LITERATURE REVIEW Chapter 2

LITERATURE REVIEW Chapter 2
Chapter 2
LITERATURE REVIEW
2.1
INTRODUCTION
The United Nations (UN) set a goal in their Millennium Declaration to reduce the
amount of people without safe drinking water by half in the year 2015 (UN, 2000). Safe
drinking water for human consumption should be free from pathogens such as bacteria,
viruses and protozoan parasites, meet the standard guidelines for taste, odour,
appearance and chemical concentrations, and must be available in adequate quantities
for domestic purposes (Kirkwood, 1998).
However, inadequate sanitation and
persistent faecal contamination of water sources is responsible for a large percentage of
people in both developed and developing countries not having access to
microbiologically safe drinking water and suffering from diarrhoeal diseases (WHO,
2002a; WHO, 2002b).
Diarrhoeal diseases are responsible for approximately 2.5
million deaths annually in developing countries, affecting children younger than five
years, especially those in areas devoid of access to potable water supply and sanitation
(Kosek et al., 2003; Obi et al., 2003; Lin et al., 2004; Obi et al., 2004).
Political upheaval, high numbers of refugees in some developing countries, and the
global appearances of squatter camps and shanty rural towns, which lack proper
sanitation and water connections, have contributed to conditions under which disease
causing microorganisms can replicate and thrive (Leclerc et al., 2002; Sobsey, 2002;
Theron and Cloete, 2002). The people most susceptible to waterborne diseases include
young children, the elderly, people suffering from malnutrition, pregnant woman,
immunocompromised individuals, people suffering from chemical dependencies and
persons predisposed to other illnesses like diabetes (Sobsey et al., 1993; Gerba et al.,
1996; Grabow, 1996; Leclerc et al., 2002; Theron and Cloete, 2002). Furthermore, an
increasing number of people are becoming susceptible to infections with specific
pathogens due to the indiscriminate use of antimicrobial drugs, which have lead to the
selection of antibiotic resistant bacteria and drug resistant protozoa (WHO, 2002c;
NRC, 2004).
Chapter 2
5
In developing countries, many people are living in rural communities and have to
collect their drinking water some distances away from the household and transport it
back in various types of containers (Sobsey, 2002). Microbiological contamination of
the water may occur between the collection point and the point-of-use in the household
due to unhygienic practices causing the water to become a health risk (Sobsey, 2002;
Gundry et al., 2004; Moyo et al., 2004).
To improve and protect the microbiological quality and to reduce the potential health
risk of water to these households, intervention strategies is needed that is easy to use,
effective, affordable, functional and sustainable (CDC, 2001; Sobsey, 2002). Many
different water collection and storage systems have been developed and evaluated in the
laboratory and under field conditions (Sobsey, 2002). In addition, a variety of physical
and chemical treatment methods to improve the microbiological quality of water are
available (Sobsey, 2002). The aim of this study was to improve the microbiological
quality of drinking water in rural households by the implementation of intervention
strategies which include the use of traditional storage containers as well as the CDC safe
storage container, with or without the addition of a sodium hypochlorite solution at the
point-of-use.
2.2
WATERBORNE DISEASES
Many infectious diseases are associated with faecally contaminated water and are a
major cause of morbidity and mortality worldwide (Leclerc et al., 2002; Theron and
Cloete, 2002). Waterborne diseases are caused by enteric pathogens such as bacteria,
viruses and parasites (Table 2.1) that are transmitted by the faecal oral route (Grabow,
1996; Leclerc et al., 2002; Theron and Cloete, 2002). Waterborne spread of infection
by these pathogenic microorganisms depends on several factors such as: the survival of
these microorganisms in the water environment, the infectious dose of the
microorganisms required to cause a disease in susceptible individuals, the
microbiological and physico-chemical quality of the water, the presence or absence of
water treatment and the season of the year (Deetz et al., 1984; Leclerc et al., 2002;
Theron and Cloete, 2002).
Chapter 2
6
Table 2.1
Waterborne pathogens and their associated diseases (Bifulco et al., 1989;
Grabow, 1996; WHO, 1996a; Guerrant, 1997; Leclerc et al., 2002; Theron
and Cloete, 2002; Yatsuyanagi et al., 2003; NRC, 2004)
Pathogen
Diseases
Campylobacter spp.
Diarrhoea and acute gastroenteritis
Enteropathogenic Escherichia coli
Diarrhoea
Escherichia coli O157:H7
Bloody diarrhoea and haemolytic uremic
syndrome
Salmonella spp.
Typhoid fever, diarrhoea
Shigella spp.
Dysentery, diarrhoea
Vibrio cholera
Cholera, diarrhoea
Yersinia spp.
Diarrhoea, gastrointestinal infections
Adenoviruses
Diarrhoea, respiratory disease, conjunctivitis
Astroviruses
Diarrhoea
Coxsackie viruses (Enterovirus)
Respiratory, meningitis, diabetes, diarrhoea,
vomiting, skin rashes
Echoviruses (Enterovirus)
Meningitis, diarrhoea, myocarditis
Enteroviruses 68-71
Meningitis, diarrhoea, respiratory diseases,
rash,
acute
enteroviral
haemorrhagic
conjunctivitis
Hepatitis viruses (A, E)
Hepatitis (jaundice), gastroenteritis
Caliciviruses
Diarrhoea, vomiting
Poliovirus (Enterovirus)
Poliomyelitis
Rotaviruses
Diarrhoea, vomiting
Small Round Structured viruses
Diarrhoea, vomiting
Cryptosporidium parvum
Cryptosporidiosis, diarrhoea
Entamoeba hystolytica
Amoebic dysentery
Giardia
Giardiasis, diarrhoea
Helminths
Dracunalis medinensis
Guinea worm (Dracunculiasis)
Emerging
opportunistic
pathogens
Actinobacter spp.
Septicemia, meningitis, endocarditis
Aeromonas spp.
Diarrhoea, gastroenteritis
Cyclospora spp.
Diarrhoea, abdominal cramping, fever
Isospora spp.
Diarrhoea
Legionella spp.
Legionnaires disease, Pontiac fever
Microsporidia spp.
Gastrointestinal infections, diarrhoea
Nontuberculosis Mycobacteria
Skin infections, cervical lymphadenitis,
nontuberculosis mycobacterium disease
Pseudomonas aeruginosa
Septicaemia, wound and eye infections
Bacteria
Viruses
Protozoan
parasites
Chapter 2
7
The survival of microorganisms such as bacteria in water environments depends on the
presence of nutrients and the water temperature (Edberg et al., 2000; Leclerc et al.,
2002). The infectious dose of some bacteria range between 107 to 108 cells, with some
enteric bacteria able to cause infections at doses as low as 101 cells (Edberg et al., 2000;
Leclerc et al., 2002). Viruses cannot replicate outside living cells, but can survive for
extended periods in the water (Raphael et al., 1985; Leclerc et al., 2002).
The
infectious dose of viruses has been established to be as low as 1 to 10 infectious
particles (Raphael et al., 1985; Leclerc et al., 2002). Enteric protozoa such as Giardia
and Cryptosporidium cannot replicate in water and are highly resistant to most
disinfectants and antiseptics used for water treatment (Leclerc et al., 2002; Masago et
al., 2002). The infectious dose for parasites depends on host susceptibility and strain
virulence (Leclerc et al., 2002; Masago et al., 2002). The infectious dose for Giardia
might be as low as 10 oocysts and for Cryptosporidium the presence of 30 oocysts
might cause an infection (Leclerc et al., 2002; Masago et al., 2002; Carlsson, 2003).
Although waterborne pathogens are distributed worldwide, outbreaks of cholera,
Hepatitis E and Dracunculiasis tend to be subjected to geographical factors (Sacks et
al., 1986; Alarly and Nadeau, 1990; Kukula et al., 1997; Kukula et al., 1999; Hänninen
et al., 2003; Hrudey et al., 2003). In the last number of years several outbreaks of
pathogenic diseases have appeared that cannot be prevented by traditional water
treatment. In 1981 a community waterborne outbreak in Colorado, USA, could be
traced to Rotavirus (Hopkins et al., 1984). In 1983 and in 1987 two community
outbreaks of waterborne Campylobacter spp were reported in the USA and Canada,
respectively (Sacks et al., 1986; Alarly and Nadeau, 1990). In 1993 in Milwaukee,
USA, 400 000 people fell ill with 54 deaths from using drinking water that was
contaminated by Cryptosporidium cysts (Hoxie et al., 1997).
In 1998, Calici-like
viruses in municipal water were responsible for an acute gastroenteritis outbreak in
Heinävesi, Finland, affecting approximately 3 000 people (Kukkula et al., 1997;
Kukkula et al., 1999). In 2000, E. coli O157:H7 was responsible for 2 300 people
falling ill in Walkerton, Canada (Hrudey et al., 2003). Recent flooding in Bangladesh
has lead to 67 718 reported cases of diarrhoea and 9 people died due to waterborne
diseases (International Water Association, 2004)
Chapter 2
8
Consequently, during the past 5 years in rural communities in South Africa, severe
outbreaks of cholera in the KwaZulu Natal, Limpopo, Eastern Cape and Mpumalanga
have been reported with confirmed cases of mortality (DOH, 2000; DOH, 2002; DOH,
2003; NICD, 2004a; NICD, 2004b). In addition, typhoid cases have been reported in
the Limpopo and the Mpumalanga Provinces during 2004 and 2005 with cases of
mortality (NICD, 2004b).
Rotaviruses have been found during 2005 to be the
responsible agent in a large outbreak of watery diarrhoea in the Northern Cape (Laprap,
2005). A report compiled by the Department of Water Affairs and Forestry (DWAF)
focussed on the waterborne diseases currently reported in South Africa by the
Department of Health (DOH), the National Laboratory Services, DWAF and Rand
Water (DWAF, 2005). In summary this report found that records in some provinces are
not well kept and although information on waterborne diseases such as Hepatitis A,
Shigella spp, cholera and typhoid fever is available, it is not reported. The report found
that the number of people infected with Hepatitis A in South Africa was 231 in 2003
and 9 503 in 2004 indicating an increase in the rate of infection (DWAF, 2005). The
report further showed that during 2003, 761 people and during 2004, 894 people were
infected with Shigella spp.
However, the data for Shigella spp are underreported
because it is not on the list of notifiable diseases (DWAF, 2005). All these statistics
confirm the need for the implementation of a national surveillance system to monitor
waterborne disease outbreaks in South Africa.
2.3
THE MICROBIOLOGICAL QUALITY OF WATER
Water supplies in developing countries are devoid of treatment and the communities
have to make use of the most convenient supply (Sobsey, 2002; Moyo et al., 2004).
Many of these water supplies are unprotected and susceptible to external contamination
from surface runoff, windblown debris, human and animal faecal pollution and
unsanitary collection methods (Chidavaenzi et al., 1998; WHO, 2000; Moyo et al.,
2004).
Detection of each pathogenic microorganism in water is technically difficult, time
consuming and expensive and therefore not used for routine water testing procedures
(Grabow, 1996).
Chapter 2
Instead, indicator organisms are routinely used to assess the
9
microbiological quality of water and provide an easy, rapid and reliable indication of the
microbiological quality of water supplies (Grabow, 1996).
