Public health benefits of water purification using recycled hemodialyzers in developing countries

Jochen G. Raimann, Joseph Marfo Boaheng, Philip Narh, Seth Johnson, Linda Donald, Hongbin Zhang, Nathan W. Levin

BACKGROUND

In rural regions with limited resources, the provision of clean water remains challenging. The resulting high incidence of diarrhea can lead to acute kidney injury and death, particularly in the young and the old. Membrane filtration using recycled hemodialyzers allows water purification. This study quantifies the public health effects. Between 02/2018 and 12/2018, 4 villages in rural Ghana were provided with a high-volume membrane filtration device (NuFiltration). Household surveys were collected monthly with approval from Ghana Health Services. Incidence rates of diarrhea for 5-month periods before and after implementation of the device were collected and compared to corresponding rates in 4 neighboring villages not yet equipped. Data of 1,130 villagers over 10 months from the studied communities were studied. Incidence rates showed a decline following the implementation of the device from 0.18 to 0.05 cases per person-month (ppm) compared to the control villages (0.11 to 0.08 ppm). The rate ratio of 0.27 for the study villages is revised to 0.38 when considering the nonsignificant rate reduction in the control villages. Provision of a repurposed hemodialyzer membrane filtration device markedly improves health outcomes as measured by diarrhea incidence within rural communities.

METHODS

Easy Water for Everyone (EWfE) is a 501(c)(3) non-profit, non-governmental organization (NGO) in the United States, Ghana (and with other countries in progress). With the help of local politicians and stakeholders a need for water purification in the estuary of the Volta River in Ghana was identified. For those living in this region the river is the main source for drinking water even though it is known to carry pathogens. Under the supervision of local committees and administrators, EWfE started to install and maintain a device in each of the villages. The chronological order was arbitrary and data collection was commenced on the islands around Ada Foah since 02/2018. Water purification method. The membrane filtration device (NUF500; NUFiltration, Israel), consists of a set of 8 hollow-fiber hemodialyzers, appropriate tubings and a faucet. These hollow fiber hemodialyzers in this project have been used as hemodialyzers once, then reprocessed and sterilized according to FDA/AAMI standards before installation into the water-purification device. Each hemodialyzer contains around 12,000 capillaries providing a membrane surface area of nearly 2 square meters per hemodialyzer. The membrane pore size is 0.003 µm, notably preventing passage of bacteria, parasites and notably also of pathogenic viruses. The output of pure water can be as high as 500 L/h when actively pumped into the device or up to 250 L/h passed into the device by gravity after being pumped into an overhead tank as used in this study. The pressure by gravity is caused by a height of about 12 feet from which the polluted water enters the eight dialyzers placed in parallel (see Fig. 1a, b).

Contaminated river water enters the inside of the capillaries (“blood” compartment) while clean water collects outside of the capillaries (“dialysate” compartment in clinical hemodialysis). Only water (and dissolved salts) passes through the pores. Organic matter that accumulates on the inside of the capillary fibers needs to be rinsed away by intermittently reversing the pressures and filtering clean water back across the membranes (backwashing) through manual pumping. It takes less than 5 min for the backflow to change from dirty to clean appearance and then regain full efficiency for providing clean water. Data collection. Following the approval of our research project, embedded in the non-profit endeavor, by Ghana Health Services, we initiated data collection with trained local community members to support our endeavor. Next to demographic data and water results before and after passing through the filter, we collected data monthly from the heads of households on self-reported diarrhea events in 8 villages during the months February through November 2018. This was a subset of villages served by EWfE. In late June 2018, the hemodialyzer filtration devices became operational in 4 of these villages so that this ongoing monthly data collection started 5 months before the installation. It was concluded 5 months after the installation of the hemodialyzer filtration device. Simultaneously the same data was collected in the 4 villages without the device. For each village and each month, the count of diarrhea events and the number of persons exposed to the data collection were analyzed to estimate the monthly diarrhea incidence rates. Monthly data were summarized for each of the two groups of villages, the control group of 4 villages never exposed to the hemodialyzer water treatment and the group of 4 villages exposed to the water treatment during their second 5 months of the 10-months study period. This approach allowed comparison of the incidence rates during the first and second 5-months periods and incidence rate ratios (second/first 5 months) for the study group and the control group. Having this concomitant data allows us, in a univariate fashion, to use village populations as their own controls and consider the potential confounding effect of seasonality.

