Abstract
This study accessed the efficiency and health risks of drinking water from different sources treated by filtration, boiling, chlorination, flocculation, and solar disinfection. The microbial quality of 45 treated water samples from boreholes, wells, and pipe-borne water was analyzed to determine treatment effectiveness and to quantify risk using quantitative microbial risk assessment. The effectiveness of each treatment method was a function of sampling sources (p < 0.05) and location (p < 0.10), chlorination and boiling being the most efficient methods (100%). Shiegella in well water samples treated by filtration and flocculation had the highest daily infection risk of 69.5 × 10−1 and 67.5 × 10−1 pppd. The annual risk of infection from Salmonella, Shigella, and Staphylococcus ranged from 7.8 × 10−1 to 1.00 pppy, exceeding the U.S. EPA annual infection benchmark (≤10−4 pppy). Salmonella, Shigella, and Staphylococcus had the highest risk of illness of 4.50 × 10−1, 3.30 × 10−1, and 9.80 × 10−1, respectively. All disease burden values exceeded the WHO disease burden benchmark (≤10−6 DALYs/pppy), with Staphylococcus and Salmonella contributing the highest disease burden of 4.71 × 10−2 and 2.13 × 10−2, DALYs/pppy. Therefore, boiling and chlorination are the best disinfection methods for the pathogens tested.
HIGHLIGHTS
Filtration of drinking water with ceramic filters (the most common LHWT method in Bamenda, Cameroon) is the least effective method in removing bacterial contaminants such as Salmonella, Shigella and Staphylococcus.
Boiling and chlorinaIon are the most effective disinfection methods for the pathogens tested.
Filtration, flocculation, and SODIS did not reduce pathogenic concentraIons below risk levels.
Bacteria contaminants in poorly-treated drinking water exceeded the WHO disease burden benchmark.
INTRODUCTION
However, simple local household water treatment (LHWT) and good storage methods can considerably reduce health risks by improving the microbial quality of drinking water. Globally, approximately 1.8 billion people use household water treatment to improve drinking water quality and thus prevent water-borne diseases (Rosa et al. 2016). In developing countries, bio-sand filters, ceramic water filters, boiling, solar disinfection (SODIS), and chemical treatment (chlorination) are the common LHWT methods (Rosa et al. 2014). Treated drinking water can be considered safe, but it may be a source of water-borne infections if not treated properly (Saturday 2016). The microbial quality of treated water can be used to determine the effectiveness of the LHWT method and the quantitative microbial risk assessment (QMRA) due to the consumption of drinking water treated by various LHWT methods.
QMRA is a tool for estimating the risks of an adverse effect of infection, illness, and /or death due to exposure to various doses of water-borne pathogens (NRC 1983). So far, the QMRA model has been recently used to assess the microbial risk of wastewater, surface water, drinking water, food, and recreational water (Howard et al. 2006; Enger et al. 2012; Membré et al. 2015; Abia et al. 2016; Kouamé et al. 2017; Ahmed et al. 2020). Recent studies on the QMRA of drinking water treated by various LHWT methods are, however, limited. For instance, Sari et al. (2019) used the QMRA model to compare the annual infection risk of drinking water after boiling, filtration, and water-refill methods in urban–slum areas. However, their study was limited only to annual infection risks of coliforms and Escherichia coli from one water source. The prevalence of water-borne diseases is continuously rising even though LHWT methods are used by many to improve drinking water quality from various sources, especially in countries experiencing water stress such as Cameroon. This shows the need for a detailed QMRA study to determine the health risks associated with the consumption of drinking water from different sources treated by various LHWT methods.
Currently, rapid population growth and high rates of urbanization in Bamenda have increased water stress and the dependency on alternative drinking water sources such as boreholes, springs, rivers, rainwater, and locally dug wells. These water sources in Bamenda have been shown to have high concentrations of pathogenic contaminants (Abendong et al. 2019; Anyang 2021; Mufur et al. 2021). Water-borne illnesses are also common though the inhabitants use ceramic filters, boiling, chlorination, or SODIS to treat drinking water (Nde 2017). These factors show the need for assessing the microbial quality and health risks of treated drinking water to recommend the best LHWT methods for the inhabitants in Bamenda. Therefore, the current study aimed to compare the bacteriological quality and QMRA of drinking water from different sources treated by various LHWT methods.
