Abstract.
Poor hand hygiene and food handling put consumers of restaurant and street food at risk of enteric disease, especially in low-income countries. This study aimed to collect hygiene indicators from a nationally representative sample of restaurants and street food vendors. The field team collected data from 50 rural villages and 50 urban administrative units (mahallas). We explored restaurant service staff, cook, and food vendor hygiene practices (N = 300 restaurants and 600 street food vendors), by observing hygiene facilities, food handling, and utensil cleaning. A qualitative assessment explored perceptions of hygiene related to food handling. During restaurant spot checks, 91% (273/300) had soap and water at handwashing location for customers but in only 33% (100) at locations convenient for restaurant staff. Among street food–vending stalls, 11% (68/600) had soap and water when observed. During 90-minute structured observations, cooks used soap to wash hands on 14/514 (3%) of occasions before food preparation, 6/82 (8%) occasions after cutting fish/meat/vegetables, 3/71 (4%) occasions before serving food, and 0/49 (0%) occasions) before hand-mashing food/salad preparation; no street food vendors washed hands with soap during these food-handling events. Most of the qualitative study participants perceived that customers select a vendor based on tastiness of the food, whereas no one mentioned the importance of food hygiene. The study demonstrates widespread poor hygiene and food-handling practices in restaurants and among food vendors. Based on our study findings, we proposed a food premises Hygiene Investigation Model to create action plans to improve food safety.
INTRODUCTION
Low-income countries, including Bangladesh, facing rapid urbanization and industrialization, are undergoing changing food consumption patterns where restaurants and street food vendors play an important role in meeting food demands. In a study conducted in 1997–1999 in Nairobi found that middle- and low-income households depend a lot on street food vendors (Mwangi et al., unpublished doctoral thesis) besides. Food vending also provides a livelihood for many workers.
Millions of people globally develop illness everyday as a result of consuming microbiologically and chemically contaminated food and water.1,2 Food safety has been declared a global and increasing public health concern by international agencies such as Food and Agriculture Organization (FAO) and WHO.1 Poor personal and environmental hygiene contribute to food contamination and result in food-borne diseases.3–5 A study conducted in 2008 in central Bangladesh found that most (68%) of the vending shops were situated on the footpath, 30% near a drain and 18% near a sewerage.6
Microbial contamination is an important health hazard associated with street foods.7 Outbreaks of food-borne disease have been linked to poor hygiene in restaurants,8 and eating food from street vendors.9 Street foods created 691 events of food poisoning outbreaks and 49 deaths from 1983 to 1992 in Shangdong Province, China.10 In Bangladesh, preparing, serving and eating foods with bare hands are common.6,11,12
Availability of water and soap at the handwashing place can improve hygiene practices,13 but restaurants’ food handlers from low-income/developing countries often have little knowledge on hand hygiene and food safety.14,15 Although some street food vendors from developing countries reported that their food preparation and selling conditions are not hygienic,16,17 we are aware of little data on food hygiene practices or facilities and perceptions of street food vendors and restaurants and their staff. Without a clear description of practices, it is difficult to identify approaches to address these risks.
This study aimed to describe a range of hygiene indicators from a nationally representative sample of restaurants and street food vendors to advocate for intervention programs, with the potential application for other low-income countries, with similar contexts. To do this, we present data collected through multiple methods, that is,. observations, spot checks, surveys to describe representative facilities and practices, and qualitative in-depth interviews guided by the Integrated Behavioral Model for Water, Sanitation and Hygiene (IBM-WASH) theoretical framework to explore respondent perceptions.
METHODS
Definitions.
We defined a restaurant as a business establishment with a permanent structure, where staff prepare and serve food and are open for at least two meal services a day. We defined a street food vendor as one sells food from a temporary static structure or mobile stall in the street/open public location.
We used the WHO/UNICEF Joint Monitoring Programme definitions of an improved toilet,18 one that hygienically separates human excreta from human contact, and improved water source.
Study population and sampling.
We conducted a national hygiene assessment to measure handwashing coverage and related water and sanitation infrastructure among households, schools, hospitals, restaurants and street food vendors from all over Bangladesh. The data were collected from January to October 2013 (http://www.wateraid.org/∼/media/ Publications/bnhbs.pdf). This article reports data on restaurants, their staff, and street food vendors.
Data collection.
