Food Control 20 (2009) 1158–1166
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/foodcont
Quantiﬁcation and variability of Escherichia coli and Staphylococcus aureus
cross-contamination during serving and consumption
of cooked thick porridge in Lungwena rural households, Malawi
S. Taulo a,b,*, A. Wetlesen a, R.K. Abrahamsen a, J.A. Narvhus a, R. Mkakosya c
Department of Chemistry, Biotechnology and Food Sciences, Norwegian University of Life Sciences, P.O. Box 5003, N-1432, Ås, Norway
Department of Environmental Health, University of Malawi, The Polytechnic, P/B 303, Blantyre 3, Malawi
Department of Microbiology, University of Malawi, College of Medicine, P/B 360, Blantyre 3, Malawi
a r t i c l e i n f o a b s t r a c t
Article history: The study investigated bacterial transfer to cooked thick porridge via ladles and hands during serving in
Received 29 June 2008 29 households in Lungwena, rural Malawi. Household stored water used for hand and ladle washing, was
Received in revised form 3 March 2009 contaminated with Escherichia coli and Staphylococcus. aureus from hands of members of the household or
Accepted 18 March 2009
from contaminated ladles used in food preparation. The results showed that hands became contaminated
with E. coli and S. aureus cells in the range 0.6–3.7 and 2.2–4.3 log10 CFU/cm2, respectively, following
washing with the contaminated water. Ladles became contaminated with 0.9–3.2 log10 CFU/cm2 of E. coli
cells whereas contamination with S. aureus on ladles ranged between 1.9 and 4.6 log10 CFU/cm2. Bacterial
Cooked thick porridge
transfer from hands to food ranged from <1 to 3.6 log10 CFU/g for E. coli and 2.1 to 4.2 log10 CFU/g for S.
Hand wash water aureus. Ladle surfaces transferred from 1.3 to 3.1 and from 1.2 to 4.3 log10 CFU/g of E. coli and S. aureus,
Hand contamination respectively, on to the food. Contamination of food by hands was signiﬁcantly (p < 0.05) higher than that
of ladles and transfer of S. aureus was signiﬁcantly (p < 0.05) higher than that of E. coli. The amount of
bacteria transferred to the recipient depended on the wash water type and bacteria type. The study
has demonstrated that although the traditional cooking of thick porridge inactivates S. aureus and E. coli,
the porridge can be contaminated with bacteria during consumption using hands and serving on to a
plate with wooden ladles.
Ó 2009 Published by Elsevier Ltd.
1. Introduction transfer to food (Bloomﬁeld & Scott, 1997; Kusumaningrum,
Van Putten, Rombouts, & Beumer, 2002; Miller, Brown, & Call,
Diarrhoeal infections acquired in the home are associated 1996; Moore et al., 2003). Transfer of pathogens to utensils
with poor hygiene practices and improper food handling may take place either by direct contact with contaminated ob-
(Kusumaningrum, Riboldi, Hazeleger, & Beumer, 2003). Food, jects or indirectly through airborne particles (Baker, Naeeni, &
water and surfaces may be contaminated with considerable Bloomﬁeld, 2003; Cogan, Slader, Bloomﬁeld, & Humphrey,
numbers of pathogenic microorganisms in household kitchens 2002; Kusumaningrum et al., 2002). Previous studies have quan-
during food serving and consumption and may thereby result tiﬁed bacterial survival and cross-contamination between hands
in serious diseases. In many situations, the hands are major and food and from utensils to food (Chen, Jackson, Chea, &
vehicles that contaminate and disseminate faecal–oral bacteria Schaffner, 2001; Dawson, Han, Cox, Black, & Simmons, 2006;
(Burt, Volel, & Finkel, 2003; Moore, Sheldon, & Jaykus, 2003). Kusumaningrum et al., 2003; Scott & Bloomﬁeld, 1990; Zhao,
On the skin surface, hands harbour a resident and/or transient Zhao, Doyle, Rubino, & Meng, 1998).
ﬂora (Aycicek, Aydogan, Kucukkaraaslan, Baysallar, & Basustao- S. aureus and E. coli are resident and transient bacteria, respec-
glu, 2004; Larson, 2001). Utensils for food preparation, such as tively, on hands and are associated with poor hygiene practices
wooden chopping boards, stainless steel and knives have also (Department of Health, 2000; Hundy & Cameron, 2002). Toxin-
been found to harbour pathogenic microorganisms, allowing producing strains of S. aureus are the leading cause of gastroen-
teritis following handling of food by persons who carry the
microorganism in their noses and skin (Le Loir, Baron, & Gautier,
* Corresponding author. Address: Department of Chemistry, Biotechnology and
2003; Portocarrero, Newman, & Mikel, 2002). These bacteria are
Food Sciences, Norwegian University of Life Sciences, P.O. Box 5003, N-1432, Ås,
Norway. Tel.: +47 6496 6238/265 1877537; fax: +47 6496 5901/265 1870578. present in about 60% of a given human population and can also
E-mail address: firstname.lastname@example.org (S. Taulo). survive on hands, knives, chopping boards and dish cloths for
0956-7135/$ - see front matter Ó 2009 Published by Elsevier Ltd.
S. Taulo et al. / Food Control 20 (2009) 1158–1166 1159
hours to days after initial contamination (Barro et al., 2006; material which has been implicated in cross-contamination (Bryan,
Kusumaningrum et al., 2002; Scott & Bloomﬁeld, 1990). E. coli 1988; Miller et al., 1996). Although food particles are removed
is transmitted via the faecal–oral route and is used as indicator from ladles, bacteria may still be attached to the surface. In Lungw-
of recent faecal contamination. The organism is naturally found ena, stored water is used for ladle dipping and 65–80% of such
in the human intestine and although most strains are harmless water from this area has been found to be contaminated by E. coli,
some, such as serotype 0157:H7 can cause serious illness. The with mean concentrations of 2–4 log10 CFU/100 ml (Osmundsen,
detection of E. coli in any sample signals the potential presence 2005; Taulo, Wetlesen, Abrahamsen, Mkakosya, & Kululanga,
of pathogenic microorganisms originating from the intestinal 2008).
tract of humans and warm-blooded animals (Brown et al., Our previous study (Taulo et al., in press), on the microbiologi-
2002; Moore & Grifﬁth, 2002). E. coli has a relatively limited cal quality of food prepared and served in rural households of
ability to survive on hands and in the environment but can sur- Lungwena found thick porridge to be the most frequently contam-
vive on hand surfaces in sufﬁcient numbers to allow transmis- inated food containing high numbers of viable E. coli, S. aureus and
sion to other surfaces (Dawson et al., 2006). Salmonella spp. Considering that most household members eat the
Lack of personal hygiene among food handlers is one of the food after washing their hands in communal hand washing water
most commonly reported practices contributing to food-borne ill- and that the food is served onto plates using ladles dipped in con-
ness and poor hand and surface hygiene is also a ‘‘signiﬁcant con- taminated water, it was hypothesised that the cooked thick por-
tributory factor” (Cogan et al., 2002; Collins, 2001; Olsen, ridge becomes re-contaminated after cooking and cooling via the
MacKinon, Goulding, Bean, & Slutsker, 2000). Previous studies con- hands and ladles. This study was carried out in order to quantify
ducted in food industries have shown that microorganisms are E. coli and S. aureus in hand and ladle wash water and assess the
transferred to the hands in the process of handling food and potential of these bacteria to cross-contaminate hands and ladles
through poor personal hygiene after visiting the lavatory, resulting and subsequently contaminate cooked thick porridge during food
in the hands being heavily contaminated with faecal-related serving and consumption.
pathogens (Barza, 2004; Taylor, Brown, Toivenen, & Holah, 2000).