In order for a microorganism to be used as an indicator organism of pollution, the
following requirements should be fulfilled (Grabow, 1986; WHO, 1993; NRC, 2004):
•
The concentration of the indicator microorganism should have a quantitative
relationship to risk of disease associated with exposure (ingestion/recreational
contact) to the water;
•
The indicator organism should be present when pathogens are present;
•
The persistence and growth characteristics of the indicator organism should be
similar to that of pathogens;
•
Indicator organisms should not reproduce in the environment;
•
The indicator organism should be present in higher numbers than pathogens in
contaminated water;
•
The indicator organism should be at least as resistant to adverse environmental
conditions, disinfection and other water treatment processes as pathogens;
•
The indicator organism should be non-pathogenic and easy to quantify;
•
The tests for the indicator organism should be easy, rapid, inexpensive, precise,
have adequate sensitivity, quantifiable and applicable to all types of water;
•
The indicator organism should be specific to a faecal source or identifiable as to the
source of origin of faecal pollution.
Although many microorganisms have desirable features to be considered as possible
indicators of faecal pollution, there is no single microorganism that meets all of these
requirements (Moe et al., 1991; Payment and Franco, 1993; Sobsey et al., 1993; Sobsey
et al., 1995). Several studies have showed the limitations of some of the current
indicator organisms, which include the following:
•
Indicator organisms may be detected in water samples in the absence of
pathogens (Echeverria et al., 1987).
•
Some pathogens may be detected in the absence of indicator organisms
(Seligman and Reitler, 1965; Thompson, 1981). Echeverria and co-workers
Chapter 2
10
(1987) have showed that Vibrio cholera (V. cholera) persists in water exposed to
solar disinfection well after E. coli was inactivated. El-Agaby and co-workers
(1988) have showed that potable water supplies in Egypt contained
bacteriophages, with zero total and faecal coliform counts, which indicated the
possible risk of the presence of human enteric viruses.
•
Thompson (1981) has showed that E. coli bacteria have a short die-off curve
with temperature playing an important role.
•
McFeters and co-workers (1986) have showed that injured coliform bacteria can
be undetected due to several chemical and physical factors and were unable to
grow on commonly used media.
•
LeChevallier and co-workers (1996) have showed that improper filtration,
temperature, inadequate disinfection and treatment procedures, biofilms and
high assimilable organic carbon (AOC) levels, could all be responsible for the
regrowth of coliform bacteria in water samples.
•
Regli and co-workers (1991) and Hot and co-workers (2003) have showed that
the prevalence of viruses in water may differ from that of indicator organisms.
Low numbers of viruses are present in water samples compared to indicator
organisms, viruses are only excreted for short periods of time while coliform
bacteria is excreted continuously, and the structure, size, composition and
morphological differences between viruses and bacteria also had an influence on
behavioural and survival patterns of these microorganisms (Regli et al., 1991;
Hot et al., 2003).
In spite of the shortcomings of indicator microorganisms, it is better to use a
combination of indicator microorganisms to give a more accurate picture of the
microbiological quality of water (DWAF, 1996; NRC, 2004). In general, every country
has its own set of guidelines for drinking water. However, most of these guidelines are
similar for different countries and the same indicator microorganisms to indicate the
presence of pathogenic microorganisms are used. The water quality guidelines for
South Africa are shown in Table 2.2.
Chapter 2
11
Table 2.2
Microbiological requirements for domestic water in South Africa
(Kempster et al., 1997; SABS, 2001)
Indicator organism
Units
Allowable
compliance
Heterotrophic plate count
Colony forming units.1 ml-1
100
Total coliform bacteria
Colony forming units.100 ml-1
10
Faecal coliform bacteria
Colony forming units.100 ml-1
1
-1
Escherichia coli
Colony forming units.100 ml
0
Somatic bacteriophages
Colony forming units.10 ml-1
1
Enteric viruses
Plaque forming units.100 l-1
1
Protozoan parasites (Giardia/Cryptosporidium)
Count.100 l-1
0
The most commonly used indicator microorganisms include heterotrophic plate counts,
total coliform bacteria, faecal coliform bacteria, E coli, faecal enterococci, C.
perfringens as well as somatic and male specific F-RNA bacteriophages (WHO, 2000).
Each of these indicator microorganisms has advantages and disadvantages which will be
discussed in more detail in the following sections.
2.3.1
Heterotrophic plate counts
Heterotrophic microorganisms or heterotrophs are naturally present in the environment
and can be found in soil, sediment, food, water and in human and animal faeces (Collin
et al., 1988; Olson et al., 1991; Standard Methods, 1995; Lillis and Bissonnette, 2001).
Broadly defined, heterotrophs include bacteria, yeasts and molds that require organic
carbon for growth (WHO, 2002c). Although generally considered harmless, some
heterotrophic microorganisms are opportunistic pathogens, which have virulence factors
that could affect the health of consumers with suppressed immune systems (Lye and
Dufour, 1991; Bartram et al., 2003). Heterotrophic microorganisms can also survive in
biofilms inside water distribution systems, water reservoirs and inside household
storage containers (Momba and Kaleni, 2002; Jagals et al., 2003).
Therefore,
heterotrophic plate counts can also be used to measure the re-growth of organisms that
may or may not be a health risk (WHO, 2002c).
Chapter 2
12
Heterotrophic Plate Count, also known as Total or Standard Plate Count includes simple
culture based tests intended to recover a wide range of heterotrophic microorganisms
from water environments (Bartram et al., 2003). Enumeration tests for heterotrophic
plate counts are simple and inexpensive giving results within 48 h to 5 days, depending
on the method, type of media and the incubation temperature used (Collin et al., 1988;
Olson et al., 1991; Standard Methods, 1995; Lillis and Bissonnette, 2001). The pour
plate, membrane filtration or spread plate methods are used routinely in various
laboratories, with either Yeast-extract agar, Plate Count Agar (PCA), Tryptone Glucose
agar or R2A agar, and incubation periods either at room temperature (25ºC) for 5 to 7
days, or at 35°C to 37°C for 48 h (Collin et al., 1988; Olson et al., 1991; Standard
Methods, 1995; Lillis and Bissonnette, 2001). Heterotrophic plate counts alone cannot
indicate a health risk and additional studies on the presence of E. coli or other faecal
specific indicator microorganisms need to be conducted to establish the potential health
risk of the water analysed (WHO, 2002c).
2.3.2
Total coliform bacteria
Total coliform bacteria are defined as aerobic or facultative anaerobic, Gram negative,
non-spore forming, rod shaped bacteria, which ferments lactose and produce gas at
35°C (Standard Methods, 1995). Total coliforms include bacteria of known faecal
origin such as E. coli as well as bacteria that may not be of faecal origin such as
Klebsiella spp, Citrobacter spp, Serratia spp and Enterobacter spp which are found in
nutrient rich water, soil decaying vegetation and drinking water with relatively high
levels of nutrients (Pinfold, 1990; Ramteke et al., 1992; WHO, 1996a).
The
recommended test for the enumeration of total coliforms is membrane filtration using
mEndo agar and incubation at 35°C to 37°C for 24 h to produce colonies with goldengreen metallic shine (Standard Methods, 1995).
In water quality studies, total coliform bacteria are used as a systems indicator, which
provides information on the efficiency of water treatment (Standard Methods, 1995).
The presence of total coliform in water samples are therefore, an indication that
opportunistic pathogenic bacteria such as Klebsiella and Enterobacter which can
multiply in water environments and pathogenic pathogens such as Salmonella spp,
Chapter 2
13
Shigella spp, V. cholera, Campylobacter jejuni, Campylobacter coli, Yersinia
enterocolitica and pathogenic E. coli may be present (DWAF, 1996; Grabow, 1996).
These pathogens and opportunistic microorganisms could cause diseases such as
gastroenteritis, dysentery, cholera, typhoid fever and salmonellosis to consumers
(DWAF, 1996; Grabow, 1996). In particular, individuals who suffer from HIV/AIDS
related complications are more at risk of being infected by these microorganisms
(DWAF, 1996).
2.3.3
Faecal coliform bacteria
Faecal coliform bacteria are Gram negative bacteria, also known as thermotolerant
coliforms or presumptive E. coli (Standard Methods, 1995). The faecal coliform group
includes other organisms, such as Klebsiella spp, Enterobacter spp and Citrobacter spp,
which are not exclusively of faecal origin (Standard Methods, 1995). Escherichia coli
are specifically of faecal origin from birds, humans and other warm blooded animals
(WHO, 1996a; Maier et al., 2000). Faecal coliform bacteria are therefore considered to
be a more specific indicator of the presence of faeces (Maier et al., 2000).
The recommended test for the enumeration of faecal coliforms is membrane filtration
using mFC agar and incubation at 44.5°C for 24 h to produce blue colored colonies
(Standard Methods, 1995). Faecal coliforms are generally used to indicate unacceptable
microbial water quality and could be used as an indicator in the place of E. coli (SABS,
2001). The presence of faecal coliforms in a water sample indicates the possible
presence of other pathogenic bacteria such as Salmonella spp, Shigella spp, pathogenic
E. coli, V. cholera, Klebsiella spp and Campylobacter spp associated with waterborne
diseases (DWAF, 1996).
Unfortunately faecal coliform bacteria exhibit species to
species variations in their respective stability and resistance to disinfection processes;
do not distinguish between faeces of human and animals origin; have low survival rates
and have been detected in water sources thought to be free of faecal pollution (Goyal et
al., 1979; Fujioka et al., 1988).
Chapter 2
14
2.3.4
Escherichia coli bacteria
Globally E. coli is used as the preferred indicator of faecal pollution (Edberg et al.,
2000).
It is a Gram negative bacterium and predominantly an inhabitant of the
intestines of warm blooded animals and humans, which is used to indicate recent faecal
pollution of water samples (Rice et al., 1990; Rice et al., 1991; WHO, 1996a; Edberg et
al., 2000). Confirmation tests for E. coli include testing for the presence of the enzyme
β-glucuronidase, Gram staining, absence of urease activity, production of acid and gas
from lactose and indole production (Mac Faddin, 1980; Rice et al., 1991; Standard
Methods, 1995).
Commercially available growth media containing the fluorogenic substrate 4-methylumbelliferyl-β-D-glucuronidase (MUG) is used for the isolation and identification of E.
coli from water samples (Shadix and Rice, 1991; Covert et al., 1992). The E. coli
bacteria hydrolyse the MUG in the media, which then fluoresces under ultraviolet light
(Shadix and Rice, 1991; Covert et al., 1992). However, false negative results on this
media have been found due to injured cells, lack of expression of the gene which codes
for the enzyme β-glucuronidase by the E. coli bacterium isolate, and non-utilization of
the MUG reagent in the media by some E. coli strains (Chang et al., 1989; Feng et al.,
1991; NRC, 2004).
2.3.5
Faecal enterococci bacteria
Faecal enterococci bacteria are found in the genus Enterococcus and include species
like Enterococcus faecalis, Enterococcus faecium, Enterococcus durans and
Enterococcus hirae (Standard Methods, 1995; WHO, 1996a). The genus Enterococcus
are differentiated from the genus Streptococcus by their ability to grow in 6.5% sodium
chloride, pH 9.6, temperatures of 45ºC and their tolerance for adverse growth conditions
(Maier et al., 2000). Faecal enterococci are spherical, Gram positive bacteria, which are
highly specific for human and animal faecal pollution (Standard Methods, 1995). Most
of the species in the Enterococcus genus are of faecal origin and is regarded as specific
indicators of human faecal pollution, although some species are found in the faeces of
animals and plant material (WHO, 1996a).
Chapter 2
15
The recommended test is membrane filtration using mEnterococcus agar and incubation
at 35°C to 37°C for 48 h to produce pink colonies (Standard Methods, 1995). Faecal
enterococci rarely multiply in polluted water environments and are more resistant to
disinfection and treatment processes than the Gram negative faecal coliform bacteria
(Standard Methods, 1995). The presence of faecal enterococci in water samples are
therefore, an indication of the health risk to waterborne diseases such as meningitis,
endocarditis and infections of the eyes, ears and skin (DWAF, 1996; Grabow, 1996).
