Managing water in the home: accelerated health gains from improved water supply
Storage and treatment of household water
1. Household Water Storage, Microbial Quality and Infectious Disease Risks
Key factors in the provision of safe household water include the conditions and practices of water collection and storage and the choice of water collection and storage containers or vessels. As shown in Table 1, numerous studies have documented inadequate storage conditions and vulnerable water storage containers as factors contributing to increased microbial contamination and decreased microbial quality compared to either source waters or water stored in improved vessels. Some studies also have documented increased risks of waterborne infectious diseases from inadequately stored water compared to water stored in an improved vessel (safe storage), treated in the home to improve microbial quality, or consumed from a quality source without storage (Table 1). Higher levels of microbial contamination and decreased microbial quality are associated with storage vessels having wide openings (e.g., buckets and pots), vulnerability to introduction of hands, cups and dippers that can carry fecal contamination, and lack of a narrow opening for dispensing water. Some studies have noted the vulnerability of storage vessels with these undesirable characteristics to fecal and other contamination without having reported microbiological data on water quality or increased levels of diarrheal disease (Miller, 1984). Other factors contributing to greater risks of microbial contamination of stored water are higher temperatures, increased storage times, higher levels of airborne particulates (dust storms), inadequate handwashing and the use of stored water to prepare weanling and other foods that also become microbiologically contaminated and contribute to increased infectious disease risks (Black et al., 1983; Dunne, 2001; Echeverria et al., 1987: Iroegbu et al., 2000; Knight et al., 1992; Luby et al., 2001a, van Steenbergen et al., 1983).
Table 1. Evidence for Increased Microbial Contamination (Decreased Microbial Quality) and Increased Infectious Disease Risks from Inadequately Stored Household Water
Location Storage Vessel Storage Times Impact on Microbial Quality? Disease Impact? Reference Rural Bangladesh Water jars 1-2 days Increased V. cholerae presence Increased (~10-fold higher) cholera rates Spira et al., 1980 Bahrain Capped plastic vessels, jars, pitchers Not reported V. cholerae present in stored but not source water Uncertain. No significant association with stored water in a case-control study Gunn et al., 1981 Calcutta, India Wide-mouth vs. narrow-necked Not reported Not measured Cholera infections 4-fold higher using wide-mouth storage vessel Deb et al., 1982 Khartoum, Sudan Clay jars ("zeers") in homes, etc. 2 days to 1 month Increased fecal indicator bacteria over time, in summer and during dust events Not Measured Hammad and Dirar, 1982 Rural Egypt Clay jar ("zir") in homes <1 to 3 days Algae growth and accumulated sediment Not detected based on protozoan infection rates Miller, 1984 Abeokuta, Nigeria Elevated tanks in hospitals Not reported Higher plate count bacteria and E. coli in tanks than in central supply Not Measured Mascher and Reinthaler, 1987 Rural Malawi Stored household water & other sources Not reported Higher fecal coliforms compared to other sources Not measured Lindskog and Lindskog, 1988 South Sudan Not reported Not reported Increased fecal bacteria levels Not Measured Mascher et al., 1988 Rangoon, Burma Buckets Up to 2 days Higher levels of fecal coliforms than source Not Measured Han et al., 1989 Urban slum and rural villages, Liberia Large containers, open or closed "A long time" Higher levels of enterobacteria in stored than source water Not Measured Molbak et al., 1989 Kurunegala, Sri Lanka Earthen pots and others Not reported Higher levels of fecal coliforms in stored unboiled water Not Measured Mertens et al., 1990 Venda, South Africa Plastic vessel ("tshigubu") 4 hours Higher levels of coliforms over time Measured; no effect Verwejj et al., 1991 Rural Africa Traditional and metal jars 24 hours or more Higher levels of total