Guzewich and Ross (1999) found that in 89% of outbreaks caused
by food contamination by food workers, pathogens were trans- 2. Materials and methods
ferred to food by workers’ hands. Proper hand washing has been
recognised as one of the most effective measures to control the 2.1. Study participants
spread of faecal–oral disease (Guzewich & Ross, 1999; Lues &
Van Tonder, 2006; Michaels et al., 2004; Montville, Chen, & Schaff- Six villages were chosen randomly from dry (three villages) and
ner, 2001). The United States Food and Drug Administration (USF- wet (three villages) areas of Lungwena. Twenty nine households
DA) Food Code 2005 stipulates that proper hand washing entails (ﬁve households from each village except one with four house-
washing the hands with running warm water, soap, friction be- holds) were randomly picked. Each household was visited once,
tween the hands for 10–15 s, rinsing, and drying with clean towels during a two months period between July and August 2006. During
or hot air (USFDA, 2005). each visit, seven samples were collected from each household:
In most of the rural communities of developing countries in hand wash water, ladle dip/wash water, hand surface swab, ladle
Africa, people eat their main meals with their hands and ﬁngers. surface swab and food samples from pot after cooking (porridge–
Many cultures have a strong tradition of washing hands before eat- pot), after contact with hands (porridge–hand), and after contact
ing, but often with cold water and without soap (Hoque, 2003; Ho- with ladles (porridge–ladle). The porridge–pot food sample was
que & Briend, 1991). Although washing of hands with water alone collected immediately after cooking when the temperature was
reduces microbial loads on hands (De Wit & Kampelmacher, 1981; still high (P65 °C). A total of two hundred and three samples were
Michaels, 2002; Michaels, Ganger, Chia-Min, & Doyle, 2003), a collected. The samples were collected in cooler boxes at approxi-
combination of sanitizers and/or detergents and warm water have mately 4–8 °C and transported within 2 h from the site to Mango-
demonstrated to be more effective (Cogan, Bloomﬁeld, & Hum- chi district hospital laboratory where they were kept at 4 °C and
phrey, 1999; Courtenay et al., 2005; Josephson, Rubino, & Pepper, analysed within 2–4 h. Some samples were taken to the University
1997; Michaels et al., 2003). In Bangladesh, during post-defecation laboratory (Malawi Polytechnic) for further analysis.
hand washing, women living in the semi-urban rub their hands
with soil, soap or ash and rinse them with safer water (tube well 2.2. Sample collection protocols
water) followed by drying with a clean cloth. This practice results
in 80% reduction of faecal coliform counts of hands over traditional 2.2.1. Hand washing water
post-defecation hand washing (Hoque, Mahalanabis, Alam, & Is- These samples were collected in order to quantify the numbers
lam, 1995). Use of soap, ash or mud as scrubbing agents has been of E. coli and S. aureus in communal hand washing water after
demonstrated to have similar efﬁciency in faecal bacterial removal washing. Before eating from the communal plate, members of
(Hoque, 2003; Hoque et al., 1995). the household (average of three members) were asked to wash
In Malawi, and Lungwena in particular, the main meal con- their hands in the same water bowl (household stored water) with-
sumed in most households is cooked thick porridge locally known out changing the water and without drying the hands, this being
as nsima. The meal is prepared by adding maize ﬂour to boiling their usual custom. This was done after the temperature of the
water until a solid dough-like porridge is reached. The cooking water was recorded using a Pyrex digital probe thermometer
temperature is assumed highly enough to kill or inactivate most (Merck, Midrand, South Africa). A 100 ml sample of the hand wash-
pathogenic microorganisms. It is common that members of the ing water was collected from each household and put into 100 ml
household eat together from one plate after washing hands in a Colilert bottles (IDXX, London, UK).
communal bowl. The water in the bowl is usually cold and is not
changed in between hand washing by different people. Shortly 2.2.2. Ladle wash/dip water
after the food is cooked, it is usually served onto a plate using woo- These samples were collected in order to quantify E. coli and S.
den ladles which are dipped in cold or warm water in between the aureus in the water which had been used for washing or dipping
serving on to the plate as this prevents the porridge sticking to the ladles prior to serving the food onto the plate. Household stored
ladle. Most of the ladles used in Lungwena are made of wood, a water was poured in the usual ladle washing bowl whose hygienic
1160 S. Taulo et al. / Food Control 20 (2009) 1158–1166
status was not assessed. The stored water was assumed to be 2.2.6. Transfer of bacteria from water to ladle surface
contaminated based on previous reports (Osmundsen, 2005; Taulo The ladles’ surfaces were disinfected in the same manner as for
et al., 2008). The water was further contaminated by dipping in the hand surface, with a high (P70 °C) water temperature (Section
unwashed ladle and the temperature of the water was recorded 2.2.4). The disinfected ladles were further dipped into ladle wash
and 100 ml water was collected in 100 ml Colilert bottle. water for 15 min to allow attachment of E. coli and S. aureus cells.
Cells were recovered as described by Dawson et al. (2006). The la-
2.2.3. Control samples (porridge–pot) dle’s inside surface (ca. 5 Â 5 cm) was swabbed with a sterile cot-
Samples of cooked thick porridge were collected soon after ton swab moistened with BPW and further swabbed (three times)
cooking and before being exposed to contaminated hands and la- with a 10 ml of 0.1% sterile peptone water using the same swab (in
dles. In addition, this sampling point could conﬁrm that such order to obtain maximum recovery). Both the peptone rinse water
foods attain high temperatures that can inactivate a wide variety and swabs were transferred into 100 ml sterile Colilert bottles.
of bacteria. In each household, the person who usually prepares
the household food and who had never received any food safety 2.2.7. Bacterial transfer from ladle surface to food (porridge–ladle)
health education was requested to prepare thick porridge using Transfer of bacteria from the contaminated ladles to the food
any type of maize/corn ﬂour that was available in the house was determined by dipping a re-sterilised ladle into the ladle wash
(the microbiological status of the ﬂour was not determined). water for 15 min to allow attachment followed by collection of
Within 2 min of preparation, a 100–150 g sample of thick por- 100–150 g of the food sample directly from the cooking pot with
ridge was collected directly from the pot (temperatures P70 °C) the ladle while the ladle was still wet (in the usual manner). The
and put into a sterile Ziploc bag (Cofresco) after recording the sample was collected directly from the pot within 4–6 min after
temperature. cooking and put into sterile Ziploc bags after recording the
2.2.4. Bacterial transfer from water to hand surface
One participant from each household was requested to decon- 2.3. Sample processing and enumeration of the microbes in thick
taminate his/her hands by washing thoroughly for 2 min in warm porridge
water using a liquid antibacterial soap solution (Unilever South
Africa, Blantyre, Malawi), followed by rinsing for 20 s with hypo- Microbiological analysis of cooked thick porridge was per-
chlorite containing 5000 ppm of available chlorine and shaking formed by aseptically collecting cooked thick porridge from the
the hand to effect drying. This was done in attempt to disinfect area that was in contact with hands’ and ladles’ surface. Approxi-
hands as demonstrated by Baker et al. (2003) and Cogan et al. mately 10 g of the sample in the Ziploc bags was transferred into
(1999, 2002) who recorded a disinfection efﬁciency of 90% and Stomacher ﬁlter bags (A.J. Seward, London, UK), diluted with
97% for contaminated hands and utensils, respectively. In addi- 90 ml sterile BPW and homogenised with Colworth Stomacher
tion, our experimental trial in the laboratory conducted on hands Blender (Lab Blender 400, Seward Medical, London, UK) for 1 min
and ladles of two participants yielded zero counts for both bacte- at 230 rpm. Tenfold dilutions were made with BPW. 1ml of each
ria. The hypochlorite was used basically to disinfect the cold rinse sample at appropriate dilutions was plated in duplicate on selected
water. Immediately after drying, the participant was requested to media, as described below.
