Latest research.
A closer look at everyday life.

Authors: Md. Mamun Miah, Kabir Hossain, Hafiz T. A. Khan, H. M. Shahadat Ali

Waterborne diseases remain a major public health challenge, especially in low- and middle-income countries like Bangladesh. This study aimed to identify household level predictors of waterborne diseases and develop an optimized, explainable machine learning framework for predicting disease risk in flood-affected communities. This cross-sectional study was conducted in two flood-affected districts in Bangladesh, using face-to-face interviews with 501 respondents selected through simple random sampling to collect household-level data. Following data preprocessing, correlation analysis, variance inflation factor assessment, mutual information analysis, and Recursive Feature Elimination with Cross-Validation (RFECV) were performed for feature selection. Five machine learning algorithms—Logistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Extreme Gradient Boosting (XGBoost) were optimized using randomized hyperparameter tuning with five-fold cross-validation. Model performance was evaluated on an independent test set using accuracy, precision, recall, specificity, F1-score, receiver operating characteristic area under the curve (ROC-AUC), bootstrap confidence intervals, and SHAP-based interpretation. The prevalence of waterborne disease in the flood affected area was found to be 63.7% (n=319). RFECV retained ten important predictors: age, gender, monthly household income, household size, flood frequency, flood severity, source of drinking water, availability of medical assistance, water purification method, and community preparedness for preventing waterborne diseases. Among the optimized machine learning models, the Random Forest classifier showed the most balanced predictive performance on the independent test set, achieving an accuracy of 61.4%, precision of 71.9%, recall of 64.1%, specificity of 56.8%, F1-score of 67.8%, and ROC-AUC of 0.643. SHAP analysis identified flood severity, community preparedness, water purification practices, and socio-economic factors as the most influential predictors. Although stacking and weighted soft voting ensemble models were additionally evaluated, neither outperformed the Random Forest. The proposed explainable framework may support targeted interventions and early warning strategies after external validation. All the analyses were conducted using Python.

Source: PLOS Water. DOI: 10.1371/journal.pwat.0000542. Abstract from the publisher’s feed; HTML formatting and repeated author line removed.

Authors: Katherine Davies, Katayi Mwila-Kazimbaya, Ian Ross, Elisabeth Tadiri, Jenala Chipungu, Robert Dreibelbis

Access to a dedicated handwashing facility (HWF) increases handwashing with soap (HWWS), an effective behaviour for preventing respiratory and diarrheal disease. Achieving universal hand hygiene will require household investment of a HWF, yet the ability and willingness of end-users to meet these costs has received little attention in stated or revealed preference studies. This two-phase voucher-based randomised pricing experiment explored whether households in peri-urban Lusaka, Zambia were willing to invest in a HWF. In Phase 1, three HWFs were tested with 60 households using two of the three HWFs for two weeks (20 households per combination). There was strong preference for a locally manufactured bucket with a tap and metal stand (“Kalingalinga bucket”) (250 ZMW, ~ $10), over high-cost (1000 ZMW, ~ $38) and low-cost (100 ZMW, ~ $4) pre-manufactured HWFs. In Phase 2, n = 160 used the Kalingalinga bucket for two weeks. At the end of the trial period, participants received a 50 ZMW gift (~$2) and a randomly assigned discount voucher (20%, 40%, 60%, 80%; 40 households per group), before deciding whether to purchase the HWF. Ninety-eight percent (39/40) of households offered an 80% discount purchased the HWF (effective price: 0 ZMW), compared with 30% (12/40) offered a 20% discount (effective price: 150 ZMW). Predictive modelling estimated that 50% of households would purchase at an effective price of 103 ZMW (~40% of the retail price). Despite declining the purchase offer, all non-purchasing households reported positive stated willingness-to-pay, suggesting latent demand. Findings highlight the need for financial interventions that make desirable HWFs affordable.

Source: PLOS Water. DOI: 10.1371/journal.pwat.0000583. Abstract from the publisher’s feed; HTML formatting and repeated author line removed.

Authors: James Harper, Rana Abdel Sattar, Tyler Kozole, Yoem Vichet, Yan Prummanh, Jonathan Stokes, Drew Capone

Safe management of fecal sludge in on-site sanitation systems remains challenging, particularly in low-resource contexts. Fecal sludge treatment plants are often non-existent, non-functional, or non-viable in these contexts due to costly and complex operational requirements. Decentralized fecal sludge treatment methods (e.g., burial on-site) are more commonly used by the 3.4 billion people who use on-site systems but can pose health and environmental hazards if not properly managed. We assessed fecal indicator bacteria in surface soils immediately before, immediately after, and 30 days after a novel service called limed on-site burial (LOB), which was recently pilot tested in rural Cambodia and improves upon existing on-site burial methods. Before-and-after analysis of surface soils at the toilet entrance and at the center and two edges of the burial locations found that Escherichia coli (E. coli) concentrations did not increase due to LOB and likely decreased immediately after (-0.23 log10, [-0.46, -0.01]) and 30 days after service (-0.76 log10, [-0.98, -0.55]). Results confirm existing recommendations of burying fecal sludge 1.0 m underground and covering with at least 0.3 m of soil to mitigate microbial hazard to humans from LOB. Standardizing and confirming the outcomes of LOB provide a pathway towards safely managed sanitation in low-resource contexts that can help challenging regions achieve Sustainable Development Goal 6.2.

Source: PLOS Water. DOI: 10.1371/journal.pwat.0000598. Abstract from the publisher’s feed; HTML formatting and repeated author line removed.