and fecal coliforms Not Measured Empereur et al., 1992 Rural Malaysia Various containers Not reported Higher levels of fecal coliforms in unboiled than boiled water Higher diarrhea risks from water unboiled or stored in wide-necked than narrow-necked containers Knight et al., 1992 Rural Zimbabwe Covered and uncovered containers 12 hours or more Higher E. coli and Aeromonas levels with storage and use Not Measured Simango et al., 1992 Trujillo, Peru Wide-mouth storage containers Not reported Higher fecal coliform levels in stored than source waters Increased cholera risks Swerdlow et al., 1992 The Philippines - - - - VanDerslice and Briscoe, 1993 Rural Bangladesh Traditional pots ("kulshis" Not Reported Increased fecal coliform levels and multiple antibiotic resistance Increased fecal coliforms and multiply antibiotic resistant flora Shears et al., 1995 Merica, Mexico Not reported Not reported Increased bacterial levels in some locales Not Measured Flores-Abuxapqui et al., 1995 Malawi, Africa - - Increased V. cholerae Increased cholera risks Swerdlow et al., 1997 Rural Trinidad Open (drum, (barrel, bucket) vs. tank or none Not reported Increased fecal bacteria levels in open vessel storage than in tank Not measured Welch et al., 2000 Abidjan, Cote d'Ivoire - - Increased E. coli levels Not Measured Dunne et al., 2001
2 Collection Methods and Storage Vessels for Household Water
As summarized in Table 1, collection, storage and use of fecally contaminated water containing excessive levels of fecal bacteria poses health risks to consumers, regardless of where or how the water have become contaminated. In some cases water is collected from a contaminated source to begin with. In other cases water is obtained from a source of high microbiological quality, including treated supplies containing residual chlorine, but it becomes contaminated in the home due to inadequate and unsanitary storage conditions that allow for the introduction and/or proliferation of disease-causing microbes. In either situation, the microbially contaminated water poses health risks that can be reduced by improved storage conditions and household treatment, as will be further documented in this report. A few studies in the literature suggest that contamination of water within households does not pose increased health risks to consumers, perhaps because these pathogens are already present in household members and their contacts. However, evidence to support this interpretation based on sound study design with adequate data and data analyses are lacking. The majority of studies suggesting such lack of risk are based on inadequate study designs, low sample sizes, measurement of waterborne microbes not adequately predictive of health risks (e.g., plate count or coliform bacteria) and/or inadequate data analyses. The majority of studies document decreases in microbial quality, including increased pathogen levels, and increased health risks from consumption of fecally contaminated and inadequately stored household water.
Since ancient times, water for household use is collected by a variety of physical methods ranging from manual (e.g., dipping), to passive (e.g., roof catchments and diversions) to mechanical (e.g., pumps), and it is stored in a variety containers. In developing countries, many of the traditional types of water collection and storage methods employing vessels of various compositions and sizes are still widely used today (CDC, 2001; Mintz et al., 1995; White et al., 1972). These include traditional pots or urns fashioned from natural materials (e.g., gourds or wood) or fabricated from clay, copper, brass and other impervious materials, and flexible bags or other vessels made of animal hides, other animal parts or fabrics treated to seal and prevent leakage. Today, other metals, including aluminum, steel and iron, as well as other materials, primarily plastics, have come into widespread use for water collection and storage in the form of buckets, jerry cans, picnic coolers and other vessel types and shapes. Cisterns and other basins are also still widely used for water collection and bulk storage near or adjacent to dwellings, as they have been since ancient times.