dip both hands in the communal hand washing water for 30 s in E. coli was cultured on 3MTM PetriﬁlmTM E. coli/Coliform Count
order to allow adsorption of the test organisms. Sampling of con- Plates (E-C plates) (3M Company, St. Paul, USA) and incubated at
taminated hands was done as described by Dawson et al. (2006). 37 °C for 24 h (Nordic Committee on Food analysis, 1993). Blue
The inside surface of part of the palms and ﬁngers (5 Â 5 cm area) colonies with entrapped gas bubbles on E-C plates were enumer-
of one hand was swabbed by cotton swabs pre-moistened in Buf- ated as presumptive E. coli. Primary identiﬁcation involved con-
fered Peptone Water (BPW; Oxoid, Basingtoke, UK). The surface ducting indole test with Kovacs reagent (Merck, Midrand, South
was then further rinsed with 3 Â 10 ml of BPW (Oxoid) using Africa) after sub-culturing E. coli colonies into tryptone water
the same swab (in order to obtain maximum recovery) and col- and incubating at 44 °C for 18–24 h (Oxoid). For detection of S.
lected into sterile stomacher bags (Stomacher 400, Seward Medi- aureus, 1 ml of each serial dilution was spread onto Staphylococcus
cals; London, UK). Finger-tips were massaged in the stomacher Express Petriﬁlms (3M Company, St Paul, USA) and incubated at
bags containing the peptone water rinse for 30 s in order to max- 37 °C for 24 h (Nordic Committee on Food Analysis, 1993). Colo-
imise the recovery of bacteria from the ﬁngernails (participants nies showing pink colour were identiﬁed and enumerated as S.
repeated the whole exercise twice). The cotton swabs and the aureus. Petriﬁlms with blue colonies were further incubated for
10 ml peptone water rinse were transferred into 100 ml sterile 60 min and those colonies that changed colour to pink after over-
Colilert bottles. laying the petriﬁlms with Staphylococci disks were also recorded
as S. aureus. Conﬁrmation was made by observing yellow colour-
2.2.5. Transfer of bacteria from hand to food (porridge–hand) ation of colonies on mannitol salt agar after 24 h of incubation at
Whilst the porridge–pot food was still hot (P65 °C), it was 37 °C (Difco Manual., 1998).
transferred directly onto the communal eating plate. All members
of the household sharing the meal were requested to start eating 2.4. Statistical analysis
the food with their hands washed in the hands contaminated water
(wherever possible). The same participant (in Section 2.2.4) was The number of bacteria isolated from contaminated hands, la-
asked to pick about 100–150 g food sample from the eating plate dles and cooked thick porridge were expressed as log10 CFU/unit.
after all the people had started eating and transferred it into sterile Bacterial counts were compared statistically using one way analy-
Ziploc bag after recording its temperature. In the case of food that sis of variance by means of Minitab version 14 (Minitab Inc., 2004:
was very hot to handle, participants attempted to collect samples PA, USA), with signiﬁcance deﬁned at the p-value of 60.05. The
from the food surface when members had not started eating. In independent student t-test was used to test for differences in bac-
the case where participants could not manage to pick hot food terial counts between two samples. Correlation between tempera-
samples, samples were collected aseptically with sterilised forceps ture and bacterial counts was conducted using regression analysis.
to just ascertain that the food did not get contaminated by external Transfer of bacteria to hand and ladle surfaces were expressed as
environment. log10 CFU/cm2 according to Dawson et al. (2006).
S. Taulo et al. / Food Control 20 (2009) 1158–1166 1161
3. Results Table 1
Geometric means of positive E. coli and S. aureus in hand and ladle washing water and
on hands and ladles after contact with waters, and in cooked thick porridge after
3.1. Occurrence of E. coli and S. aureus contact with washed hands and ladles in rural households of Lungwena, Malawi.
Sample Bacterial counts (Log10 CFU/unit)
Results in Fig. 1 show incidences of the E. coli and S. aureus for
various sample types analysed. Bacteria were not detected in por- E. coli S. aureus
ridge–pot, the food samples which were collected at high temper- Mean Min. Max. Mean Min. Max.
atures (P70 °C) within 2 min after cooking. The results show that, Hand wash watera 3.2 ± 0.8 2.0 4.4 4.1 ± 0.5 3.3 4.9
during consumption and serving of the food with hands and ladles, Hand surfaceb 2.8 ± 0.9 0.6 3.7 3.2 ± 0.6 2.2 4.3
E. coli and S. aureus were present on hand and ladle surfaces and in Ladle wash watera 2.9 ± 0.8 1.0 3.9 3.7 ± 0.6 2.3 5.0
the cooked food, indicating that both hand and ladle wash water Ladle surfaceb 2.3 ± 0.8 0.9 3.2 3.0 ± 0.8 1.9 4.6
Porridge–handc 2.3 ± 0.8 0.60 3.6 3.1 ± 0.6 2.1 4.2
caused contamination. The highest incidence of E. coli (22%/29– Porridge–ladlec 2.1 ± 0.7 1.3 3.1 3.2 ± 0.9 1.2 4.3
76%) and S. aureus (26%/29–90%) was in hand washing water and
incidences of S. aureus were signiﬁcantly (p < 0.05) higher than Bacterial counts expressed as log10 CFU/ml.
Bacterial counts expressed as log10 CFU/cm2.
those of E. coli. Generally, hand and ladle surfaces that were posi- c
Bacterial counts expressed as log10 CFU/g.
tive for E. coli were also positive for S. aureus. All S. aureus contain-
ing wash waters contaminated the disinfected hands and ladles,
while 91% (20/22) and 89% (17/19) of E. coli contaminated hand counts than hand wash water. There was great variation in bacte-
and ladle wash waters resulted in contamination of the disinfected rial counts in hand and ladle wash waters between households as
hands and ladles, respectively. demonstrated by large standard deviations and wide bacterial
In only three households, hand and ladle wash water did not count ranges. Bacterial counts on hand and ladles surfaces did
contain either bacteria. Subsequent dipping of hands and ladles not differ signiﬁcantly. Food that had been touched with hands
in these water samples did not result in contamination and neither contained marginally higher (2.8 ± 0.8 log10 CFU/g) bacterial counts
was bacterial growth observed in food samples served and con- than food served with ladles (2.6 ± 1.0 log10 CFU/g) but the differ-
sumed by such hands and ladles. However, E. coli was not detected ences were not signiﬁcant. It was not possible to observe transfer
on the hands of two participants who had previously washed their of E. coli and S. aureus cells from contaminated hands to food in se-
hands in water that contained E. coli. In the case of ladle wash ven and six samples, respectively. Similarly, we were not able to
waters, E. coli was not detected in six samples and ladles washed observe transfer of E. coli and S. aureus cells from contaminated la-
in these waters did not contain the organism and nor did food dles to food in eight and 13 samples, respectively (data not shown).