2.3.6
Clostridium perfringens bacteria
Clostridium perfringens is a Gram positive, sulphite reducing anaerobic, rod shaped,
spore forming bacteria normally present in faeces of humans and warm blooded animals
(Standard Methods, 1995). However, C. perfringens are also found in soil and water
environments (WHO, 1996a).
The spores can survive much longer than coliform
bacteria and are highly resistant to water disinfection and treatment processes (Standard
Methods, 1995). Clostridium perfringens are therefore used as an indicator of faecal
pollution to indicate the potential presence of enteric viruses, which may include
Enteroviruses, Adenoviruses and Hepatitis viruses as well as the cysts and oocysts of
protozoan parasites such as Giardia, Entamoeba and Cryptosporidium in treated
drinking water (Payment and Franco, 1993). The enumeration test includes membrane
filtration using specific medium (e.g. mCP or Perfringens selective OPSP medium with
supplements) and incubation 35°C to 37°C for 48 h at in micro-aerophillic conditions to
produce black colonies (Standard Methods, 1995).
2.3.7
Bacteriophages
Bacteriophages are viruses, which specifically infect bacteria (Grabow, 2001).
Bacteriophages have been suggested as useful indicators to predict the potential occurrence
of enteric viruses in water (Grabow et al., 1984; Leclerc et al., 2000). The survival of
bacteriophages is affected by the densities of the host and the bacteriophages in the water
sample (Grabow, 2001). In addition, the association of the bacteriophage with solids and
the presence of organic matter in the water sample could influence the attachment of the
bacteriophages to the host bacterium (Grabow, 2001). Several studies have shown that
ultra violet light, temperature, pH of the water, and ion concentrations in the water could
Chapter 2
16
affect the survival of bacteriophages in water (Brion et al., 2002; Schaper et al., 2002b;
Allwood et al., 2003). Bacteriophages show higher resistance to environmental stress
compared to bacterial indicators such as total coliforms and faecal coliforms and assays for
bacteriophages can be conducted quickly, economically and quantitatively (Vaughn and
Metcalf, 1975; Havelaar et al., 1993). There are several bacteriophages that can be used
as indicator organisms which includes the somatic bacteriophages, Bacteroides fragilis
HSP40 bacteriophages and male specific F-RNA bacteriophages (Grabow, 2001).
2.3.7.1 Somatic bacteriophages
The somatic bacteriophages are a heterogeneous group of organisms that absorbs to
bacterial receptors for infection and replication on the cell wall of the laboratory host strain
E. coli WG5 (Leclerc et al., 2000).
Somatic bacteriophages are therefore, used as
indicators of the potential presence of enteric viruses in water (Grabow, 2001). These
bacteriophages can serve as models for the assessment of the behaviour of enteric viruses
in water treatment and disinfection processes (Grabow, 2001). The double layer plaque
assay is generally used to detect somatic bacteriophages (ISO, 2000; Mooijman et al.,
2001). However, somatic bacteriophages are not specific to E. coli, and may infect and
replicate in other species of the Enterobacteriaceae family, which includes the total
coliform group (Leclerc et al., 2000).
Somatic bacteriophages are therefore, not
considered a specific indicator for faecal pollution (Leclerc et al., 2000).
2.3.7.2 Bacteroides fragilis HSP40 bacteriophages
Bacteroides bacteria are present in high numbers in human faeces (Leclerc et al., 2000).
Bacteroides is a strict anaerobic, Gram negative, non-spore forming bacterium which is
rapidly inactivated by oxygen levels in water, and needs complex growth media with
antibiotics to inhibit the interference from other intestinal microorganisms (Leclerc et
al., 2000). The Bacteroides fragilis HSP40 bacteriophages are a relatively homogeneous
group that do not multiply in the environment (Havelaar, 1993; Jagals et al., 1995; Puig et
al., 1999). In some countries, Bacteroides fragilis HSP40 bacteriophages is present in
relatively low numbers in human faeces (Havelaar, 1993; Jagals et al., 1995; Bradley et al.,
1999; Puig et al., 1999). Although this bacteriophage has been shown to be highly
Chapter 2
17
specific for human faeces, tests are complicated and labour intensive (ISO, 2001; Sinton
et al., 1998).
2.3.7.3 Male specific F-RNA bacteriophages
The male specific F-RNA bacteriophages have small hexagonal capsomers without tails,
are approximately 30 nm long with a single RNA genome (Leclerc et al., 2000). Male
specific F-RNA bacteriophages have been recommended as useful models for
monitoring the behaviour of human enteric viruses in water treatment processes because
of their size and structure, which are similar to those of the Enteroviruses (Lewis, 1995;
Leclerc et al., 2000; Grabow, 2001). These bacteriophages are relatively resistant to
disinfectants, sunlight, heat- and water treatment processes (Leclerc et al., 2000).
Male specific F-RNA bacteriophages specifically attach to the sex pili of the host
bacterium [E. coli HS(pFamp)R or Salmonella typhimirium WG49] in temperatures
higher than 30°C (Havelaar and Hogeboom, 1984; Debartolomeis and Cabelli, 1991).
The F-pilli are short tube-like protrusions produced by certain bacteria for the transfer of
nucleic acid to other bacteria of the same or closely related species and are only produced
by the bacteria in the log growth phase which is usually above 30ºC (Havelaar et al., 1993;
Woody and Cliver, 1995).
These bacteriophages are assayed according to an
International Standardization Method (ISO, 1995; Mooijman et al., 2002).
Male
specific F-RNA bacteriophages belong to the family Leviviridae, which contains two
genera, the Leviviridae and the Alloleviviridae. Both these genera contain distinct
subgroups (Watanabe et al., 1967; Furuse et al., 1979), which is useful in genotyping
assays where specific probes are used to distinguish between animal (subgroups I and
IV) and human (subgroups II and III) faecal pollution (Osawa et al., 1981; Furuse,
1987; Beekwilder et al., 1996).
2.4
HUMAN AND ANIMAL FAECAL POLLUTION IN WATER
Water polluted with human and animal faeces may contain potentially pathogenic
microorganisms that can cause diseases in consumers (Sobsey et al., 1993; Gerba et al.,
1996; Grabow, 1996; Leclerc et al., 2002; Theron and Cloete, 2002).
The most
commonly used faecal indicator microorganisms which include the total coliform
Chapter 2
18
bacteria, thermotolerant coliform bacteria, E. coli and faecal enterococci bacteria, are
found in both human and animal faeces, but do not differentiate between the origins of
faecal pollution (Sinton et al., 1998). Human viral pathogens such as Calicivirus,
Hepatitis E virus, Reoviruses, Rotaviruses, somatic bacteriophages and male specific FRNA bacteriophages also infect other animals which can serve as reservoirs (NRC,
2004).
Consequently, these animals can be important potential sources of
contamination of water sources because the release of microorganisms into aquatic
environments by animal hosts could lead to human exposure (NRC, 2004).
Poor
communities in developing countries share their water sources with cattle and other
domestic animals, therefore, the risk of waterborne transmission of zoonotic pathogens
to humans, increases (Pournadeali and Tayback, 1980; Meslin, 1997; Sinton et al.,
1998; Franzen and Muller, 1999; Slifko et al., 2000; Enriquez et al., 2001; Hoar et al.,
2001; Leclerc et al., 2002; Theron and Cloete, 2002; Hackett and Lappin, 2003).
However, water contaminated with human faeces is regarded as a greater risk to human
health since it is more likely that it would contain human specific enteric pathogens
(Sinton et al., 1998). Although various microbial and chemical indicators have been
described to identify the origin of faecal pollution in water supplies, different levels of
success have been obtained (Sinton et al., 1998; Gilpen et al., 2002; Gilpen et al.,
2003).
2.4.1
The use of microorganisms to determine the origin of faecal pollution
Several microorganisms have been suggested and tested to distinguish between human
and animal faecal pollution in domestic drinking water supplies (Wheather et al., 1980;
Mara and Oragui, 1985; Tartera and Jofre, 1987; Gavini et al., 1991; Arango and Long,
1998; Sinton et al., 1998; Gilpen et al., 2002). Various factors can have an effect on the
specificity of microorganisms that can be used as indicators to determine the origin of
faecal pollution, such as: (1) specific bacteria, viruses and protozoan parasites can have
multiple hosts (not species specific) (Sinton et al., 1998; Gilpen et al., 2002); (2)
different microorganisms can have similar biochemical reactions in the environment,
especially within the same species or genus (Sinton et al., 1998; Gilpen et al., 2002) and
(3) interspecies gene transfer may occur which include small pieces of DNA (eg.
plasmids and integrons) and transposons that are carried from one bacteria to another
Chapter 2
19
during sexual and asexual reproduction of bacterial cells (Sinton et al., 1998; Gilpen et
al., 2002).
Microorganisms that have been used in assays to determine the origin of faecal
pollution include total coliforms, faecal coliforms, faecal streptococci/enterococci,
Bacteroides spp, Bacteroides fragilis HSP40 bacteriophages, Pseudonomas aeruginosa,
Bifidobacterium spp, Rhodococcus coprophilus, male specific F-RNA bacteriophages
and specific human enteric viruses (Wheather et al., 1980; Mara and Oragui, 1985;
Tartera and Jofre, 1987; Gavini et al., 1991; Arango and Long, 1998; Sinton et al.,
1998; Gilpen et al., 2002).
2.4.1.1 The ratio of faecal coliform bacteria to faecal streptococci bacteria
The ratio between faecal coliform (FC) and faecal streptococci/enterococci (FS) counts
in water is an old method used in several earlier studies to determine the origin of faecal
pollution (Wheather et al., 1980; Mara and Oragui, 1985; Tartera and Jofre, 1987;
Gavini et al., 1991; Arango and Long, 1998; Sinton et al., 1998; Gilpen et al., 2002).
This method is based on the fact that faecal streptococci/enterococci are more abundant
in animal faeces than in human faeces while faecal coliforms are more abundant in
human faeces than in animal faeces (Sinton et al., 1998). The test stipulates that a
FC:FS ratio greater than 4 is indicative of human faeces and a FC:FS ration of less than
7 is indicative of animal faecal pollution (Sinton et al., 1998).
The limitation of this method is the variable survival rates of some faecal streptococci
species, which make this test unreliable (Wheather et al., 1980; Mara and Oragui, 1985;
Tartera and Jofre, 1987; Gavini et al., 1991; Arango and Long, 1998; Sinton et al.,
1998; Gilpen et al., 2002). Sinton and Donnison (1994) have showed that Enterococcus
faecalis survives longer than Enterococcus faecium which survives longer than
Enterococcus durans which survives longer than Streptococcus equines and
Streptococcus bovis in water environments.
Chapter 2
20
2.4.1.2 The ratio of faecal coliform to total coliform bacteria
Faecal coliforms constitute a subset of total coliforms but grow and ferment lactose with
the production of gas and acid at 44.5°C within 24 h (DWAF, 1996). The ratio of faecal
coliforms to total coliforms is used to show the percentage of total coliforms that
comprises of faecal coliforms which comes from the gut of warm blooded animals
(Sinton et al., 1998). If the faecal coliforms to total coliforms ration exceeds 0.1 it may
suggests the presence of human faecal contamination (Sinton et al., 1998). However,
this method only shows the possibility of faecal pollution but do not distinguish
between human and animal faecal matter (Bartman and Rees, 2000).
Another
disadvantage of this assay is that some faecal coliforms can multiply in soils in tropical
regions and give a false positive result for water pollution (Bartman and Rees, 2000).
