Abstract
INTRODUCTION
Water chlorination is the most widely used means of providing safe water in humanitarian operations. Current emergency water chlorination guidelines (e.g. Sphere Guidelines, MSF Public Health Engineering) are based on conventions from municipal piped water systems, not on field evidence from humanitarian settings. As a result, they do not reliably ensure water safety at the point of consumption in refugee and internally displaced persons (IDP) settlements. In response to this gap, we developed the Safe Water Optimization Tool (SWOT), a web-based platform that enables field workers to upload water quality monitoring data to the cloud where it is analysed using numerical modelling and machine-learning (artificial neural networks) techniques to generate a site-specific, evidence-based, water chlorination target. In a proof-of-concept study, we aimed to assess the effectiveness of using the SWOT for ensuring water safety at the point of consumption compared to using current chlorination guidelines.
METHODS
In ten water supply networks operated by MSF in the Camp 1 area of Kutupalong-Balukhali expansion site in Cox’s Bazaar District of Bangladesh, we measured free residual chlorine (FRC) levels as water exited taps at 34 public water distribution points (tapstands). After water users had collected water from tapstands, we accompanied them back to their shelters, where we marked each unique parcel of water. We returned after 4-18 hours of household storage and use to measure FRC again from each parcel of water. Between July and December 2019, we collected 2095 paired observations of upstream tapstand FRC, downstream household FRC, and time elapsed between the two measurement events. We used the data to train the SWOT's analytical models and generate and assess a site-specific water chlorination target.
ETHICS
This study was approved by the MSF Ethics Review Board, the York University Ethics Review Board, and the Ethical Review Committee of the Centre for Injury Prevention and Research Bangladesh.
RESULTS
The SWOT recommended that tapstand FRC be set to 0.7 mg/L to provide household water safety (defined as ≥0.2 mg/L FRC at the point of consumption) up to 12 hours post-distribution, compared with the 0.2-0.5 mg/L range recommended by current guidelines. With the SWOT FRC target delivered at tapstands, 88.3% (278 of 315) of household observations had FRC ≥0.2 mg/L at 12 hours post-distribution versus 36.9% (117 of 317) with the current guideline range.
CONCLUSION
The SWOT can utilize field water quality monitoring data to generate a site-specific chlorination target that outperforms current guidelines with respect to ensuring household water safety. It has the potential to improve monitoring, reporting, and accountability for water safety and to yield public health benefits. Further trials focusing on feasibility and usability are underway in Nigeria (MSF) and Tanzania (UNHCR & NRC). Future work will focus on diversifying SWOT applications to surface water treatment, water trucking, and medical facilities in emergency settings, and for intermittent water systems in non-emergency settings. We recommend that the SWOT be implemented and evaluated for improving water safety in future water supply interventions by MSF and other agencies.
CONFLICTS OF INTEREST
None declared.
