor migrants, due to lack of data. The aim of this study was to examine the spatiotemporal patterns of human-mosquito interactions and evaluate the effectiveness and suitability of long-lasting insecticidal nets (LLINs) in settings involving lowland resident populations, seasonal migrant workers and highland communities.

Methods: Concurrent human and vector behavior data were collected from high-transmission lowlands (residents and temporary migrant workers) and vulnerable highlands populations. Hourly human behavior observations (HBOs), which examined LLIN use, indoor versus outdoor human presence and sleeping patterns, were paired in a crossover design with mosquito sampling using US Centers for Disease Control light traps (CDC LT) as a proxy for mosquito biting behavior. The study was conducted during the peak (October-December 2022) and minor (March-May 2023) malaria transmission seasons ('peak' and 'minor') for a total of 368 nights. In the highlands, four villages consisting of eight households per village were selected for surveillance; in the lowlands, four villages consisting of two resident villages and two farm sites with migrant workers, with eight households/structures per village or farm, were used for data collection. CDC LT and HBO data were integrated to evaluate HBO-adjusted human biting rates (HBO-adjusted HBR) of Anopheles mosquitoes.

Results: In the highland villages, residents predominantly engaged in indoor activities, with their peak activity overlapping with the peak biting hours (1800-2200 hours). A substantial proportion of inhabitants slept indoors without LLINs in the peak and minor seasons (42.8% and 39.2%, respectively). Highland residents were significantly more exposed to malaria vectors indoors (88.4% peak, 88.6% minor) than outdoors during both transmission seasons. In lowland villages, both resident and seasonal migrant worker populations exhibited predominantly outdoor activity, particularly during peak biting hours (1800-2300 hours). Both residents and temporary migrants were significantly more exposed to Anopheles mosquitoes outdoors (resident: 65.0% peak, 67.1% minor; migrant: 70.5% peak, 80.0% minor) than indoors during both transmission seasons. LLIN usage was minimal and offered limited protection, with < 16.63% of person-time spent under nets by resident populations and 10.7% by migrant workers.

Conclusions: Malaria control in Ethiopia requires context-specific strategies tailored to diverse ecological settings that consider the impact of human behavior on exposure to Anopheles mosquitoes. Limited LLIN effectiveness, human activities coinciding with peak biting times and minimal LLIN usage create significant protection gaps. Comprehensive control necessitates supplemental tools addressing exposure in all locations and times. In the Ethiopian highlands, where indoor activities predominate, increased LLIN usage combined with targeted indoor residual spraying could reduce transmission. In lowland areas, both residents and seasonal migrant workers face relatively higher outdoor exposure risks, requiring additional measures, such as topical and spatial repellents. We recommend implementing data-driven, hyperlocal approaches based on specific human-vector interactions to enhance malaria control effectiveness across the Ethiopian highlands and lowlands.

Keywords: Ethiopia; Highlands; Human behavior; Long-lasting insecticidal nets; Lowlands; Malaria; Seasonal migrant workers; Vector behavior.

Fig. 1

Map of Ethiopia showing the locations and administrative districts included in this study. Map provided by Esayas et al. [8]

Fig. 2

Entomological sampling details in the 4 highland villages, 2 lowland—resident population villages and 2 lowland—seasonal migrant workers camps, northwestern Ethiopia. CDC-LT, US Centers for Disease Control and Prevention light trap; HBOs, human behavioral observations; HHs, households; ITS2, internal transcribed spacer region 2; Cox1, cytochrome c oxidase gene

Fig. 3

Proportion of human population in the highland villages observed sleeping or awake, inside or outside and under or not under LLINs (bars), superimposed with Anopheles hourly HBR (lines) from 1800 to 0600 hours. a Peak malaria transmission season (October–December 2022), b minor malaria transmission season (March–May 2023). Outdoor exposure categories (outdoors awake, and asleep without bed-nets) were combined into a single category as LLIN use was not observed outdoors. HBR, Human biting rate; LLINs, long-lasting insecticide nets

Fig. 4

Proportion of human population in the resident population villages observed sleeping or awake, inside or outside and under or not under LLINs (bars), superimposed with Anopheles hourly HBR (line) from 1800 to 0600 hours. a Peak malaria transmission season (October–December 2022), b minor malaria transmission season (March–May 2023) in the lowlands. Outdoor exposure categories (outdoors awake, and asleep without bed-nets) were combined into a single category as LLIN use was not observed outdoors. HBR, Human biting rate; LLINs, long-lasting insecticide nets

Fig. 5

Proportion of human population in the camps of seasonal migrant workers in the lowlands observed sleeping or awake, inside or outside and under or not under LLINs (bars), superimposed with Anopheles hourly HBR (solid line) from 1800 to 0600 hours. a Peak malaria transmission season (October–December 2022), b minor malaria transmission season (March–May 2023). Outdoor exposure categories (outdoors awake, and asleep without bed-nets) were combined into a single category as LLIN use was not observed outdoors. HBR, Human biting rate; LLINs, long-lasting insecticide nets

Fig. 6

Hourly HBO-adjusted HBR (from 1800 to 0600 hours) showing both LLIN-based protection and exposure to Anopheles in the highland villages during the peak malaria transmission season (a) and during the minor malaria transmission season (b). c, d Pie chart showing proportional vector exposure by activity in the highland villages during the peak season (c) and during the minor season (d), over the course of a night. Outdoor exposure profiles consist of people awake and asleep without LLINs. HBO, Human behavioral observations; HBR, human biting rate; LLINS, long-lasting insecticide nets

Fig. 7

Hourly HBO-adjusted HBR (from 1800 to 0600 hours) showing both LLIN-based protection and exposure to Anopheles in the lowland—resident population villages during the peak malaria transmission season (a) and during the minor malaria transmission season (b). c, d Pie chart showing proportional vector exposure by activity in the lowland—resident population villages during the peak season (c) and during the minor season (d), over the course of a night. Outdoor exposure profiles consist of people awake and asleep without LLINs. HBO, Human behavioral observations; HBR, human biting rate; LLINS, long-lasting insecticide nets

Fig. 8

Hourly HBO-adjusted HBR (from 1800 to 0600 hours) showing both LLIN-based protection and exposure to Anopheles in the lowland—seasonal migrant workers camps during the peak malaria transmission season (a) and during the minor malaria transmission season (b). c, d Pie chart showing proportional vector exposure by activity in the lowland—seasonal migrant workers camps during the peak season (c) and during the minor season (d), over the course of a night. Outdoor exposure profiles consist of people awake and asleep without LLINs. HBO, Human behavioral observations; HBR, human biting rate; LLINS, long-lasting insecticide nets