OBJECTIVE: To investigate the microbial quality of foods sold on streets of Accra and factors predisposing to their contamination.
METHODS: Structured questionnaires were used to collect data from 117 street vendors on their vital statistics, personal hygiene, food hygiene and knowledge of foodborne illness. Standard methods were used for the enumeration, isolation, and identification of bacteria.
FINDINGS: Most vendors were educated and exhibited good hygiene behaviour. Diarrhoea was defined as the passage of ³3 stools per day) by 110 vendors (94.0%), but none associated diarrhoea with bloody stools; only 21 (17.9%) associated diarrhoea with germs. The surroundings of the vending sites were clean, but four sites (3.4%) were classified as very dirty. The cooking of food well in advance of consumption, exposure of food to flies, and working with food at ground level and by hand were likely risk factors for contamination. Examinations were made of 511 menu items, classified as breakfast/snack foods, main dishes, soups and sauces, and cold dishes. Mesophilic bacteria were detected in 356 foods (69.7%): 28 contained Bacillus cereus (5.5%), 163 contained Staphylococcus aureus (31.9%) and 172 contained Enterobacteriaceae (33.7%). The microbial quality of most of the foods was within the acceptable limits but samples of salads, macaroni, fufu, omo tuo and red pepper had unacceptable levels of contamination. Shigella sonnei and enteroaggregative Escherichia coli were isolated from macaroni, rice, and tomato stew, and Salmonella arizonae from light soup.
CONCLUSION: Street foods can be sources of enteropathogens. Vendors should therefore receive education in food hygiene. Special attention should be given to the causes of diarrhoea, the transmission of diarrhoeal pathogens, the handling of equipment and cooked food, hand-washing practices and environmental hygiene.
Keywords Food services; Food contamination; Food handling; Food microbiology; Diarrhea/etiology; Risk factors; Ghana (source: MeSH, NLM).
Mots clés Restauration; Contamination alimentaire; Traitement aliments; Microbiologie alimentaire; Diarrhée/étiologie; Facteur risque; Ghana (source: MeSH, INSERM).
Palabras clave Servicios de alimentación; Contaminación de alimentos; Manipulación de alimentos; Microbiología de alimentos; Diarrea/etiología; Factores de riesgo; Ghana (fuente: DeCS, BIREME).
Bulletin of the World Health Organization 2002;80:546-554.
Foodborne illness is a major international health problem and an important cause of reduced economic growth (1). There is concern about, for example, chemical contamination, Escherichia coli O157:H7 infections, the use of antibiotics in animal rearing and the transfer of antibiotic resistance to human pathogens (2).
The problems of food safety in the industrialized world differ considerably from those faced by developing countries. Whereas traditional methods are used for marketing fresh produce in the latter countries, food processing and packaging are the norm in industrialized countries. In developing countries a large proportion of ready-to-eat food is sold on the streets. Table 1 shows some of the foods sold on the streets in Ghana.
The term "street food" refers to a wide variety of ready-to-eat foods and beverages sold, and sometimes prepared, in public places. As with fast food, the final preparation occurs when meals are ordered by customers. Street food may be consumed where it is purchased or can be taken away and eaten elsewhere.
The consumption of street food is common in many countries where unemployment is high, salaries are low, work opportunities and social programmes are limited, and where urbanization is taking place. Street food vendors benefit from a positive cash flow, often evade taxation, and can determine their own working hours. In selling snacks, complete meals, and refreshments at relatively low prices, they provide an essential service to workers, shoppers, travellers, and people on low incomes. People who depend on such food are often more interested in its convenience than in questions of its safety, quality and hygiene.
The hygienic aspects of vending operations are a major source of concern for food control officers. For example, stands are often crude structures, and running water may not be readily available. Also toilets and adequate washing facilities are rarely available. The washing of hands, utensils, and dishes is often done in buckets or bowls. Disinfection is not usually carried out, and insects and rodents may be attracted to sites where there is no organized sewage disposal. Finally food is not adequately protected from flies and refrigeration is usually unavailable.
We have previously studied risk factors for acute and persistent diarrhoea in an urban slum in Accra, Ghana (3). About 60% of 951 mothers supplemented their children's diet with street food. The children had an increased risk of both acute and persistent diarrhoea. There were higher levels of contamination in the street food given to these children than in food cooked at home. In the same area we evaluated also the following: the role of street food vendors in the transmission of diarrhoeal pathogens; the microbial quality of street foods; and factors that could predispose food to contamination with bacteria.
We have already published data on the role of street food vendors in the transmission of diarrhoeal pathogens (4). In the present paper we report our findings on the microbial quality of street foods and the risk factors for food contamination.
Prior to starting the investigation, we explained its aims to head teachers and representatives of the food vendors in the study area. The street food vendors were recruited into the study after the owners of the vending sites had given their approval and the vendors were assured total confidentiality.
Study population and sample collection
The study was conducted in Nima-Kotobabi-Pig Farm-Accra New Town sub-area. Vendors of street foods in the major streets, markets and two schools were recruited. There were approximately 120 vending sites; the persons in charge of the vending points (subsequently referred to as the vendors) were the main subjects of study. A structured questionnaire was used to gather data on vital statistics, personal and food hygiene practices and knowledge of foodborne pathogens.
Samples of about 50 g of each food on sale were placed in separate sterile containers and transported to the laboratory on ice within two hours of collection. All samples were examined the same day.
Portions of food weighing 10 g were diluted at 1:10 with 90 ml of phosphate-buffered saline (Oxoid Dulbecco A BR 14a, UNIPATH (Oxoid), Basingstoke, England). Further tenfold serial dilutions were made and examined by means of the pour-plate method (57).
Total bacterial counts were made by means of plate count agar (Oxoid CM463). Counts of Enterobacteriaceae, Staphylococcus aureus, and Bacillus cereus were made using violet red bile lactose agar (Oxoid CM107), BairdParker agar (Oxoid CM275) and B. cereus selective agar (Oxoid CM617 and SR99), respectively (57). After appropriate incubation, dilutions with 30300 colonies were selected and counted. The number of colony-forming units per g (cfu/g) of food was calculated by multiplying the number of bacteria by the dilution.
Isolation and identification of Enterobacteriaceae
Portions of food weighing 10 g were inoculated into buffered peptone water (Oxoid CM509) for resuscitation of metabolically injured bacteria. After overnight incubation of the inoculates, the entire buffered peptone water was decanted and centrifuged at 11 000 rpm for 30 minutes in a refrigerated centrifuge (Hitachi 20PR-259, Tokyo, Japan). The deposit was streaked on Salmonella/Shigella agar (BBL 11597, BBL Microbiological Systems, MD, USA), xyloselysine deoxycholate agar (Oxoid CM469) and MacConkey agar (Oxoid CM7) for the detection of Salmonella, Shigella, E. coli, and other Enterobacteriaceae.
Portions of the pellet were inoculated into selenitelactose broth (Eiken 42001, Tokyo, Japan) for selective enrichment of Salmonella and Shigella and were streaked on Salmonella/Shigella agar and xyloselysine deoxycholate agar for the detection of isolated colonies. All incubations were conducted at 37 oC under aerobic conditions. Suspected colonies were identified using standard biochemical methods. API 20E kits (bio-Mérieux SA, Marcy-l'Étoile, France) were used for further identification. Specific Salmonella antisera (O grouping and Vi-sera (Denka Seiken No. 21041, Tokyo, Japan)), Shigella antisera (III) (Denka Seiken No. 14530) and E. coli (I) O sera (Denka Seiken No. 24506) were used for further typing.
Bacteria identified as E. coli were stored individually at 70 oC in tryptic soy broth with 40% (v/v) glycerol. They were tested for heat-labile and heat-stable enterotoxin plasmids by means of DNADNA hybridization (8). Enteroaggregative E. coli were tested using HEp2 tissue culture cells (9).
Data handling and analysis
The data were analysed using EpiInfo version 6 software. The values obtained for cfu/g of food were transformed into log10 values. Foods were classified as having no-to-low risk of transmitting pathogenic bacteria if the total count was less than 5.0 log10 cfu/g and if the counts of Enterobacteriaceae, B. aureus and S. aureus were less than 3.0 log10 cfu/g. A classification of medium-to-high risk corresponded to values of at least 5.0 log10 cfu/g for total counts and of at least 3.0 log10 cfu/g for Enterobacteriaceae, B. cereus and S. aureus.
Cross-tabulations were made of the levels of the various bacteria tested and the responses obtained in the interviews. The significance of any observed differences was determined using c2 test and Student's t-test. The MannWhitney rank sign test was performed for samples that violated a parametric test. Statistical significance was set at P £ 0.05. In order to determine the effect of the knowledge and practices of vendors on the microbial quality of street foods, we calculated confidence intervals after factors and levels of contamination were cross-tabulated.
In Accra the street food trade was conducted by children aged ³10 years and by women aged £ 52 years. A total of 38 of the vendors (33.3%) had received no formal education; nevertheless, they exhibited good hygiene behaviour as indicated by the high prevalence of hand-washing and personal care (Table 2). Diarrhoea was defined by 110 (94%) vendors as the passage of three or more stools per day but no vendors associated diarrhoea with bloody stools (Table 3). A total of 21 vendors (17.9%) associated diarrhoea with germs but none was aware that dirty hands were a risk factor for diarrhoea (Table 3).
Table 4 shows that various people bought street foods held in open or protected structures. The surroundings of the vending sites were predominantly clean; only four (3.4%) were classified as very dirty.
Food-handling practices that could affect the microbial quality of street foods are indicated in Table 5 and Table 6. Among these practices were the following: cooking food well in advance of consumption; exposure of food to flies; working with food at ground level; and handling food.
A total of 511 menu items, classified as breakfast/snack foods, main dishes, soups and sauces, and cold dishes were examined for aerobic mesophilic bacteria (total counts), B. cereus, S. aureus and Enterobacteriaceae. Mesophilic bacteria were identified in 356 food samples (69.7%); and B. cereus, S. aureus and Enterobacteriaceae were found in 28 (5.5%), 163 (31.9%), and 172 (33.7%), respectively.
Table 7 shows that the microbial quality of most foods was within acceptable limits, i.e. 10 cfu/g for total counts, 10 cfu/g for Enterobacteriaceae, S. aureus and B. cereus, and zero tolerance for other pathogens. Samples of salads, macaroni, fufu, omo tuo, and red pepper had unacceptable levels of contamination. Shigella sonnei was isolated from macaroni, Salmonella arizonae from light soup with meat, and enteroaggregative E. coli from macaroni, tomato stew, and rice (Table 8).
Despite poor environmental hygiene the street foods were, in general, microbiologically safe (Table 7). Similar results were reported for salads and gravies in Johannesburg (10).
Breakfast and snack foods
Breakfast and snack foods, e.g. koko and koose (bean cake), were the least contaminated; only a few samples contained mesophilic bacteria. There were no B. cereus, S. aureus, or Enterobacteriaceae in any of these foods. Porridge was prepared in the early hours of the morning and sold within 23 hours at 5090 oC, a temperature range over which most vegetative bacteria do not survive. Most of the porridge samples were prepared from maize dough that had undergone lactic acid fermentation. This traditional food preparation technology produces an environment that inhibits various diarrhoeal pathogens (11). The contamination of porridge possibly occurred during serving with a spoon and calabash kept in a bowl of water in which the vendor washed her hands and cleaned cups and eating bowls. Koose was prepared on site by frying at 165175 oC, a temperature range that kills most bacteria. Koose was served hot, another factor accounting for the relatively low levels of bacteria in this product. If care is taken during serving, microbial quality can be markedly improved.
A large proportion of main dishes was contaminated with unacceptable levels of bacteria: fufu, akpler, rice, and wankye were particularly heavily contaminated. These foods were handled excessively after cooking. The preparation of fufu involved pounding the staple after cooking by means of a mortar and pestle and turning the stiff dough with the bare hands, which were occasionally washed in a container of water. Akpler was rolled into balls after cooking, a process that involved using the hands and a calabash or bowl and water, which were possible sources of contamination of the cooked product. The scooping of rice and wankye into bowls or polythene bags was a major influence in the contamination of these foods, as was demonstrated when microbial levels were cross-tabulated with data relating to this factor (Table 6). Contamination was reduced where vendors sold food from the cooking pots. These findings confirm a previous observation (12) that contamination from utensils was possible during serving but not from storage containers, since leftovers were generally stored in the original cooking pots.
Soups and sauces
Soups and sauces appeared to be even more contaminated with enteroaggregative E. coli, which were isolated from tomato stew and shito. These are usually prepared the day before consumption. If reheating is inadequate, bacterial contaminants can multiply. Fish and meat can be additional sources of pathogens. Table 8 shows that Salmonella arizonae was found in soup. High levels of faecal bacteria and various enteric pathogens have been found in chicken, beef and goat meat (13, 14). The preparation of food long before its consumption, storage at ambient temperature, inadequate cooling and reheating, contaminated processed food, and undercooking were identified as the key factors in the handling of food that contributed to food poisoning outbreaks in England and Wales (14).
A wide range of menu items including formula foods, milk, soups and stews were examined in Peru (15), and levels of 09.0 log10 cfu/g of food reported. The proportion of contaminated infant foods was 11/66 (16.6%) (5.09.0 log10 cfu/g), while 4/221 (1.8%) of soups and stews, had similar levels of bacteria of the type indicating faecal contamination. Other ingredients, e.g. vegetables, were also possible sources of bacterial contaminants since vendors often used cheap supplies in order to maximize their profits. Inadequate cooking results in the survival of bacterial pathogens; cooking utensils can also add to the bacterial load.
Sauces, such as red pepper sauce, are made from fresh vegetables and eaten without having been heated. All bacteria introduced in the ways indicated above survive and multiply if held for prolonged periods at ambient temperature.
Macaroni and salads
Macaroni and salads carried the greatest risk of transmitting diarrhoeal pathogens. In general, bacterial counts were high in the foods, and Shigella sonnei and pathogenic E. coli were isolated from macaroni. This was not unexpected because, after cooking, tomato stew was stirred into the macaroni. Serving was performed using bare hands as this food was slippery and the use of a spoon or fork might have been difficult. The consumption of pansit (rice noodles with shrimps, meat and vegetables), a similar street food, has been associated with cholera in Manila (16).
The high levels of contaminants in salads were not unexpected. In Ghana all types of water are used for watering vegetables, especially those grown in the cities where there are not many natural bodies of water. Shigella flexneri and E. coli have been isolated from lettuce in Accra (17). Lettuce and tomatoes were contaminated with faecal bacteria, Salmonella and Shigella. Animal waste, e.g. chicken droppings, was used as manure for these vegetables (18). These findings call for a careful analysis of events occurring before crops are planted, of production and of packaging and the distribution of fresh produce, as well as of the preparation and sale of street foods.
The use of a fork or spoon to serve food reduced the level of contamination, while the use of bare hands resulted in an increase. Both spoons and bare hands were used to serve rice and wankye. Indeed, 36% of vendors served food with their bare hands. Vendors were carriers of a variety of bacterial enteropathogens, including Salmonella typhi (4). Defective personal hygiene can facilitate the transmission of these pathogens via food to humans. The serving stage is a critical point in the street food industry.
Enteropathogens can survive on the hands for three hours or longer. Diarrhoeal pathogens on the hands of mothers can be transmitted to infants (3). E. coli was detected in hand washings of high-income and low-income mothers in India at levels of 7.0 ± 4.2 log10 cfu/ml and 9.0 ± 5.7 log10 cfu/ml, respectively (19). In Peru, E. coli was detected in 11 of 78 mothers' hand washings (15). In Thailand, enterotoxigenic E. coli (ETEC) was detected in 6 of 42 mothers' hands and in 50 of 37 children's hands. The samples were from homes where children were suffering from ETEC diarrhoea. In most instances the type isolated from diarrhoea cases corresponded to that isolated from hands (20).
The present study identified a number of factors that could reduce the risk of contamination (Table 6). An example is the sale of food in chop-bars. These are usually wooden structures, which may be completely or partially enclosed. There is an area for food preparation, and most foods sold in these facilities are freshly prepared or reheated.
That food sold in schools carried the highest risk of transmitting bacteria was cause for concern. A study on foods sold in Nigerian schools showed unacceptable levels of bacteria (21). In addition E. coli and S. flexneri were isolated from some samples. Vendors appeared to be selling substandard food to schoolchildren because of their low purchasing power and their lack of knowledge on food safety. Children were also more interested in satisfying their hunger than in the quality of the food sold to them.
A lack of knowledge of the definition and causes of diarrhoea were important risk factors. None of the vendors associated dirty hands with the transmission of diarrhoeal pathogens. This explained why people commonly used their hands to serve cooked food. The container in which the food was served was also important and the use of paper and leaves increased the risk of contamination. The paper used for holding food was usually newsprint of questionable origin. Leaves were wiped with a piece of cloth and there was no disinfection. Clearly, the possibility existed that microflorae on leaves and microbes acquired through poor handling were transferred to foods.
Certain risk factors might not have been directly linked to food preparation. Environmental hygiene and the vendor's appearance did not show any significant relationship with the levels of contamination, implying that there was no risk if direct contact could not be made with food. In this connection it should be noted that the handling of food at ground level increased the risk of contamination because dust could easily be blown on to food thus handled. Pathogens can be passed mechanically by flies (22). Salmonella typhimurium and Shigella can multiply in the gut of the housefly and can be excreted for weeks or longer (22). There is consequently a risk of contamination associated with the exposure of food to flies (Table 6).
The use of soap to wash utensils and crockery reduced the levels of bacteria. Gram-negative bacilli such as S. typhi are fairly susceptible to soap made from saturated fatty acids but resistant to soap made from unsaturated fatty acids. Most microorganisms die after coming into contact with soap but their susceptibilities vary. For example, the vegetative cells of Bacillus mesentericus are inhibited within 20 sec of exposure but S. aureus survives for 20 minutes (23).
An epidemiological link between street foods and diarrhoea has been reported (3, 24, 25). Certain foods sold by street vendors, e.g. poultry, pork, beef, fish and rice, have been implicated in food poisoning outbreaks (2628).
Our findings indicate the need for stricter implementation of the food sanitation code and the licensing of street food vendors. Public health authorities should intensify efforts to monitor conditions of sanitation and hygiene in establishments serving food and drink to the public.
The public, and, in particular, consumer organizations, could play key roles in the food control system by calling attention to deficiencies and constructively supporting national food quality control systems. Wherever consumer pressures exist there is heightened awareness of food problems and both the authorities and the food industry are under pressure to make improvements.
The street food traders in Accra are mainly women. They could play major roles in food control programmes. Women in Africa, Asia and other parts of the world are involved in a wide variety of food-processing activities. They are also charged with the responsibility of providing food for their households. The success of any food control programme depends on women, since the transfer and implementation of ideas takes place through them (4).
In the present study, however, there was a high illiteracy rate (33.3%) among women working as street food vendors (Table 2 and Table 3). Their educational level appeared not to affect their knowledge of the definition of diarrhoea but was strongly associated with knowledge of the mode of transmission of enteric pathogens. The education of these women (29) by means of messages based on our findings is essential in programmes aimed at improving the microbial quality of street foods. Special attention should be given to the following: causes of diarrhoea; transmission of diarrhoeal pathogens; handling of food after cooking and of equipment used for cooking and serving; hand washing; use of soap; and environmental hygiene. Action along these lines can be expected to improve the safety of street foods and thereby to heighten consumer protection (30).
The people of the Ayawaso District Assembly are thanked for their tireless efforts to ensure the success of this study. The hospitality of the street food vendors who kindly agreed to participate in the study is highly appreciated. We also acknowledge the hard work of Mr Emmanuel Ansah Yeboah, field manager, the drivers of the Noguchi Institute for Medical Research, Ms Vivienne Agyiri, Mr Emmanuel Amaquandoh and all the technical staff of the Institute's Bacteriology and Virology Units. Professor Nkrumah and Dr H. Kamiya are thanked for their encouragement at every stage of the study. Funding was provided by the Japan International Cooperation Agency, the Ministry of Health of Japan and the Government of Ghana.
Conflicts of interest: none declared.
La nourriture vendue dans les rues à Accra au Ghana présente-t-elle un danger pour la santé ?
OBJECTIF: Etudier la qualité microbiologique de la nourriture vendue dans les rues à Accra ainsi que les facteurs prédisposant à la contamination microbienne.
MÉTHODES: Des données ont été recueillies auprès de 117 vendeurs des rues à l'aide de questionnaires structurés portant sur l'état civil des personnes interrogées, leur hygiène personnelle et leur connaissance des maladies d'origine alimentaire. Des méthodes normalisées ont été utilisées pour le dénombrement, l'isolement et l'identification des bactéries.
RÉSULTATS: La plupart des vendeurs étaient éduqués et avaient une hygiène satisfaisante. La diarrhée a été définie par 110 vendeurs (94 % de l'échantillon) comme étant un nombre de selles égal ou supérieur à trois par jour mais aucun n'a associé la diarrhée à la présence de sang dans les selles : 21 seulement (17,9 %) ont fait le lien entre la diarrhée et une infection bactérienne. D'une manière générale, les lieux de vente étaient propres mais quatre d'entre eux (3,4 %) ont été classés comme très sales. La préparation des mets longtemps à l'avance, l'exposition aux mouches, et le fait de manipuler les aliments au ras du sol et à main nue ont été identifiés comme des facteurs de contamination potentiels. Des études ont été effectuées sur 511 mets se répartissant entre aliments pour le petit déjeuner ou pour la collation, plats principaux, soupes et sauces et plats froids. Des bactéries mésophiles étaient présentes dans 356 échantillons (69,7 %) : Bacillus cereus a été isolé dans 28 d'entre eux (5,5 %), et Staphylococcus aureus dans 163 (31,9 %). Cent soixante douze échantillons (33,7 %) contenaient des entérobactéries. La qualité microbiologique de la plupart des aliments restait dans des limites acceptables, mais des échantillons de salade, de macaroni, de foufou, d'omo tuo et de piment rouge présentaient des niveaux de contamination excessifs. Des souches de Shigella sonnei et d'Escherichia coli entéroaggrégatif ont été isolées dans des macaronis, du riz et des préparations à base de tomate et Salmonella arizonae a été mis en évidence dans de la soupe claire.
CONCLUSION: Les aliments vendus dans la rue peuvent être des sources d'entéropathogènes. Les vendeurs devraient recevoir une formation en hygiène alimentaire. Une attention particulière devrait être accordée à toutes les causes de diarrhée, aux mécanismes de transmission des pathogènes diarrhéiques, à la manipulation des ustensiles et des aliments cuits, au lavage des mains et à l'hygiène du milieu.
Estudio de la inocuidad de los alimentos de venta callejera en Accra, Ghana
OBJETIVO: Se evaluaron la calidad microbiana de los alimentos vendidos en las calles de Accra y los factores que favorecían su contaminación.
MÉTODOS: Se usaron cuestionarios estructurados para reunir datos de 117 vendedores callejeros respecto a sus estadísticas vitales, higiene personal, higiene alimentaria y conocimiento de las enfermedades transmitidas por los alimentos. La enumeración, el aislamiento y la identificación de las bacterias se llevaron a cabo empleando métodos estándar.
RESULTADOS: La mayoría de los vendedores poseían estudios y mostraban un buen comportamiento higiénico. En total 110 vendedores (94,0%) definieron la diarrea como la evacuación de ³ 3 heces al día, pero ninguno asoció la diarrea a la presencia de heces sanguinolentas; sólo 21 (17,9%) asociaban la diarrea a la presencia de gérmenes. Los alrededores de los sitios de venta estaban limpios, pero cuatro de los sitios (3,4%) fueron clasificados como muy sucios. La cocción de los alimentos mucho antes de su consumo, su exposición a las moscas y el hecho de manipularlos a nivel del suelo y con las manos eran factores de riesgo frecuentes de contaminación. Se examinaron 511 artículos, clasificados como desayuno/aperitivo, platos principales, sopas y salsas, y platos fríos. Se detectaron bacterias mesofílicas en 356 alimentos (69,7%): 28 (5,5%) contenían Bacillus cereus, 163 (31,9%) Staphylococcus aureus , y 172 (33,7%) enterobacterias. La calidad microbiana de la mayoría de los alimentos entraba dentro de lo aceptable, pero diversas muestras de ensaladas, macarrones, fufu, omo tuo y pimentón picante presentaban niveles inadmisibles de contaminantes. Se aislaron Shigella sonnei y Escherichia coli enteroagregativa a partir de macarrones, arroz y guiso de tomate, y Salmonella arizonae a partir de muestras de sopa ligera.
CONCLUSIÓN: Los alimentos de venta callejera pueden ser fuente de enteropatógenos, de ahí la necesidad de enseñar a los vendedores nociones de higiene alimentaria. Se debe prestar especial atención a las causas de diarrea, la transmisión de patógenos diarreicos, el manejo del equipo y los alimentos cocinados, las prácticas de lavado de manos y la higiene del medio.
1. The role of food safety in health and development. Report of the Joint FAO/WHO Expert Committee on Food Safety. Geneva: World Health Organization; 1983.
2. Background paper: Developing a food safety strategy. WHO Strategic Planning Meeting. Geneva: World Health Organization 2001.
3. Mensah P. Persistent diarrhoea in Ghana. (Report submitted to Japan International Cooperation Agency; 1997).
4. Mensah P, Owusu-Darko K, Yeboah-Manu D, Ablordey A, Nkrumah FK, Kamiya H. The role of street food vendors in the transmission of enteric pathogens. Ghana Medical Journal 1999;33:19-29.
5. Harrigan FW, McCance ME. Laboratory methods in food microbiology. London: Academy Press;1968.
6. Thatcher PS, Clark DS. Microorganisms in foods: their significance and methods of enumeration. Volume 1. Toronto: University of Toronto Press; 1968.
7. Balows A,Hausler WJ Jr., Herrmann KL, Isenberg HD, Shadomy HJ editors. Manual of clinical microbiology. Washington (DC): American Society of Microbiology; 1991.
8. Tamatsukuri S, Yamamoto K, Shibata S, Leano F, Honda T, Miwatani T. Detection of heat-labile enterotoxin gene in enterotoxigenic Escherichia coli by densitometric evaluation using highly specific enzyme-linked oligonucleotide probes. European Journal of Infectious Diseases 1991;10:1048-55.
9. Nataro JP, Kaper JB, Robins-Browne R, Prado V, Vial P, Levine MM. Patterns of adherence of diarrheagenic Escherichia coli to HEp-2 cells. Paediatric Infectious Diseases Journal 1987;6:829-31.
10. Kubheka LC, Mosepye FM, von Holy A. Microbial survey of street-vended salad and gravy in Johannesburg City, South Africa. Food Control 2001;12:127- 31.
11. Mensah P, Tomkins AM, Drasar BS, Harrison TJ. Antimicrobial effect of fermented Ghanaian maize dough. Journal of Applied Bacteriology 1991;70:302-10.
12. Desmarchelier PM, Apiwathnasom C, Vilainerun D, Watson C, Johan MR, Ahmad Z, et al. Evaluation of safety of domestic food preparation in Malaysia. Bulletin of the World Health Organization 1994;72:877-84.
13. Sackey BA, Mensah P, Collison E, Sakyi-Dawson E. Campylobacter, Salmonella, Shigella and Escherichia coli in live and dressed poultry from metropolitan Accra. International Journal of Food Microbiology 2001;71:21-8.
