Evaluation of microbial safety knowledge, attitude and practice of street food vendors and consumers in Can Tho City, Vietnam

Street foods have become much more popular in Vietnam. However, the information on the safety of street food sold in Vietnam and food knowledge safety as well as awareness of both Vietnamese vendors and consumers is very limited. The aim of this study was to evaluate the level of knowledge, attitude, and practice in food safety of street food consumers and vendors who volunteered to participate in the study in Can Tho city, Vietnam. This study was conducted in two parts. In the first part, 838 food consumers and 598 food handlers were interviewed via a questionnaire to assess their food safety knowledge and attitude. The second part focused on the observation of the food handling practice of the vendors by means of a checklist. Results indicated that consumers and vendors generally exhibited average food safety knowledge and attitude levels. The consumers and vendors were determined to have higher levels of food safety attitude than knowledge, whilst trained consumers and vendors had better food safety knowledge and attitude compared to untrained consumers and vendors. The observation study showed that the majority of the vendors had unhygienic practice i.e. handled food with bare hands (84.5%), handled money while serving food (92.3%) and without washing hands after handling money (93.3%), etc. These results justify the need for training the vendors about food safety knowledge and practice and further additional studies of street foods and their potential risk of foodborne illness.


Introduction
Street foods are defined as ready-to-eat food and beverages prepared and/or sold by vendors and handlers especially in streets and similar public places for immediate consumption or consumption at a later stage without further processing or preparation (FAO, 2013). It was estimated that there are 2.5 billion people who consumed street food all around the world daily (FAO, 2007). Street foods are highly appreciated by consumers due to their flavors, attractive, diverse, convenient, low cost, cultural and social heritage links (Ekanem, 1998;Chukuezi, 2010;Aluko et al., 2014;da Silva et al., 2014).
However, street food vendors are very often poor, uneducated and show little concern towards the safe handling of foods (Mensah et al., 2002;Lues et al., 2006). Consequently, some serious concerns do exist about the safety of street food (Muinde and Kuria, 2005;Rheinländer et al., 2008;FAO, 1995). The concerns have been realized as street-vended foods are associated with poor microbiological quality and potential cause of outbreaks around the world (Dawson and Canet, 1991;Bryan et al., 1992;Aluko et al., 2014).
To date, the food safety knowledge, attitude, and practice of food handlers (including street food vendors) in several countries have been reported in several studies. Food safety and sanitary practice at car parks in Nigeria showed only 16.9% of vendors always wash hands after using the toilet and only 3.8% always refrigerate leftover cooked food; 61.9% of the vendors had no formal training about food safety (Aluko et al., 2014). In addition, observation street food vendors in Port-au-Prince, Haiti found that flies and animals were around the street food stall (60%) and no accessibility to potable water (65%) (Samapundo et al., 2015). It is explained that poor quality of raw materials, improper food preparation, lack of awareness about food safety, etc. are possibly the most common cause of food poisoning outbreaks (Liu et al., 2014). However, for vendors theoretical training is not related to the attitude and practice of food handlers (da Cunha et al., 2014).
In addition, the consumer's knowledge and attitude toward food safety differ according to demographic and socio-economic factors. Consumers were however determined to have a lower level of food safety knowledge than vendors (Samapundo et al., 2015). Although the specific regulations for street food vendors including locations, facilities, environment around the stall and personal hygiene have been issued by the Ministry of Health in their revised Law of food safety in 2012 (30/2012/TT-BYT). Until now, there is no formal study on the food safety knowledge, attitude and practice of vendors and consumers of street foods, particularly in Can Tho city.
As other developing countries, Vietnam is largely known for its delicious and diverse street foods which do not only serve the eating habits of urban residents, especially in low and medium-income areas in Can Tho city but also attract the curiosity of tourists. There are about 5000 street food vendors in Can Tho city (population of 1.2 million and 1408 km 2 ). The street food in Can Tho city accounts for up to 95% of urban consumer purchases (mainly by more than 60 thousand students studying in universities located in Can Tho city); 51% consuming for the main dish daily and 82% for breakfast (based on a local report). Despite the alarming food poisoning outbreaks due to consuming street foods, at the moment very few studies are available about the level of food safety knowledge, attitude, and practice of both vendors and consumers of street foods in Can Tho city, the biggest city in Mekong Delta of Vietnam. Therefore, this study is aimed to evaluate the level of food safety knowledge and attitude of both the vendors and consumers, and to access the food safety practice of the vendors in order to raise the safety awareness of street foods in Vietnam.