served with such ladles. Incidences of E. coli and S. aureus in the In general, bacterial transfer from hand wash water to food via
food samples (45% and 31%; 69% and 31%) occurred as a result of hands was consistently higher than transfer from ladle wash water
serving and consuming the food with contaminated ladles and to food via ladles. S. aureus was transferred in signiﬁcantly higher
hands, respectively. The frequency of occurrence of both bacteria quantities than E. coli.
in ladle wash/dip waters, ladle surfaces and ladle contaminated
foods (porridge–ladle) was signiﬁcantly (p < 0.05) lower than that 3.3. Sampling temperature and bacterial counts of wash waters
of hand washing waters, hand surfaces and hand contaminated
foods (porridge–hand). Fig. 2 shows bacterial counts distributed into different temper-
ature ranges at which water samples were collected. Hand and la-
3.2. Bacterial transfer dle washing waters were sampled at mean temperatures of
32.86 ± 9.85 and 34.83 ± 13.81 °C, respectively. A comparison of
Table 1 summarizes geometric means and ranges of bacterial sampling temperatures showed no signiﬁcant difference between
counts for positive samples of wash waters, surfaces and foods. La- hand wash water and ladle wash water. Of the wash water sam-
dle wash water contained signiﬁcantly (p < 0.05) lower bacterial ples, 27 samples were collected at temperatures 630 °C, demon-
Frequencies of positive samples (%)
Bacteria counts (Log10 CFU/ml)
80 HWW. E.coli
70 LWW E. coli
LWW. S. aureus
20 30 40 50 60 70 80
Sample types Fig. 2. Bacterial counts of E. coli and S. aureus in samples of hand wash water
(HWW) and ladle wash water (LWW) at different temperatures of the waters at
Fig. 1. Incidences (%) of E. coli and S. aureus in samples collected in 29 households of sampling in 29 Lungwena households, Malawi. The zero log CFUs represent
Lungwena, Malawi. zero CFU/ml detected.
1162 S. Taulo et al. / Food Control 20 (2009) 1158–1166
strating frequent use of cold water for hand and ladle washing. than 64 °C. As a result of this, we decided to collect the rest of the
High temperatures (P35 °C) were recorded in some hand and ladle samples at lower temperatures (a deviation from the normal prac-
wash waters and further inquiry revealed that the water had been tice) since we assumed that bacteria may not be detected at such
boiled or left in the sun prior to use. In the case of the relatively high temperatures. As expected, all 20 samples did not contain
high hand wash water temperatures, participants had to wait for any of the organisms. All the samples collected at a temperature
the water to cool down before washing their hands. S. aureus was range between 30 and 40 °C showed the presence of both organ-
detected in the 27 out of 58 samples, while E. coli was detected isms. At this temperature range, E. coli and S. aureus counts ranged
in 23 out of 58 samples collected at temperatures 630 °C. Fourteen 1–3 and 2.8–4 log10 CFU/g, respectively. Only three samples were
samples were collected at a temperature range between 31 and collected at a temperature range between 40 and 650, out of which
40 °C and all the samples contained S. aureus cells, while 13 con- two were positive for both organisms. Neither of the organisms
tained E. coli cells at the same temperature range. Only 17 positive was detected at temperatures >47 °C in both types of food samples.
samples were detected at temperatures P40 °C, with ﬁve and eight
of the samples containing E. coli and S. aureus, respectively. Neither
E. coli nor S. aureus were detected in wash waters over 45 °C. 4. Discussion
3.4. Sampling temperature and bacterial counts of food samples The transmission of bacteria from water to food via hands dur-
ing normal consumption and serving of the food with hands and la-
Fig. 3 presents bacterial counts results distributed into different dles, respectively, is important because it can lead to food-borne
temperature ranges at which food samples were collected. Tem- disease outbreaks. Most transfer experiments have been per-
peratures of porridge–hand were lower (38.76 ± 12.25 °C) than formed with inoculated products with high numbers of an indica-
those of porridge–ladle (64.48 ± 21.16 °C). The high temperatures tor bacterium (Chen et al., 2001; Zhao et al., 1998). This is possibly
recorded for porridge–hand were from samples that were collected one of the very few published studies reporting the transfer of bac-
when participants had not started eating as the food was hot to teria from contaminated wash water to food (cooked thick por-
handle. However, some participants attempted to pick three hot ridge) via hands and wooden ladles surfaces in a natural and
porridge–hand samples (at temperature range 39–45 °C) while realistic food preparation environment. It should be noted that
cooling the food by blowing on it and transferred the samples into Lungwena is a Muslim community whose religion entails mechan-
sterile bags at a very short period of time (<5 s). For the purpose of ical cleaning of their anal region using hands while pouring water
comparisons hot samples (seven samples) that could not be picked from a container after defecation rather than using toilet papers
by hand were also collected aseptically with sterilised forceps (personal observation). Washing of the hands with soap, ash or
(temperatures >51 °C). There were no detectable cells of the organ- mud after toilet use is not often practiced. Hoque (2003) explains
isms in these samples, conﬁrming that there was no contamination that if hands are not properly washed at this time, faecal pathogens
from equipment and environment and that cooking inactivated the are likely to be transmitted to water, food and other households
organisms. Of the 22 samples collected by hands, 13 were tested objects and ultimately to the mouth. Shojaei, Shooshtaripoor, and
positive for E. coli, while S. aureus was detected in 20 samples. Most Amiri (2005) further speculated that in such communities, foods
of the positive porridge–hand samples were collected at tempera- are more likely to be contaminated with faecal matter during prep-
ture ranges between 25 and 40 °C and this possibly demonstrates aration or dissemination by the food handler.
the temperature at which the food is taken using hands. At this Results concerning the occurrence of contamination in hand
temperature range, E. coli counts ranged 1–3 log10 CFU/g whereas and ladle wash waters (Fig. 1) are in agreement with those of Barro
counts of S. aureus ranged 2–4 log10 CFU/g. et al. (2006), while those for hand surface contamination (69% and
In the case of porridge–ladle, it was observed that most of the 90% for E. coli and S. aureus, respectively) are similar to those of
samples (20 samples) were being collected at temperatures higher Lues and Van Tonder (2006) who found E. coli and S. aureus to oc-
cur in 40% and 88% of hands of food handlers. Bacterial counts (E.
coli) on washed hands (2.8 ± 0.9) are in close agreement with
counts reported by Courtenay et al. (2005), De Wit and Kampelm-
acher (1981), and Hansen and Knochel (2003) whereas numbers of
4 Variable E. coli cells transferred to the foods by hands are fairly supported
by those reported by Chen et al. (2001) who studied the transfer
Bacterial counts (Log10 CFU/g)
Porridge-ladle E. coli of Enterobacter aerogenes (an organism with similar characteristics
to those of E. coli) from washed hands to lettuce in a laboratory.
Signiﬁcant variations were observed in wash waters bacterial lev-
els between households (Table 1). This possibly indicates the ex-
treme diversity of hygienic practices undertaken during the hand
washing, the number of participants who washed hands in the
water in each household and differences in water treatment
(boiled and non boiled wash waters). It was not possible to detect
1 E. coli and S. aureus in porridge–pot which is cooked at high tem-
perature in Lungwena, re-afﬁrming the ﬁndings of Mosupye and
von Holy (2000).