RESULTS

The results of water testing showed coliform bacteria at 558 CFU/100 mL in the source water (Volta River) and zero CFU in the filtrate water at the beginning of our installations in the villages of Big Ada. We studied 8 villages (4 were designated control villages and 4 were study villages) in rural Ghana. Table 1 shows the population characteristics of the study arms. Of the village populations studied in this cohort study, 11% and 8% were younger than 5 years of age and notably showed a remarkably high proportion of villagers (96% and 99%) had to resort to open defecation. Monthly diarrhea incidence rates averaged 0.18 counts per exposure month during the baseline period of the study villages and 0.11 for the same 5 months of the control group. During the first 5 months after the installation of the hemodialyzer filtration device, the rate reduced to 0.05, yielding a rate ratio for the study group of 0.28. For the control group the second 5 months gave an average rate of 0.08, showing modest non-significant reduction from the prior 5 months period with a rate ratio of 0.73 (Table 2). Figure 2a and b show the monthly data for the two periods in both village groups. The control villages of the same region and during the same calendar months allow consideration of a seasonal effect on the diarrhea incidence in the study group. Thus, using the incidence rate ratio for the second 5 months over the first 5 months gives a seasonally adjusted rate ratio of 0.38 (0.28/0.73), which translates to a diarrhea incidence rate that is reduced by 62% following initiation of the hemodialyzer filtration device in the study villages.

DISCUSSION

In many countries microbiologically contaminated water is the underlying cause of gastrointestinal disease, mainly diarrhea, associated with deleterious consequences such as acute kidney injury resulting in a high mortality rate, particularly in weaned children younger than five and the elderly. Our data, collected in 4 rural communities in the Ada-East distric of Greater Accra Region in Ghana, before and after the implementation of a hemodialyzer membrane filtration device to produce clean drinking water, shows a substantially reduced risk (rate) of self-reported diarrhea by 72%. This is a major public health outcome particularly since diarrhea is well known to be associated with deleterious consequences such as acute kidney injury and death, particularly in younger children and the elderly. This finding is striking and the rigorous analytic design where each community serves as their own control allows for drawing solid conclusions. Studying and comparing our data to that of a control group which presented only with modest reduction in the incidence of diarrhea over the same time period, corroborates an effect that can be attributed to implementation of our approach. The only modest reduction of diarrhea incidence in the control villages also reduces concerns of seasonality in the incidence rates confounding our interpretation.

Conclusion

Our study using monthly surveys of diarrhea in households may be considered soft data, however the magnitude of a relative 72% reduction in the incidence of diarrhea per monitored population is strikingly large. It is also corroborated by many mothers reporting spontaneously a sudden virtual absence of diarrhea in their children after availability of the dialyzer-filtered water. The marked reduction in the diarrhea incidence is likely due to using sterile water instead of using river water polluted with known pathogens, such as E. coli, as the main source of drinking water. It may be expected that combination of installing a membrane filtration device and combining it with WASH initiatives will have a strong amplified effect as compared to clean water provision alone. This however remains to be shown in further prospective research. The hemodialyzer membrane filtration device used in this study was clearly associated with a substantial reduction in the incidence of self-reported diarrhea compared to the prior period and compared to a control group without the device. Use of repurposed hemodialyzers, that had already saved lives once in their initial purpose in renal replacement therapy, can again serve as an affordable means of water purification to again save lives within entire communities. Our hemodialyzer membrane filtration approach using hollow fibers with pore size as tight as 0.003 µm in the a surface-maximizing configuration used in the technology of the device described in this paper is highly effective and unique. This renders it not only eligible but potentially highly effective to allow the world population to successfully accomplish the United Nations’ Sustainable Development Goal 6.