Sample size
The study was conducted in five districts (i.e. Ninh Kieu, Cai Rang, O Mon, Phong Dien and Binh Thuy) in Can Tho city, Vietnam from November 2016 to October 2018. In the study, 838 street food consumers and 598 vendors were involved in the survey. Besides, 677 street food vendors were observed to check their food handling practice.

Food safety knowledge and attitude surveys
The questionnaire used to assess the food safety knowledge and attitude of street food vendors and consumers were based on previous studies by Angelillo et al. (2001), Bolton et al. (2008), Ansari-Lari et al. Soares et al. (2012) and Samapundo et al. (2015). The questionnaire was translated into Vietnamese and pre-tested with 30 people to ensure its clear and properly structured content. All questions were standardized before the adoption of the final version for the survey. The questionnaire was classified into three main sections including demographic information, food safety knowledge, and food safety attitude.
The demographic section included gender, age, educational level, location, income, employment status, food safety training for both consumers and vendors and additionally the length of time spent in vending and types of vending by the vendors. The food safety knowledge section (including 18 questions) was designed to assess the awareness of the vendors and consumers to food poisoning pathogens, food hygiene, cross-contamination, personal hygiene, proper cleaning, etc. The food safety attitude questionnaire (including 16 questions) was organized to test how much consumers and vendors understand about food safety. The vendors operating around industrial zones, schools, universities, parks and busy street pavements in five districts were selected for the survey. The consumers in and around these areas which are frequented the most by local inhabitants of Can Tho city were asked to voluntarily participate in the survey. When the participants are not clear enough to fill in the questionnaire themselves, they were aided by the researchers during the survey.
These sections of knowledge and attitude were designed with three possible answers: "true", "false" and "do not know". Each correct answer considered as one point whilst no point was given for incorrect answers or when the respondent indicated that they did not know the answer. The score was then converted to 100 based on a maximum possible score of 18 and 16 points for knowledge and attitude, respectively. Scores less than 50 were considered as poor level, scores between 50 and 75 were considered as average (adequate) level whilst scores higher than 75 were considered as a good level of food safety knowledge as well as attitude.

Food handling observation checklist
A checklist was used to assess the food safety practice of street food vendors. The checklist was a combination of different checklists used in previous studies by Samapundo et al. (2015), Samapundo et al. (2016), Muinde and Kuria (2005), Chukuezi (2010) and some minor changes following the Law of food safety issued by Ministry of Health in 2012 (30/2012/TT-BYT) were made before adoption for the final version. The checklist contained five main sections: i) information on facilities ii) the environment around the stall iii) personal hygiene iv) food storage (cool/refrigeration storage at the vending sites) and v) utensils (the use of soap and clean water). The absence or presence of each component of the checklist was recorded. The selection of participants for this part of the study was based on the same methodology as for the selection of vendors for the food safety knowledge and attitude surveys.

Statistical analysis
The data obtained from the surveys and the checklists were computed by Microsoft Excel version 2013 (Microsoft, U.S.A.). The data set was then exported from Excel to SPSS Statistics version 20 (SPSS Inc., Chicago, U.S.A.) for further statistical analysis based on an α = 0.05. For comparisons of the scores, cut-off points of <50, 50-75 and >75 were used.

Demographic characteristics
The demographic characteristics of the 838 consumers who participated in this study showed that there was 535 female (63.8%) and 303 males (36.2%). The mean age of the participants was 26.4±10.35 years. The age of consumers ranged from 15 to 67 years old. With regards to the level of education attained, almost (84.1%) of the consumers had a high school or higher education. Half of the respondents were located at Ninh Kieu district (52.1%), the biggest district in Can Tho city. Most of the consumers were students (51.8%) and employed (42.6%). The income of consumers each month was below or equal to 3 million VND (66.2%), indicating that most of the consumers that participated in this study are students while only 6.9% of the consumers had income above 8 million VND. In addition, the majority of the consumers who participated in this study (77.9%) did not have any food safety training.
Of the 598 vendors who participated in this study, the majority of them were women (84.1%). This is a reflection that the street food stalls were mostly manned by women. In most South Asian cultures, women play a major role in food handling and cooking. Specifically, women in poor economics can get income by means of vending street foods. These findings are in agreement with other studies with women vendors of 72.5% in Ho Chi Minh City, Vietnam (Samapundo et al., 2016), 75.9% in Accra, Ghana (Kunadu et al., 2016) and 87.5% in Port-au-Prince, Haiti (Samapundo et al., 2016). The mean age of the vendors was 41.3±12.21 years with a wide age range of 16 to 83 years. Over 70% of vendors in this study had at least secondary education, this showed a better educational status among Vietnamese vendors compared with other findings reported in Haiti (Samapundo et al., 2015) where 65% had primary education while 25% had secondary education. 44.5% of the vendors have income below or equal 3, 48% have income between 3 and 8, and only 7.5% have income above 8 million VND per month. The majority (65.9%) of the vendors in this study did not have any food safety training. Other studies have reported even higher levels of untrained street food vendors including those in Ghana (Kunadu et al., 2016), in Brazil (Rebouças et al., 2017) and Haiti (Samapundo et al., 2015). Almost half (49.3%) of the vendors have below five years of experience in vending foods and 9.9% have above 20 years of experience. The main vending type was stationary (84.6%).

Food safety knowledge of consumers and vendors
The results of the survey which was carried out to determine the effect of sex, age, education, location, income, employment status, and food safety training on food safety knowledge of the consumers are shown in Table 1. The consumers had a mean food safety knowledge score of 63±16, indicating average knowledge of food safety. About 16.3% (137/838) of the consumers had an inadequate level of food safety knowledge (scores <50) whilst 24.2% (203/838) of the consumers had a good level of food safety knowledge (scores >75). No significant difference (p>0.05) was found between the food safety knowledge of female and male consumers. However, food safety knowledge was significantly different (p<0.05) with age groups of 15-20 years compared to others (>20 years). The food safety knowledge of consumers was dependent on education level (p<0.05). The food safety knowledge of consumers located in Ninh Kieu district was significantly higher than those in Cai Rang district (p<0.05) whereas those in both Ninh Kieu and Cai Rang did not differ significantly from the other districts (p>0.05). As mentioned above, most of the consumers were students and employed as well as half of the consumers were located in Ninh Kieu district; as a result, the students and employed consumers had a higher level of food safety knowledge than unemployed consumers (p<0.05). No significant difference was observed among income levels (p>0.05). In similarity to the trend observed for employment status, the consumers who were trained on food safety had a significantly higher score than those untrained (p<0.05), i.e. 69 and 61 respectively. In contrast to our findings, other studies have indicated that an increase in food safety knowledge occurs with age and that women have higher levels of food safety knowledge compared to men (Bruhn and Schutz, 1999). These findings were in good agreement with those reported by Samapundo et al. (2015) and Samapundo et al. (2016) where it was indicated that no statistical difference (p>0.05) was found in the food safety knowledge of consumers based on gender in Port-au-Prince, Haiti and Ho Chi Minh City, Vietnam. The results of the survey done to determine the effect of sex, age, education, location, income, food safety training, length of time spent vending, and types of vending on food safety knowledge of the vendors are shown in Table 2. Approximately 24.4% (146/598) of the vendors who participated in the study had a good level (score of >75), 57% (341/598) had an average level (scores between >50 and <75) and 18.6% (111/598) had an inadequate level of food safety knowledge (scores of <50). Amongst the studies that have used the same scoring methods, the mean food safety knowledge score of the street food vendors in Can Tho was found to be higher than that of food handlers from Ho Chi Minh City, Vietnam (Samapundo et al., 2016) and Haiti (Samapundo et al., 2015). There were no significant differences (p>0.05) in food safety knowledge on the basis of gender, age, educational level attained, income, and length of time spent vending. The same findings have also been observed in Brazil (Auad et al., 2019) and Haiti (Samapundo et al., 2015) where no statistical difference was found between the food safety knowledge of the vendors based on gender, age, and the educational level attained. In this study, significant differences occurred between the food safety knowledge levels of vendors who had received food safety training and those who had not (p<0.05), with the untrained vendors attaining lower scores (60±16) compared to trained vendors (67±15). It indicated that the training of vendors may have a positive influence on their food safety knowledge. The food safety knowledge of food vendors also had significant difference (p<0.05) among studied districts, the lowest score was seen in Cai Rang and Binh Thuy districts where there are no universities or colleges, more frequent vendors may be of low income and low FULL PAPER  (2010) and Odonkor et al. (2011) studies conducted on street foods in Nigeria and Ghana, respectively.

Food safety attitude of consumers and vendors of street food
The results of the survey to determine the food safety attitude of the consumers are shown in Table 3. The consumers had a mean food safety attitude score of 72, which indicated that they generally had an average awareness of food safety. The range of scores was between 19 and 100. With regards to the distribution of scores, it can be seen in Table 3 that 60.3% (505/838) of the consumers had an average food safety attitude and 34.5% (289/838) had a very good food safety attitude. No statistical difference (p>0.05) was found between the food safety attitude of the consumers based on gender and income. However, statistically significant differences (p<0.05) occurred between the food safety attitude of consumers based on their age, educational level, location, employment status, and food safety training.  Consumers who only had higher education were determined to have higher scores than those who went to high and secondary school (p<0.05) however no statistical difference was observed about their food safety attitude when compared to those who had primary school or illiterate (p>0.05). A higher educational level would have a positive effect on food safety attitude (Samapundo et al., 2016;Ma et al., 2019). In contrast, cooking and household tasks would gain the experiences and attitude for consumers who are less educated. However, the educational level has been determined to have no bearing on the food safety knowledge and attitude of consumers in Handan, China (Ma et al.,

2019).
The results of the survey to determine the food safety attitude of the vendors are shown in Table 4. The mean food safety attitude score of the vendors was 73 and the scores ranged from 25 to 100. Only 2.2% of the vendors had an inadequate food safety attitude (scores <50); whilst 97.8% had at least an average food safety attitude (scores of ≥50). Even a lower proportion of the vendors (32.3%) had good food safety attitude compared to consumers (34.5%). No statistical difference (p>0.05) was found in the food safety attitude of the vendors based on gender, age, level of education, income, length of time spent and types of vending. These findings FULL PAPER   (Baş et al., 2006).
It can be seen the detail responses of the consumers and vendors to the food safety knowledge questionnaire in Table 5 and 6, respectively. The majority of the consumers did not know that Salmonella spp. (55.7%), hepatitis A virus (63.8%) and Staphylococcus aureus (63.6%) are responsible for foodborne diseases. As shown in Table 5, almost half (49.7%) of the consumers failed to identify the demographic groups at the greatest risk of foodborne diseases. In contrast, 54.9% of the consumers knew that bloody diarrhea can be transmitted by food. 81.9% of the consumers recognized that AIDS cannot be transmitted by food. 92.2% knew that it is necessary to take leave from work if they had infectious skin diseases and 82.6% knew that microorganisms can be found in the skin, mouth, and nose of healthy handlers. The majority of the consumers were also aware of the critical role of washing hands (97.1%) and proper cleaning of utensils (97.7%) with regards to the prevention of foodborne diseases. 95.4% knew that the FULL PAPER  use of gloves is important in reducing the risk of contamination. More than half (53.5%) of the consumers were aware that eating and drinking in the workplace lead to an increase in the risk of foodborne disease.
In comparison to the consumers, the majority of the vendors (Table 6) did not know that hepatitis A (63.2%), Salmonella (65.4%) and S. aureus (70.2%) were foodborne pathogens. In agreement with this finding, vendors in Brazil and Haiti also failed to identify these foodborne pathogens (Soares et al., 2012;Samapundo et al., 2015). However, most of the vendors (57.2%) knew that bloody diarrhea can be transmitted through food. Compared to the consumers, a higher proportion of the vendors (77.1%) knew that AIDS could not be transmitted by food. 88.1% of the vendors also knew that it is necessary to take leave from work during an infectious disease of the skin. Vendors (89%) knew that microbes could be found on the skin, in the nose, and mouth of healthy handlers while 57.2% of the vendors did not know that abortion could be induced by foodborne diseases. About 81.8% wrongly believed that washing utensils with detergent would leave them free of contamination. Most of the vendors were aware of the importance of washing hands (97.8%), proper cleaning (97.8%) and the use of gloves (94.3%) in the prevention of foodborne diseases. About 55.7% of the vendors were not able to identify children, pregnant women and the elderly as being at equal risk of foodborne diseases. The majority of the vendors (68.1%) responded correctly that swollen cans may be a possible host of microorganisms. Vendors (57.9%) knew that food prepared in advance and stored at room temperature after four hours increases the risk of foodborne disease. Slightly more vendors than consumers knew that the reheating of cooked foods can contribute to food poisoning (76.6% vs. 67%) and slightly more consumers than vendors knew that the preparation of food in advance could lead to food poisoning (67% vs. 57.9%).
The food safety attitude response of the consumers and vendors are shown in Table 7 and 8, respectively.
Half of the consumers (62.5%) wrongly believed that a well-cooked food is free of microbial hazards and that the ideal place to store raw meat was the bottom shelf of the refrigerator (61.9%). Most of the consumers (83.7%) thought that eggs should be washed as soon as possible after purchase and 62.8% thought that chicken should not be thawed in a bowl of cold water. About 50.9% of consumers incorrectly understand that defrosted foods can be refrozen. Almost all of them (94.6%) were able to identify wearing masks as an important practice to reduce the risk of food contamination. The low percentage of correct answers (36.8%) was observed for the question concerning whether or not well-cooked food is free of microbial hazards. In agreement with the findings of other studies conducted in Ghana, South Africa and Haiti (Lues et al., 2006;Donkor et al., 2009;Samapundo et al., 2015), most of the street food vendors (94.5%) who participated in this study were aware of the importance of separating cooked and raw foods to prevent contamination. Figure 1 shows the characteristics of the observed stalls. Based on the observation of facilities, 72.7% of vendors prepared food on-site. The finding was similar to previous studies conducted in Ho Chi Minh City, Vietnam (Samapundo et al., 2016) and in China (Ma et al., 2019) (Mensah et al., 2002), Uganda (Muyanja et al., 2011) and Kenya (Muinde and Kuria, 2005), which found that potable water was not readily available at street vending sites. Therefore, vending sites can contribute to street food contamination with microorganisms (Tambekar et al., 2011). 50.1% of the stalls were not protected from the sun, dust, and wind. Dust potentially carries pathogens and therefore it may become a vector for their transmission to prepared foods. These findings are in agreement with the observations studied in Uganda, Sudan, Nigeria, and Kenya ( Abdalla et al., 2009;Chukuezi, 2010;Muyanja et al., 2011). About 79% (535/677) of the stalls had a clean environment i.e. far from rubbish, wastewater, toilet facilities, and open drains. In contrast, Muinde and Kuria (2005) reported that only 15% of street food had a clean environment around the stall in Kenya. According to the FAO, the place of food preparation should be kept clean at all times and should be far from any source of contamination such as rubbish, wastewater, dust, and animals (FAO, 1995).

Observed food handling habits
In addition to facilities and environment around the stalls, the personal hygiene of vendors will have a positive impact on the safety of street food. Results showed that only 12.4% of the vendors wore an apron when handling, preparing and serving of food and 64% of their hair was covered when handling, preparing, and serving of food. Other studies in various developing countries have also observed that a low percentage of street food vendors use aprons and gloves while handling, preparing, and serving food (Subratty et al., 2004;Muinde and Kuria, 2005;Lues et al., 2006;Chukuezi, 2010;Samapundo et al., 2015). Omemu and Aderoju (2008) also reported varying cleaning practice among food vendors, with the majority of them usually ignoring personal hygiene.
In agreement to the observations that have been made in Kenya (Muinde and Kuria, 2005), Nigeria (Omemu and Aderoju, 2008;Chukuezi, 2010) and Haiti (Samapundo et al., 2015), 84.5% of the vendors in Can Tho handled food with bare hands and 92.3% handled money while serving the food; however, only 6.7% washed their hands thereafter. Accordingly to Khairuzzaman et al. (2014), foods and ingredients may be subjected to cross-contamination from unwashed hands. It has also been reported by European Commission (1997) that food handlers should avoid handling food with bare hands and handling money at the same time. Additionally, according to the WHO, food should be preferably handled with clean tongs, forks, spoons or disposable gloves (FAO/WHO, 1999). In this study, 30.1% of the operators were breathing or coughing in their hands, picking their noses with their hands and continued working without washing their hands. 39% of the food operators wore jewelry and only 13.4% covered the jewelry adequately. This action is considered as a source of contamination because food handlers should have, good appearance, healthy body, clean clothes, hygienic hands, short nails, no jewelry and without beard or mustache (Trafialek et al., 2018).
It was observed that cooked food products were not kept cool (61.3%). This is an important finding as inadequate cooling is one of the key factors that contribute to the occurrence of food poisoning outbreaks.
It has been reported that storing raw materials at ambient temperature (29-32 o C) encouraged the growth of foodborne pathogens which may pose risks to consumers (Badrie et al., 2004). In the stalls, the utensils were washed clean in soapy water (69.7%) and covered (69.7%). This practice of washing the dishes seems to be common in other countries such as the Philippines and South Africa (Azanza et al., 2000;Lues et al., 2006).

Conclusion
This study aims to improve the food safety status of street food through the evaluation of food safety knowledge and attitude of both vendors and consumers to street foods, and the practice of food vendors in Can Tho city, Vietnam. Results indicated that consumers and vendors generally exhibited an average food safety knowledge and attitude. Although Vietnam's street food regulations were adopted in 2012, the present study shows this had not yet transformed the practice of food vendors. Therefore, there is an urgent need to organize formal training in food hygiene and food safety. To control the risk of foodborne disease or foodborne outbreak, an effort should be made to improve the environmental conditions and infrastructure i.e. providing clean protected structures, potable water and waste disposal facilities. Also, the high frequency of checklist about the hygienic practice of street food vendors can be planned in order to strengthen the safety of street food.