How Are Insecticide-Treated Bednets Used in Ugandan Households? A Comprehensive Characterization of Bednet Adherence Using a Remote Monitor

Remote bednet use monitoring to describe patterns of use and exposure to female Anopheles mosquitoes in an Ugandan cohort

Background

Long lasting insecticide-treated bednets (LLINs) are the most widely used tool for preventing malaria. There has been a plateau in progress in the highest burden African countries since 2015, leading to questions about the effectiveness of LLINs. In this study, remote LLIN use monitors were deployed in a cohort in Eastern Uganda to explore how LLIN use interacts with mosquito exposure.

Methods

The SmartNet study included 20 households from May to October 2019. SmartNet devices recorded, every 15 min, whether an LLIN was unfurled or folded up. Unannounced visits were used to assess SmartNet accuracy. Risk factors associated with poor LLIN use were assessed using generalized linear equations. Female Anopheles exposure was estimated by combining hourly probabilities of exposure from human landing catches and measures of density from biweekly CDC light traps in participants rooms. Mosquito exposure averted by LLINs was quantified using SmartNet measurements and age-related differences were estimated using generalized linear equations, adjusting for relevant covariates and household clustering.

Results

96 individuals contributed 5,640 SmartNet observation nights. In 126 unannounced visits, SmartNet had an area under the curve of 0.869 in classifying whether the LLIN was up or down. The rate of non-use was 13.5% of nights (95% CI: 12.6-14.3%). Compared to children under 5, non-use was 1.8 times higher (95% CI: 1.6-2.1; p < 0.001) in children 5-15 years and 2.6 times higher (95% CI: 2.2-3.1; p < 0.001) in participants aged 15-<30years. There was no difference between children under 5 years and adults > 30 years. LLIN use averted 50.3% of female Anopheles mosquito exposure (95% CI: 40.0-60.0%), with decreasing point estimates of efficacy across age groups: from 61.7% (95% CI: 42.6-80.7%) in children under 5 years to 48.0% (95% CI: 29.1-66.8%) in adults over 30.

Conclusions

Objective monitors are accurate and can feasibly be deployed to obtain data about LLIN use. LLINs provided protection from only 50% of female Anopheles mosquito exposure in this cohort and protection was dependent upon age. In assessing the role of LLINs in malaria prevention it is crucial to consider the dynamics between mosquito exposure and LLIN use behaviors.

A qualitative study of the acceptability of remote electronic bednet use monitoring in Uganda

BACKGROUND

Distribution of long-lasting insecticide treated nets (LLINs) is the most widely used intervention for the prevention of malaria but recall and social desirability biases may lead to challenges in accurately measuring use of bednets. SmartNet is a remote electronic monitor that provides objective measurements of bednet use over weeks at a time. Assessing local acceptability is important when implementing innovative global health technologies such as SmartNet. This study draws on established models such as the Technology Acceptance Model (TAM) and Theoretical Framework of Acceptability (TFA) to assess acceptability of SmartNet in Ugandan households.

METHODS

Semi-structured qualitative interviews were conducted at weeks one and six following installation of SmartNet in ten households in Western Uganda. Heads-of-households answered open-ended questions addressing the main acceptability domains of the TFA and TAM models (i.e. perceived ease of use, ethicality, etc.). Responses were digitally recorded, transcribed, coded and analyzed using a thematic analysis approach.

RESULTS

Seven out of ten households interviewed reported no difference in use between SmartNet and a standard LLIN. Households stated the large size, soft fabric, and the efficacy of SmartNet relative to a standard LLIN contributed to perceived usefulness and perceived ease of use. Opportunity costs of the novel monitoring system expressed by households included difficulty washing nets and dislike of blinking lights on the device. Barriers to SmartNet use focused on questions of the ethics of bednet use monitoring, discomfort with technical aspects of the device and a poor understanding of its function amongst others in the community. However, explaining SmartNet to other community members resolved these concerns and often resulted in interest and acceptance among peers.

CONCLUSION

Objective monitoring of bednet use with SmartNet appears acceptable to these households in Uganda. Use of SmartNet seems to be similar to behaviors around use of standard LLINs. Viewpoints on many aspects of SmartNet were generally favorable. Concerns around ethicality of bednet monitoring are present and indicate the need for continuing community education. The device will continue to be optimized to make it more acceptable to users and to accurately reflect standard LLIN use to improve our understanding of prevention behaviors in malaria endemic settings.

Evaluation of an accelerometer-based monitor for detecting bed net use and human entry/exit using a machine learning algorithm

Background

Distribution of long-lasting insecticidal bed nets (LLINs) is one of the main control strategies for malaria. Improving malaria prevention programmes requires understanding usage patterns in households receiving LLINs, but there are limits to what standard cross-sectional surveys of self-reported LLIN use can provide. This study was designed to assess the performance of an accelerometer-based approach for measuring a range of LLIN use behaviours as a proof of concept for more granular LLIN-use monitoring over longer time periods.

Methods

This study was carried out under controlled conditions from May to July 2018 in Liverpool, UK. A single accelerometer was affixed to the side panel of an LLIN and participants carried out five LLIN use behaviours: (1) unfurling a net; (2) entering an unfurled net; (3) lying still as if sleeping; (4) exiting from under a net; and, (5) folding up a net. The randomForest package in R, a supervised non-linear classification algorithm, was used to train models on 20-s epochs of tagged accelerometer data. Models were compared in a validation dataset using overall accuracy, sensitivity and specificity, receiver operating curves and the area under the curve (AUC).