In this multi-methods study, we conducted a quantitative survey to assess restaurant and street food vendor hygiene facilities and practices, and in addition conducted a qualitative assessment to explore restaurant staff and street food vendors’ perceptions and barriers to maintain hygiene. Our qualitative data collection was guided by the IBM-WASH theoretical framework, which integrates the contextual, psychosocial and technological factors, influencing water, sanitation, and hygiene behavior.19 Contextual factors are related to the individual, setting, and/or environment that can influence behavior change and adoption of new technologies. The psychosocial factors influence behavioral outcomes and technology adoption, which comprise behavioral, social, or psychological determinants. The technological factors attributes of a technology, product, or device that influence its adoption and sustained use. The framework provides a systematic guideline to design, development, implement, and evaluate WASH interventions.
To ensure uniform understanding among all data collectors, we provided in-house training, including role-play exercises, and pretesting in the field with the data collection instruments for seven days in total.
Quantitative data collections were led by an epidemiologist, and qualitative data collection were led by a sociologist. Field research assistants (N = 30) with considerable experience using WASH instruments (questionnaires, spot check surveys, and structured observations) collected quantitative data; all had at least a bachelor’s degree. Field research assistants were directly supervised by research officers (N = 5), who had greater experience on WASH and survey data collection and had at least a post graduate qualification. Four anthropologists collected the qualitative data. After each day of data collection, the data were cross-checked during daily team meetings to discuss data quality issues and field-related challenges, if any.
Quantitative data collection.
We used quantitative methods to determine representative facilities and practices. As part of the Bangladesh National Hygiene Baseline Survey, where we aimed to measure the national coverage of handwashing with soap in various venues, we included street food vendors and restaurants. To calculate sample size, we used the indicator “street food vendor served food with bare hands” from a recent survey which detected this practice among 43% of food vendors.6 Sample sizes were calculated to have sufficient power to compare rural and urban areas; thus, we assumed a 10% difference in this indicator between rural and urban areas, with 80% power, an α of 0.05 with a design effect of 12 and estimated that 864 premises would be required. Therefore, we aimed to sample 875, where approximately one-third would be restaurants.
To select 50 rural villages and 50 urban administrative units (mahallas) as clusters, the quantitative team used probability proportional to size (PPS) sampling. We used the National Population and Housing Census 2011 data to select rural clusters, and the 2006 Urban Health Survey data to select urban clusters.20,21
When conducting the household survey for the broader National Hygiene Survey, the quantitative field team asked residents of each enrolled household for the most popular restaurants in their locality selling readymade foods. The field team enrolled three most commonly mentioned restaurants from each of the 100 clusters. From each of these 300 restaurants, we collected data from one service staff and one cook. The survey team also listed the six most popular food vendors from each of 100 clusters, thus 600 street food vendors in total.
The research team identified different categories of food vendors through a brain-storming session with 80 research assistants who have experience conducting surveys in both urban and rural sites all over Bangladesh and prepared a list of street food vendor categories. We listed street food vendors according to those that were likely to have hand contact during food preparation and serving and according to popularity. The research team divided those vendors into six categories based on their food preparation, serving, and storage methods and enrolled them in the study accordingly. Those six categories were tea-bakery snacks vendors, chotpoti-fried snacks vendors, beverage-butter milk vendors, precut fruits-pickles vendors, pitha-flattened bread vendors, and mixed puffed rice-Bombay mix vendors (further details of terminology and food types included in Supplemental Appendix).
The quantitative field team collected data in the following sequence: first, they conducted structured observations (for 90 minutes during the rush time; either in the morning or evening) and recorded restaurant staff and street food vendor hygiene practices related to hands, foods, and utensils. Subsequent to structured observations, the quantitative team conducted spot checks to observe water-, sanitation-, and hygiene-related facilities and water/food storage methods as part of the survey. After the observation work was completed, field staff observed hand cleanliness and conducted interviews using standard questionnaires. During hand cleanliness observation, the field team conducted direct inspection of palm and finger pads of both hands and coded them as unclean if any visible dirt was seen.
Qualitative study.
We used qualitative approaches to explore perceptions of hygiene related to food handling among restaurant service staff, cooks, and street food vendors among participants who were not enrolled in the quantitative survey data collection. The field team purposively selected 16 (eight urban and eight rural) clusters from the 100 surveyed clusters. To ensure geographical variation from urban sites, we purposively selected seven metropolitan clusters from seven divisions (largest administrative unit). From the underserved Chittagong Hill Tract clusters, we randomly selected one district town (Khagrachori) as a further cluster.
For rural sites, we first divided rural clusters into seven administrative divisions and randomly selected one cluster from each of seven divisions. There are some river bank (char) areas in Bangladesh, which differ from other rural areas in terms of socioeconomic, and water- and sanitation-related infrastructural conditions.22,23 From the river bank (char) clusters, we randomly selected one district (Bhola) as a further cluster.