0 Despite the fact that all the hands wash waters were collected
when people hand washed, E. coli and S. aureus cells were not de-
20 30 40 50 60 70 80 90 100 tected in six and three samples, respectively. This can be explained
Samplng temperatures by the fact that participants from those households may have
washed their hands well in advance because they anticipated the
Fig. 3. Bacterial counts of E. coli and S. aureus in food after eating with hands
(porridge–hand) and in food after serving with ladles (porridge–ladle) at different
purpose of our study. The occurrence of E. coli and S. aureus in high
temperatures in the food at sampling in 29 Lungwena households, Malawi. The zero numbers especially in the wash water was not unexpected as our
log CFUs represent zero CFU/g detected. previous studies (Taulo et al., 2008) and those of previous
S. Taulo et al. / Food Control 20 (2009) 1158–1166 1163
researchers, Roberts et al. (2001), and Osmundsen (2005) have aureus from surfaces when there is moisture compared to E. coli
shown that in most households of Malawi, stored water is contam- which forms bio-ﬁlms, making E. coli attach ﬁrmly on to ladle sur-
inated even before it is used as wash water. Roberts et al. (2001) faces. As mentioned before, S. aureus often occurs as clumps of cells
found faecal coliform in the range between 1 and 3 log10 CFU/ml, that provide a high chance of detachment from the surfaces. The
while Osmundsen (2005) found E. coli counts ranging between 2 clumps also protect the organism from extrinsic factors such as
and 4 log10 CFU/ml. This ﬁnding is also observed by previous temperatures (Kusumaningrum et al., 2003).
researchers (Chen et al., 2001; Mattick et al., 2003). E. coli is an indicator of fresh faecal contamination and is asso-
Contamination of the ladle wash water was lower than that of ciated with poor hygienic practices. De Wit and Rombouts (1992)
the hand wash water possibly because the hand wash water was explains that E. coli is normally absent from hands. The presence
contaminated by several dirty hands. In some households, ladles of E. coli in wash water and in hand rinse samples in our study re-
were dipped in waters whose temperatures were greater than that ﬂects among other things, hygienic practices in connection with
of hand wash water. This ﬁnding stresses the importance of using toilet (Moore & Grifﬁth, 2002; Shojaei et al., 2005). S. aureus exists
heated water for dipping in ladles before the ladle is used for serv- on skin, in noses and on wounds of humans and the presence of
ing food. The higher numbers of S. aureus compared to E. coli in this organism indicates improper hygienic practices in food prepa-
wash water observed in our study possibly relates either to the fact ration and consumption, and possible cross-contamination. Its
that S. aureus is ubiquitous in the environment or that S. aureus of- presence on food products is regarded as a public health hazard,
ten occurs as clumps of cells that protect the organism from extrin- due to the ability of toxin-producing strains of this organism to
sic factors such as temperatures. E. coli, however, has a relatively cause staphylococcal intoxication at concentration levels of
limited ability to survive on the surfaces and in the environment 5 log10 CFU/g (Atanassova, Meindl, & Ring, 2001; Kloos & Banner-
(Kusumaningrum et al., 2003). The above may also explain the man, 1999; Rørvak & Granum, 1999).
presence of S. aureus in hand wash water samples which were In Malawi, diarrhoeal diseases in under-ﬁve children have been
tested negative for E. coli. If the wash water is not boiled prior to associated with Cryptosporidium species (Morse et al., 2007). Sal-
the washing, bacterial presence is a mixture of the initial micro- monella and Shigella species are often isolated in patients with
ﬂora in the stored water in the homes and further contamination bloody stools whereas Vibrio cholera is not uncommon during the
during washing of soiled hands and ladles as both bacteria can re- rainy season (Ministry of Health, 1985). Isolation of these bacteria
side on the hands and utensil surfaces. includes a faeces to hand spread and indicating poor hygiene prac-
Hands are probably the main vector for transfer of pathogens in tices (Hatakka, Bjorkroth, Asplund, Maki-Petays, & Korkeala, 2000).
the home and microbes such as S. aureus are likely to occur on In our study, the isolation of E. coli and S. aureus possibly indicates
hands of people who are carriers (Bloomﬁeld, 2003). Although potential presence of other human pathogenic bacteria that cause
household members washed their hands before consuming the diarrhoeal disease in Lungwena. The levels of S. aureus and E. coli
food, both organisms remained on the palms and ﬁngers, re-afﬁrm- cells recorded in the porridge in this study were below the infec-
ing the results of Patrick, Findon, and Miller (1997). It is likely that tive dose. If the food is consumed straight away, the numbers of
the food was cumulatively contaminated with each contaminated S. aureus and E. coli found in this study may not be a problem. How-
hand following washing with contaminated water as samples were ever, households that cook the food well in advance or eat leftover
taken after all the members had started eating. Our ﬁndings that food can be at risk as such practices can allow the growth of organ-
hands transferred signiﬁcantly higher bacterial counts on to food isms to infective doses.
than ladles (utensils) re-afﬁrms that the hands are the major vehi- Many viable E. coli and S. aureus cells were detected in samples
cles that contaminate and disseminate faecal–oral bacteria as re- with temperatures between 21 and 40 °C (Figs. 2 and 3), tempera-
ported by previous researchers (Burt et al., 2003; Chen et al., tures which possibly represent hand wash water temperatures as
2001; Moore et al., 2003). This demonstrates a potential risk of well as temperatures at which most main meals are eaten using
hands over ladles in transmitting diarrhoeal diseases in Lungwena. hands in Lungwena. At this temperature range, these bacteria are
Hand washing before eating the main meal is a routine practice able to grow and thereby pose a potential risk of food-borne dis-
and is considered more important than at any other time, for ease. The results of food samples which were collected with ladles
example after visiting the toilet. This practice could be a potential at low temperatures demonstrate the potential of the ladles to
risk to the Lungwena community as hands can accumulate large transfer pathogens on to the food when the food is served at such
numbers of bacteria on the surface and transfer them on to food low temperatures. On the other hand, hot food (temperatures over
when eating the main meal. 64 °C) served on to a plate with contaminated ladles may not
Unclean serving utensils have been found to contaminate transmit the bacteria. In the case of consumption by hand, the re-
cooked food with E. coli and S. aureus during meal serving (Barro sults have shown that people do not take food when it is very hot
et al., 2006; Mosupye & von Holy, 2000). Most of the households but wait until the food cools down. Under such cold temperatures
were seen using cooking pots also as bowls and also used for cook- of the food, the bacteria easily transfer on to the food. This demon-
ing other food items and therefore could possibly have retained strates the dangers of use of contaminated hands when taking
food residues if not properly cleaned. Proliferation of bacteria in thick porridge meals whose temperatures are relatively low. None
food residues has been established by many previous researchers of the foods with temperatures >45 °C were taken using hands and
(Kusumaningrum et al., 2003; Mattick et al., 2003). Wooden ladles, such foods did not show presence of the bacteria.
just like wooden chopping/cutting boards are porous and contain It was surprising to observe that neither of the organisms was
crevices and abrasions. Such structures can easily promote forma- detected at temperatures of P50 °C. Small quantities of organisms
tion of bio-ﬁlms, especially when the ladles are not cleaned prop- may have been transferred into the wash water and onto food sam-
erly and allowed to stay for some days. On surfaces of such ples when the temperatures were above the organisms’ optimum
utensils, bio-ﬁlms dry-up and shield bacteria. Following exposure growth temperature. It is however, likely that the bacteria could
to water, the bio-ﬁlms are likely to detach and slough off from not be detected by the methods used in this study as heat-stressed
the surfaces together with bacteria (Rusin, Orosz-Coughlin, & Ger- bacteria cannot always be detected by cultivation, but still pose
ba, 1998). The above reason may also explain the lower transfer of threat (Robine, Boulangé-Peterman, & Delangerè, 2002). The log
the bacteria to foods by ladles compared to the transfer from hands linear bacterial death kinetics theory assumes that inactivation of
to food. Four ladles that did not show presence of E. coli contained bacteria results from random hits of key targets within the micro-
S. aureus and this can be attributed to the ease of detachment of S. bial cells and that when values of extrinsic stress factors are out-
1164 S. Taulo et al. / Food Control 20 (2009) 1158–1166
side the range of growth, bacterial cells may be damaged (József, tion method we applied in the ﬁeld, the hand and ladle disinfec-
2001). A study on heat shocking of E. coli by Kaur, Ledward, Park, tion procedure followed in this study may not have completely
and Robson (1998), demonstrated that heating at temperatures be- inactivated the bacteria, as Baker et al. (2003) and Cogan et al.
tween 42 and 50 °C for 0–5 min can reduce organism by 1.3– (1999,2002) demonstrated that the procedure followed can
1.4 log CFU/ml. In this study of Kaur et al. (1998), the microbial achieve a disinfection efﬁciency of about 90% and 97% on hands
concentration was higher (8 log10 CFU/ml) than those found in and utensils, respectively contaminated with >10–1000 CFU Sal-
our study. Bunning, Crawford, Tierney, and Peeler (1990) and Bré- monella spp., an organism which is more difﬁcult to remove
and, Fardel, Flandrois, Rosso, and Tomassone (1997), also observed from contaminated surfaces. The zero CFU/cm2 for hand and la-
that when E. coli and S. aureus undergoes an unfavourable increase dle surfaces decontamination demonstrated in our experiment
in temperature (between 49 and 53 °C) for a given duration (dura- was based on a mean concentration of <2 log CFU/cm2 for both
tion stress), a death phase followed by lag phase occurs. These bacteria and this can be regarded as ‘‘mild contamination”. In
reactions may explain the absence of the bacteria in samples col- addition, laboratory experiments may not be reproduced in the
lected at temperatures greater than 45 °C. ﬁeld. We cannot therefore rule out that when we were carrying
It should however be noted that studies dealing with bacterial out the re-contamination exercise, some of the hands and ladles
stresses have demonstrated that after thermal shock, bacteria are had background ﬂora due to inadequate disinfection and pres-
capable of reversibly entering a viable-but non-culturable state ence of any cuts, abrasions or damage to the skin and ladle sur-
(Caro, Got, Lesne, Binard, & Baleux, 1999). Bunning et al. (1990) faces since we did not inspect the hands and ladles of the
and Bréand et al. (1997), explain that when the bacteria recover volunteers as stipulated by Sattar et al. (2001). We did not also
from the duration stress and become exposed to increasing tem- establish the background ﬂora of the stored water to conﬁrm
peratures, they can grow to infective doses. Some of the food sam- that such water contained background ﬂora but depended on
ples that were exposed to contaminated surfaces at temperatures our previous results (Taulo et al., 2008) and those of Osmundsen
above 45 °C and did not show presence of viable cells, can never- (2005).
theless be a potential risk to consumers in the event that the non-
culturable organisms become resuscitated and grow to infective
doses following temperature drop or increase.
The results of this study suggest that wash waters that are
The study furnished a number of important conclusions.
microbiologically contaminated can serve as reservoirs of bacteria
Firstly, properly cooked porridge is microbiologically safe and
that can easily transfer to hand and ladle surfaces through direct
should therefore not pose a health risk if hygienically handled.
contact, which in turn can be easily transferred to foods. Because
Secondly, communal food eating practices may inﬂuence bacteri-
of the contamination of these waters with hands during washing
ological quality of cooked food. Thirdly, the microbiological con-
of several hands of different degree of contamination, and wash-
tent of cooked porridge increased signiﬁcantly to high and
ing/dipping of dirty ladles during serving the food onto the plate,
possibly hazardous levels when the food was consumed with
concentrations of organisms are higher than what one would ﬁnd
hands washed in contaminated water, with S. aureus being con-
in ordinary household stored water in Lungwena as reported by
sistently higher than E. coli. The Lungwena baseline census con-
Osmundsen (2005). The signiﬁcantly higher bacterial counts in
ducted in 2004 (unpublished) revealed that most of the
the hand wash water than on the hands as well as absence of bac-
households in Lungwena (98%) do not own refrigerators. People
teria on the hands that were washed with water that did not con-
store their leftover foods under ambient temperatures. Consider-
tain the bacteria suggest that the presence of bacteria on the hand
ing that S. aureus cells multiply faster under such temperatures, it
of the person who participated in the experiment was from con-
is likely that the low levels detected in the present study may
taminated wash water. Hand washing may not necessarily result
grow to infective doses following optimum growth temperature
in a signiﬁcant improvement in the bacterial load on the hand sur-
and acidic storage conditions as demonstrated by Rørvak and
faces as transfer can be either from the hand to the water or from
Granum (1999). Though we did not develop a model for predict-
the water to the hands. Members of the households whose hands
ing probability of a risk, we can suggest that in Lungwena, the
are relatively clean may in fact contaminate their hands from
probability of getting food-borne infection during consumption
washing them in communal hand washing water. This calls for
of the main meal is very high because the food contains a combi-
change in the communal hand washing habits to one that involves
nation of levels of bacteria from hands and ladles. Quantitative
using a vessel with narrow opening to pour water over the hands
data obtained in this study can assist in providing data for the
rather than the traditional way of scooping water directly from
risk characterisation and application of effective risk management
the communal hand washing bowl with a hand and rubbing the
strategies aimed at reducing transmission of food-borne infec-
tions in Malawi and in particular, Lungwena.
The presence of bacteria on ladles that had been disinfected
prior to dipping and also the absence of bacteria on ladles
washed/dipped in the water that did not contain the organisms,
indicates that contamination of the ladles occurred as a result of
This work was supported by the Norwegian Program for Devel-
use of contaminated water. The presence of bacteria on ladles
opment, Research and Higher Education (NUFU).
could easily serve to transmit food-borne diarrhoeal disease be-
cause the ladles are also used to serve other cold food supplements
soon after serving thick porridge. The importance of contaminated
ladles compared to hands as a vehicle for transfer of organisms Atanassova, V., Meindl, A., & Ring, C. (2001). Prevalence of Staphylococcus aureus and
leading to exposure will depend to a greater extent on consumer Staphylococcal enterotoxins in raw pork and uncooked smoked ham, a
comparison of classical culturing detection and RFLP-PCR. International Journal
behaviour in the kitchen. The proportion of consumers that wash
of Food Microbiology, 68(1), 105–113.
their utensils after using them on contaminated food as well as Aycicek, I. F., Aydogan, H., Kucukkaraaslan, A., Baysallar, M., & Basustaoglu, A. C.
the efﬁciency of the washing procedures needs to be included in (2004). Assessment of the bacterial contamination on hands of hospital food
evaluating the exposure routes. handlers. Journal of Food Control, 15(14), 253–259.
Baker, J., Naeeni, M., & Bloomﬁeld, S. F. (2003). The effects of cleaning and
It should be noted that although our laboratory experimental disinfection in reducing Salmonella contamination in a laboratory model
trial showed that both bacteria were inactivated by the disinfec- kitchen. Journal of Applied Microbiology, 95, 1351–1360.
S. Taulo et al. / Food Control 20 (2009) 1158–1166 1165
Barro, N., Abdoul, B. R., Savadogo, A., Outtara, C. A. T., IIboudo, A. J., & Traore, A. S. Kloos, W. E., & Bannerman, T. L. (1999). Staphylococcus and Micrococcus. In P. R.
(2006). Hygienic assessment of dish washing waters, utensils, hands and pieces Murray, E. J. Baron, M. A. Pfaller, F. C. Tenover, & R. H. Yolken (Eds.), Manual of
of money from street food processing sites in Ouagadougou (Burkina Faso). clinical microbiology (pp. 264–282). Washington, DC: American Society for
African Journal of Biotechnology, 5(11), 1107–1112. Microbiology.
Barza, M. (2004). Efﬁcacy and tolerability of C1O2 – generating gloves. Clinical Kusumaningrum, H. D., Riboldi, G., Hazeleger, W. C., & Beumer, R. R. (2003). Survival
Infectious Diseases, 38, 857–863. of food-borne pathogens on stainless steel surfaces and cross-contamination to
Bloomﬁeld, S. (2003). Home hygiene: A risk approach. International Journal of foods. International Journal of Food Microbiology, 85(3), 227–236.
Hygiene and Environment, 209(5), 1–8. Kusumaningrum, H. D., Van Putten, M. M., Rombouts, F. M., & Beumer, R. R. (2002).
Bloomﬁeld, S. F., & Scott, E. (1997). Cross-contamination and infection in the Effects of antibacterial dishwashing liquid on food-borne pathogens and
domestic environment and the role of chemical disinfectants. Journal of Applied competitive microorganisms in kitchen sponges. Journal of Food Protection,
Microbiology, 83(1), 1–9. 65(1), 61–65.
Bréand, S., Fardel, G., Flandrois, J. P., Rosso, L., & Tomassone, R. (1997). A model Larson, E. (2001). Hygiene of the skin. When is clean too clean. Emerging Infectious
describing the relationship between lag time and mild temperature increase Disease, 7(2). <http://www.cdc.gov/ncidod/eid/vol7no2/Larson.htm>. Accessed
duration. International Journal of Food Microbiology, 38(3), 157–167. 17.11.2008.
Brown, M. H., Gill, C. O., Hollingsworth, J., Nickelson, R., Seward, S., & Sheridan, J. J. Le Loir, Y., Baron, F., & Gautier, M. (2003). Staphylococcus aureus and food poisoning.
(2002). The role of microbiological testing in systems for assuring the safety of Genetics and Molecular Research, 2(1), 63–76.
beef. International Journal of Food Microbiology, 62(1–5), 7–16. Lues, J. F. R., & Van Tonder, I. (2006). The occurrence of indicator bacteria on hands
Bryan, F. L. (1988). Risk of practices, procedures and processes that lead to and aprons of food handlers in the delicatessen sections of a retail group. Food
outbreaks of food-borne diseases. Journal of Food Protection, 51(8), 663–667. Control, 18(4), 326–332.
Bunning, V. K., Crawford, R. G., Tierney & Peeler, T. (1990). Thermotolerance of Mattick, K., Durham, K., Hendrix, M., Slader, J., Grifﬁth, C., Sen, M., et al. (2003). The
Listeria monocytogenes and Salmonella typhimurium after sub-lethal heat shock. microbiological quality of washing- up water and the environment in domestic
Applied and Environmental Microbiology, 62(10), 1029–1035. and commercial environment. Applied Microbiology, 94(5), 842–848.
Burt, M. B., Volel, C., & Finkel, M. (2003). Safety of vendor-prepared foods: Michaels, B. (2002). Hand washing: An effective tool in the food safety arsenal. Food
Evaluation of ten processing mobile food vendors in Manhattan. Public Health Quality, 45/46(Sept/Oct), 49–53.
Report, 118(5), 470–476. Michaels, B., Ganger, V., Chia-Min, M., & Doyle, M. (2003). Use limitations of
Caro, A., Got, P., Lesne, J., Binard, S., & Baleux, B. (1999). Viability and virulence of alcoholic sanitizer as part of food service hand hygiene program. Food Service
experimentally stressed nonculturable Salmonella typhimurium. Journal of Technology, 3, 71–80.
Applied and Environmental Microbiology, 65(7), 3229–3232. Michaels, B., Keller, C., Blevins, M., Paoli, G., Ruthman, T., Todd, E., et al. (2004).
Chen, Y., Jackson, K. M., Chea, F. P., & Schaffner, D. W. (2001). Quantiﬁcation and Prevention of food worker transmission of foodborne pathogens: Risk
variability analysis of bacterial cross-contamination rates in common food assessment and evaluation of effective hygiene intervention strategies. Food
service tasks. Journal of Food Protection, 64(1), 72–80. Service Technology, 4, 31–49.
Cogan, T. A., Bloomﬁeld, S., & Humphrey, T. J. (1999). The effectiveness of hygiene Miller, A. J., Brown, T., & Call, J. E. (1996). Comparison of wooden and polyethylene
procedures for prevention of cross-contamination from chicken carcasses in the cutting boards: potential for the attachment and removal of bacterial from
domestic kitchen. Letters in Applied Microbiology, 29, 354–358. ground beef. Food Protection, 59(8), 854–858.
Cogan, T. A., Slader, J., Bloomﬁeld, S. F., & Humphrey, T. J. (2002). Achieving hygiene Ministry of Health, Malawi. (1985). Control of diarrhoeal diseases. Malawi
in the domestic kitchen: the effectiveness of commonly used cleaning Epidemiological Quarterly, 1–2.
procedures. Journal of Applied Microbiology, 92, 885–892. Minitab Inc., (2004). Minitab for windows release 14. (2004). Minitab Inc.
Collins, J. E. (2001). Impact of changing consumer lifestyles on the emergence/re- Pennsylvania State College.
emergence of food-borne pathogens. Emerging Infectious Diseases, 3(4), 1–13. Montville, R., Chen, Y., & Schaffner, D. W. (2001). Determination of bacterial cross-
Courtenay, M., Ramirez, L., Cox, B., Han, I., Jiang, X., & Dawson, P. (2005). Effects of contamination rates from hand to food through a glove barrier. Food Protection,
various hand hygiene regimes on removal and/or destruction of Escherichia coli 64(6), 845–849.
on hands. Food Service Technology, 5, 77–84. Moore, G., & Grifﬁth, C. (2002). A comparison of surface methods for detecting
Dawson, P., Han, I., Cox, M., Black, C., & Simmons, L. (2006). Resident time and food coliforms on food contact surfaces. Food Microbiology, 19(1), 65–73.
contact time effects on transfer of Salmonella Typhimurium from tile, wood and Moore, C. M., Sheldon, B. W., & Jaykus, L. A. (2003). Transfer of Salmonella and
carpet: Testing the ﬁve second rule. Journal of Applied Microbiology, 102(4), Campylobacter from stainless steel to romaine lettuce. Food Protection, 66(12),
De Wit, J. C., & Kampelmacher, E. H. (1981). Some aspects of microbial Morse, T. D., Nichols, R. B. A., Grimason, A. M., Campbell, B. M., Tembo, K. C., & Smith,
contamination of hands of workers in food industries. Zentralblatt Fur H. V. (2007). Incidence of Cryptosporidium species in paediatric patients in
Bcteriologie Mikrobiologie and Hygiene, 172, 290–400. Malawi. Journal of Epidemiology and Infection, 135, 1307–1315.
De Wit, J. C., & Rombouts, F. M. (1992). Faecal micro-organisms on the hands of Mosupye, F. M., & von Holy, A. (2000). Microbiological hazard identiﬁcation and
carriers. Escherichia coli as model for Salmonella. Zentralblatt Fur Hygiene and exposure assessment of street food vending in Johannesburg, South Africa. Food
Umweltmedizin, 193, 230–236. Microbiology, 61(2), 137–145.
Department of Health, South Africa (2000). Guidelines for environmental health Nordic Committee on Food Analysis, (1993). Coliform bacteria and E. coli in foods.
ofﬁcers on the interpretation of microbiological analysis data of food. Pretoria: Determination by the Plate Count Method with Petriﬁlm TM plates. No. 147, NMKL,
Directorate of Food Control, Government Printer. Esbo, Finland. Nordisk Metodik Komite for Livsmedel.
Difco Manual. (1998). Detroit/MI: Difco Laboratories. Olsen, S., MacKinon, l., Goulding, J., Bean, N., & Slutsker, L. (2000). Surveillance for
Guzewich, J., & Ross, M. (1999). Evaluation of risks related to microbiological food-borne disease outbreaks United States, 1993–1997. Morbidity and
contamination of ready-to-eat foods by food preparation workers and the Mortality Weekly Report, 49, 1–51.
effectiveness of interventions to minimize those risks. http:// Osmundsen, B. (2005). Drinking water safety; a study of collection and storage
www.cfsan.fda.gov/ear/rterisk.html. Accessed 20.10.2008. practices of drinking water in Lungwena. Masters Thesis, Norwegian University
Hansen, T. B., & Knochel, S. (2003). Image analysis methods for evaluation of speciﬁc of Life Sciences, Department of Chemistry, Biotechnology and Food Sciences, Ås.
and non-speciﬁc hand contamination. Journal of Applied Microbiology, 94, 483–494. Norway.
Hatakka, M., Bjorkroth, K. J., Asplund, K., Maki-Petays, N., & Korkeala, H. J. (2000). Patrick, D. R., Findon, G., & Miller, T. E. (1997). Residual moisture determines the
Genotypes and enterotoxicity of Staphylococcus aureus isolated from the hands level of touch associated bacteria transfer following hand washing.
and nasal cavities of ﬂight-catering employees. Journal of Food Protection, Epidemiology and Infection, 119(3), 319–325.
63(11), 1487–1491. Portocarrero, S. M., Newman, M., & Mikel, B. (2002). Staphylococcus aureus survival,
Hoque, B. A. (2003). Hand washing practices and challenges in Bangladesh. staphylococcal enterotoxin production and shelf stability of country-cured
International Journal of Environmental Health Research, 13, S81–87. hams manufactured under different processing procedures. Meat Science, 62(2),
Hoque, B. A., & Briend, A. (1991). A comparison of local hand washing agents in 267–273.
Bangladesh. Journal of Tropical Medicine and Hygiene, 94, 61–64. Roberts, L., Chartier, Y., Chartier, O., Malenga, G., Toole, M., & Rodka, H. (2001).
Hoque, B. A., Mahalanabis, D., Alam, M. J., & Islam, M. S. (1995). Post-defecation Keeping clean water clean in a Malawi refugee camp: A randomized
hand washing in Bangladesh: Practice and efﬁciency perspectives. Journal of intervention trial. Bulletin of World Health Organisation, 79(4), 280–287.
Public Health, 109(1), 15–24. Robine, E., Boulangé-Peterman, L., & Delangerè, D. (2002). Assessing bactericidal
Hundy, R. L., & Cameron, S. (2002). An outbreak of infections with a new Salmonella properties of materials the case of metallic surfaces in contact with air.
phage type linked to a symptomatic food handler. Journal of Communicable Microbiological Methods, 49(3), 225–234.
Diseases Intelligence, 26(4), 562–567. Rørvak, L. M., & Granum, P. E. (1999). Staphylococcus aureus. In M. Gran, A.
Josephson, K. L., Rubino, J. R., & Pepper, I. L. (1997). Characterisation and Wetlesen, A. N. Mutukumira, G. Rukure, & J. A. Narvhus (Eds.), Occurrence of
quantiﬁcation of bacterial pathogens and indicator organisms in household pathogenic bacteria in raw milk, cultured pasteurized milk and naturally soured
kitchens with and without the use of disinfectant cleaner. Journal of Applied milk produced at small scale dairies in Zimbabwe (pp. 539–544). Amsterdam:
Microbiology, 83(6), 737–750. Elsevier Science.
József, F. (2001). Physical methods of food preservation. In M. P. Doyle, L. R. Beuchat, Rusin, P., Orosz-Coughlin, P., & Gerba, C. A. (1998). Reduction of faecal coliform,
& T. J. Montville (Eds.), Food microbiology: Fundamentals and frontiers coliform and heterotrophic plate count bacteria in the household kitchen and
(pp. 567–591). Washington DC: ASM Press. bathroom by disinfection with hypochlorite cleaners. Applied Microbiology,
Kaur, J., Ledward, D. D., Park, R. W. A., & Robson, R. L. (1998). Factors affecting the 85(5), 819–828.
heat resistance of Escherichia coli 0157:H7. Letters in Applied Microbiology, 26(4), Sattar, S. A., Springthorpe, S., Mani, S., Gallant, M., Nair, R. C., & Scott, E. (2001).
325–330. Transfer of bacteria from fabrics to hands and other fabrics: development and
1166 S. Taulo et al. / Food Control 20 (2009) 1158–1166
application of a quantitative method using Staphylococcus aureus as a model. Taulo, S., Wetlesen, A., Abrahamsen, R., Mkakosya, R., & Kululanga, G. (2008a).
Journal of Applied Microbiology, 90, 962–970. Microbiological quality of water and associated management practices and
Scott, E., & Bloomﬁeld, S. (1990). The survival and transfer of microbial risks at source, transport and storage points in a predominant Moslem
contamination via clothes, hands and utensils. Journal of Applied Bacteriology, community of Lungwena, Malawi. African Journal of Microbiology Research,
68(3), 271–277. 2(5), 131–137.
Taylor, J. H., Brown, K. L., Toivenen, J., & Holah, J. T. (2000). A microbiological
Shojaei, H., Shooshtaripoor, J., & Amiri, M. (2005). Efﬁcacy of simple hand-washing evaluation of warm air hand driers with respect to hand hygiene and the
in reduction of microbial hand contamination of Iranian food handlers. washroom environment. Journal of Applied Microbiology, 89, 910–919.
International Journal of Food Research, 39, 525–529. USFDA, (2005). Food Code, 2005. <http://www.cfsan.fda.gov/.dms/fc05-toc.html>.
Taulo, S., Abrahamsen, R., Wetlesen, A., Kululanga, G., Mkakosya, R., & Grimason, A. Accessed 11.11.2008.
Microbiological hazard identiﬁcation and exposure assessment of food Zhao, P., Zhao, M., Doyle, J. R., Rubino, J. R., & Meng, J. (1998). Development of a
prepared and served in rural households of Lungwena, Malawi. Journal of model for the evaluation of microbial cross-contamination in the kitchen. Food
Food Microbiology (in press). Protection, 61(8), 960–963.