Some of the key factors influencing the impact of storage vessels and conditions on household water quality are: (1) portability and ease of use, based on capacity, size, shape, weight, presence of handles, (2) durability, weight and other properties related to resistance and longevity, (3) presence of a coverable (preferably screw-cap) opening for filling and cleaning access but small enough to reduce the potential for introducing contaminants by contaminated hands, dipping utensils and other vehicles (e.g., airborne dust), vectors, or other sources, (3) ability to withdraw water in a sanitary manner, such as via a tap, spigot, spout or other narrow orifice, and (4) presence and accessibility of documentation describing how to properly use the container for water treatment and sanitary storage. The advantages and disadvantages of different types of water collection and storage containers in relation to the development of systems for safe storage and use of household water have been reviewed and summarized by the US Centers for Disease Control and Prevention and their collaborators (Mintz et al., 1995; Reiff et al., 1995; USA CDC, 2001). The key findings and recommendations of their investigations and experiences are summarized in Table 2.
Table 2. Alternative Household Water Storage Vessels: Advantages and Disadvantages of Different Designs and Materials
Type of Vessel Protected Opening for Filling and Cleaning Size or Volume Material/ Cleanability/Composition Compatible with Use Protected Dispenser (Spigot, Spout, etc.) Shape/Weight/ Portability Pot, Jug or Urn Varies; some yes, some no Varies; usually 4-40L Varies/ Varies/Varies No, often; Yes, some Varies/Varies/ Moderate-High Bucket No Varies: usually 4-40L Plastic or Metal/High/Varies No Cylindrical/ Light/Moderate-High Cooking Pot Yes (lid)
No (no lid)
Varies: usually 4-20L Metal or Clay/High/High No Cylindrical/
Gourd (Calabash) Yes Varies, usually 1-10 liters Plant fruit/moderate/moderate Yes, usually Globular or elliptical, with a curved neck Flexible Bags, Flagons, etc. Yes Varies; typically 1-10L Animal hide or bladder; canvas, rubber, plastic, etc./ Varies/Varies Yes Elliptical, oval or rectangular/Light/High Storage Drum or Barrel No Varies, often 200 L (55 gal.) Metal/Moderate/High No Cylindrical/Heavy/Low Cistern or Basin No, typically Varies; often large (>200L) Varies: concrete, metal, clay/ Low- moderate/High Often No Cylindrical; Rectangular/ Heavy/Low Plastic Beverage Bottle Yes, if cap is available Usually 1-2 L Plastic/ High-Moderate/Varies by type of plastic and use conditions Yes, narrow mouth Cylindrical/ Light/High Jerry Can Yes Usually 4-40 L Metal; Plastic/ Medium/varies Yes, narrow mouth Rectangular/Light/High CDC Vessel Yes 20L Plastic/ High/High for chlorina-tion Treatment; low for solar Treatment Yes, spigot Rectangular/ Light/High Oxfam Vessel (a) Yes 14L Plastic/ High/High for chlorin-ation Treatment; low for solar Treatment Yes, spigot Cylindrical/ Light/High
(a) Oxfam vessel is used primarily for emergency water storage and delivery. But, vessels of similar size and shape have been used for household water collection and storage worldwide.
The most desirable water storage vessels for many household treatment and storage options are: (1) between 10-25 liters capacity, rectangular or cylindrical with one or more handles and flat bottoms for portability and ease of storage, (2) made of lightweight, oxidation-resistant plastic, such as high-density polyethylene or polypropylene, for durability and shock resistance, (3) fitted with a 6-9 cm screw-cap opening to facilitate cleaning, but small enough to discourage or prevent the introduction of hands or dipping utensils, (4) fitted with a durable, protected and easily closed spigot or spout for dispensing water, and (5) provided with pictorial and/or written instructions for use affixed permanently to the container, as well as an affixed certificate of approval or authenticity. The cost of water storage vessels is also an important consideration, as they must be affordable or be subsidized. Locally available buckets, pots, urns, jerry cans, barrels, used beverage containers and flexible bags and flagons are usually low in cost and readily available. However, only some of these, in particular jerry cans, some plastic beverage containers, some urns and some flexible vessels, have properties and characteristics that are preferred or desirable as readily transported water storage vessels. Others, such as some buckets, cooking pots, some plastic beverage containers and other cylindrical vessels are less desirable for household water storage, but may be suitable for water collection and transport, especially if they are lightweight, have protective lids and are composed of easily cleaned materials (e.g., plastics).