Anthropologists conducted 32 in-depth interviews from both urban and rural sites (64 in total), which we assumed would be sufficient to reach data saturation according to qualitative data collection methods and the team’s experience.24 For the in-depth interviews, we used a guideline that included a range of topics covering our study objectives; these included hand hygiene, water, and sanitation facilities; current practices of hand, water, and food hygiene; perceptions related to hygiene; and barriers to maintaining hygiene. In addition, we collected respondent sociodemographic characteristics.
Among selected restaurants, anthropologists conducted two in-depth interviews (one with a cook and one with a restaurant manager) in each of 16 clusters for 32 interviews in total. Among street food vendors, anthropologists conducted two in-depth interviews in each of 16 clusters for 32 interviews in total. During qualitative data collection, among vendors from each of six food categories, we conducted at least five interviews. In addition, we enrolled two further vendors for interviews from category 2 (Chotpoti-fried snacks vendors), as we found a greater number from that category, and these were considered the most popular.
Data analysis.
For quantitative data analysis, we calculated frequencies to describe food vendor and restaurant characteristics, handwashing and sanitation facilities for staff and customers, and current hygiene practice. We reported frequencies with 95% CI for categorical variables, and means and SDs for continuous variables that were normally distributed. Because we used PPS to select clusters from among 86,925 rural and 10,552 urban clusters, we used a sampling weight to calculate national estimates adjusting for rural/urban balance, calculated as f = 1/F, where F is the total number of population clusters. All percentages and means reported are weighted national estimates. We used generalized estimating equations25 to compare WASH facilities and related practices for restaurants versus food vendors, while accounting for the clustered design.
The in-depth interviews lasted 70 to 90 minutes and were recorded using a digital audio recorder. The audio recordings were transcribed verbatim in Bengali. Based on thematic content, anthropologists coded the transcripts, prepared a summary of codes for each interview, and identified suitable quotes. Summary codes were translated into English and individual summaries were compiled into a master summary report, followed by thematic analysis.
We analyzed these data using the IBM-WASH framework and arranged the study findings according to framework dimensions and levels. Where applicable, we analyzed data according to the contextual, psychosocial, and technological dimensions at the community, behavioral, interpersonal, individual, and food premises levels, related to behavior in infrastructure-restricted settings.
Developing a restaurant and food vendor hygiene investigation model.
During data analysis, we considered that the observed and surveyed hygiene facilities and related practices could guide a practical framework for use during premises inspections. Accordingly, we have developed a “Restaurant and Food Vendor Hygiene Investigation Model,” based on Sustainable Development Goal (SDG) #6: Ensure availability and sustainable management of water and sanitation for all.26 To date, indicators for the goal have been developed for “universal settings” (households, schools, health facilities, workplaces, and public spaces)27 and separately for health-care facilities,28 but not specifically for restaurants or vendors. Our model includes a framework of observed premises facilities and recommended actions for restaurant owners/managers and food vendors to improve food hygiene (Figure 1).
Figure 1.
Restaurant and food vendor hygiene investigation model: Mapping factors affecting hygiene practices in restaurants and food vendors: steps needed to address Sustainable Development Goals (SDG) #6.
Ethical considerations.
We explained the research study objectives clearly to survey (including structured observation) and interview participants. Before taking part in the study, these participants provided written informed consent, which was documented by a signature. The protocol for this study was reviewed and approved by the icddr,b Institutional Review Board.
RESULTS
We present our study findings according to the themes: sociodemographic characteristics; hand hygiene facilities and related current practices; hand hygiene perceptions; reported barriers to maintain hygiene; and other facilities and water/food hygiene practices. The integrated presentation helped to focus on the overall study findings from two different yet related population groups; staff from restaurants and street food vendors.
Sociodemographic characteristics.
In total, this study enrolled and collected data from 300 restaurants and 600 street food vendors from 53 of the 64 districts in Bangladesh. Three-quarters of the restaurant managers were also the restaurant owners. Almost all (97%, 581/600) of the food vendors and restaurant managers (99%, 298/300) and the majority (81%, 242/300) of restaurant cooks were male. We found that 13% (41/300) of restaurant managers’ and 52% (311/600) of food vendors had no formal education (Table 1). Of our 64 qualitative study respondents, the majority (59) interviewed in this study were male. Among the female respondents interviewed, all were involved with food preparation in the restaurants and street food vending businesses, and none of them were involved in restaurant ownership or management (Table 2).
Table 1.
Respondent demographic information and business characteristics
| Indicators | Restaurant (N = 300) | Food vendor (N = 600) | ||
|---|---|---|---|---|
| n (%*) | 95% CI or interquartile range | n (%*) | 95% CI | |
| Gender of respondents (male) | ||||
| Owner/manager | 298 (99) | (98, 100) | 581 (97) | (95, 98) |
| Customer service staff | 295 (98) | (97, 100) | – | – |
| Cook | 242 (81) | (76, 86) | – | – |
| Median age of respondents | ||||
| Owner/manager | 40† | (19) | 35† | (35, 38) |
| Customer service staff | 28† | (18) | – | – |
| Cook | 35† | (18) | – | – |
| Education of respondents—managers | ||||
| No formal education | 41 (13) | (9, 18) | 311 (52) | (48, 56) |
| Median years of formal education | 259 (5)† | (5, 6) | – | – |
| Relation of manager respondents with the business owners | ||||
| Self | 225 (75) | (68, 80) | 574 (96) | (94, 97) |
| Son/daughter/spouse of owner | 26 (9) | (5, 12) | 8 (1) | (0, 2) |
| Relative of owner | 12 (4) | (2, 7) | – | – |
| Others—salaried | 37 ( 22) | (8, 18) | 18 (3) | (1, 5) |
| Relation of service staff respondents with the business owners | ||||
| Self | 54 (18) | (13, 22) | – | – |
| Son/daughter/spouse of owner | 26 (9) | (5, 12) | – | – |
| Relative of owner | 21 (7) | (4, 9) | – | – |
| Others—salaried | 199 (66) | (61, 73) | – | – |
| Relation of cook respondents with the business owners | ||||
| Self | 63 (21) | (15, 26) | – | – |
| Son/daughter/spouse of owner | 34 (11) | (7, 15) | – | – |
| Relative of owner | 17 (6) | (3, 8) | – | – |
| Others—salaried | 186 (62) | (56, 70) | – | – |
| Nature of area/location‡ | ||||
| Bazaar | 164 (53) | (46, 61) | 180 (29) | (23, 34) |
| Street gathering location | 85 (30) | (23, 37) | 161 (28) | (23, 33) |
| Bus station | 51 (17) | (11, 23) | 80 (13) | (9, 18) |
| Near school | – | – | 166 (28) | (23, 33) |
| Business season (food vendors) | ||||
| Seasonal | – | – | 91 (15) | (12, 18) |
| Year round | – | – | 509 (85) | (82, 88) |
| Business mobility (food vendors) | ||||
| Semiambulant/mobile locations | – | – | 255 (42) | (38, 47) |
| Stipulated location | – | – | 345 (58) | (53, 62) |
| Hours open each day (median) | 16† | (15, 16) | 8 | (8, 9) |
| Mean number of customers per day | 221 | (193, 249) | – | – |
| Mean customer number that could be accommodated at one time | 25 | (23, 27) | – | – |
| Ownership of restaurant building | ||||
| Self | 59 (19) | (14, 25) | – | – |
| Rented | 241 (81) | (75, 86) | – | – |
| Materials of restaurant building | ||||
| Roof—tin | 243 (80) | (74, 86) | – | – |
| Roof—concrete | 51 (18) | (12, 23) | – | – |
| Floor—concrete | 236 (80) | (74, 85) | – | – |
| Floor—katcha (not concrete) | 64 (20) | (15, 26) | – | – |
| Wall—tin | 103 (33) | (27, 40) | – | – |
| Wall—cement | 181 (61) | (55, 68) | – | – |
| Wall—straw/tarpaulin/wood | 15 (5) | (2, 8) | – | – |
* Weighted.
† Not weighted.
‡ Nature of area/location was a single answer and spot-checked for close proximity; question was not asked.
Table 2.
Sociodemographic characteristics of qualitative respondents
| Characteristic | Food preparers, N = 16 | Restaurant owners/managers, N = 16 | Street food vendors, N = 32 | All, N = 64 |
|---|---|---|---|---|
| n | n | n | n (%) | |
| Gender | ||||
| Male | 14 | 16 | 29 | 59 (92) |
| Age (years) | ||||
| 20–30 | 4 | 4 | 10 | 18 (28) |
| 30–40 | 3 | 4 | 7 | 14 (22) |
| 40–50 | 8 | 6 | 9 | 23 (36) |
| Older than 50 | 1 | 2 | 6 | 9 (14) |
| Income from food vending–associated profession: Taka (US $) | ||||
| 0–5,000 (0–61) | 3 | 2 | 9 | 14 (22) |
| 5,000–10,000 (61–122) | 12 | 4 | 21 | 37 (58) |
| 10,000–15,000 (122–183) | 1 | 5 | 2 | 8 (13) |
| 15,000–20,000 (183–244) | 0 | 0 | 0 | 0 (0) |
| 20,000–25,000 (244–305) | 0 | 2 | 0 | 2 (3) |
| Above 25,000 (305) | 0 | 3 | 0 | 3 (5) |
| Education | ||||
| Had no formal education | 7 | 5 | 21 | 33 (52) |
| Primary education | 6 | 6 | 4 | 16 (25) |
| Secondary education | 3 | 4 | 6 | 13 (20) |
| Above secondary | 0 | 1 | 1 | 2 (3) |
Note: US$1 = Taka. 81.96 (November 2013).
Hand hygiene facilities and related current practices.
Almost all (298/300, 99%) restaurants had a handwashing location for customers inside the restaurants with water present, whereas 91% (273/300) had both soap and water present at the handwashing location. However, only 42% (124/300) of restaurants had water present at the food preparation area, whereas one-third (100/300, 30%) had both soap and water at the food preparation area (Table 3).
Table 3.
Availability of soap and hand cleanliness (spot check)
| Indicators | Restaurant (N = 300) | Food vendor (N = 600) | ||
|---|---|---|---|---|
| n (%*) | 95% CI | n (%*) | 95% CI | |
| Handwashing agents available in food preparation area | ||||
| Water only | 124 (42) | (36, 48) | – | – |
| Water and soap | 100 (34) | (28, 40) | – | – |
| No specific food preparation area | 172 (58) | (51, 64) | – | – |
| Handwashing location† for customers | ||||
| Water only | 298 (99) | (98, 100) | 190 (32) | (27, 36) |
| Water and soap | 273 (91) | (88, 95) | 68 (11) | (8, 14) |
| Respondents’ hands appeared clean‡ | ||||
| Service staff/food vendors | 156 (52) | (46, 59) | 205 (34) | (30, 39) |
| Cooks | 106 (35) | (29, 41) | – | – |
* Weighted.
† Specific place for handwashing for customers, such as tube well, basin, tap, drum with tap, and bucket/piped/tank/container and mug together.
‡ No visible dirt over palms, finger pads, and over/under finger nails; question was not asked.
At the time of spot checks, about one-third of the 600 street food–vending stalls (190/600, 32%) had a handwashing location with water only, whereas only 11% (68/600) had soap and stored water (Table 3). Street food–vending stalls with a stipulated location were more likely to have water at handwashing stations (46%, 159/345) than ambulant street food vendors (12%, 31/245: P < 0.001).
During structured observations, handwashing with soap among restaurant cooks was low; 14/514 (3%) of occasions before food preparation, 6/82 (8%) occasions after cutting fish/meat/vegetables, 3/71 (4%) occasions before serving food, and 0/49 (0%) occasions before using hands for mashing food/salad preparation. No street food vendors were observed to wash hands with soap during these food-handling events (Table 4).
Table 4.
Observed handwashing behavior
| Observed to wash hands with soap during related events | Restaurant | Food vendor | ||||
|---|---|---|---|---|---|---|
| Staff | Cook | |||||
| n/N (%*) | 95% CI | n/N (%*) | 95% CI | n/N (%*) | 95% CI | |
| Before eating | 51/243 (21) | (14, 29) | 8/55 (14) | (5, 24) | 1/241 (0) | (0, 2) |
| Before food preparation | 6/151 (4) | (2, 8) | 14/514 (3) | (1, 4) | 1/3,691 (0) | (0, 0.2) |
| Before serving food | 25/2,335 (1) | (0, 2) | 3/71 (4) | (1, 12) | 4/3,678 (0) | (0, 0.2) |
| Before mashing food/salad preparation | 2/142 (1) | (0, 5) | 0/49 (0) | (0, 0) | 0/1,096 (0) | (0, 0.3) |
| After cleaning human/animal feces | 1/4 (25) | (1, 81) | – | – | 0/29 (0) | (0, 12) |
| After cleaning/removing wastage/leftover | 17/82 (20) | (11, 30) | 4/36 (12) | (0, 23) | 2/27 (7) | (1, 24) |
| After cutting fish/meat/raw vegetables | 2/28 (7) | (1, 24) | 6/82 (8) | (2, 14) | 0/26 (0) | (0, 13) |
| After cleaning utensils | 78/1,309 (6) | (4, 75) | 9/63 (14) | (5, 24) | 4/574 (1) | (0, 2) |
| After cleaning bench, table, chair, and floor | 38/855 (4) | (3, 6) | 1/38 (3) | (0, 14) | 3/264 (1) | (0, 3) |
| After cleaning cough/sneezing/nose/eyes/mouth inside | 0/64 (0) | – | 1/32 (3) | (0, 16) | 0/104 (0) | (0, 4) |
| After defecation/cleaning a child after defecation | – | – | 1/1 (100) | (2, 100) | 1/5 (20) | (1, 72) |
* Weighted; event was not observed.
During cleanliness checks, a little more than half (52%, 156/300) of the restaurant service staff member's hands appeared clean. Few staff who were directly handling food had clean hands; 35% (106/300) of the restaurant cooks’ and 34% (205/600) of the food vendors’ hands appeared clean during observation.
Hand hygiene perceptions.
In the qualitative study, the majority (45 of 64) of the respondents perceived that customers select a vendor based on tastiness of the food, whereas no one mentioned the need to maintain hygiene during food handling. A mixed puffed rice seller from a rural area said:
Nobody can make foods as tasty as me. That’s why customers don’t buy foods from other vendors if I am present over there.
Most of the qualitative study respondents (51 of 64) reported that during food preparation and serving, they did not touch anything dirty that would contaminate their hands. A small proportion (9 of 64) of the respondents considered “hand to water contact” while washing vegetables and utensils as reasonable handwashing.
Reported barriers to maintain hygiene.
In the qualitative study, some street food vendors (five of 32) reported that carrying soap and water is not possible during food service periods as they frequently move location. Therefore, washing hands with soap is not possible when selling food. A street food vendor from a rural area stated:
We always move from one place to another, so carrying water and soap is a hassle for me, therefore I can’t wash hands with soap during vending.
Some restaurant managers and food preparers (n = 8) stated that fast-paced work because of high volume of customer-directed activities was a barrier to wash hands during food handling. Restaurant managers reported that customers often do not have patience to wait after placing orders. That is why during rush hours, restaurant staff cannot wash hands with soap frequently. A restaurant service staff from an urban area said:
We can’t wash hands with soap as we have to serve customers so frequently. If we can’t serve quickly then customers get angry. Customers should have patience on this issue.
A few restaurant managers and street food vendors (n = 4) reported that soap is not affordable for them because of insufficient income. A restaurant manager from an urban area said:
We people are poor. Soap is sometimes available in my restaurant and sometimes not. When there is no soap we wash our hands with water only.
Other facilities and water/food hygiene practices.
Only 12% (35/300) of restaurants had improved toilets within their premises, whereas 82% (247/300) had no facilities (Table 5). More than two-thirds of restaurants (70%, 210/300) and food vendors (62%; 375/600) used tube wells as a source of drinking water. However, tap water supplied by the Water Supply and Sewerage Authority was the most common drinking water source in urban restaurants (37%, 56/150). Twenty-four percent of the food vendors had no water source for drinking or did not offer drinking water to customers (Table 5). Drinking water was stored for both customers and staff by 50% (150/300) of restaurants and 55% (332/600) of street food vendors, but in covered containers which appeared to be clean among only 7% (22/300) of restaurants and 14% (83/600) of vendors (Table 5).
Table 5.
Water and sanitation facilities, and related hygiene practices
| Indicators | Restaurant (N = 300) | Food vendor (N = 600) | ||
|---|---|---|---|---|
| n (%*) | 95% CI | n (%*) | 95% CI | |
| Toilet within premises | ||||
| Improved toilets† | 35 (12) | (8, 16) | – | – |
| Unimproved toilets | 18 (6) | (3, 8) | – | – |
| No facilities | 247 (82) | (77, 87) | – | – |
| Source of drinking water by category | ||||
| Shallow tube well‡ | 141 (46) | (37, 54) | 244 (40) | (33, 47) |
| Deep tube well/Tara pump | 69 (23) | (16, 30) | 128 (20) | (16, 26) |
| Tap water inside restaurant/food-vending structure | 70 (25) | (18, 32) | 40 (7) | (4.0, 10) |
| Tap water outside restaurant/food-vending structure | 1 (0) | (0.0, 2) | 31 (5) | (2.5, 8.4) |
| Filter (ceramic/other filter, which is refilled by a plastic jar; considered as “not improved” according to JMP definition) | 19 (7) | (3, 11) | 11 (2) | (0, 4) |
| No water source/not applicable | – | – | 145 (24) | (20, 30) |
| Stored drinking water in a container | 150 (50) | (43, 57) | 332 (56) | (51, 60) |
| Stored drinking water in a covered and clean§ container | 22 (7) | (4, 11) | 83 (14) | (11, 17) |
| Stored reused water for cleaning utensils | ||||
| Poured water on the utensils | 44 (15) | (11, 19) | 46 (8) | (5, 10) |
| Dipped utensils inside the stored water | 122 (40) | (33, 48) | 258 (44) | (39, 48) |
JMP = Joint Monitoring Programme.
* Weighted.
† Improved toilet according to JMP: flush or pour-flush to piped sewer system, septic tank, pit toilet, ventilated improved pit (VIP) toilet, pit toilet with slab, and composting toilet.
‡ Less than 250 feet deep.
§ No visible staining with black, green, or yellow spots inside the container.
During structured observations, 40% (122/300) of restaurants and 44% (258/600) of food vendors dipped utensils into stored reused water for cleaning (Table 5, Figure 2). At the time of spot checks, depending on food type, between 3% and 25% of foods in restaurants, and between 5% and 45% of foods served by street vendors were kept covered (Table 6).
Figure 2.
Cleaning utensils in stored reused water. This figure appears in color at www.ajtmh.org.
Table 6.
Food items observed stored in a covered and clean pot/container for sale (spot check)
| Indicators | Restaurant | Food vendor | ||
|---|---|---|---|---|
| n/N (%†) | 95% CI | n/N (%†) | 95% CI | |
| Food items kept in a covered and clean* pot/container for sale (spot check) | ||||
| Sweets/curd/milk | 17/69 (25) | (13, 36) | 1/3 (33) | (1, 91) |
| Rice, lentils, and vegetable mix | 68/285 (23) | (18, 29) | 6/14 (42) | (18, 66) |
| Meat/egg | 38/248 (15) | (10, 20) | – | – |
| Fish | 36/246 (14) | (10, 19) | – | – |
| Vegetables | 25/198 (12) | (8, 17) | – | – |
| Lentil soup | 27/243 (11) | (7, 15) | – | – |
| Mashed food—potato, fish, egg, shrimp, spices, lentil, and vegetables | 3/44 (7) | (1, 19) | 0/6 (0) | (0, 0) |
| Salad | 7/121 (6) | (2, 10) | – | – |
| Flatbread | 7/174 (4) | (1, 7) | 1/24 (4) | (0, 21) |
| Fried pastry filled with eggplant, lentils, potato, onion, shrimp (Piyaju, beguni, samusa, and singara) | 5/150 (3) | (0, 6) | 6/134 (5) | (1, 8) |
| Curry (fish, lentil, meat, eggs, and vegetables) | – | – | 10/27 (38) | (19, 56) |
| Tea, biscuits, and dry cake | – | – | 16/45 (36) | (22, 50) |
| Fushka/chotpati/golgoppa (boiled diced potatoes, onions, chillies, and chickpeas with grated eggs on top with roasted spice powder) | – | – | 30/118 (26) | (17, 34) |
| Sliced raw fruits | – | – | 1/93 (1) | (0, 6) |
| Pickles (variety of dried fruits and vegetables kept in oil which are sweet, sour, and spicy) | – | – | 8/67 (12) | (4, 20) |
| Puffed rice with chilies and oils/nuts | – | – | 24/191 (13) | (8, 18) |
* No visible dirt inside or outside the containers that contained food for sale.
† Weighted; was not observed.
DISCUSSION
In this study, we found limited facilities and poor hygiene and food-handling practices in restaurants and food vendor stalls across Bangladesh. The presence of water and soap at food preparation areas was very low. Poor practices were predominantly explained by the technological dimension at the community and structural level. Lack of facilities affects rates of handwashing with soap,13 confirmed during structured observations especially at street food–vending stalls, where both water and soap were rarely present. Very few restaurant food handlers were observed to wash hands with soap during food handing, and none of the street food vendors washed hands with soap at these key times. These low rates were similar to those from households when preparing food; among 500 intervention and 500 control households, only 1% of food preparers washed hands with soap before food preparation29; out of 85 opportunities related to food preparation, participants from households washed hands with soap two times11; of 12 opportunities related to food preparation, none washed hands with soap.12
Moreover, street food vendors stated that carrying water is a burden and, therefore, an important barrier that limits good food hygiene practices, which was also supported by behavior change theory.30,31 A street food study in Ghana similarly identified low availability of running water–compromised hygiene practices.32 Thus, foods prepared by Bangladeshi street food vendors and restaurants are likely vehicles for pathogens, especially as bare hands are commonly used when preparing and serving food.6,11,12
As facilities are critical to facilitate practice, placement of handwashing agents and water in the food preparation area (in restaurants) could enhance convenience and improve practices.30,31 Hand hygiene is difficult to achieve if adequate water is not available; therefore, promoting a low-cost hand rub for use at key times might be a good option, especially for ambulant street food vendors. The psychosocial dimension at individual level was a component of the IBM-WASH model that comprises self-efficacy, knowledge, perceived threat, and disgust. The psychosocial dimension at the individual level largely explained participants’ perceptions, for example, where participants perceived that they did not touch anything dirty during food handling and that “hand to water contact” was reasonable handwashing. Such perceptions, similar to those reported by household food preparers,12,33 could hamper effective hand hygiene. The contextual dimension at individual and behavioral level suggests additional barriers to maintaining hygiene: restaurant managers and food preparers