MATERIALS AND METHODS
Study area
The Bamenda Town, found in the North West region of Cameroon, lies between latitude 5°43′ to 7°10′ north and longitude 9°35′ to 11°12′ east and at an altitudinal range of 1,200 to 1,865 m above sea level (Mufur et al. 2021). The town has a surface area of about 3,125 ha (AchoChi 1998) and is characterized by very steep slopes (Ndenecho 2004). The climate is equatorial, characterized by a rainy season that lasts for 7 months (from April to October) and a short dry season that lasts for 5 months (November to March) (Mufur et al. 2021). The average annual rainfall is 2,670 mm, and the mean annual temperature is 25°C. The population of the town has been on the rise for several decades now, making it the third-largest city in Cameroon (Wirba 2020). Presently the town has a population of over 600,000 inhabitants and is divided into three subdivisions (Bamenda I, II, and III) with city councils that manage local developments (Figure 1). The water consumption rate per household is mostly below 50 l/day, with very few households using up to 200 l/day. Pipe-borne water is the main source of water (65%), boreholes 18.2%, wells 8.3%, and streams/rivers 8.3% (Wirba 2020). Because of the inconsistency and unreliability of the pipe-borne water supply, coupled with water stress, the population uses these other sources as coping strategies (Chiaga 2019).
Sample collection
Water samples were obtained from field sites by stratified random sampling. The three subdivisions of Bamenda were used as a basis for stratification into Bamenda I (Up-station), Bamenda II (Old-town), and Bamenda III (Nkwen). Forty-five water samples were collected randomly from wells, taps, and boreholes into labeled 1-l sterile bottles. The water samples were put in a cooling flask and transferred to the Science Laboratory of the Catholic University of Cameroon for bacteriological analysis. The physical properties of water were measured on-site.
Household water treatment methods
The collected water samples were independently treated with the following methods to determine the effectiveness of each treatment method.
Filtration method: A popular and commonly used ceramic water filter was used in the filtration process, after which bacteriological tests were conducted. The pore size of the ceramic filter was between 0.3 and 3 μm. For each filtered sample, 0.25 l was used for the microbiological tests to identify bacteria that were not removed by filtration (Sari et al. 2019).
Boiling method: One liter of each water sample was boiled at 100°C in a stainless-steel pot for 30 min. After cooling, the samples were tested for microbes that were not destroyed by heat treatment (Sari et al. 2019).
Chlorination: Two drops (0.1 ml) of 10% sodium hypochlorite solution (commercial liquid laundry bleach solutions; ‘Eau de javel’), NaClO were put into 1 l water samples, swirled, covered, and allowed to stand for 30 min before drinking (Lantagne et al. 2006). As it dissolved in water, hypochlorous acid (HClO) was formed, which diffused through the bacterial cell wall destroying the membrane proteins. The samples were then tested for any residual pathogen.
Flocculation method: Five gram of alum, Al2(SO4)3‧14H2O, were put into 1 l water samples and vigorously mixed to dissolve thoroughly. It was then swirled (slow mixed) and allowed to stand for 30 min to form a coagulum, and sediment negatively charged the microorganisms to collide and clump together into larger, more easily removable clots or ‘flocs’.
SODIS: Labeled sterilized transparent plastic bottles were filled with 1 l of each water sample and kept on a zinc rooftop to expose it to full sunlight for 5–6 h. That is, from 9 am to 3 pm on a sunny day (CDC 2008). This exposure allows the bottled water to receive solar radiation intensity of 500 W/m2 (equivalent to 5 h of mid-latitude sunshine in summer). The water samples were later tested for the presence of pathogenic water-borne bacteria.
Isolation and identification of bacteria in drinking water before and after treatment
The Gram stain test was used to identify the Gram-positive bacteria from the Gram-negative bacteria. After Gram staining, the Gram-positive cocci were tested for catalase and coagulase production. This test separated pathogenic Staphylococcus aureus from other non-pathogenic Staphylococci species. The sulfide indole motility (SIM) medium test was used to identify Gram-negative bacteria. SIM differentiates Gram-negative members of Enterobacteriaceae by their ability to reduce sulfur, produce indole, or move.