Another consideration of household water storage vessels is their compatibility with household water treatment methods. In some cases, water treatment takes place in the collection and storage vessel or the treated water is delivered to the storage vessel. The design and composition of the vessel should be compatible with these tasks and also protect water quality. In some household water treatment systems, multiple containers are needed, for example, one for raw, untreated water and another for treated water. The materials of which the vessel is composed must be compatible with the physical and chemical agents used for water treatment. In the case of treatment chemicals, such as oxidant disinfectants (e.g., chlorine), the vessel must not exert excessive oxidant demand or result in chemical reactions forming excessive concentrations of toxic disinfection by-products. In the case of solar or heat treatments, the vessel must be capable of withstanding high temperatures, and depending on the type of solar treatment, they must allow the penetration of UV radiation and/or the absorption of heat energy.
Overall, the properties of household water collection, treatment and storage vessels must be compatible with the intended uses (collection, treatment and storage), meet the daily water volume needs of the household, be practical and manageable for the users (women, men or children) and be socio-culturally acceptable.
3. Water Treatment Methods - Overview and Historical Perspective
The various physical and chemical methods for water treatment at the household level or point-of-use are summarized in Tables 3 and 4, respectively. These methods are listed along with categorizations (listed as high medium and low) of their availability and practicality, technical difficulty, cost and microbial efficacy. Availability, practicality and technical difficulty are considered on a worldwide basis, including availability, practicality and technical difficulty for use at the household level. Cost is categorized as low, medium and high on a worldwide basis including the poorest people. Categories for annual household cost estimates in US dollars are less than $10 for low, >$10-100 for moderate and >$100 for high. Clearly, these cost categories will be different for different economic situations in different regions and countries of the world. The categories for microbial efficacy are based on estimated order-of-magnitude or log10 reductions of waterborne microbes by the treatment technology. The categories are <1 log10 (<90%) is low, 1 to 2 log10 (90-99%) is moderate and >2 log10 (>99% is high). The values of these categories also may differ in different situations and settings, but they are intended to distinguish among the various water treatment technologies available for use at the household level. On this basis, clear differences are discernable in the available candidate technologies for household water treatment.
Most of the methods or processes to purify water and make it safe for drinking and other potable purposes can be historically traced to ancient versions of them used since recorded history (Baker, 1948; Jahn, 1980). The practice of many of these water purification methods since ancient times has been documented by pictorial, written and archaeological records from a variety of original sources and recounted by scholars and historians of water treatment and water quality (Baker, 1948). Although the ancients may not have been aware of how such treatments improved the microbiological quality of water, they apparently were aware of and appreciated the benefits of these methods in making the water more healthful by reducing disease and improving its aesthetic qualities. Recorded in ancient history are the physical methods of sedimentation, filtration, boiling or heating, and exposure to sunlight (UV irradiation and heating), and the chemical methods of coagulation or adsorption with alum, lime, and plant extracts, adsorption with carbon (charcoal), clay and plant materials, and exposure to germicidal metals such as silver and copper. However, the development and use of chlorine and other chemical oxidants, such as ozone and chlorine dioxide, for water disinfection are more recent developments, dating back only to the mid-nineteenth century or later, when modern chemistry emerged as a science. Two of the earliest methods of generating chlorine, electrolyzing brine (NaCl) to produce sodium hypochlorite and reacting lime with chlorine gas to produce bleaching powder (calcium hypochlorite), are still widely used today. They are the basis for some of the most promising systems to produce chlorine for water treatment at the household level.
Most of the physical and chemical methods for on-site or point-of-use treatment of household water in developing countries are also employed in developed countries, either at point-of-use or in community (municipal) water treatment systems, using the same or similar technologies (AWWA, 1999; LeChevallier and Au, 2000). In developed countries, a number of point-of-use treatment technologies not widely employed in community water systems also have been employed, including various filters, adsorbents, ion exchange resins and softeners (Geldreich and Reasoner, 1990). Key differences in the application of these technologies in developing countries compared to developed countries are in the availability and affordability of the materials and the need to adapt the technologies to local conditions and personal or community preferences. Furthermore, point-of-use or point-of-entry treatment devices or systems in developed countries are often being applied to waters already subjected to extensive treatment, including disinfection, or withdrawn from high quality water sources. Hence, such waters are already likely to be relatively safe or low risk with respect to microbial quality and waterborne disease risks without point-of-use or point-of-entry treatment. In many developing countries as well as in many settings in developed countries, point-of-use, point-of-entry and household treatment often must be applied to water that is microbiologically contaminated. Therefore, the treatment requirements to achieve acceptable microbiological quality can be substantial and only some technologies or unit process will be capable of meeting this objective.
Table 3. Physical Methods for Water Treatment at the Household Level
Method (a) Availability and Practicality Technical Difficulty Costa Microbial Efficacy (b) Boiling or heating with fuels Varies (c) Low-Moderate Varies (c) High Exposure to Sunlight High Low-Moderate Low Moderate UV Irradiation (lamps) Varies (d) Low-moderate Moderate-high (d) High Plain Sedimentation High Low Low Low Filtration (e) Varies (e) Low-Moderate Varies (e) Varies (f) Aeration Moderate Low Low Low (g)
(a) Categories for annual household cost estimates in US dollars are less than $10 for low, >$10-100 for moderate and >$100 for high.
(b) Categories for microbial efficacy are based on estimated order-of-magnitude or log10 reductions of waterborne microbes by the treatment technology. The categories are <1 log10 (<90%) is low, 1 to 2 log10 (90-99%) is moderate and >2 log10 (>99% is high).
(c) Depends on heating method as well as availability and cost of fuels, which range from low to high.
(d) Depends on availability of and type of lamps, housings, availability and cost of electricity, as well as operation and maintenance needs (pumps and system cleaning methods).
(e) Different filtration technologies are available. Some (e.g., membrane filtration) are recommended for emergency water treatment). Practicality, availability, cost and microbial efficacy depend on the filter medium and its availability: granular, ceramic, fabric, etc.
(f) Depends on pore size and other properties of the filter medium, which vary widely. Some are highly efficient (>>99% or >>2log10) for microbial removals.
(g) Aeration (oxygenation) may have synergistic effects with other water treatments, such as solar disinfection with sunlight or with other processes that may oxidize molecular oxygen.
Table 4. Chemical or Physical-Chemical Methods for Water Treatment at the Household Level
Method Availability and Practicality Technical Difficulty Cost (a) Microbial Efficacy (b) Coagulation-Flocculation or Precipitation Moderate Moderate Varies Varies (c) Adsorption (charcoal, carbon, clay, etc.) High to moderate Low to moderate Varies Varies with adsorbent (d) Ion exchange Low to Moderate Moderate to high Usually High Low or moderate Chlorination High to Moderate Low to Moderate Moderate High Ozonation Low High High High Chlorine Dioxide Low Varies (e) High High Iodination (elemental, salt or resin) Low Moderate to High High High Acid or base treatment with citrus juice, hydroxide salts, etc. High Low Varies Varies Silver or Copper High Low Low Low Combined systems: chemical coagulation-flocculation, filtration, chemical disinfection Low to Moderate Moderate to High High High
(a) See footnote to Table 3.
(b) See footnote to Table 3
(c) Varies with coagulant, dose, mixing and settling conditions and pH range.
(d) Microbial adsorption efficiency is low for charcoal and carbon and high for some clays.
(e) On-site generation of gas is difficult but chemical production by acidifying chlorate or chlorite is simple if measuring devices and instructions are provided.