2.4.1.3 Bacteroides bacteria and Bacteroides HSP40 bacteriophages
Bacteroides bacterial species are among the numerous bacteria in human faeces and is
also found in low numbers in animal faeces (Maier et al., 2000). The bacterium does
not survive for long periods outside the human body making the detection of
Bacteroides difficult (Wheather et al., 1980; Mara and Oragui, 1985; Tartera and Jofre,
1987; Gavini et al., 1991; Arango and Long, 1998; Sinton et al., 1998; Gilpen et al.,
2002).
However, the Bacteroidis fragilis HSP40 bacteriophage strain is a highly specific
indicator for human faecal pollution (Grabow, 2001) but is only present in low numbers
in human sewage (Wheather et al., 1980; Mara and Oragui, 1985; Tartera and Jofre,
1987; Gavini et al., 1991; Arango and Long, 1998; Sinton et al., 1998; Gilpen et al.,
2002). The assays used for the Bacteroides bacteria and the Bacteroides fragilis HSP40
bacteriophages are expensive, complicated, time consuming and require specialised
equipment and skilled labour (Wheather et al., 1980; Mara and Oragui, 1985; Tartera
and Jofre, 1987; Gavini et al., 1991; Arango and Long, 1998; Sinton et al., 1998; Gilpen
et al., 2002).
Chapter 2
21
2.4.1.4 Pseudomona aeruginosa bacteria
Pseudonoma aeruginosa bacteria are present in 16% of human adults but occur rarely in
lower animals (Sinton et al., 1998; Gilpen et al., 2002). Unfortunately this bacterium is
present in water, soil and sewage samples and can rapidly die-off in aquatic
environments and is therefore not a suitable candidate to determine the source of faecal
pollution (Wheather et al., 1980; Mara and Oragui, 1985; Tartera and Jofre, 1987;
Gavini et al., 1991; Arango and Long, 1998; Sinton et al., 1998; Gilpen et al., 2002).
2.4.1.5 Bifidobacterium spp
Bifidobacteria spp are strickly anaerobic, Gram positive bacteria present in the gut of
humans and animals (Nebra et al., 2003). Species such as Bifidobacteria adolescentis
are specific to humans while species such as Bifidobacteria thermophilum are specific
to animal faeces (Nebra et al., 2003). It is difficult to differentiate between the species
based on biochemical and microbiological analysis, which complicates the
interpretation of the results (Wheather et al., 1980; Mara and Oragui, 1985; Tartera and
Jofre, 1987; Gavini et al., 1991; Arango and Long, 1998; Sinton et al., 1998; Gilpen et
al., 2002).
2.4.1.6 Rhodococcus coprophilus bacteria
Rhodococcus coprophilus is a Gram positive, aerobic nocardioform actinomycete which
forms a fungus-like mycelium that breaks up into bacteria-like pieces (Sinton et al.,
1998). The bacteria contaminate grass and when eaten by herbivores these bacteria-like
pieces are found in the herbivore dung (Jagals et al., 1995; Sinton et al., 1998).
Rhodococcus coprophilus has never been found in human faeces and is therefore used
as an indicator of animal faecal pollution (Jaggals et al., 1995). The disadvantage of
this bacterium is the long growth time of 21 days (Wheather et al., 1980; Mara and
Oragui, 1985; Tartera and Jofre, 1987; Gavini et al., 1991; Arango and Long, 1998;
Sinton et al., 1998; Gilpen et al., 2002). Saville and co-workers (2001) have designed a
PCR protocol to detect this organism in faecal specimens of animals, which showed
potential to be used as a routine laboratory test, but more studies are needed to evaluate
this detection technique.
Chapter 2
22
2.4.1.7 Male specific F-RNA bacteriophages
Male specific F-RNA bacteriophages are a homogeneous group of microorganisms
belonging to the Family Leviviridae (Leclerc et al., 2000). This family comprise of four
subgroups, those predominating in humans (groups II and III), and those predominating
in animals (groups I and IV) (Leclerc et al., 2000). Genotyping with specific probes or
serotyping with specific antisera can be used to classify male specific F-RNA
bacteriophages into one of the four distinct subgroups (Beekwilder et al., 1996). The
application of these assays makes it possible to distinguish between environmental
contaminations from human or animal faecal origin (Beekwilder et al., 1996). Grouping
is based on serological and physico-chemical properties of each subgroup (Leclerc et al.,
2000). However, antisera necessary for serotyping are expensive, not readily available
and some isolates are difficult to serotype (Furuse et al., 1978; Havelaar et al., 1986).
Genotyping of F-RNA bacteriophages are based on molecular techniques, which
include specific oligonucleotide probes and nucleic acid hybridisation (Hsu et al., 1995;
Beekwilder et al., 1996). Hsu and co-workers (1995) investigated genotyping with nonradioactive oligonucleotide probes as an alternative to serotyping for the grouping of
male specific F-RNA bacteriophages.
Beekwilder and co-workers (1996) also
described a method which identifies male specific F-RNA bacteriophages quantitatively
by a plaque hybridisation assay. Comparison of genotype and serotype results showed
that genotyping is a more effective and technically feasible method for the grouping of
male specific F-RNA bacteriophages (Hsu et al., 1995; Beekwilder et al., 1996).
Several studies have suggested that male specific F-RNA bacteriophage subgroup
classification, especially subgroups II and III that predominates in human faeces, will
not always distinguish between human and pig faecal contamination due to similar
dietary and living conditions of pigs as well as exposure of the pigs to human faecal
wastes (Osawa et al., 1981; Havelaar et al., 1990; Hsu et al., 1995). Consequently, a
small percentage of overlapping between the serotypes and their expected animal
sources were found with studies showing that animal samples might contain all 4
serotypes (NRC, 2004). In addition, Schaper and co-workers (2002) have showed that
human samples contained serotypes I and IV that is mainly associated with animal
hosts. Despite these results, various studies have used genotype and serotype analysis
Chapter 2
23
successfully to distinguish between faecal pollution of either human or animal origin
(Osawa et al., 1981; Havelaar et al., 1990; Hsu et al., 1995; Beekwilder et al., 1996;
Schaper et al., 2002a). Rose and co-workers (1997) have used reverse transcriptase
polymerase chain reaction (RT-PCR) to isolate male specific F-RNA bacteriophages
from polluted marine waters. However, a study conducted by Schaper and Jofre (2000)
comparing RT-PCR followed by southern blotting with plaque hybridisations on male
specific F-RNA bacteriophages in sewage samples, indicated that RT-PCR was less
sensitive than plaque hybridisation analysis to identify the various F-RNA
bacteriophages present in the sewage water samples. Therefore, genotyping of male
specific F-RNA bacteriophages using nucleic acid hybridisation seems to be the
microbial method of choice to distinguish between human and animal origin of faecal
pollution (Schaper and Jofre, 2000).
2.4.1.8 Human enteric viruses
Human enteric viruses associated with waterborne diseases include Adenoviruses,
Caliciviruses, Enteroviruses, Hepatitis A virus and Rotaviruses (Grabow, 2001).
Although excreted in high numbers in faeces by infected individuals, these viruses may
be present in low numbers in environmental samples due to dilution (Grabow, 2001).
The detection of specific human enteric viruses can be used to confirm the presence of
human faecal pollution (Grabow, 2001). Since the detection of viruses is mostly based
on molecular techniques, it is not a cost-effective method to include in routine
monitoring of water (Tartera and Jofre, 1987; Gavini et al., 1991; Arango and Long,
1998; Sinton et al., 1998; Gilpen et al., 2002; NRC, 2004).
Viability of viruses can also not be indicated by molecular techniques and additional
cell culture techniques should be included, thereby further increasing the cost and
labour (Grabow, 2001; Gilpen et al., 2002). However, all viruses are not able to grow
in cell cultures (Grabow, 2001). In addition these techniques are labour intensive and
skilled personnel are required (Tartera and Jofre, 1987; Gavini et al., 1991; Arango and
Long, 1998; Sinton et al., 1998; Gilpen et al., 2002; NRC, 2004).
Chapter 2
24
2.4.1.9 Multiple antibiotic resistant analyses
Resistant bacteria have the ability to survive exposure to antibiotics or disinfectants and
through rapid multiplication pass their resistant genes on to other pathogenic as well as
to non-pathogenic bacteria (Sergeant, 1999). These antibiotic resistant genes are often
associated with transposons (genes that can easily move from one bacterium to another
bacterium or by bacteriophages) (Sergeant, 1999). Many bacteria also possess integrons
and plasmids, which are small pieces of DNA that accumulate new genes (Sergeant,
1999). Over a period of time, a bacterium can build up a whole range of resistant genes,
which is referred to as multiple resistances, which may be passed on within a genus or
species to other strains or species (Sergeant, 1999).
The multiple antibiotic analysis (MAR) includes the use of antibiotic resistance patterns
of specific microorganisms to differentiate between phenotypes within a specific genus
(Krumperman, 1983; Sergeant, 1999). In E. coli, Salmonella spp and Shigella spp, a
chromosomal locus is used to determine the intrinsic levels of these organisms for their
susceptibility to structurally different antibiotics and disinfectants (Krumperman, 1983).
Over expression of this chromosomal locus due to mutations or chemical induction,
produces a range of new bacterial phenotypes within a bacterial species (Krumperman,
1983). Bacteria isolated from humans have different MAR profiles than isolates from
domestic animals (Krumperman, 1983; Hair et al., 1998; Sergeant, 1999). Individual
bacterial isolates can be classified into phenotypic groups when the MAR profiles are
combined with discriminant statistical analyses (eg. a variation of multivariant analysis
of variance) (Krumperman, 1983; Hair et al., 1998; Sergeant, 1999). However, MAR
studies are time consuming, complicated and expensive.
In addition, antibiotic
resistance encoded on plasmids can be lost during isolation and there are constant
population shifts in antibiotic resistance (Sergeant, 1999).
2.4.1.10
Deoxy Ribonucleic Acid based profiles of microorganisms
The microbial Deoxy Ribonucleic Acid (DNA) based profile approach provide genomic
profiles of microbial communities and are used to identify the genus, species,
subspecies and strains of microorganisms (Turner et al., 1996; Nebra et al., 2003). The
DNA based profile techniques used to distinguish between microbial genus and species
Chapter 2
25
include ribotyping, Internal Transcribed Spacer-Polymerase Chain Reaction (ITS-PCR),
tRNA-PCR and 16S rRNA sequencing (Nebra et al., 2003). The DNA based profile
techniques used to distinguish between microbial subspecies and strains include
Amplified Ribosomal DNA Restriction Analysis (ARDRA), Enterobacterial Repetitive
Intergenic
Consensus-Polymerase
Chain
Reaction
(ERIC-PCR),
plasmid
or
chromosomal restriction-fragment-length-polymorphism (RFLP), Internal Transcribed
Spacer-sequencing (ITS-sequencing) and Pulsed Field Gel Electrophoresis (PFGE)
methods (Nebra et al., 2003). These DNA profiling methods are expensive, labour
intensive, require skilled personnel, need specialised equipment and are therefore not
used routinely (Turner et al., 1996; Nebra et al., 2003; Wei et al., 2004).
Although several microbiological methods have been proposed and tested to determine
the origin of faecal contamination, many of these microorganisms have proved to be
difficult to use in routine laboratory procedures because of the type of equipment
required, the cost and the skill necessary to perform the assay (Sinton et al., 1998;
Gilpen et al., 2002; Gilpen et al., 2003).
Genotyping of male specific F-RNA
bacteriophages seems to be the most promising microbiological method presently
available to distinguish between human and animal faecal pollution of water supplies in
rural communities based on results obtained by various studies on animal and human
faeces (Osawa et al., 1981; Havelaar et al., 1990; Hsu et al., 1995; Beekwilder et al.,
1996; Schaper et al., 2002a).