Procedure to determine the effectiveness of treatment
Quantitative microbial risk assessment
The QMRA was used to determine the annual infection or disease risk resulting from exposure to pathogens in treated water samples (NRC 1983; Haas et al. 2014). The identified bacteria in water samples treated by various local household methods were considered for the QMRA. The QMRA comprises four steps: hazard identification, exposure assessment, dose–response assessment, and risk characterization (Rose & Haas 1999; Medema & Ashbolt 2006) as described below.
Hazard identification
This step identifies the microbial agent and its associated adverse health effects on a given population. In the current study, Salmonella typhi, Shigella, and S. aureus were identified and used in estimating the exposure of the inhabitants to these pathogens after consuming locally treated water. These bacteria were selected because, in Bamenda, the prevalence of typhoid fever, dysentery, pneumonia, and skin infections is high, and they are the causative infectious agents (Fonyuy 2014; Biosengazeh et al. 2020).
Exposure assessment through locally treated drinking water
Dose–response assessment
Organism . | Parameter . | Type of model . | Reference . |
---|---|---|---|
Dose–response model | |||
Salmonella typhi | α = 1.75 × 10−1 N50 = 1.11 × 106 PDi = 0.45 | beta-Poisson | Abia et al. (2016) |
Shigella sp. | α = 0.162; N50 = 1.127 × 103 PDi = 0.35 | beta-Poisson | Van Lier et al. (2016) |
S. aureus | r = 7.64E-08 PDi = 1 | Exponential | Rose & Haas (1999); Busgang et al. (2018) |
Disease burden | |||
Salmonella typhi | 0.049 | Havelaar et al. (2012) | |
Shigella sp. | 0.026 | Van Lier et al. (2016) | |
S. aureus | 0.049 | Havelaar et al. (2012) |
Organism . | Parameter . | Type of model . | Reference . |
---|---|---|---|
Dose–response model | |||
Salmonella typhi | α = 1.75 × 10−1 N50 = 1.11 × 106 PDi = 0.45 | beta-Poisson | Abia et al. (2016) |
Shigella sp. | α = 0.162; N50 = 1.127 × 103 PDi = 0.35 | beta-Poisson | Van Lier et al. (2016) |
S. aureus | r = 7.64E-08 PDi = 1 | Exponential | Rose & Haas (1999); Busgang et al. (2018) |
Disease burden | |||
Salmonella typhi | 0.049 | Havelaar et al. (2012) | |
Shigella sp. | 0.026 | Van Lier et al. (2016) | |
S. aureus | 0.049 | Havelaar et al. (2012) |
Risk characterization
Data analysis
Data for evaluating the efficiency of the various LHWT methods were analyzed using the IBM Statistics SPSS version 21.0 (SPSS, Chicago, USA). A one-way analysis of variance was used to compare the means at a significant level of p < 0.05. Microsoft Excel, version 2013, was used to plot graphs and to determine the efficiency of treatment methods.
RESULTS
Physical properties
Table 2 shows the physical properties of the 45 water samples measured during collection. The temperature ranged from 22.3 to 24.2°C. Nkwen well water had a brown color and was highly turbid.
Location . | Source . | Color . | Odor . | Mean turbidity (NTU) . | Mean temperature (°C) . | Mean pH . |
---|---|---|---|---|---|---|
Up-station (Bamenda I) | Pipe-borne | Colorless | Odorless | 1.7 | 23.5 | 6.2 |
Covered-well | Colorless | Odorless | 1.5 | 24.1 | 5.5 | |
Borehole | Colorless | Odorless | 0.6 | 23.9 | 6.2 | |
Old-town (Bamenda) | Pipe-borne | Colorless | Odorless | 2.1 | 22.3 | 6.2 |
Covered-well | Colorless | Yes | 1.5 | 23.5 | 5.1 | |
Borehole | Colorless | Odorless | 0.4 | 23.8 | 5.8 | |
Nkwen–Bamenda III | Pipe-borne | Colorless | Odorless | 1.7 | 22.7 | 6.2 |
Uncovered-well | Brown | Yes | 2.1 | 23.4 | 5.6 | |
Borehole | Colorless | Odorless | 0.5 | 24.2 | 5.5 |
Location . | Source . | Color . | Odor . | Mean turbidity (NTU) . | Mean temperature (°C) . | Mean pH . |