14. Roberts D. Factors contributing to outbreaks of food poisoning in England and Wales 1970-1979. Journal of Hygiene 1982;89:491-8.
15. Black RE, Lopez de Romana G, Brown KH, Bravo N, Bazalar OG, Kanashiro HC. The incidence and etiology of infantile diarrhea and major routes of transmission in Huascar, Peru. American Journal of Epidemiology 1989;189: 785-99.
16. Lim-Quizon MC, Benabaye RM, White FM, Dayrit MM, White ME. Cholera in metropolitan Manila: foodborne transmission via street vendors. Bulletin of the World Health Organization 1994;72:745-9.
17. Kotoku EK. Health hazards (microbiological) associated with salad (lettuce) purchased from street stalls in Accra [dissertation]. Legon: University of Ghana; 1978.
18. Mensah P, Armar-Klemesu M, Hammond AS, Haruna A, Nyarko R. Bacterial contaminants in lettuce, tomatoes, beef and goat meat from metropolitan Accra. Ghana Medical Journal 2001;35:1-6.
19. Mathur R, Reddy V. Bacterial contamination of infant foods. Indian Journal of Medical Research 1983;77:342-6.
20. Echeverria P, Taylor DN, Seriwatana J, Leksomboon U, Chaicumpa W, Tirapat C, et al. Potential sources of enterotoxigenic Escherichia coli in homes of children in Thailand. Bulletin of the World Health Organization, 1987;65:207-15.
21. Olukoya DK, Bakare SB, Abayomi O. Microbiological evaluation of food samples sold to primary school children in Lagos, Nigeria. Journal of Tropical Paediatrics 1991;37:266.
22. Levine OS, Levine MM. Houseflies (Musca domestica) as mechanical vectors of shigellosis. Reviews of Infectious Diseases 1991;13:688-96.
23. Bannan EA, Judge LF. Bacteriology studies relating to hand washing. The inability of soap bars to transit bacteria. American Journal of Public Health 1965;55:915-22.
24. Heinze JE, Yackovich F. Washing with contaminated bar soap is unlikely to transfer bacteria. Epidemiology and Infection 1988;101:135-42.
25. Ericsson CD, Pickering LK, Sullivan P, DuPont HL. The role of location of food consumption in the prevention of travellers' diarrhoea in Mexico. Gastroenterology 1980;79:812-916.
26. Tjoa WS, DuPont HL, Sullivan P, Pickering LK, Holguin AH, Olarte J, et al. Location of food consumption and travellers' diarrhoea. American Journal of Epidemiology 1977;106:61-6.
27. El-Sherbeeny MR, Saddik MF, Bryan FL. Microbiological profile of foods served by street vendors in Egypt. International Journal of Food Microbiology 1985;2:355-64.
28. Jiwa SF, Krovacek K, Wadstrom T. Enteropathogenic bacteria in food and water from an Ethiopian community. Applied and Environmental Microbiology 1981;41:1010-9.
29. Umoh VJ, Odoba MB. Safety and quality evaluation of street foods sold in Zaria, Nigeria. Food Control 1999;10:9-10.
30. Abdussalam M, Kaferstein FK. Safety of street foods. World Health Forum 1993;14:191-4.
1 Professor of Medical Bacteriology and Head, Bacteriology Unit, Noguchi Memorial Institute for Medical Research, University of Ghana, PO Box LG581, Legon/Accra, Ghana. Correspondence should be sent to this author at St Hilda's College, Oxford, OX4 1DY, England (email: email@example.com).
2 Chief Research Assistant, Bacteriology Unit, Noguchi Memorial Institute for Medical Research, University of Ghana, Legon/Accra, Ghana.
3 Technologist, Bacteriology Unit, Noguchi Memorial Institute for Medical Research, University of Ghana, Legon/Accra, Ghana.
4 Principal Research Assistant, Bacteriology Unit, Noguchi Memorial Institute for Medical Research, University of Ghana, Legon/Accra, Ghana.
Ref. No. 00-0982