2.4.2
The use of chemicals to determine the origin of faecal pollution
Several chemical indicators have been used to identify the source of faecal pollution in
various water supplies (Sinton et al., 1998; Gilpen et al., 2002; Gilpen et al., 2003).
However, expensive equipment and high concentrations of the chemical in the water
sample is needed for accurate identification of the origin of faecal pollution (Sinton et
al., 1998; Gilpen et al., 2002; Gilpen et al., 2003).
2.4.2.1 Direct chemical indicators
Direct chemical indicators include chemicals present in the faeces, e.g. faecal sterols,
uric acid and urobilin (Sinton et al., 1998; Gilpen et al., 2002; Gilpen et al., 2003). The
Chapter 2
26
breakdown products of sterols are stanols (Leeming et al., 1996). Leeming and coworkers (1996) have conducted tests on human and animal faeces and especially on
sterols and stanols and found that stanols produced in animals were distinctively
different than the stanols formed in humans.
Faecal sterol cholesterol is reduced in the gut of humans to coprostanol and in the gut of
animals to epicoprostanol (Leeming et al., 1996). These compounds can be found in the
environment as cholestanol (Leeming et al., 1996). Coprostanol is used exclusively as a
marker of human faecal pollution (Leeming et al., 1996).
Plant derived 24-
ethylcholestrol is reduced to 24-ethylpicoprostanol in the intestinal tract of herbivores
and found in the environment as 24-ethylcholestanol (Leeming et al., 1996). The 24ethylcoprostanol is used as an exclusive marker of animal faecal pollution (Leeming et
al., 1996).
2.4.2.2 Indirect chemical indicators
Indirect chemical indicators are specific for human faecal contamination (Sinton et al.,
1998; Gilpen et al., 2002; Gilpen et al., 2003). These chemicals are associated with
faecal discharge in wastewater and septic tank discharges (Sinton et al., 1998; Gilpen et
al., 2002; Gilpen et al., 2003).
Fluorescent whitening agents (FWA) and sodium
tripolyphosphate (STP) present in washing powders, long chain alkylbenzenes (LAB)
present in commercial detergents and polycyclic aromatic hydrocarbons have been used
as indirect indicators of human faecal pollution (Sinton et al., 1998; Gilpen et al., 2002;
Gilpen et al., 2003).
Although different studies have described the use of these microbiological and chemical
indicators, it is apparent that no single chemical determinant could reliably distinguish
human from animal faecal contamination (Jagals et al., 1995; Sinton et al., 1998). It
seems that the use of a combination of these determinants may provide the best solution
for identifying the origin of faecal pollution in water environments (Jagals et al., 1995;
Sinton et al., 1998).
Chapter 2
27
2.5
SOURCE WATER SUPPLIES
The World Health Organization (WHO) classifies source water supplies as either
improved or unimproved (WHO, 2000; Gundry et al., 2004). Improved water sources
include public standpipes, household connections, boreholes, protected dug wells,
protected springs, boreholes and springs connected via a pipe system to a tap, as well as
rainwater collection (WHO, 2000; Gundry et al., 2004). Unimproved water sources
include unprotected wells, unprotected springs, vendor-provided water, rivers as well as
tanker truck provision of water (WHO, 2000; Gundry et al., 2004).
Several studies carried out in developing countries have determined the microbiological
quality of these improved and unimproved water sources and depending on the water
source, different results were obtained (Pournadeali and Tayback, 1980; Obi et al.,
2002; Sobsey et al., 2003; Gundry et al., 2004; Obi et al., 2004). Studies conducted in
Iran (Pournadeali and Tayback, 1980) and in northern Sudan (Musa et al., 1999) have
both showed that water at communal taps were microbiologically of a better quality
than untreated irrigation canal water. Contrary to these findings, a study in Burma (Han
et al., 1989) has showed that tube well and shallow well water supplies were
microbiologically of a better quality than municipal tap water and pond water source
supplies.
In South Africa, studies in the Limpopo Province (Verweij et al., 1991) have showed
that communal standpipes were microbiologically less contaminated than borehole and
unprotected spring water sources. Another study in the rural Kibi area of the Limpopo
Province of South Africa (Davids and Maremane, 1998), have indicated that spring and
borehole water sources were microbiologically less contaminated than river water
sources.
In addition three recent studies conducted in the Vhembe region of the Limpopo
Province in South Africa indicated that rivers and fountains used by rural communities
for domestic water were all contaminated by enteric pathogens including E. coli,
Plesiomonas shigelloides, V. cholera, Enterobacter cloacae, Shigella spp, Salmonella
spp, Aeromonas hydrophila, Aeromonas caviae and Campylobacter spp (Obi et al.,
2002; Obi et al., 2003; Obi et al., 2004). Escherichia coli isolates obtained from the
Chapter 2
28
different rivers during this study were typed using molecular techniques to determine
the presence of virulent genes (Orden et al., 1999; Kuhnert et al., 2000; Obi et al.,
2004). Enterotoxigenic E. coli isolates (11.8%) contained heat stable and heat labile
genes; Shigatoxin producing E. coli (4.4%) isolates contained stx1 and stx2 genes;
Necrotoxigenic E. coli (35.6%) contained cnf1 and cnf2 genes and Enteropathogenic E.
coli (34.1%) isolates contained BfpA and EaeA genes (Obi et al., 2004). Necrotoxigenic
E. coli may play a role in possible zoonotic transmission since it has been shown that
human and animal strains share similar serogroups and carry the same genes coding for
fimbrial and afimbrial adhesion (Mainil et al., 1999). All of these studies indicated that
the water sources used by communities in developing countries are microbiologically
contaminated and pose a health risk to the consumers (Pournadeali and Tayback, 1980;
Obi et al., 2002; Sobsey et al., 2003; Gundry et al., 2004; Obi et al., 2004).
2.5.1
Water collection from the source water supply
In most developing countries, women are responsible for the collection of water
(Sobsey, 2002). The work involved in fetching the water may differ in each region, it
may vary according to the specific season, it depends on the time spent queuing at the
source, the distance of the household from the source and the number of household
members for which the water must be collected (WHO, 1996b; WHO, 1996c). Water
for domestic use is collected either by dipping the container inside the water supply Fig
2.1), collecting rainwater from a roof catchment system (Fig 2.2) or by using different
types of pumps connected to the water supply system (Fig 2.3) (Sobsey, 2002). The
transportation of the water from the source water supply could be either by a
wheelbarrow (Fig 2.4), a donkey cart (Fig 2.5), a motor vehicle (Fig 2.6), using a rolling
system (Fig 2.7) or by carrying the container by hand or on the head (Fig 2.8) (CDC,
2001). A common practice often seen in rural areas was the use of leaves or branches
with leaves to stop water slopping out during transit in wide-neck storage and transport
containers (Fig 2.9) (Sutton and Mubiana, 1989). Consequently, a study by Sutton and
Mubiana (1989) has showed that these leaves can be an additional source of coliform
bacteria to the drinking water.
Chapter 2
29
Figure 2.1:
Water collection by rural people in the Vhembe region of the Limpopo
Province of South Africa: Dipping containers inside theprimary water
source
Figure 2.2:
Water collection by rural people in the Vhembe region of the Limpopo
Province of South Africa:
Collecting rain water from the roof of the
household
Chapter 2
30
Figure 2.3:
Water collection by rural people in the Vhembe region of the Limpopo
Province of South Africa: Ground water pumped to a communal tap
Figure 2.4:
Water transportation by rural people in the Vhembe region of the
Limpopo Province, South Africa: Use of a wheelbarrow
Chapter 2
31
Figure 2.5:
Water transportation by rural people in the Vhembe region of the
Limpopo Province, South Africa: Use of a donkey cart
Figure 2.6:
Water transportation by rural people in the Vhembe region of the
Limpopo Province, South Africa: Use of a motor vehicle
Chapter 2
32
Figure 2.7:
Water transportation by rural people in the Vhembe region of the
Limpopo Province, South Africa: Use of a rolling drum
Figure 2.8:
Water transportation by rural people in the Vhembe region of the
Limpopo Province, South Africa: Use of hands and head
Chapter 2
33
Figure 2.9:
Methods used by rural people in the Vhembe region of the Limpopo
Province, South Africa to stop water from spilling while in transport: Use
of leaves/branches
Water sources could be some distance away from the households, particularly in rural
areas (WHO, 1996b; WHO, 1996c). In studies conducted in Malawi, Kenya, Uganda
and Tanzania (Lindskog and Lundqvist, 1989; White et al., 2002), it was found that if
the water taps were situated closer to the dwelling, the amount of water
collected/person/day increases from 9.7 to 15.5 litres.
Studies in Mosambique
(Cairncross and Cliff, 1987) showed that households collect on average 11.1 litres of
water/person/day if the source is less than 300 m from the dwelling, while the
households who have to walk more than 4 km collected on average 4.1 litres of
water/person/day. In Lesotho, Esrey and co-workers (1992) made a rough estimate of
10 litres of water/person/day based on direct observations of households in rural
communities. Studies in rural communities in the Limpopo Province of South Africa
(Verweij et al., 1991) showed that on average 11.4 litres of water/person/day was
collected if the source was close to the household, compared to an average of 8.6 litre of
water/person/day if the sources were more than 1 km from the household.
The
Department of Water Affairs and Forestry in South Africa recommends 25
Chapter 2
34
litre/person/day from a source within a distance of 200 m from the dwelling (DWAF,
1994) and the WHO estimates a minimum of 20 litres of water/person/day is sufficient
(WHO, 1996b), while Gleick (1998) recommends 50 litres of water/person/day is
efficient. These studies indicated that more water was collected per person per day if
the source was closer to the dwelling (White et al., 2002; Lindskog and Lundqvist,
1989; Verweij et al., 1991).
Very few studies have investigated the microbiological quality of water during
collection and transportation. In a study in Rangoon, Burma (Han et al., 1989) the
water at the source and during collection were analysed and indicated that the faecal
coliform counts in the collection samples were higher than the counts in the source
water samples (Han et al., 1989). The increase in faecal contamination of the water in
the collection containers after collection from the source could have been due to
unhygienic handling of the water and posed a potential health risk of diseases to the
consumers (Sobsey, 2002). In a study in Sri Lanka (Mertens et al., 1990) it was found
that only 5% of tube well water samples were contaminated if the pump was sterilised
prior to collection of the sample compared to 50% if the pump was not sterilised. This
implied that the taps were contaminated by hands or animals during collection (Mertens
et al., 1990).
In another study in rural communities in South Africa (Verweij et al., 1991), water
samples were taken immediately after collection from communal taps and unprotected
borehole and springs. Special precautions were taken to prevent contamination during
collection, which included rinsing of the container before filling, using a calabash to
scoop water from the source and demarcation of a special area for water collection
(Verweij et al., 1991). The results from this study indicated no significant difference
between faecal coliform counts at the source and immediately after collection of the
water (Verweij et al., 1991). The drawbacks of this study however included the sample
size (only 8 households were studied), and inadequate information given regarding who
collected the water samples e.g. a technician or a woman from the study households
(Verweij et al., 1991). A study carried out in a Malawi refugee camp has found that
hands are primarily responsible for contamination of collected water because the
women rinses the container with small amounts of water using their hands to rub around
the container opening in an effort to clean it (Roberts et al., 2001). A study by Dunker
Chapter 2
35
(2001) has concluded that rural communities in South Africa spent little time on proper
cleaning of the collection containers, especially if water has to be collected more that
once a day.
These studies have shown that although the microbiological quality of the source water
could be classified as safe for domestic purposes, the water collected by the households
from these sources, become contaminated after collection (Sobsey, 2002). The origin of
the contamination includes: transport and unhygienic collection and handling practices
such as dirty utensils, dirty hands and unclean storage containers (Dunker, 2001;
Sobsey, 2002).
2.5.2
Interventions to improve source water supplies
Various intervention strategies to improve the water at the source have been described
in the literature (Sobsey, 2002). These improvements can include the building of
reservoirs, building protective structures around boreholes and fountains, providing
communities with communal taps closer to the dwelling and the treatment of the water
source with a disinfectant (Sobsey, 2002). A study in Shangai (Xian-Yu and Hui-Gang,
1982) have showed that continuous chlorination rather than periodic chlorination of
wells is more reliable, safes time and labour and showed a reduction in the mortality
rates due to enteric diseases from 13.7 per 100 000 people to 1.1 per 100 000 people.
However, Jensen and co-workers (2002) have found that in rural areas of Parkistan,
where public water supply systems was chlorinated, no reduction in diarrhoea incidence
in children from these villages were found compared to diarrhoea incidence in children
from villages where the people used untreated ground water supplies.
Different interventions can be implemented to improve the microbiological quality of
the source water supply. A study in rural Malawi (Lindskog and Lindskog, 1988) has
showed that communal piped water supplies situated within a distance of 400 m from a
specific household, improved the microbiological water quality used for drinking
because people collected water more often and did not store water which could have
become contaminated during storage. A 3 year study by Ghannoum and co-workers
(1981) in Libya have showed that the installation of water treatment plants did reduce
the incidence of bacillary and amoebic dysentery between 10% and 50%, but not
Chapter 2
36
Giardia infections. However, studies carried out in peri-urban communities in South
Africa (Genthe et al., 1997; Jagals et al., 1999) have showed that although the
households were supplied with good quality water complying with South African
drinking water specifications (DWAF, 1996), the water in the household storage
containers had increased levels of indicator microorganisms.
This implied that
secondary contamination was introduced after the water collection.
Consequently,
many of these studies have indicated that improvements at the water source are useless
as water is contaminated during collection and storage in households due to poor
sanitation practices.
2.6
POINT-OF-USE WATER SUPPLIES IN THE HOUSEHOLD
Source water contamination is likely to have a wide effect on the community because it
can introduce new pathogens in the home environment (Sobsey, 2002). However,
several studies have reported that the microbiological quality of the water deteriorate
after collection, during transport and during storage at the point-of-use due to secondary
contamination factors (Rajasekaran et al., 1977; El Attar et al., 1982; Han et al., 1989;
Lindskog and Lindskog, 1989; Sandiford et al., 1989; Blum et al., 1990; Henry and
Rahim, 1990; Mertens et al., 1990; Pinfold, 1990; Verweij et al., 1991; Simango et al.,
1992; Swerdlow et al., 1992; Shears et al., 1995; Kaltenhaler and Drasar, 1996; Genthe
et al., 1997; Jensen et al., 2002; Wright et al., 2004).
Due to the distances and
unavailability of piped water supplies on the dwelling or inside the households in many
developing regions of the world, people are forced to store their drinking water (Sobsey,
2002).
Transmission of microorganisms inside the household can occur through several routes
(Briscoe, 1984; Roberts et al., 2001). The most important transmission routes include
water, food, person-to-person contact, unhygienic behaviour (eg. intra-household
transmission of faeces), the storage conditions of the water storage containers at the
point-of-use and the abstraction conditions of water from the storage container (Briscoe,
1984; Roberts et al., 2001). In addition, a number of studies (as shown in Table 2.3)
suggested that inadequate storage conditions increased the risk of contamination, which
can lead to infectious diseases.
Chapter 2
37
Table 2.3
Summary of studies indicating increased microbiological contamination of
stored water and the associated infectious disease risk due to inadequate
storage conditions (Sobsey, 2002)
Study
Area
Bangladesh
Storage
container
Water jars
Storage
time
1-2 days
Bahrain
Capped plastic
vessels, jars,
pitchers
Clay jars
(zeers)
Not
reported
Egypt
Clay jars (zir)
<1- 3 days
India
Wide mouth vs
narrow neck
Not
reported
Burma
Buckets
Up to 2
days
Liberia
Large
containers,
open or closed
Long time
Sri Lanka
Earthen pots
and others
Not
reported
South
Africa
Africa
Plastic
container
Traditional
and metal jars
4 hours
Malaysia
Various
containers
Not
reported
Higher levels of faecal
coliforms in unboiled than
boiled water
Zimbabwe
Covered and
uncovered
containers
Wide mouth
containers
12 hours
or more
Bangladesh
Traditional
pots
Not
reported
Trinidad
Open drum,
barrel, bucket
vs tank or
none
Not
reported
Higher E. coli and
Aeromonas levels with
storage and use
Higher faecal coliform
levels in stored waters
than source waters
Increased faecal coliform
levels and antibiotic
resistance
Increased faecal bacteria
levels in open storage
vessels than tank
Sudan
Peru
Chapter 2
2 days – 1
month
24 hours
and more
Not
reported
Impact on
Microbial quality
Increased Vibrio cholera
presence
Vibrio cholera present in
stored water and not in
source water
Increased faecal indicator
bacteria over time, in
summer and during dust
events
Algae growth and
accumulated sediment
Not measured
Higher levels of faecal
coliform bacteria than
sources
High levels of
enterobacteria in stored
samples compared to
sources
High levels of faecal
coliforms in unboiled
stored water
Higher coliform levels
over time
High total and faecal
coliform levels
Disease
Impact
Increased cholera rates
Uncertain
Reference
Spira et al.,
1980
Gunn et al.,
1981
Not measured
Hammad
and Dirar,
1982
Not detected
Miller, 1984
Cholera infections
fourfold higher in wide
mouth storage vessels
Not measured
Deb et al.,
1986
Han et al.,
1989
Not measured
Molbak et
al., 1989
Not measured
Mertens et
al., 1990
Measured; no effect
Verweij et
al., 1991
EmpereurBisonette et
al., 1992
Knight et
al., 1992
Not measured
Higher diarrhoea risks
from water unboiled or
stored in wide neck
than narrow neck
containers
Not measured
Simango et
al., 1992
Increased cholera risks
Swerdlow et
al., 1992
Increased faecal
coliforms and multiple
antibiotic resistant flora
Not measured
Shears et al.,
1995
Welch et al.,
2000
38
Some studies showed an increase in the number of V. cholera in stored water (Spira et
al., 1980; Gunn et al., 1981), while other studies indicated an increase in faecal coliform
bacteria and enterobacteriaceae (E. coli and Aeromona spp) in the stored water (Deb et
al., 1986; Hammad and Dirar, 1982; Han et al., 1989; Molbak et al., 1989; Mertens et
al., 1990; Verweij et al., 1991; Empereur-Bisonette et al., 1992; Knight et al., 1992;
Simango et al., 1992; Swerdlow et al., 1992; Shears et al., 1995; Welch et al., 2000).
The geometric design of household water storage containers could play an important
role in ensuring that the stored drinking water does not become contaminated during
storage (Sobsey, 2002). Many different types and sizes of traditional storage containers
(Fig 2.10 and 2.11) are commonly used in developing countries such as the nomadic
people of Sudan which uses a container made from animal hide called a girba (Musa et
al., 1999) and communities in Africa which use traditional African clay pots or urns
(Patel and Isaacson, 1989; Sutton and Mubiana, 1989; Sobsey, 2002).
Figure 2.10:
Typical 25 litre water storage containers and buckets used for point-of-use
water storage by rural people in the Vhembe region of the Limpopo
Province, South Africa
Chapter 2
39
Figure 2.11:
Typical 200 litre water storage container used for point-of-use water
storage by rural people in the Vhembe region of the Limpopo Province,
South Africa
The material of the container is also important because the chemical material of the
storage container could be conducive to bacterial growth and survival of potentially
pathogenic microorganisms if contamination of the water occurs. This was shown in a
study conducted by Patel and Isaacson (1989), which showed that Vibrio cholera 01
survived longer in corroded iron drums than in new iron drums.
The studies in Table 2.3 have showed that water can be stored between 4 h and 1 month
at the point-of-use. Faechem and co-workers (1983) indicated that the time of storage
was important, with the highest increase in faecal contamination occurring if the storage
time was longer than 10 h.
Similar observations were reported by other studies,
especially if the storage periods were longer than 12 h (Han et al., 1989; Mertens et al.,
1990; Verweij et al., 1991; Simango et al., 1992 Ahmed and Mahmud, 1998; Momba
Chapter 2
40
and Kaleni, 2002). These studies have showed that the microbiological quality of water
deteriorates during long storage times and increased the risk of the transmission of
waterborne diseases.
Other factors, which could contribute to the contamination of the water during storage at
the point-of-use, included unsanitary and inadequately protected (open, uncovered,
poorly covered) containers (Dunker, 2001). Many of the studies listed in Table 2.3 had
either uncovered containers, containers with wide openings or buckets, which were used
as storage containers. Storage containers need to be covered at all times to prevent flies,
animals (Fig 2.12) and small children from touching the water (Fig 2.13) (Sobsey,
2002). It was noted by Jensen et al., (2002), that containers with openings of less than
10 cm were less contaminated with coliform bacteria than those with wider openings.
Water was poured from these containers, while water was dipped out with hands and
utensils where containers with wider openings were used. However, a study by El Attar
and co-workers (1982) showed no differences in water quality between containers that
were covered versus those that were uncovered.
Figure 2.12:
Possible contamination route of stored drinking water in rural households
in the Vhembe region of the Limpopo Province, South Africa: animals
licking containers while containers are filled with water
Chapter 2
41
Figure 2.13:
Possible contamination route of stored drinking water in rural households
in the Vhembe region of the Limpopo Province, South Africa: small
children touching water storage containers which are not closed
Human faecal pollution from children and adults who do not wash their hands after
being to the toilet can contribute to secondary contamination of household stored
drinking water (DeWolf Miller, 1984; Dunker, 2001). Several studies have indicated
that E. coli can survive for 10 min, Klebsiella spp for up to 2.5 h (Casewell and Phillips,
1977) and Shigella sonnei and faecal enterococci for up to 3 h (Knittle, 1975; Pinfold,
1990) on unwashed hands, which could contaminate food and water in the household.
Finally, inadequate cleaning measures of the storage containers could lead to the
formation of biofilms (Fig 2.14) which could harbour potentially pathogenic and
opportunistic microorganisms such as total coliforms, faecal coliforms, E. coli, somatic
and F-RNA bacteriophages, C. perfringens, Salmonella spp and Helicobacter pylori
(Bunn et al., 2002; Jensen et al., 2002; Momba and Kaleni, 2002; Sobsey, 2002). These
Chapter 2
42
indicator and pathogenic microorganisms could survive longer than 48 h in biofilms
inside household drinking water storage containers and pose a potential risk factor for
humans consuming this water (Bunn et al., 2002; Jensen et al., 2002; Momba and
Kaleni, 2002; Sobsey, 2002).
Figure 2.14:
Possible contamination route of stored drinking water in rural households
in the Vhembe region of the Limpopo Province, South Africa: biofilm
formation inside a 25 litre water storage container
The studies mentioned in this section clearly showed that contamination of water
occurred during collection and storage at the point-of-use and does contribute to the risk
of disease transmission and possibly the spread of anti-microbial resistant genes (Shears
et al., 1995; Sobsey, 2002). Therefore, the focus must be on point-of-use interventions
rather than water source interventions because point-of-use interventions will be more
effective in the removal and inactivation of potential disease causing microorganisms
introduced during collection and storage inside a family cohort.
Chapter 2
43
2.6.1
Interventions to improve point-of-use water supplies in the household
Point-of-use interventions must improve the water used for drinking at the household
level (Sobsey, 2002).