---|---|---|---|---|---|---|
Up-station (Bamenda I) | Pipe-borne | Colorless | Odorless | 1.7 | 23.5 | 6.2 |
Covered-well | Colorless | Odorless | 1.5 | 24.1 | 5.5 | |
Borehole | Colorless | Odorless | 0.6 | 23.9 | 6.2 | |
Old-town (Bamenda) | Pipe-borne | Colorless | Odorless | 2.1 | 22.3 | 6.2 |
Covered-well | Colorless | Yes | 1.5 | 23.5 | 5.1 | |
Borehole | Colorless | Odorless | 0.4 | 23.8 | 5.8 | |
Nkwen–Bamenda III | Pipe-borne | Colorless | Odorless | 1.7 | 22.7 | 6.2 |
Uncovered-well | Brown | Yes | 2.1 | 23.4 | 5.6 | |
Borehole | Colorless | Odorless | 0.5 | 24.2 | 5.5 |
The number of colony-forming units counted
Table 3 shows the data obtained from bacteria viable plate count, where LCBT is the level of contamination before treatment and LCAT is the level of contamination after treatment. Twenty-five out of the 45 (55.56%) water samples collected for the studies before treatment were seen to be contaminated. Only 20 water samples (44.44%) were free of pathogens and suitable for drinking as they met the WHO guidelines for quality drinking water (Table 3). Water samples from wells were the most contaminated, with samples from Nkwen having the highest bacteria load (298 CFU/ml). Water samples from boreholes had the lowest level of contamination, with the highest bacterial load of 4 CFU/ml before treatment (4 CFU/ml) (Table 3). After disinfection, 86.67% of the 45 treated samples met the WHO drinking water standard, and 13.33% remained contaminated (not safe for drinking). Well water samples treated by filtration were the most highly contaminated (66.67%) (Table 3).
Source . | Site (5 samples per site) . | LCBT/(CFU/ml) . | LCAT/(CFU/ml) . | ||||
---|---|---|---|---|---|---|---|
Filtering . | Boiling . | Chlorine . | Flocculation . | SODIS . | |||
Pipe-borne | Up-station | 0 | 0 | 0 | 0 | 0 | 0 |
Nkwen | 0 | 0 | 0 | 0 | 0 | 0 | |
Old-town | 57 | 17 | 0 | 0 | 0 | 0 | |
Well | Up-station | 31 | 7 | 0 | 0 | 0 | 0 |
Nkwen | 298 | 145 | 0 | 0 | 8 | 73 | |
Old-town | 109 | 13 | 0 | 0 | 0 | 0 | |
Borehole | Up-station | 4 | 0 | 0 | 0 | 0 | 0 |
Nkwen | 0 | 0 | 0 | 0 | 0 | 0 | |
Old-town | 0 | 0 | 0 | 0 | 0 | 0 | |
WHO standard | ≤ 0.01 CFU/ml (1 CFU/100 ml) |
Source . | Site (5 samples per site) . | LCBT/(CFU/ml) . | LCAT/(CFU/ml) . | ||||
---|---|---|---|---|---|---|---|
Filtering . | Boiling . | Chlorine . | Flocculation . | SODIS . | |||
Pipe-borne | Up-station | 0 | 0 | 0 | 0 | 0 | 0 |
Nkwen | 0 | 0 | 0 | 0 | 0 | 0 | |
Old-town | 57 | 17 | 0 | 0 | 0 | 0 | |
Well | Up-station | 31 | 7 | 0 | 0 | 0 | 0 |
Nkwen | 298 | 145 | 0 | 0 | 8 | 73 | |
Old-town | 109 | 13 | 0 | 0 | 0 | 0 | |
Borehole | Up-station | 4 | 0 | 0 | 0 | 0 | 0 |
Nkwen | 0 | 0 | 0 | 0 | 0 | 0 | |
Old-town | 0 | 0 | 0 | 0 | 0 | 0 | |
WHO standard | ≤ 0.01 CFU/ml (1 CFU/100 ml) |
Effectiveness of disinfection methods
Risk of pathogenic infection
The concentration of Salmonella typhi, Shigella sp., and S. aureus was highest in well water samples from Nkwen, disinfected by filtration, flocculation, and SODIS, and maximum concentration values were 87.1, 56.2, and 76.3 CFU/100 ml, respectively (Table 4). Filtered water samples had the highest risk of infection. However, infection risk per day varied with the source of drinking water. The risk of infection per day due to Shigella sp. in well water was high with maximum values of 6.95 × 10−1 and 6.75 × 10−1 pppd in filtered and flocculated water samples, while Salmonella typhi and S. aureus had a low infection risk, with maximum values of 1.55 × 10−1 and 1.16 × 10−2 pppd, respectively (Table 4). The infection risk of pathogens per day for children was not significantly different from the infection risk of pathogens per day for adults.
Treatment method . | Water source . | Mean concentration of Bacteria CFU/100 ml . | Quantity of water consumed/children (ml) . | Quantity of water consumed/adult (ml) . | Exposure dose/children . | Exposure dose/adult . | P(inf)/children (pppd) . | Pinf/adult (pppd) . |
---|---|---|---|---|---|---|---|---|
Filtration | Salmonella typhi | |||||||
Pipe-borne water Old-town | 17.2 | 1,000 | 2,000 | 17,200 | 34,400 | 1.55 × 10−1 | 2.43 × 10−2 | |
Well water Nkwen | 89.1 | 1,000 | 2,000 | 89,100 | 178,200 | 6.16 × 10−1 | 1.19 × 10−1 | |
Shigella sp. | ||||||||
Well water Up-station | 7.3 | 1,000 | 2,000 | 7,300 | 14,600 | 6.29 × 10−1 | 6.69 × 10−1 | |
Well water Old-town | 13.2 | 1,000 | 2,000 | 13,200 | 26,400 | 6.64 × 10−1 | 6.99 × 10−1 | |
S. aureus | ||||||||
Well water Nkwen | 56.2 | 1,000 | 2,000 | 56,200 | 112,400 | 4.30 × 10−3 | 8.60 × 10−1 | |
Flocculation | Shigella sp. | |||||||
Well water Nkwen | 8.1. | 1,000 | 2,000 | 8,100 | 16,200 | 6.36 × 10−1 | 6.75 × 10−1 | |
SODIS | S. aureus | |||||||
Well water Nkwen | 76.3 | 1,000 | 2,000 | 76,300 | 152,600 | 5.80 × 10−3 | 1.16 × 10−2 |
Treatment method . | Water source . | Mean concentration of Bacteria CFU/100 ml . | Quantity of water consumed/children (ml) . | Quantity of water consumed/adult (ml) . | Exposure dose/children . | Exposure dose/adult . | P(inf)/children (pppd) . | Pinf/adult (pppd) . |
---|---|---|---|---|---|---|---|---|
Filtration | Salmonella typhi | |||||||
Pipe-borne water Old-town | 17.2 | 1,000 | 2,000 | 17,200 | 34,400 | 1.55 × 10−1 | 2.43 × 10−2 | |
Well water Nkwen | 89.1 | 1,000 | 2,000 | 89,100 | 178,200 | 6.16 × 10−1 | 1.19 × 10−1 | |
Shigella sp. | ||||||||
Well water Up-station | 7.3 | 1,000 | 2,000 | 7,300 | 14,600 | 6.29 × 10−1 | 6.69 × 10−1 | |
Well water Old-town | 13.2 | 1,000 | 2,000 | 13,200 | 26,400 | 6.64 × 10−1 | 6.99 × 10−1 | |
S. aureus | ||||||||
Well water Nkwen | 56.2 | 1,000 | 2,000 | 56,200 | 112,400 | 4.30 × 10−3 | 8.60 × 10−1 | |
Flocculation | Shigella sp. | |||||||
Well water Nkwen | 8.1. | 1,000 | 2,000 | 8,100 | 16,200 | 6.36 × 10−1 | 6.75 × 10−1 | |
SODIS | S. aureus | |||||||
Well water Nkwen | 76.3 | 1,000 | 2,000 | 76,300 | 152,600 | 5.80 × 10−3 | 1.16 × 10−2 |
Disease burden
The risk of infection per day due to Shigella sp. in well water was high, with maximum values of 6.96 × 10−1 and 6.75 × 10−1 pppd in filtered and flocculated water samples (Table 5). The annual risk of infection due to Salmonella typhi, Shigella sp., and S. aureus ranged from 7.80 × 10−1 to 1.00 pppy. The maximum risk of illness was 4.50 × 10−1, 3.30 × 10−1, and 9.80 × 10−1 for Salmonella typhi, Shigella sp., and S. aureus, respectively (Table 5). S. aureus and Salmonella typhi contributed the highest disease burden with a maximum value of 4.71 × 10−2 and 2.13 × 10−2 DALYs/pppy (Table 5). The annual infectious risk and disease burden of S. aureus and Salmonella typhi in water samples treated by filtration, flocculation, and SODIS exceeded the U.S. EPA annual infection benchmark (≤10−4 pppy−1) and the WHO disease burden benchmark (≤10−6 DALYs/pppy) (U.S. EPA 2005; WHO 2008) (Table 5). There was no significant difference between the annual risk of infection, risk of illness, and disease burden values for adults and children. S. aureus had the highest probability of the infection developing into a symptomatic illness (9.8 × 10−1), while Shigella sp. had the lowest probability of symptomatic infection (3.0 × 10−1) (Table 5).
Treatment method . | Water source . | P(inf)/children . | P(inf)/adult . | Pa(inf)/children (pppy-1) . | Pa (inf)/adult (pppy) . | P(ill)/children . | P(ill)/adult . | DB/children (DALYs/pppy) . | DB/adults (DALYs/pppy) . |
---|---|---|---|---|---|---|---|---|---|
Filtration | Salmonella typhi | ||||||||
Pipe-borne water –Old-town | 1.55 × 10−1 | 2.43 × 10−2 | 1.00 | 9.90 × 10−1 | 4.50 × 10−1 | 4.40 × 10−1 | 2.07 × 10−2 | 2.13 × 10−2 | |
Well water Nkwen | 6.16 × 10−2 | 1.19 × 10−1 | 1.00 | 1.00 | 4.50 × 10−1 | 4.50 × 10−1 | 2.07 × 10−2 | 2.07 × 10−2 | |
Shigella sp. | |||||||||
Well water Up-station | 6.29 × 10−1 | 6.69 × 10−1 | 1.00 | 1.00 | 3.50 × 10−1 | 3.30 × 10−1 | 9.10 × 10−3 | 8.58 × 10−3 | |
Well water Old-town | 6.64 × 10−1 | 6.99 × 10−1 | 1.00 | 1.00 | 3.50 × 10−1 | 3.00 × 10−1 | 9.10 × 10−3 | 7.80 × 10−3 | |
S. aureus | |||||||||
Well water Nkwen | 4.50 × 10−3 | 8.60 × 10−3 | 7.90 × 10−1 | 9.60 × 10−1 | 7.90 × 10−1 | 9.60 × 10−1 | 2.87 × 10−3 | 4.52 × 10−2 | |
Flocculation | Shigella sp. | ||||||||
Well water Nkwen | 6.36 × 10−1 | 6.75 × 10−1 | 1.00 | 1.00 | 3.50 × 10−1 | 3.30 × 10−1 | 9.10 × 10−3 | 8.58 × 10−3 | |
SODIS | S. aureus | ||||||||
Well water Nkwen | 5.80 × 10−3 | 1.16 × 10−2 | 8.8 × 10−1 | 9.8 × 10−1 | 8.8 × 10−1 | 9.8 × 10−1 | 3.65 × 10−2 | 4.71 × 10−2 |
Treatment method . | Water source . | P(inf)/children . | P(inf)/adult . | Pa(inf)/children (pppy-1) . | Pa (inf)/adult (pppy) . | P(ill)/children . | P(ill)/adult . | DB/children (DALYs/pppy) . | DB/adults (DALYs/pppy) . |
---|---|---|---|---|---|---|---|---|---|
Filtration | Salmonella typhi | ||||||||
Pipe-borne water –Old-town | 1.55 × 10−1 | 2.43 × 10−2 | 1.00 | 9.90 × 10−1 | 4.50 × 10−1 | 4.40 × 10−1 | 2.07 × 10−2 | 2.13 × 10−2 | |
Well water Nkwen | 6.16 × 10−2 | 1.19 × 10−1 | 1.00 | 1.00 | 4.50 × 10−1 | 4.50 × 10−1 | 2.07 × 10−2 | 2.07 × 10−2 | |
Shigella sp. | |||||||||
Well water Up-station | 6.29 × 10−1 | 6.69 × 10−1 | 1.00 | 1.00 | 3.50 × 10−1 | 3.30 × 10−1 | 9.10 × 10−3 | 8.58 × 10−3 | |
Well water Old-town | 6.64 × 10−1 | 6.99 × 10−1 | 1.00 | 1.00 | 3.50 × 10−1 | 3.00 × 10−1 | 9.10 × 10−3 | 7.80 × 10−3 | |
S. aureus | |||||||||
Well water Nkwen | 4.50 × 10−3 | 8.60 × 10−3 | 7.90 × 10−1 | 9.60 × 10−1 | 7.90 × 10−1 | 9.60 × 10−1 | 2.87 × 10−3 | 4.52 × 10−2 | |
Flocculation | Shigella sp. | ||||||||
Well water Nkwen | 6.36 × 10−1 | 6.75 × 10−1 | 1.00 | 1.00 | 3.50 × 10−1 | 3.30 × 10−1 | 9.10 × 10−3 | 8.58 × 10−3 | |
SODIS | S. aureus | ||||||||
Well water Nkwen | 5.80 × 10−3 | 1.16 × 10−2 | 8.8 × 10−1 | 9.8 × 10−1 | 8.8 × 10−1 | 9.8 × 10−1 | 3.65 × 10−2 | 4.71 × 10−2 |
where P (inf) - probability of infection, Pa (inf) - probability of annual infection, P (ill) - probability of illness, and DB - disease burden
DISCUSSION
Bacteriological quality of water treated by LHWT methods