Results

The five-category model had overall accuracy of 82.9% in the validation dataset, a sensitivity of 0.681 for entering a net, 0.632 for exiting, 0.733 for net down, and 0.800 for net up. A simplified four-category model, combining entering/exiting a net into one category had accuracy of 94.8%, and increased sensitivity for net down (0.756) and net up (0.829). A further simplified three-category model, identifying sleeping, net up, and a combined net down/enter/exit category had accuracy of 96.2% (483/502), with an AUC of 0.997 for net down and 0.987 for net up. Models for detecting entering/exiting by adults were significantly more accurate than for children (87.8% vs 70.0%; p < 0.001) and had a higher AUC (p = 0.03).

Conclusions

Understanding how LLINs are used is crucial for planning malaria prevention programmes. Accelerometer-based systems provide a promising new methodology for studying LLIN use. Further work exploring accelerometer placement, frequency of measurements and other machine learning approaches could make these methods even more accurate in the future.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12936-022-04102-z.

Quantifying and characterizing hourly human exposure to malaria vectors bites to address residual malaria transmission during dry and rainy seasons in rural Southwest Burkina Faso

Background

To sustain the efficacy of malaria vector control, the World Health Organization (WHO) recommends the combination of effective tools. Before designing and implementing additional strategies in any setting, it is critical to monitor or predict when and where transmission occurs. However, to date, very few studies have quantified the behavioural interactions between humans and Anopheles vectors in Africa. Here, we characterized residual transmission in a rural area of Burkina Faso where long lasting insecticidal nets (LLIN) are widely used.

Methods

We analysed data on both human and malaria vectors behaviours from 27 villages to measure hourly human exposure to vector bites in dry and rainy seasons using a mathematical model. We estimated the protective efficacy of LLINs and characterised where (indoors vs. outdoors) and when both LLIN users and non-users were exposed to vector bites.

Results

The percentage of the population who declared sleeping under a LLIN the previous night was very high regardless of the season, with an average LLIN use ranging from 92.43 to 99.89%. The use of LLIN provided > 80% protection against exposure to vector bites. The proportion of exposure for LLIN users was 29–57% after 05:00 and 0.05–12% before 20:00. More than 80% of exposure occurred indoors for LLIN users and the estimate reached 90% for children under 5 years old in the dry cold season.

Conclusions

LLINs are predicted to provide considerable protection against exposure to malaria vector bites in the rural area of Diébougou. Nevertheless, LLIN users are still exposed to vector bites which occurred mostly indoors in late morning. Therefore, complementary strategies targeting indoor biting vectors in combination with LLIN are expected to be the most efficient to control residual malaria transmission in this area.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-021-10304-y.

Changing malaria fever test positivity among paediatric admissions to Tororo district hospital, Uganda 2012-2019

BACKGROUND

The World Health Organization (WHO) promotes long-lasting insecticidal nets (LLIN) and indoor residual house-spraying (IRS) for malaria control in endemic countries. However, long-term impact data of vector control interventions is rarely measured empirically.

METHODS

Surveillance data was collected from paediatric admissions at Tororo district hospital for the period January 2012 to December 2019, during which LLIN and IRS campaigns were implemented in the district. Malaria test positivity rate (TPR) among febrile admissions aged 1 month to 14 years was aggregated at baseline and three intervention periods (first LLIN campaign; Bendiocarb IRS; and Actellic IRS + second LLIN campaign) and compared using before-and-after analysis. Interrupted time-series analysis (ITSA) was used to determine the effect of IRS (Bendiocarb + Actellic) with the second LLIN campaign on monthly TPR compared to the combined baseline and first LLIN campaign periods controlling for age, rainfall, type of malaria test performed. The mean and median ages were examined between intervention intervals and as trend since January 2012.

RESULTS

Among 28,049 febrile admissions between January 2012 and December 2019, TPR decreased from 60% at baseline (January 2012-October 2013) to 31% during the final period of Actellic IRS and LLIN (June 2016-December 2019). Comparing intervention intervals to the baseline TPR (60.3%), TPR was higher during the first LLIN period (67.3%, difference 7.0%; 95% CI 5.2%, 8.8%, p < 0.001), and lower during the Bendiocarb IRS (43.5%, difference - 16.8%; 95% CI - 18.7%, - 14.9%) and Actellic IRS (31.3%, difference - 29.0%; 95% CI - 30.3%, - 27.6%, p < 0.001) periods. ITSA confirmed a significant decrease in the level and trend of TPR during the IRS (Bendicarb + Actellic) with the second LLIN period compared to the pre-IRS (baseline + first LLIN) period. The age of children with positive test results significantly increased with time from a mean of 24 months at baseline to 39 months during the final IRS and LLIN period.

CONCLUSION

IRS can have a dramatic impact on hospital paediatric admissions harbouring malaria infection. The sustained expansion of effective vector control leads to an increase in the age of malaria positive febrile paediatric admissions. However, despite large reductions, malaria test-positive admissions continued to be concentrated in children aged under five years. Despite high coverage of IRS and LLIN, these vector control measures failed to interrupt transmission in Tororo district. Using simple, cost-effective hospital surveillance, it is possible to monitor the public health impacts of IRS in combination with LLIN.