Determinants of malaria transmission and bed net utilization among households around Lake Tana, northwest Ethiopia

No Thumbnail Available



Journal Title

Journal ISSN

Volume Title


Addis Ababa University


Background: Malaria still attacks millions of people predominantly in sub Saharan Africa, including Ethiopia. The contribution of house proximity to water bodies, the role of migration, malaria information and child age on malaria transmission dynamics have not yet been examined in detail in northwest Ethiopia. Identification of malaria vectors in meso-endemic settings with a high potential of introduced malaria has not also been done in the study setting. The ownership and use of LLINs is one of the primary prevention strategies for malaria transmission control and elimination. In Ethiopia, evidences showed the presence of major gaps between LLIN ownership and compliance. However, the reasons that influenced consistent utilization of LLINs have not yet explored qualitatively in Amhara region, particularly in the study area. Identifying factors for imported and locally acquired malaria, malaria vectors, and exploring barriers of consistent utilization of LLINs are useful to facilitating the effort for malaria elimination. Objectives: This study aimed at investigating the contribution of house proximity to a water body, individual and household-level drivers of imported and locally acquired malaria, malaria vector species and barriers of consistent LLINs utilization in meso-endemic villages around Lake Tana, northwest Ethiopia. Methods: A combination of approaches, including matched case-control study designs, entomological survey and phenomenological qualitative method were undertaken in 12 kebeles around Lake Tana, northwestern Ethiopia. Health facility-based matched case-control studies involving 303 and 81 matched pairs among all age groups and under-five children were conducted from October 10, 2016, to June 30, 2017, respectively. Anopheles mosquito larvae sampling was done from April 30, 2017, to June 24, 2017, from 7 potential breeding sites in the study area using the standard dipping method. A phenomenological qualitative approach comprising 23 community interviewees and 38 key informants was conducted from April to June 2017. Households for cases and controls, household proximate water bodies, health centers and larvae survey sites were geo-referenced. A semi-structured questionnaire was used to collect data from households. Collected larvae were reared to adulthood at an insectary. Each adult was individually identified at the species and species complex levels using standard adult mosquito morphological identification keys. In-depth interviews were carried out to explore the local contexts and reasons that influenced consistent use of LLINs. Quantitative data were entered and X cleaned in EpiData Version 3.1 software. Qualitative data were managed and analyzed using NVivo Version 10 software. Principal component analysis was performed to estimate household wealth. Stratified analyses were used to confirm confounding and effect modification. Multivariable conditional logistic regression models were fitted to identify independent predictors of malaria occurrence. Thematic analysis was used for the interview data to code and identify themes. Statistical analyses were performed using Excel® , SPSS statistical software version 22 and STATA version 14.0. The maps were plotted using ArcGIS 10.5. Results: Of 303 malaria cases, 59 (19.5%) were imported malaria whereas 244 (80.5%) were locally acquired. Travel to malarious lowlands in the preceding month (adjusted mOR = 7.23; 95% CI: 2.38-22.00), household member‘s travel to malarious lowlands (adjusted mOR = 3.04; 95% CI: 1.14-8.08), and inadequate malaria information (adjusted mOR = 1.56; 95% CI: 1.02- 2.37) were observed as independent predictors of malaria occurrence in the study area. Stratified analyses confirmed that travel to malarious lowlands had a confounding effect on malaria occurrence in the study area. Children aged 36-59 months (adjusted mOR = 3.88; 95% CI: 1.55-9.74) were at increased risk of locally acquired malaria among under-five children. Presence of water body within a 500m radius from a house was not a statistically significant predictor of local malaria. The malaria vectors, Anopheles gambiae complex, An. pharoensis and An. funestus group were identified in malaria meso-endemic villages around Lake Tana. The qualitative investigation explored wrong perceptions about LLINs, malaria and mosquitoes; bedbug infestation; structural inconvenience; unintended uses; weather conditions; problem in distribution of nets; and socio-cultural and economic factors as critical barriers for consistent utilization of LLINs. Killing ability of nets against arthropods other than mosquitoes reportedly made use of LLINs a favored malaria prevention method despite their ineffectiveness after 3 months. Circular-shaped nets were preferred due to their compatibility with varied sleeping structures. Unintended uses were often associated with local needs and seldom linked with social issues and deficiencies in information about malaria and LLINs. Collateral benefits were considered equally important, principally in terms of short-term disinfestations of bedbugs. Conclusions: In this study, travel to malarious lowlands and inadequate malaria information determined the occurrence of malaria infection in villages around Lake Tana. Children aged 36- XI 59 months old continued to be at increased risk of local malaria in these villages where An. gambiae complex is present. Non-consistent bed net use in villages around Lake Tana was associated with inconvenient bed net design and early damage; non-potency of the insecticide against other arthropods; facilitation of bedbug infestation; unintended uses; wrong perceptions about malaria, mosquitoes, and LLINs; and inadequate follow-up regarding LLINs utilization. Recommendations: National malaria control & elimination programmers and health authorities should design improved effective and integrated interventions that consider imported malaria cases. Malaria information that involves IEC and SBCC should be given to optimize coverage and adequacy of information by considering the educational status of the community. Increment of long-acting insecticide spraying coverage and frequency, use of repellents until sleep under nets and equal management of both child age groups while applying LLINs utilization should be considered to reduce local malaria in children aged 36-59 months. Vector control interventions should be designed in accordance with malaria vector species present in a locality. The provision of nets that are structurally compatible with sleeping spaces and the delivery of adequate information about malaria and its vectors, and the properties and proper uses of LLINs to the community indicate substantial accumulation of knowledge that may guide effective LLIN programs in the study area.



Locally acquired malaria, Malaria information, Travel, Matched analysis, Under-five children, Malaria vectors, LLINs, Consistent use, Circular nets, Bedbug infestation, Ethiopia