According to the WHO guidelines, safe drinking water should contain no pathogens to avoid water-borne diseases, and pathogenic safety levels should be less than 1 CFU/100 ml of water (WHO 2006). In this study, less than half of the drinking water samples met these criteria (Table 3), underpinning the need for a household treatment of drinking water in Bamenda. However, the choice of an appropriate LHWT method depends on the source of water since the bacteria load in the current study varied with treatment method and water source (Table 3). The significant difference (p ≤ 0.05) between the effectiveness of LHWT methods in the current study was probably due to the different levels of exposure of water sources to microbes (Rosa et al. 2014). Pipe-borne water may not be considered safe for consumption since samples from Old-town had low microbial concentrations (Table 3), probably due to the presence of contaminants in storage tanks or pipes during transportation (Katukiza et al. 2014). Borehole sources had the lowest contamination levels (Table 3), confirming findings by Njunda and colleagues at the Cameroon Development Corporation (Njunda 2013). Contrarily, locally dug well water sources were the most contaminated (Table 3), aligning with the results of Magha (2021) in Bamenda. The high contamination of exposed well water sources may be attributed to possible wash-off from fecal contaminants from neighboring toilets or sewage from poor sewer systems. Treatment methods differ in their effectiveness in preventing water-borne pathogens (Figure 4). For instance, chemical treatment with sodium hypochlorite and heat treatment by boiling showed no bacteria prevalence for tested pathogens, indicating their high effectiveness (100%) for use in treating water unless resources are unavailable. This result corroborates findings established by Okwadha & Ahmed (2017). However, chlorination may be more cost-effective for poor households. According to Mohamed et al. (2016), decisions related to scaling up LHWT practices are affected by factors such as effectiveness, adherence (correct, consistent, and sustained use), and the cost of achieving the desired aim of reducing pathogens.
SODIS was ineffective in disinfecting well water samples from Nkwen (Table 3), probably due to high bacteria load, high turbidity, and high concentration of dissolved organic matter. Normally, for solar treatment to be efficient, turbid water must be made clear by filtration or sedimentation. Little to no sunlight on rainy days also reduces the effectiveness of SODIS and prolongs disinfection time, making disinfection by solarization difficult for consistent use in the rainy season. Moreover, ultraviolet radiation catalysis the degradation of plastics to Bisphenol-A (BPA), which can diffuse into treated water in plastic bottles to pose a carcinogenic risk (Seachrist et al. 2016). Furthermore, recent research findings have proven the occurrence of carcinogenic microplastics, phthalates, and alkylphenol in plastic bottled water (Amiridou & Voutsa 2011; Gambino et al. 2022). These factors greatly limit or discourage the use of solarization method of disinfection. In the current study, filtration had the lowest effectivity (Figure 4), though the effectiveness of filtration may depend on the filtering apparatus. The current study used ceramic filters with pore sizes of 0.3–3 μm, commonly and frequently used in treating water locally by most households in Bamenda.
Health risk assessment of treated water
The approach of QMRA is a comparatively new approach in identifying health risks associated with pathogens in drinking water (Bentham & Whiley 2018). In the current study, the overall health risks (Tables 4 and 5) due to Salmonella typhi, Shigella sp., and S. aureus in locally treated water may be ascribed to the low infectious dose of pathogens (Katukiza et al. 2014), since all pathogenic concentrations above the WHO benchmark (≤1 CFU/100 ml) posed a high annual risk of infection (Table 4). The overall high risks of illness due to S. aureus, Shigella sp., and Salmonella in the current study are congruent with other studies on drinking water (Busgang et al. 2015; Ahmed et al. 2020). However, these findings were based on untreated water samples. Water samples treated by filtration contributed the highest overall risks due to the presence of all three pathogens, indicating that health risks may also vary with local household treatment methods.
Disease burden of pathogens in treated water
Diseases related to inadequate water, sanitation, and hygiene contribute to a huge burden of disease in low-income countries (Ahmed et al. 2020). The disease burden of all three pathogens in treated water samples exceeded the WHO disease burden benchmark. Katukiza et al. (2014) also found a high burden of water-borne diseases in a low-income population who use surface water as the main source of drinking. Filtered well water samples contributed to the highest disease burden (Table 5), probably due to the high prevalence of all three pathogens, putting the proportion of the population that depends on well water as a drinking source and the use of artificial filters at risk of water-borne disease. Thus, to minimize the overall risk of illness from infection by tested pathogens, users should treat well water by boiling or chlorination. However, filtration, solarization, and flocculation can be effective if combined with other treatment methods. Salmonella typhi was present in both well and pipe-borne water sources and contributed to the highest proportion of disease burden (Table 5). Globally, about 212 million cases of typhoid, with 129,000 deaths, are reported yearly, with children and young adults being the vulnerable groups (Steele et al. 2016). Several studies in Cameroon have reported a high prevalence of typhoid fever due to Salmonella sp infection (Njoya et al. 2021; Ndip et al. 2022). The global burden of Shigella and other water-borne illnesses is mostly unknown because most of the cases go unreported (Zahid 2018).
The current study is, however, limited because it did not consider the uncertainty of risk estimates as in the case of Monte Carlo risk analysis. Also, the prevalence of Salmonella sp. in treated water samples suggests fecal contamination and, therefore, the possibility of contamination by other fecal pathogens such as Vibro cholerae, E. coli, Campylobacter, and Cryptosporidium bacteria with time and availability. However, based on available resources, the current study investigated only three water-borne pathogens. Also, previous studies by Fonyuy (2014) and Biosengazeh et al. (2020) show high prevalence of illnesses such as typhoid fever, dysentery, pneumonia, and skin infections caused by Salmonella typhi, Shigella sp., and S. aureus in Bamenda probably due to contaminated water, since most households used local water treatment methods to reduce disease risk especially during water stress, when water supply is unreliable or scarce.
CONCLUSION
For the pathogen tested, chlorination and boiling were the most efficient disinfection methods, regardless of the water source. The overall health risks for both children and adults were not significantly different. Drinking water from well sources treated by filtration with a ceramic filter of pore size 0.3–3 μm contributed to the highest health risks. Salmonella typhi and Shigella sp. had approximately 1.00 annual probability risk of infection. S. aureus had the highest probability risk of developing into illness (9.80 × 10−1), while Salmonella typhi had the lowest (4.50 × 10−1). The overall probability risk of infection and disease burden exceeded the U.S. EPA annual infection and the WHO disease burden benchmark. The filtration, flocculation, and SODIS did not reduce pathogenic concentrations below risk levels. Although the boiling and chlorination methods did not pose risks for the pathogens tested, the study did not address a wide range of pathogens.
ACKNOWLEDGEMENTS
The authors wish to acknowledge Dr Polycarp Ndikvu Chia, the Head of the Department of Biochemistry of the Catholic University of Cameroon, Bamenda, for laboratory assistance. The authors also acknowledge the travel support received within the Global-SDG-Campus project funded by the DAAD.
FUNDING
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
DATA AVAILABILITY STATEMENT
All relevant data are included in the paper or its Supplementary Information.
CONFLICT OF INTEREST
The authors declare there is no conflict.