This can be achieved by educating household members to
improve their hygienic behaviour, by improving the water storage container and by
appropriate treatment of the stored water (Dunker, 2001). All of these interventions will
be discussed in the following sections.
2.6.1.1
Improving the point-of-use water supply by improving hygienic
practices in the household
Basic hygiene practices such as hand washing was shown to be an effective intervention
in the reduction of diarrhoea in developing countries (Curtis et al., 2000; Trevett et al.,
2005). A study in Burma (Han and Hlaing, 1989) showed a 30% reduction in diarrhoeal
incidence if people washed their hands after defecation, prior to food preparation.
Studies in Indonesia (Wilson et al., 1991) and Bangladesh (Shahid et al., 1996) have
showed an 89% and 66% reduction of diarrhoea respectively after hand washing was
introduced.
However, factors like the distance from the washing area and the frequency of hand
washing do affect the influence of the intervention on the disease outcome (Faechem,
1984; Hoque et al., 1995). Faechem (1984) has showed that soap and water together
removes 100% of inoculated bacteria while water alone removed less bacteria. Hoque
and co-workers (1995) has showed that soap, ash and soil were equally effective handwashing reagents, however, drying wet hands on clothing, resulted in recontamination
of the hands.
Proper education should therefore be given to people from rural
communities to promote the correct hygiene practices and these communities should be
informed on the transmission risk and the causes of waterborne diseases (Dunker,
2001).
Chapter 2
44
2.6.1.2
Improving the point-of-use water supply by using an improved
storage container
The United States Centres for Disease Control and Prevention (CDC) and the Pan
American Health Organization (PAHO) have studied and reviewed the advantages and
disadvantages of different types of water collection and storage containers from studies
carried out in various regions of the world. These two organisations have written
guidelines for the most desirable container to be used by households for drinking water
storage. The guidelines include the following (Mintz et al., 1995; Reiff et al., 1996;
CDC, 2001):
•
The container must have a capacity of 15 to 25 litres, rectangular or cylindrical
with one or more handles and flat bottoms for portability and ease of storage;
•
Should be made of lightweight, oxidation-resistant plastic, such as high-density
polyethylene or polypropylene, for durability and shock resistance;
•
Should be fitted with a 6 to 9 cm screw-cap opening to facilitate cleaning, but
small enough to discourage or prevent the introduction of hands or dipping
utensils;
•
Should have a durable, protected and preferably easily closed spigot or spout for
dispensing water;
•
Should have an affixed certificate of approval or authenticity;
•
Should be affordable to the user.
Based on these guidelines, the CDC and PAHO designed a 20 litre container to decrease
the risk of contamination during storage (Fig 2.15) (Mintz et al., 1995; Reiff et al.,
1996; CDC, 2001; Sobsey, 2002). Together with the use of a sodium hypochlorite
solution, this container has proved effective in several studies carried out in different
developing countries in Africa, Europe and South America as indicated in Table 2.4
(CDC, 2001; Sobsey, 2002).
Chapter 2
45
Figure 2.15:
The CDC safe storage container designed by the CDC and PAHO in the
USA for point-of-use treatment
Several of the studies mentioned in Table 2.4, have investigated the reduction of
disease, especially the reduction of diarrhoea during the intervention phase (Semenza et
al., 1998; Quick et al., 1999; Mong et al., 2001; Quick et al., 2002; Sobsey et al., 2003).
The results from all of these studies showed that the diarrhoea incidences were reduced
between 20% and 85%, while cholera incidence were reduced by 90% during a cholera
outbreak in Madagascar (Semenza et al., 1998; Quick et al., 1999; Mong et al., 2001;
Quick et al., 2002; Sobsey et al., 2003). Unfortunately most of these studies have only
used E. coli and thermotolerant indicator bacteria to assess the microbiological quality
of the stored household water (Semenza et al., 1998; Quick et al., 1999; Mong et al.,
2001; Quick et al., 2002; Sobsey et al., 2003).
However, none of these studies
investigated the survival of pathogenic microorganisms in the CDC safe storage
container nor have any study investigated the origin of the faecal contamination in the
CDC safe storage container. Although, the incidence of diarrhoea decreased during the
intervention studies, little information is available on the origin or the causative
microorganism of the diarrhoeal diseases (Sobsey, 2002).
Chapter 2
46
Table 2.4
Efficacy of chlorination and water storage in the CDC safe storage container to disinfect household water, reduce waterborne diseases
and improve the microbiological quality of water (Sobsey, 2002)
Location
Uzbekistan
Guatamala
Guinea-Bisseau
Bolivia
Parkistan
Madagascar
Zambia
Bolivia
and
Bangladesh
Chapter 2
Water and
service level
Household
On site and off plot
Mixed sources
Street vendor water
Off plot
Mixed sources
Oral rehydration
solution
Off plot
Ground water or
Surface water
Household
On site
Ground water
Treatment
Storage vessel
Disease reduction
(%)
85% diarrhoea
Free chlorine
CDC safe storage
container
Free chlorine
CDC safe storage
container
No data
Free chlorine
CDC safe storage
container
No data
Electrochemical
oxidant (mostly
free chlorine)
CDC safe storage
container
44% diarrhoea
Household
On site and off plot
Municipal
Household
Free chlorine
CDC safe storage
container
No data
Free chlorine
(traditional
vessel)
CDC safe storage
container or
traditional vessel
90% cholera (during
outbreak)
Household
Off plot or on site
Not reported
Ground water
Household
Onsite
Shallow groundwater
and municipal water
Free chlorine
CDC safe storage
container or
traditional vessel
48% diarrhoea
Free chlorine
CDC safe storage
container or
traditional vessel
20.8% diarrhoea
Significant microbe
decrease?
No
But based on small number of
samples
Yes
E. coli positive counts
decrease from >40 to <10%
Yes
Mean E. coli positive counts
decrease from 6200 to 0
counts.100 ml-1
Yes
E. coli positive counts
decrease from 94 to 22%;
median E. coli counts from
>20 000 to 0
Yes
Thermotolerant
coliforms
counts decrease by 99.8%
Yes
Median E. coli positive counts
decrease from 13 to 0
counts.100 ml-1
Yes
E. coli positive counts
decrease from 95 to 31%
Yes
E. coli counts decreased in
intervention households
Intervention
Reference
Water intervention
only
Semenza et al., 1998
Water intervention
and Santation and
Health intervention
Water intervention
and Santation and
Health intervention
Sobel et al., 1998
Daniels et al., 1999
Water intervention
and Sanitation and
Health intervention
Quick et al., 1999
Water intervention
and Santation and
Health intervention
Water intervention
and Santation and
Health intervention
Luby et al., 2001
Mong et al., 2001
Water intervention
and Santation and
Health intervention
Quick et al., 2002
Water intervention
and Health
intervention
Sobsey et al., 2003
5
The studies in Table 2.4 have also included additional interventions together with the
CDC safe storage container and sodium hypochlorite solution interventions.
The
additional interventions included sanitation and health interventions where people were
informed and educated on hygiene and handling practices (Sobel et al., 1998; Daniels et
al., 1999; Quick et al., 1999; Luby et al., 2001; Mong et al., 2001; Quick et al., 2002;
Sobsey et al., 2003). Generally all of these studies have showed that proper education
will influence the compliance with point-of-use interventions (Sobsey, 2002). People
should be made aware and educated on the benefit of using interventions to improve the
microbiological quality of the household drinking water.
2.6.1.3 Improving the point-of-use water supply by chemical or physical treatment
Several physical and chemical treatments have been developed and tested under various
field conditions in several countries as interventions to improve the water at the pointof-use (Sobsey, 2002; Nath et al., 2006). However, many of these treatments are not
suitable for conditions in rural communities. The various advantages and disadvantages
with regards to the use of some of these treatment interventions in rural regions will be
discussed in the following sections.
2.6.1.3.1
Physical treatment methods
Physical treatment methods include boiling, heating, settling, filtration and exposure to
ultraviolet radiation from sunlight (Gilman and Skillikorn, 1985; Mintz et al., 1995;
Conroy et al., 1996; CDC, 2001; Sobsey, 2002). Boiling is widely used since it is easy
to use and effective in destroying bacteria, viruses and protozoa in all types of water
(Sobsey, 2002). However, the collection of firewood is time consuming, could lead to
deforestation and is an expensive method for general use (Gilman and Skillicorn, 1985;
Barau and Merson, 1992). A further concern is that water is often transferred to storage
containers for cooling and thus can become re-contaminated (Sobsey, 2002).
Solar disinfection such as the SOLAIR and SODIS systems, which makes use of plastic
water collection bottles which is left in the sun, have been widely tested in rural African
communities (Conroy et al., 1996; Conroy et al., 1999; Meyer and Reed, 2000; Conroy
et al., 2001). Both these systems inactivates pathogens by disinfecting small quantities
Chapter 2
47
of water for consumption, requires relative clear water (turbidity< 30 NTU) and the
effectiveness of the inactivation is dependant on exposure times (Conroy et al., 1996;
Conroy et al., 1999; McGuigan et al., 1999; Meyer and Reed, 2000; Conroy et al.,
2001; Rijal and Fujioka, 2001; Sobsey, 2002; Mascher et al., 2003; Oates et al., 2003).
Sedimentation and settling is used for very turbid water (Sobsey, 2002). The turbidity
is usually due to the presence of sand particles (mud) (Sobsey, 2002). After the water is
collected, the container is left undisturbed for a few hours (Sobsey, 2002). The large
dense particles (sands and silts) together with large microorganisms will settle out
(sediment) due to the effect of gravity (Sobsey, 2002). The upper cleaner water is
carefully removed without disturbing the sedimented particles (Sobsey, 2002).
Unfortunately sedimentation is not very effective in reducing microbial pathogens in
stored household water (Sobsey, 2002).
Filtration is a widely used method to remove particles and some microorganisms from
water samples (Potgieter, 1997; Sobsey, 2002).
Several types of filter media and
filtration processes are available for household treatment of water (Sobsey, 2002).
However, the effective removal of microorganisms, the cost and the availability of the
filter media in developing countries varies from easy to moderate to difficult (Sobsey,
2002). Granular type of filters include bucket filters, barrel or drum filters and roughing
filters and filter cisterns which can rapidly reduce turbidities and enteric bacteria by
>90% and larger parasites by >99% efficiency, and enteric viruses by 50% to 90%
(Sobsey, 2002; Clasen and Bastable, 2003).
Slow sand filters, fibre, fabric and
membrane filters, porous ceramic filters and diatomaceous earth filters are alternative
filters that have been tested and used for household water treatment in developing
countries (Sobsey, 2002; Clasen and Bastable, 2003). Many of these studies have
showed to reduce turbidity by 90% and bacteria by 60%, although the cost of the filters
is high (Sobsey, 2002; Clasen and Bastable, 2003). A study by Clasen and co-workers
(2004) in Bolivia, indicated a reduction of diarrhoea of 70% and a 100% reduction of
thermotolerant coliforms in households using ceramic filters compared to control
households not using ceramic filters. Unfortunately, little information is available on
the effectiveness of these filter systems in the reduction of viruses from household water
(Sobsey, 2002).
Chapter 2
48
2.6.1.3.2
Chemical treatment methods
Various chemical methods are available for the treatment of drinking water at the
household level and include methods such as coagulation-flocculation, precipitation,
adsorption, ion exchange and chemical disinfection with agents such as sodium
hypochlorite (Gilman and Skillicorn, 1985; Mintz et al., 1995; Conroy et al., 1996;
CDC, 2001; Sobsey, 2002).
Unfortunately most of these methods are expensive,
requires technical skilled persons, regular monitoring, specific materials and the
efficacy varies (Sobsey, 2002). Chemical disinfectant agents have proved to be the
most successful types of treatment and include free chlorine (which will be discussed in
more detail), chloramines, ozone and chlorine dioxide (Sobsey, 2002).
Several factors might play a role in the effectiveness of a chemical disinfectant. These
factors include pH, turbidity, temperature, degree of microbial contamination and the
contact time of the disinfectant to the water and microorganisms (LeChevallier et al., 1981;
Reiff et al., 1996).
According to Reiff and co-workers (1996), an ideal chemical
disinfectant should have the following qualities:
•
The disinfectant must be reliable and effective in the inactivation of pathogens
under a range of conditions likely to be encountered;
•
The disinfectant must provide an adequate residual concentration in the water as
to assure safe microbial quality throughout the storage period;
•
The disinfectant must not introduce nor produce substances in concentrations
that may be harmful to health, nor otherwise change the characteristics of the
water so as to make it unsuitable for human consumption;
•
The disinfectant must be reasonable safe for household storage and use;
•
The disinfectant must have an accurate, simple and rapid test for measurement
of the disinfectant residual in the water, which can be performed, when required;
•
The disinfectant must have an adequate shelf life without significant loss of
potency;
•
The disinfectant must have a cost that is affordable for the household.
A chemical disinfectant that has been used effectively since 1850, is chlorine (sodium
hypochlorite) (White, 1999). During a cholera outbreak in London, chlorine was used
Chapter 2
49
to disinfect water supplies (White, 1999).
During the 1890’s, Europe used
hypochlorites against epidemics of typhoid (White, 1999).
Only in the early 20th
Century Great Britain and New Jersey City began treatment of potable water supplies
on a continuous basis. Since then chlorine has become the most widely used water
treatment disinfectant because of its potency, ease of use and cost effectiveness (White,
1999).
Chlorine reacts with water to form hypochlorous acid (HOCl) and hydrochloric acid
(HCl) (Carlsson, 2003). The HOCl dissociates further into a hypochlorite ion (OCl-)
and a hydrogen atom (H+) which are commonly referred to as the free chlorine residual
(Carlsson, 2003). The main problem to overcome when chemical treatment is used is
the differences in resistance of bacteria, viruses and parasites to these chemical
disinfectants (Sobsey, 1989; Sobsey, 2002). The resistance of waterborne microbes to
be inactivated by chemical disinfectants is influenced by several factors: (1) their
physical status; (2) their physiological status; (3) the presence of microorganisms within
microbial aggregates (clumps); and (4) microorganisms embedded within other matrices
such as a membrane, a biofilm, another cell, or fecal matter (Sobsey, 1989; Sobsey,
2002). The microorganisms could be protected against chemical disinfectants and by
the oxidant demand of the material in which they are located (Sobsey, 1989; Sobsey,
2002). Consequently it has been showed that bacteria are more susceptible to chlorine
than viruses or enteric parasites (Sobsey, 1989; Sobsey, 2002).
In bacterial cells the free residual chlorine reacts with various structures on the bacterial
cell (Carlsson, 2003). The free residual chlorine can also kill the microorganism by
disrupting the metabolism and protein synthesis, to decrease respiration, glucose
transport and adenosine triphosphate levels and to cause genetic effects by modification
of the purine and pyirimidine basis (LeChevallier and Au, 2004). In viruses the free
residual chlorine targets mainly the nucleic acid and do not have a noticeable effect on
the protein coat (Carlsson, 2003). This means that viruses containing a protein coat are
more resistant to the effect of free residual chlorine (Carlsson, 2003). Free chlorine
residual is not very effective against parasites because of the tough outer coat, which
makes them very resistant to the action of hypochlorous acid (Carlsson, 2003).
Therefore, parasites need to be exposed for longer times to the free chlorine to be
inactivated (Venczel, 1997; Carlsson, 2003). Studies have showed that Giardia lamblia
Chapter 2
50
cysts are inactivated at 1 mg.l-1 free chlorine in water with a pH of 6 to 7 and at
temperatures of 5°C only after 1 to 2.5 h (USEPA, 1989) and Giardia muris cysts under
the same conditions are only inactivated after exposure of 10 h (USEPA, 1989).
Studies have showed that the use of free chlorine residual together with the CDC safe
storage container (Table 2.4) has improved the microbiological quality of the water and
reduced the prevalence of diarrhoea (Quick et al., 1996; Luby et al., 1998; Macy and
Quick, 1998; Semenza et al., 1998; Quick et al., 1999). The CDC recommends the
addition of either a 0.5% or a 1.0% stabilized concentration of sodium hypochlorite
solution to obtain a free chlorine residual between 0.5 and 1.5 mg.l-1 after 60 min
(WHO, 1996a; CDC, 2001; Dr R Quick, CDC, Atlanta, USA, personal communication).
In South Africa, the DOH’s recommendations do not specify the free chlorine residual
concentration. However, the DOH do recommend the addition of 5 ml of a 3.5%
stabilized concentration of sodium hypochlorite solution to a 20 or 25 litre storage
container (Appendix C) (Mr H Chabalala, Department of Health, Pretoria, personal
communication).
In addition, several studies have showed that the use of some chemical disinfectants
resulted in the formation of chemical by-products such as trihalomethanes,
haloacetonitriles, chlorinated aldehydes, chlorinated acetones, chlorinated phenols and
chlorinated acetic acids (WHO, 1996a; Carlsson, 2003). Some of these by-products are
potentially hazardous (carcinogenic and mutagenic) (WHO, 1996a; Carlsson, 2003).
However, the health risk posed by these by-products is small in comparison to the
health risk caused by waterborne pathogenic and opportunistic microorganisms (WHO,
1996a; Carlsson, 2003).
Although various point-of-use interventions have been proposed, the interventions
selected for a particular community must be tailored for the needs of the community and
consider the resources available to the community (Nath et al., 2006).
The ideal
solution will be to provide these communities with treated municipal tap water in the
dwelling to eliminate storage of the water. However, this is not possible in many
developing countries due to economical constraints. In the meantime, interventions at
the point-of-use should focus on point-of-use treatments that are cost effective, easy to
obtain and easy to use (Sobsey, 2002). The rural communities of the Vhembe region in
Chapter 2
51
South Africa could benefit from point-of-use interventions such as the use of the CDC
safe storage container together with a sodium hypochlorite solution to improve the
quality of household drinking water (Sobsey, 2002).
2.6.2
Sustainability of point-of-use interventions
The microbiological effectiveness of household interventions at the point-of-use has
been indicated by several studies (Sobsey, 2002; Fewtrell et al., 2005). However,
questions on acceptability, affordability, long term utilization and sustainability of
household treatments must still be answered (Nath et al., 2006). Only one published
study on the sustainability of a point-of-use water treatment system could be obtained
from the literature: Conroy and co-workers (1999) found that one year after the
completion of a solar disinfection intervention in Masaai communities, almost all
households were still using the intervention. The lack of adequate follow up studies on
the long term utilization and sustainability of household treatments therefore, needs to
be addressed in order to determine the success of point-of-use treatment systems.
2.7
SUMMARY
In South Africa almost 80% of the population are living in rural communities without
adequate water and sanitation infrastructures (Statistics South Africa, 2003). Many of
the communities have to share water sources with cattle and domestic animals (Dunker,
2001). Communal standpipes provide water on infrequent time schedules and the
majority of communal standpipe water is untreated. The Vhembe region is situated in
the Limpopo Province of South Africa. The Vhembe region was a former homeland for
the Venda people in South Africa before the 1994 elections and known as the Venda
homeland. In the Vhembe region, the majority of rural communities are povertystricken, lack access to potable water supplies and rely mainly on water sources such as
rivers, streams, ponds, springs and boreholes for their daily water needs (Davids and
Maremane, 1998; Obi et al., 2002; Obi et al., 2004). Water from these sources is used
directly by the inhabitants and the water sources are faecally contaminated and devoid
of treatment (Nevondo and Cloete, 1991; Davids and Maremane, 1998; Obi et al., 2002;
Obi et al., 2004). Consequently, a significant proportion of residents are exposed to
potential waterborne diseases (Central Statistics, 1995).
Chapter 2
52
A pilot study, which consisted of a questionnaire survey, was conducted initially to
serve as a background study before the initiation of this study. The purpose of the pilot
study was to obtain information concerning the baseline microbiological quality of the
source water and the storage containers as well as to observe sanitation and hygiene
practices of rural people in the Vhembe region. Many of the households in rural areas
of South Africa do not have individual connections to treated, piped water supplies.
These households typically store water in the household. The stored water is vulnerable
to contamination from handling during collection, transport and storage. Results from
the pilot study indicated the need for education aimed at diseases associated with
polluted water supplies and the improvement in the sanitation and hygienic behaviours
of the household members during water collection and storage at the point-of-use.
Based on the results obtained from the pilot study it was evident that intervention
strategies at the point-of-use in the rural communities were needed as interim solutions
to prevent waterborne diseases and improve the microbiological quality of domestic
stored drinking water.
The literature study has showed that depending on water collection and storage
practices, deterioration of the microbiological quality of the water may occur before the
water is actually consumed, mostly due to secondary contamination at the point-of-use.
Reviews by Sobsey (2002) and Gundry and co-workers (2004) suggested that more
point-of-use intervention field studies must be conducted. The bacteriological evidence
in their studies showed that improved storage containers may be effective at reducing
microorganisms in stored water if the sources were of good microbiological quality or
uncontaminated. However, many of the point-of-use interventions mentioned in the
literature review, especially the physical and chemical treatment interventions, are
impractical because of costs and sustainability and therefore not suitable for
impoverished rural households in developing countries such as South Africa (Sobsey,
2002; Gundry et al., 2004). In addition, the literature study has also showed that
improving the microbiological quality of water before consumption would reduce
diarrhoeal disease together with sanitation and hygiene education (Mertens et al., 1990;
Hoque et al., 1995). However, many of the studies have used indicator microorganisms
to assess the effectiveness of interventions. The literature review has indicated that
most of the currently used indicator microorganisms used to evaluate the
microbiological quality of water have shortcomings and will only give an indication of
Chapter 2
53
the potential risk associated with the transmission of waterborne diseases (Moe et al.,
1991; Payment and Franco, 1993; Sobsey et al., 1993; Sobsey et al., 1995).
Several potentially pathogenic microorganisms in water polluted by human and animal
faeces could cause diarrhoeal diseases in consumers (Sobsey et al., 1993; Gerba et al.,
1996; Grabow, 1996; Leclerc et al., 2002; Theron and Cloete, 2002). Little information
on the origin of faecal contamination in the traditional and CDC safe storage containers
are presently available.
Literature has showed that microbiological and chemical
indicators can be used to distinguish between human and animal faecal pollution in
water (Jagals et al., 1995; Sinton et al., 1998). However, no single microorganism or
chemical determinant could reliably distinguish human from animal faecal
contamination and therefore, the use of a combination of chemical and microbial
determinants together may provide the best solution for identifying the origin of faecal
pollution at the point-of-use (Jagals et al., 1995; Sinton et al., 1998).
Consequently, the literature study has indicated that the best interventions available that
will be applicable to conditions in rural communities in South Africa included the use of
the CDC safe storage container together with a chemical treatment such as sodium
hypochlorite solution.
The aim of this study was therefore to improve the
microbiological quality of drinking water in rural households at the point-of-use by the
implementation of intervention strategies which included the use of traditional storage
containers as well as the CDC water storage container, with or without the addition of a
sodium hypochlorite solution. The results obtained from this study would be used to
provide information to the DOH and DWAF, which can be used in future water and
health policy formulations to prevent waterborne outbreaks in these rural communities.
Chapter 2
54
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement