Salmonella and Campylobacter Contamination of Ready-to-Eat Street-Vended Pork Meat Dishes in Antananarivo,Madagascar: A Risk for the Consumers?

Abstract

Street-food vending has been increasing in many developing countries and particularly in Madagascar since 2000. Gastroenteric diseases cause 37% of all deaths each year, and 50% of children <5 years are infected with intestinal pathogens. However, there has been little information regarding the incidence of street-food-related diseases, or foodborne pathogens in pork, which is the most commonly eaten meat, along with chicken. Thus, the aim of this study was to investigate the safety of traditional ready-to-eat street-vended pork dishes and to assess the association of restaurant characteristics and cooking practices with Salmonella and Campylobacter contamination of these meals. Sixty street-restaurants were studied from March 2012 to August 2012 in Antananarivo. A questionnaire was submitted to the managers, and samples of ready-to-eat pork dishes were bought. Salmonella spp. were isolated in 10% of the 60 street-restaurants studied and in 5% samples of pork dishes. The most prevalent serovars isolated were Salmonella Typhimurium (44%) and Senftenberg (33%). Campylobacter was not detected. Only 4 of the 43 variables tested in the screening analysis were significantly associated with Salmonella spp. contamination of the street-restaurants. The risk for a restaurant to be Salmonella positive decreased when there were specific premises for the restaurant and when the staff was wearing specific clothes when working. Conversely, that risk increased when the temperature of ready-to-eat pork was <52°C and when tablecloths were used in the restaurant.

Introduction

Foodborne diseases encompass a wide spectrum of illnesses and are a growing public health problem worldwide; they are also a significant cause of reduced economic growth worldwide (White et al., 1997; Opsteegh and van der Giessen, 2011). The global burden of foodborne diseases and its impact on development and trade is currently unknown in both industrialized and developing countries (Majowicz et al., 2010). However, developing countries tend to suffer from the largest share of the burden of foodborne diseases (Kaferstein, 2003).

Salmonellosis is a widespread foodborne zoonosis in many countries (Wegener and Baggesen, 1996; Lo Fo Wong, 2002; Iyer et al., 2013); people of all ages are affected, and the incidence is highest in children. In the developing world, human immunodeficiency virus infection and AIDS are important risk factors for nontyphoidal salmonellosis and bacteremia (Obi and Bessong, 2002); and several studies have shown in infected African adults with documented bloodstream infections that nontyphoidal salmonellae were isolated from up to 35% of adults (Hohmann, 2001). In tropical countries, investigations have also confirmed that infections caused by Campylobacter spp. may be as serious as those by Salmonella spp., both in frequency and severity of symptoms (Coker et al., 2002).

Contaminated food is the usual source of human infections and, besides eggs and poultry, pig meat has also been considered as an important source of human salmonellosis and campylobacteriosis since the early 1990s (Mead, 1993; Bailey et al., 2001). Thus, reducing Salmonella and Campylobacter contamination of pork products would reduce the risk of foodborne disease to consumers.

Street-vended foods are ready-to-eat (RTE) foods prepared and sold by vendors on streets or similar public places (Ekanem, 1998). Informal food vending has been increasing in many developing countries, mainly due to economic and social difficulties and urbanization (Hanashiro et al., 2005). Street-foods provide a source of readily available, inexpensive, nutritional meals, while providing a source of income for the vendors (Mensah et al., 2002). However, the vendors' lack of training on basic food hygiene raises concerns over the safety of these street-vended foods (Moy et al., 1997). Stands are often crude and inefficient structures, where toilets and adequate washing facilities are rarely available. The washing of hands, utensils, and dishes is often done in the same bucket or bowl. Disinfection is not always carried out properly, and insects and rodents may be attracted to sites where there is no organized sewage disposal.

In Madagascar, gastroenteric diseases cause 37% of all infection and parasite-related deaths each year (WHO, 2013), and a recent study showed that nearly 50% of children <5 years tested positive for intestinal pathogens and 10% of this group died of diarrhea-related illness (WHO, 2011; Randremanana et al., 2012). However, in this country, where street-food vending is common, there has been little information regarding the incidence of street-food-related diseases, or foodborne pathogens in pork (Rasamoelina-Andriamanivo et al., 2013; Temmam et al., 2013).

The aim of this study was to investigate the safety of traditional RTE street-vended pork dishes and to assess the association of restaurant characteristics and cooking practices with Salmonella and Campylobacter contamination of these meals.

Materials and Methods

Sampling method

The study was conducted during 5 months from March 15, 2012 until August 15, 2012 in 13 districts of Antananarivo (Ambohimanarina, Anosibe, Besarety, Andravoahangy, Analakely, Ambodivona, Fasankarana, Ankatso, Mahamasina, Anosy, 67Ha, Isotry, and Ampefiloha) and involved 60 local street-restaurants (4–5 restaurants per district). The restaurants were chosen randomly throughout the city's districts (once in a district, a coin was thrown; when the coin landed heads, the surveyors were going to the right, and when tails came up, they were going to the left and so on). Prior to any investigation, we explained the aim of the study to the manager of the restaurant. Then, the selection was based on the manager's willingness to cooperate during the study. Total confidentiality was assured to the vendors, and only four declined to participate in our study.

Questionnaire survey

Each restaurant was visited once. Data concerning kitchen and restaurant characteristics; pork pieces and meat handling; ingredients and vegetables management; food storage conditions; knife and utensil use; control of rodents, insects, and other domestic animals; cooking procedures; use of leftovers; restaurant staff; and cleaning and disinfection procedures were collected from a questionnaire administered to each vendor (Supplementary Table S1; Supplementary Data are available online at www.liebertpub.com/fpd). A team of two professional surveyors (one public health assistant and one technician) from the laboratory (Merieux Fundation and Centre de Coopération Internationale en Recherche Agronomique pour le Développement) administered the 30-min questionnaire. This questionnaire was pretested in a preliminary study carried out in five restaurants. The final questionnaire had 58 questions, and 78% were closed-ended questions. As far as possible, the answers were cross-checked by direct observation from the team and corrected if necessary. Only managers were interviewed. When the manager was not fluent in French (administrative language), the public-health assistant translated the questions into Malagasy, the widely spoken national language. Two to three restaurants were visited on every sampling day. During each visit, three pork dishes were bought randomly, placed in separate sterile containers, and transported within 2 h to the laboratory in a cool box with freeze packs. These pork dishes were traditional Malagasy meals such as ravitoto sy henakisoa (pork stew with cassava leaves), henakisoa sy voanjobory (pork and Bambara peas), or henakisoa sy tsaramaso (pork and beans). Pork meat temperatures were recorded at the time of sampling using a portable thermometer, previously calibrated and sterilized before any use.

Sample collection

In each participating restaurant, a sample of three meals including pork meat was collected. Separate spoons were used to collect each sample from the dish into a whirlpack/plastic bag, shipped with an ice pack for 2 h before delivery to the lab for immediate culture. Each meal received a unique code, and 178 samples composed the collection.

Laboratory investigations

The pork meat contents of each dish were aseptically cut into small pieces (1–1.5 cm²) to increase the surface area (Castillo et al., 1993).

Salmonella isolation and identification

Pork samples, weighing 10 g, were homogenized with a kitchen blender (Vista, France) with buffered peptone water (AES Laboratoire, Combourg, France) in 1:10 sample/broth ratio at 37°C for 18–20 h. One milliliter of this pre-enrichment broth was used to inoculate 10 mL of Muller-Kauffmann tetrathionate broth (AES Laboratoire) and 100 μL of the pre-enriched broth was used to inoculate a Modified Semi-solid Rappaport Vassiliadis (MSRV) (Merck, Nogent sur Marne, France) agar plate. Both of the media were incubated at 37°C and 41.5°C, respectively, for 24 h. Plating was previously achieved by streaking cultures, from a migration zone on MSRV above 20 mm, onto Salmonella-Shigella (SS) agar plates and subsequently by streaking culture from tetrathionate broth onto xylose lysine desoxycholate (XLD) agar plates (AES Laboratoire). The SS and XLD agar plates were incubated at 37°C for 24 h. Typical Salmonella colonies were confirmed by biochemical assays on Kligler-Hajna medium, ONPG medium, and lysine decarboxylase, and then serotyped by a slide agglutination test using Salmonella polyvalent O and H antisera (Diagnostic Pasteur, Paris, France).

Campylobacter isolation and identification

Pork samples (10 g) were added to 90 mL of Preston broth with Preston antibiotic supplement (Oxoid Laboratory, England) and incubated at 42°C for 24 h under microaerophilic conditions (CampyGen; Oxoid Laboratory, England). Each sample was then streaked onto Virion plates (Mueller Hinton agar; Merck, Germany; Bacto agar; Difco Laboratory, United States; with 5% of defibrinated horse blood; AES Laboratoire, France) and onto Karmali plates (Oxoid Laboratory, England). Plates were incubated at 42°C under microaerophilic conditions for 48 h. Isolates were identified by direct observation (specific motility under microscope) and using a commercial identification method (API Campy^®; bioMérieux, France) and further submitted to a multiplex polymerase chain reaction to identify their species (Denis et al., 2001).

Definition of outcome variable

The observation unit was the restaurant; samples of the three pork dishes were taken in each restaurant and analyzed. A restaurant was declared contaminated by Salmonella only if one or more RTE pork dishes tested positive. The outcome variable was thus dichotomous (contaminated versus noncontaminated restaurant).

Definition of explanatory variables

Supplementary Table S2 lists the variables under study. All variables were coded categorically. The number of categories per variable was limited, such that frequency rates of categories were >10%. These variables were selected from a preliminary step aimed at lowering the chance of obtaining results affected by multicollinearity in the dataset (Dohoo et al., 1996). Correlations between possible explanatory variables were checked (chi square). For all bilateral relationships between variables evidencing strong structural collinearity, one of the two variables of interest (the one most related to the outcome variable) was chosen. The variable “Cleanliness of clothing for restaurant staff” was strongly correlated with “Specific work clothing for the staff” and “Frequency of cleaning of hands in the kitchen.” Only “Cleanliness of clothing for restaurant staff” has been kept for further analysis because it was the variable more related to Salmonella status.

Statistical procedure

A two-stage procedure was used to assess the relationship between explanatory variables and the Salmonella status of the restaurant. Logistic regression was used according to the method described by Hosmer and Lemeshow (2000). In the first stage, a univariable analysis was performed to relate Salmonella contamination of the restaurant to each explanatory variable. Only factors associated with Salmonella contamination (Pearson chi-square test, p<0.25) were included in a full model for multivariate analysis (Supplementary Table S2) (Mickey and Greenlands, 1989). The second stage involved a logistic multiple-regression model that included all factors that passed the first screening test. The contribution of each factor to the model was tested using a likelihood ratio chi-square test through a stepwise procedure (backward and forward). At the same time, the simpler models were compared to the full model by the Akaike information criterion (Akaike, 1974). This process was continued automatically until a model was obtained with all factors significant at p<0.05. Goodness-of-fit of the final model was assessed using Pearson chi square and the Deviance and Hosmer-Lemeshow tests (Hosmer and Lemeshow, 2000). Interactions were not tested (because of the small sample sizes). All the statistical analyses were performed with R software (Development Core Team, 2013), and p-values <0.05 were considered as significant.

Results

Salmonella spp. were isolated in 10% of the 60 street-restaurants studied and in 5% samples of pork dishes. The most prevalent serovars isolated were Salmonella Typhimurium (44%) and Senftenberg (33%). Campylobacter was not detected. Only 4 of the 43 variables tested in the screening analysis were significantly associated with Salmonella spp. contamination of the street-restaurants (Table 1). The risk for a restaurant to be Salmonella positive decreased when there were specific premises for the restaurant and when the staff was wearing specific clothes when working. Conversely, that risk increased when the temperature of RTE pork was <52°C and when cloths were put on the restaurant's tables.

Table 1.

Final Logistic Regression Model for Risk Factors of Salmonella Contamination in 60 Street-Restaurants in Antananarivo, Madagascar, 2012

		Logistic regression model ^b
Selected variables and p-values from the model	Percentage of Salmonella- positive restaurants ^a	Odds ratio	95% CI	p-Value
Specific premises for the restaurant				0.01
No	33.4	1	—
Yes	5.8	0.03	0.0025–0.29
Specific clothes for the staff				0.05
No	17.1	1	—
Yes	0	0.05	0.004–0.57
Use of tablecloths				0.03
No	6.5	1	—
Yes	21.4	8.83	1.24–62.2
Observed temperature (°C) in pork dishes				0.05
<52.5°C	12	5.41	1.25–35.22
>52.5°C	0	1	—

Salmonella-contaminated street-restaurant for ready-to-eat pork dish before consumption. Based on the total number of restaurants in the class.

Intercept=0.875, model deviance=53.25, alkaline information criterium (AIC)=65.25, model df=6 (p<0.01).

Discussion

One bias might have been introduced because of the required willingness of restaurant managers to cooperate during the study, but all the street-restaurants have been chosen randomly in the districts of Antananarivo, and because the objectives of the study were well explained, very few managers declined to participate. Furthermore, the use of questionnaires for data collection could also have introduced bias, but most questions were objective and closed and a detailed description for each of the categories was provided for subjective questions. Moreover, the study duration and the data collection by only six persons trained for this work would have contributed to repeatability of the results.

We used a pre-enrichment step followed by selective enrichment in semisolid medium for the bacteriological analysis of Salmonella, which is the most effective method for detecting common Salmonella in meals (Davies et al., 1997); the global procedure slightly differed from the ISO standard because some media and consumables were difficult to obtain. For Campylobacter investigation, we cut the meat in small pieces to increase the surface as recommended by Castillo et al. (1993). We only used Preston broth for the enrichment, whereas Bolton broth has the property of recovering stressed cells; thus, we may have missed some isolates. However, in other studies in street-restaurants (e.g., in Trinidad), no Campylobacter spp. in RTE foods have been detected (Gibbons et al., 2006). These bacteria are known to be extremely heat sensitive, and cooking seemed to be the major cause of destruction of Campylobacter spp. in the pork dishes in Madagascar.

In our study, we found that 10% of the street-restaurants and 5% of the pork dishes were positive for Salmonella spp. Similar results have already been shown for RTE dishes and retail meats in other countries, particularly in developing countries, such as Cuba, where Caballero et al. (1998) found that 3.5% of the meat products sold in the street were positive for Salmonella; in Trinidad, 34.4% of RTE processed meats were found Salmonella positive (Gibbons et al., 2006). In 2013, Manguiat and Fang (2013) detected Salmonella in 66% and 48% of hot grilled pork in Taiwan and in the Philippines, respectively, but the method of cooking was different. The most-prevalent Salmonella serovars in pork dishes were Salmonella Typhimurium and Senftenberg; Salmonella Typhimurium is classically recovered from pigs and pork everywhere in the world (Boyen et al., 2008) as well as in the Indian Ocean region (Cardinale et al., 2010). Salmonella Senftenberg is regularly isolated from poultry (Henry et al., 2012) but also from pigs (Wong et al., 1998). These findings suggested that these Salmonella serovars could survive during the cooking procedures or could be brought back onto pork dishes by cross-contamination; Salmonella Senftenberg is believed to be the most resistant Salmonella to heat treatment (Elliott and Heiniger, 1965).

Indeed, a core temperature of RTE pork <52°C was associated with a higher risk of Salmonella contamination. Heat treatment is still considered the essential means of eliminating foodborne pathogens from raw meat and meat products (Juneja et al., 2001), and inadequate time/temperature exposure during cooking remains a contributing factor in food-poisoning outbreaks. Salmonella is sensitive to heat, and these organisms are killed at temperatures of 65°C or above (Guthrie, 1991). In Madagascar, pork is generally boiled for >2 h, and generally boiling water even for only 15 min eliminates Salmonella from meat (Osaili et al., 2007). Thus, in most cases, RTE pork dishes might be contaminated after cooking and before being served; in fact, foods are generally cooked in the early morning and they are stored at ambient temperature, waiting for the rush hours. During that time, pork dishes are not always separated from raw meat (pork and poultry) and vegetables and they are only reheated a few minutes at a low temperature (50–55°C) when the consumers are coming, anytime during the day. Therefore, cross-contamination should be prevented by avoiding any contact with raw meat, vegetables, utensils, hands, or any surfaces previously contaminated. This method of food recontamination during postcooking has been highlighted several times in RTE foods in Ghana (Mensah et al., 2002), Senegal (Cardinale et al., 2005), in South Africa (Mosupye and Von Holy, 2000), and Trinidad (Badrie et al., 2003).

We also observed that the risk for a restaurant to be Salmonella positive was decreased when there were specific premises for the restaurant. Indeed, structures within food establishments are generally soundly built of durable materials and are easy to maintain, clean, and, where appropriate, able to be disinfected (WHO, 1996). Temporary premises such as mobile sales, street-vending vehicles, or even tents in the streets are not easily cleaned and maintained and they are not designed to avoid contaminating food and harboring pests. Furthermore, most of the time, food meals sold in these stores are prepared and cooked earlier and they are brought there to be sold. Then, no further cooking of food may occur at the point of sale and the foods are held at ambient temperature, which promotes the growth of mesophilic microorganisms and most pathogens (Jay et al., 2005).

We also found that the risk of contamination was decreased when the staff was wearing specific clothes while working in the restaurant. These results corroborated the data reported by Mensah et al. (2002) and Manguiat and Fang (2013) in identifying personnel as mechanical vehicles of Salmonella cross-contamination from raw meat, vegetables, or environment via utensils or other tools (Hanashiro et al., 2005). The strong correlation of that variable with cleanliness of clothing stressed that defective personal hygiene could increase the risk of pathogen transmission from humans to food. Restaurant staff frequently handled vegetables, raw pork or chicken, and cooked meat without in-between cleaning, and this could explain cross-contamination of pork meals. Moreover, as Isara et al. (2010) confirmed, humans may be Salmonella spp. carriers, even if they do not exhibit any symptoms, and poor personal hygiene of food handlers could be responsible for the contamination of food by Salmonella cells.

Finally, the risk of Salmonella contamination was increased when cloths were put over the restaurant's tables. That risk factor was rather unexpected since a tablecloth is considered a form of protective material designed to provide a shield between the tabletop and the physical items placed on top of it, to preserve the quality of the tabletop. However, in these street-restaurants in Madagascar, cotton is widely employed as the material in the production of tablecloths, and one disadvantage of cotton tablecloths is that they get dirty and gather organic debris easily due to spillage. Furthermore, cotton tablecloths are difficult to clean. As these tablecloths are infrequently removed, they might be an efficient cross-contamination tool between successive meals and successive consumers.

Conclusions

This study showed that only a few RTE pork meals were contaminated with Salmonella, possibly due to the heat treatment, but some incorrect food hygiene and safety practices still remain to be improved to prevent contamination efficiently in the street-restaurants. Therefore, street-vendors must be sensitized and trained continuously on food hygiene and sanitation, emphasizing the vendors' responsibility toward customers' health using the Malagasy traditional ideology of tody (automatic return of good and bad).

Footnotes

Acknowledgments

This study was conducted in the framework of AnimalRisk-OI, a research program on emerging animal diseases in the Indian Ocean, funded by FEDER POCT (European Union, Regional Council of Reunion, and the French government).

Disclosure Statement

No competing financial interests exist.

References

Akaike

. A new look at statistical model identification. In: IEEE Transactions. 19 Transaction Automatic Control. Tokyo, Japan: Institute of Statistical Mathematics, 1974, pp. 716–722.

Badrie

, Joseph

, Chen

. An observational study of food safety practices by street vendors and microbiological quality of street-purchased hamburger beef patties in Trinidad, West Indies. Int J Food Safety, 2003; 3:25–31.

Bailey

, Stern

, Fedorka-Cray

, Craven

, Cox

, Cosby

, Ladely

, Musgrove

. Sources and movement of Salmonella through integrated poultry operations: A multistate epidemiological investigation. J Food Prot, 2001; 64:1690–1697.

Boyen

, Haesebrouck

, Maes

, Van Immerseel

, Ducatelle

, Pasmans

. Non-typhoidal Salmonella infections in pigs: A closer look at epidemiology, pathogenesis and control. Vet Microbiol, 2008; 130:1–19.

Caballero

, Carrera

JAV

, Lengomin

MEF

. Evaluacion de la vigilancia microbiologica de alimentos que se venden en las calles [Street-vended food safety assessment]. Rev Cubana Aliment Nutr, 1998; 12:7–10. (In Spanish.)

Cardinale

, Gros-Claude

JDP

, Tall

, Gueye

, Salvat

. Risk factors for contamination of ready-to-eat street-vended poultry dishes in Dakar, Senegal. Int J Food Microb, 2005; 103:157–165.

Cardinale

, Maeder

, Porphyre

, Debin

. Salmonella in fattening pigs in Reunion Island: Herd prevalence and risk factors for infection. Prev Vet Med, 2010; 96:281–285.

Castillo

MML

, Sanchez

, Rodriguez

, Quinones

EIR

, Lugo

, Vazquez

. Recuperacion de Campylobacter jejuni ATCC 29428 a partir de carne de cerdo frita y de pollos rostizados inoculados [Recovery of Campylobacter jejuni ATCC 29428 from fried pork and roasted inoculated chicken]. Rev Latino-Am Microbiol, 1993; 35:15–18. (In Spanish.)

Coker

, Isokpehi

, Thomas

, Amisu

, Obi

. Human campylobacteriosis in developing countries. Emerg Infect Dis, 2002; 8:237–244.

10.

Davies

, Nicholas

RAJ

, Mclaren

, Corkish

, Lanning

, Wray

. Bacteriological and serological investigation of persistent Salmonella enteritidis infection in an integrated poultry organisation. Vet Microbiol, 1997; 58:277–293.

11.

Denis

, Refregier-Petton

, Laisney

, Ermel

, Salvat

. Campylobacter contamination in French chicken production from farm to consumers: Use of a PCR assay for detection and identification of Campylobacter jejuni and Camp. coli . J Appl Microbiol, 2001; 91:255–267.

12.

Dohoo

, Ducrot

, Fourichon

, Donald

, Hurnik

. An overview of techniques for dealing with large numbers of independent variables in epidemiologic studies. Prev Vet Med, 1996; 29:221–239.

13.

Ekanem

. The street food trade in Africa: Safety and socio-environmental issues. Food Control, 1998; 9:211–215.

14.

Elliott

, Heiniger

. Improved temperature-gradient incubator and maximal growth temperature and heat resistance of Salmonella . Appl Microbiol, 1965; 13:73–76.

15.

Gibbons

, Adesiyun

, Seepersadsingh

, Rahaman

. Investigation for possible source(s) of contamination of ready-to-eat meat products with Listeria spp. and other pathogens in a meat processing plant in Trinidad. Food Microbiol, 2006; 23:359–366.

16.

Guthrie

. Salmonella. Boca Raton, FL: CRC Press Inc., 1991.

17.

Hanashiro

, Morita

, Matte

, Torres

. Microbiological quality of selected street foods from a restricted area of Sao Paulo city, Brazil. Food Control, 2005; 16:439–444.

18.

Henry

, Granier

, Courtillon

, Lalande

, Chemaly

, Salvat

, Cardinale

. Salmonella enterica subsp enterica isolated from chicken carcasses and environment at slaughter in Reunion Island: Prevalence, genetic characterization and antibiotic susceptibility. Trop Animal Health Production, 2012; 45:317–326.

19.

Hohmann

. Nontyphoidal salmonellosis. Clin Infect Dis, 2001; 32:263–269.

20.

Hosmer

, Lemeshow

. Applied Logistic Regression. New York: Wiley, 2000.

21.

Isara

, Isah

, Lofor

PVO

, Ojide

. Food contamination in fast food restaurants in Benin City, Edo State, Nigeria: Implications for food hygiene and safety. Public Health, 2010; 124:467–471.

22.

Iyer

, Kumosani

, Yaghmoor

, Barbour

, Azhar

, Harakeh

. Escherichia coli and Salmonella spp. in meat in Jeddah, Saudi Arabia. J Infect Dev Countries, 2013; 7:812–818.

23.

Jay

, Loessner

, Golden

. Modern Food Microbiology. New York: Springer, 2005.

24.

Juneja

, Eblen

, Ransom

. Thermal inactivation of Salmonella spp. in chicken broth, beef, pork, turkey, and chicken: Determination of D- and Z-values. J Food Sci, 2001; 66:146–152.

25.

Kaferstein

. Food safety as a public health issue for developing countries. In: Food Safety in Food Security and Food Trade. Washington, DC: International Food Policy Research Institute, 2003.

26.

Lo Fo Wong

DMA

, Dahl

, van der Wolf

, Swanenburg

. Epidemiology and control measures for Salmonella in pigs and pork. Livestock Production Sci, 2002; 76:215–222.

27.

Majowicz

, Musto

, Scallan

, Angulo

, Kirk

, O'Brien

, Jones

, Fazil

, Hoekstra

, International Collaboration on Enteric Disease 'Burden of Illness' Studies. The global burden of nontyphoidal Salmonella gastroenteritis. Clin Infect Dis, 2010; 50:882–889.

28.

Manguiat

, Fang

. Microbiological quality of chicken- and pork-based street-vended foods from Taichung, Taiwan, and Laguna, Philippines. Food Microbiol, 2013; 36:57–62.

29.

Mead

. Problems of producing safe poultry: Discussion paper. J R Soc Med, 1993; 86:39–42.

30.

Mensah

, Yeboah-Manu

, Owusu-Darko

, Ablordey

. Street foods in Accra, Ghana: How safe are they?. Bull World Health Org, 2002; 80:1–14.

31.

Mickey

, Greenlands

. A study of confounder-selection criteria on effect estimation. Am J Epidemiol, 1989; 129:125–137.

32.

Mosupye

, Von Holy

. Microbiological hazard identification and exposure assessment of street food vending in Johannesburg, South Africa. Int J Food Microbiol, 2000; 61:137–145.

33.

Moy

, Hazzard

, Kaferstein

. Improving the safety of street-vended foods. World Health Stat Q, 1997; 50:124–131.

34.

Obi

, Bessong

. Diarrhoeagenic bacterial pathogens in HIV-positive patients with diarrhoea in rural communities of Limpopo Province, South Africa. J Health Popular Nutr, 2002; 20:230–234.

35.

Newell

, Koopmans

, Verhoef

, Duizer

, Aidara-Kane

, Sprong

, Opsteegh

, Langelaar

, Threlfall

, Scheutz

, van der Giessen

, Kruse

. Food-borne diseases—The challenges of 20 years ago still persist while new ones continue to emerge. Int J Food Microbiol, 2010; 139(suppl 1):S3–S25. [Erratum in Int J Food Microbiol 2011;145:493–493.]

36.

Osaili

, Griffis

, Martin

, Beard

, Keener

, Marcy

. Thermal inactivation of Escherichia coli O157: H7, Salmonella, and Listeria monocytogenes in breaded pork patties. J Food Sci, 2007; 72:M56–M61.

37.

Randremanana

, Randrianirina

, Gousseff

, Dubois

, Razafindratsimandresy

, Hariniana

, Garin

, Randriamanantena

, Rakotonirina

, Ramparany

, Ramarokoto

, Rakotomanana

, Ratsitorahina

, Rajatonirina

, Talarmin

, Richard

. Case-control study of the etiology of infant diarrheal disease in 14 districts in Madagascar. PLoS ONE, 2012; 7:e44533.

38.

Rasamoelina-Andriamanivo

, Porphyre

, Jambou

. Control of cysticercosis in Madagascar: Beware of the pitfalls. Trends Parasitol, 2013; 29:538–547.

39.

Temmam

, Besnard

, Andriamandimby

, Foray

, Rasamoelina-Andriamanivo

, Heraud

, Cardinale

, Dellagi

, Pavio

, Pascalis

, Porphyre

. High prevalence of hepatitis E in humans and pigs and evidence of genotype-3 virus in swine, Madagascar. Am J Trop Med Hyg, 2013; 88:329–338.

40.

Wegener

, Baggesen

. Investigation of an outbreak of human salmonellosis caused by Salmonella enterica ssp. enterica serovar Infantis by use of pulsed field gel electrophoresis. Int J Food Microbiol, 1996; 32:125–131.

41.

White

, Baker

, James

. Strategies to control Salmonella and Campylobacter in raw poultry products. Rev Sci Technique (Int Office Epizootics), 1997; 16:525–541.

42.

[WHO] Essential Safety Requirements for Street-Vended Food. Report no. 96.7, Division of Food and Nutrition. Geneva: WHO, 1996.

43.

WHO. DoMaHI Global Burden of Disease and Death Estimates. Geneva: WHO, 2011.

44.

WHO. DFID-AHP Madagascar: Health Profile. Geneva: WHO, 2013.

45.

Wong

, Linton

, Gerrard

. Reduction of Escherichia coli and Salmonella Senftenberg on pork skin and pork muscle using ultraviolet light. Food Microbiol, 1998; 15:415–423.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB