Contamination of Weaning Foods: Organisms, Channels,
and Sequelae
by Zeinab E. M. Afifi,* Shafika S. Nasser,* Sayad Shalaby,* and Samir A. E. Atlam**
*Public Health Department, Kasr El Aini Faculty of Medicine, Cairo, Egypt
**Public Health Department, Tanta Faculty of Medicine, Tanta, Egypt
Summary
A study was carried out to identify the microbial contaminants of weaning foods in an Egyptian
village and the sources of contamination. All 300 households containing infants (< 24 months) were
visited. Information regarding food preparation and household sanitation was gathered. Samples of
weaning foods (270) were collected and analysed for the presence of Escherichia coli, Bacillus cereus,
Shigella and parasites. The first two pathogens were detected in 43.7 per cent and 21.4 per cent of
samples respectively. The others were not detected in any sample. The risk of contamination by E.
coli and B. cereus was significantly associated with the presence of dung and/or refuse in the house,
lack of indoor latrine, non-use of latrine by children, weaning foods not freshly prepared, uncovered
storage of foods, and the presence of a case of diarrhoea in the house.
Introduction
In the developing world, the time of weaning is
considered a critical period in a child's life. Many
studies have reported that the incidence of diarrhoeal
disease is especially high after weaning is initiated.'"3
Because food is a good medium for microbial growth,
the introduction of unhygienically prepared weaning
foods exposes the child to enteropathogens.4"5
Contamination of weaning foods occurs through
different channels before, during, and after preparation.6
Many enteropathogens have been detected in weaning
foods in different parts of the world.7"9 The most
commonly encountered pathogen is E. coli; other
pathogens include Shigella* Vibrio cholerae,B'9 and
Salmonella.
Very few data are available regarding the microbiological quality of weaning foods in Egypt. The
objective of this study was to explore whether weaning
foods fed to infants in a rural Egyptian community were
contaminated or not, the pathogens involved, and
sources of contamination.
Materials and Methods
All 300 infants (0-24 months) registered in the health
office of Kombera village were visited at home. Data
concerning child feeding and home sanitation were
Acknowledgement
The authors express gratitude to the NCDDP in Egypt for
approving and sponsoring the study.
Correspondence: Dr Zeinab E. M. Afifi, 33 Baramka Street, ElHay El Saleie, Nas City, Cairo, Egypt.
Journal of Tropical Pediatrics
Vol.44
December 1998
collected by interviewing the mother or carer. Samples
were collected from foods prepared for the infant on that
day. In 265 cases one food was prepared. In 14 cases two
foods were prepared.
The samples collected were transferred immediately,
in ice, to the laboratories of the Nutrition Institute in
Cairo. They were analysed for the presence of E. coli,
Shigella, B. cereus, and parasites. The following
techniques were utilized:
• E. coli: enrichment on MacConkey broth medium;
positive tubes were subject to Indole Methyl-red,
Voges-Proskauer citrate tests and bacterial count per
gram.
• Shigella: three loops (5 mm) of enriched broth culture
were inoculated into three selective agar media; xyloselysine agar, tegritol 7, and desoxy cholate-citrate agar.
The media were incubated at 35°-37°C for 24 h.
Suspected colonies were examined microscopically.
• B. cereus: serially diluted food samples were cultured
on horse-blood agar. The occurrence of a or ^
haemolysis indicated the presence of the organism.
• Parasites: supernatants of diluted centrifuged food
samples were examined microscopically.
Infants were revisited after one week and enquiry was
made about the occurrence of diarrhoea subsequent to
the consumption of sampled foods. The relative risk
(RR) was calculated and used to quantify the risk of
contamination associated with all variables. 10
Results
E. coli and B. cereus were detected in 43.7 per cent and
21.4 per cent of the food samples analysed. Shigella and
© Oxford University Press 1998
335
Z.E.M AFIFIET AL.
parasites were not detected in any sample. Table 1 shows
the distribution of households according to sanitation. In
57 per cent of cases the source of water was surface
pumps. Indoor latrines were available in 63.3 per cent of
houses. They were used by children in only 29 per cent of
cases. Refuse and animal dung were present in 70.7 and
60.7 per cent of houses respectively. Only 6.7 per cent of
houses were free from flies.
Table 2 shows the set of factors that were significant in
determining the occurrence of contamination by the two
detected pathogens. About half (48.3 per cent) of the
children who consumed E. co//-contaminated food
suffered diarrhoea. The percentage suffering diarrhoea
after eating B. cerens-contaminated food was 49.1 per
cent.
Discussion
Microbiological analysis showed that 43.7 per cent of
food samples were contaminated with E. coli, an indicator
of faecal contamination, and 21.4 per cent were contaminated with B. cereus. Similar findings have been
reported from Guatemala and Bangladesh."12
Contamination was associated significantly with food
preparation practices. Fresh preparation of food
decreased the risk of contamination with E. coli and B.
cereus by 1.7 and 5.6 times respectively (p < 0.01). A
multiplication of faecal coliforms was observed to occur
when there was a delay of more than 4h between
preparation and consumption of weaning food.8 Storage
of food in uncovered containers increased the risk of
contamination by the two organisms by 1.5 and 1.8 times
respectively (p < 0.05).
In addition to food preparation practices, contamination was also significantly determined by household
sanitary conditions. The lack of an indoor latrine and the
presence of refuse and/or animal dung in the house
increased the chances of E. coli contamination of
weaning foods by 1.5, 1.6, and 1.6 times respectively
(p < 0.0)) The relative risks associated with the same set
and with B. cereus contamination were 2.6, 3.0, and 5.6
times respectively.
The presence of diarrhoea case(s) in the house almost
TABLE 1
TABLE 2
Risk factors associated with contamination of weaning
foods
RR (p value)
Risk factors
Environmental
Animal dung
Refuse
Indoor latrine
Latrine use by
• children
Food preparation
Unfresh/fresh
Food storage
Covered/uncovered
Diarrhoea case in
the house
336
No
%
Presence of animal dung
Refuse in the house
Indoor latrine available
Latrine use by children
Source of water
Surface pump
Piped water
Storage of water
Presence of houseflies
182
212
268
87
60.7
70.7
89.3
29.0
171
129
154
280
57.0
43.0
51.3
93.3
B. cereus
1.6 (< 0.001)
1.6 (< 0.01)
1.5 (< 0.05)
5.6 (< 0.001)
3.0 (< 0.001)
2.6 (< 0.001)
1.5 (< 0.01)
2.6 (< 0.05)
1.7 (<0.01)
5.6 (< 0.001)
1.5 (< 0.01)
1.8 (< 0.02)
1.6 (< 0.05)
2.1 (< 0.001)
doubled the risk of contamination by E. coli and B.
cereus of weaning foods prepared on the same day.
Associated relative risks were 1.6 and 2.1 respectively
(p < 0.05). Also, the ingestion of E. coli- and B. cereuscontaminated foods was significantly associated with
subsequent diarrhoea occurrence (p < 0.01). The
observed significant associations continued to exist
after controlling for the confounding effect of home
cleanliness, source of water, presence of indoor latrine,
and animal dung inside the house.
Under such conditions, strategies to reduce the
contamination of baby foods and prevent diarrhoea
should focus on household sanitation and cleanliness,
keeping animals outside residential premises, preparation of weaning foods shortly before consumption, and
keeping them covered. Last, but not least, the importance
of personal hygiene and proper food sanitation, especially in the presence of diarrhoea cases in the house, are
to be emphasized.
References
1. Barrel RAE, Rowland MGM. Infant foods as a potential
source of diarrhoeal illness in rural West Africa. Trans Roy
Soc Trap Med Hyg 1979; 73: 85-9.
2. Mata L. The Children of Santa Mana Caque: A Prospective
Sanitary conditions in the houses studied
Environmental factor
E. coli
3.
4.
5.
6.
Field Study of Health and Growth. MIT Press, Cambridge,
MA, 1978.
Rowland MGM, McCollum JPK. Malnutrition and gastroenteritis in Gambia. Trans Roy Soc Trop Med Hyg 1977;
71: 199-203.
Motarjemi Y, Kaferstein F, Moy G, Quevedo F. Contaminated weaning food: a major risk factor for diarrhoea
and associated malnutrition. Bull World Health Org 1993;
71:79-92.
Memorandum from a JHUAVHO meeting. Research on
improving infant feeding practices to prevent diarrhoea or
reduce its severity. Bull World Health Org 1989; 67: 27-33.
Jelliffe DP. Community and sociopolitical considerations
of breast feeding in breast feeding mothers. Ciba
Journal of Tropical Pediatrics
Vol. 44
December 1998
Z.E.M. AFTFI ETAL.
Foundation Symposium No 45. Elsevier, Amsterdam,
1976.
7. Black RE, Brown KH, Bckker S. Contamination of
weaning foods and transmission of enterotoxigenic E. coli
diarrhoea in children in rural Bangladesh. Trans Roy Soc
Trop Mcd Hyg 1982; 76: 259-64.
8. Henry FJ, Patwasy Y, Huttly SR, Aziz KM. Bacterial
contamination of weaning foods and drinking water in rural
Bangladesh. Epidem Inf 1990; 104: 79-85.
9. Oo KN, Han AM, Aye T, Hlaing T. Bacteriologic studies of
Journal of Tropical Pediatrics
Vol. 44
December 1998
food and water consumed by children in Myanmar 1: the
nature ofcontamination.JDiarrhoealDis Res 1991;9:87-90.
10. Daniels SR, Grenberg RS, Ibrahim AM, el al. Etiologic
research in epidemiology. J Pedialr 1983; 102: 494-504.
11. Caparelli E, Mata L. Microflora of maize prepared as
tortillas. Appl Microbiol 1979; 29: 802-6.
12. Black RE, Brown KH, Becker S. Longitudinal studies of
infectious diseases and physical growth of children in rural
Bangladesh. II. Incidence of diarrhoea and association with
known pathogens. Am J Epidemiol 1982; 115: 315-24.
337