stated that heavy workload and the need to quickly serve food to customers made it difficult to wash hands frequently. A good proportion (37%) of urban restaurants used municipal supplied tap water for drinking, a potential vehicle of Escherichia coli34 and other fecal organisms. In rural areas where tube well water is predominant,35 source water contamination is generally low but increases after collection.36 Most of the restaurants and street food vendors kept their drinking water in an uncovered container, providing additional opportunities for the spread of enteropathogens when this water is used as a food ingredient, or for washing of foods, utensils, and hands.37–39 Simply using a water reservoir that has a lid and tap can reduce water contamination.36
This study has limitations. First, data collectors observed practices. Such observation after disclosing the objective when taking consent can influence the practice in the presence of observer.40 However, in this context, handwashing was so rare, that even if these practices are slightly overestimated, they still demonstrate that recommended practices are uncommon. Second, this study did not include microbiological testing of food samples to specify microbial risk to customers. Other studies of restaurant and household food in Bangladesh suggest that many foods will be contaminated, especially if stored at warm ambient temperature and for long periods.6,41,42 Future microbial investigations of environmental and food samples could help prioritize specific hygiene practices. We asked food vendors and restaurant staff what they thought were important issues for customers. It would have been useful to ask customers directly to determine what motivates them to purchase from or avoid restaurants and food vendors.
As this study covered most Bangladesh districts (53/64), these nationally representative data provided estimates of national handwashing coverage in restaurants and among street food vendors, which can be used to advocate for and guide policy to improve food safety. In the absence of internationally recognized SDG 6 indicators for food premises, based on our spot checks and observation findings, we developed a framework (Figure 1) to assess restaurants and vendors; this framework would benefit from efforts to apply it in other settings. In the framework, we proposed pragmatic actions, applicable to the contexts we observed, which can affordably be implemented. The proposed actions have the potential to reduce food and drinking water contamination for customers, including some of the practical initiatives from interventions that members of our research group have developed for shared urban slum sanitation,43 household handwashing with soap,44 and safe drinking water storage.36
In addition to better characterizing practices and risks, efforts to improve food safety in Bangladesh and similar low-income countries could benefit from developing and piloting feasible restaurant- and food vendor–level intervention. Such an intervention should include a system that supplies sufficient water and soap to support hygienic practices. Besides, a social and behavioral change communication strategy to increase awareness among food handlers and as well as customers may aid the hygiene intervention, as a study conducted in Brazil found that ensuring hand hygiene facilities/materials was not sufficient.45 Because food vendors are business men seeking to earn profits by meeting the demands of customers, increasing demand for hygienic food is also likely to be an important part of an effective intervention.
Supplemental materials
Acknowledgments:
This study was funded by the WaterAid Bangladesh. We wish to acknowledge the study respondents and data collection team. icddr,b is thankful to the Governments of Bangladesh, Canada, Sweden, and the United Kingdom for providing core/unrestricted support.
Note: Supplemental appendix appears at www.ajtmh.org.
REFERENCES
- 1.WHO , 1999. Food Safety: An Essential Public Health Issue for the New Millennium. Geneva, Switzerland: Food Safety Programme, Department of Protection of the Human Environment, Cluster of Sustainable Development and Healthy Environments. [Google Scholar]
- 2.Havelaar AH, et al. 2015. World health organization global estimates and regional comparisons of the burden of foodborne disease in 2010. PLoS Med 12: e1001923. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Mathee A, Von Schirnding YE, Byrne J, De Beer M, Letlape A, Hobbs C, Swanepoel F, 1996. The greater Johannesburg healthy foods/markets programme. Urban Health Newsl 1996 Mar (28):39–47. [PubMed] [Google Scholar]
- 4.Gasem MH, Dolmans WM, Keuter MM, Djokomoeljanto RR, 2001. Poor food hygiene and housing as risk factors for typhoid fever in Semarang, Indonesia. Trop Med Int Health 6: 484–490. [DOI] [PubMed] [Google Scholar]
- 5.Nasrolahei M, Mirshafiee S, Kholdi S, Salehian M, Nasrolahei M, 2017. Bacterial assessment of food handlers in Sari city, Mazandaran Province, north of Iran. J Infect Public Health 10: 171–176. [DOI] [PubMed] [Google Scholar]
- 6.Faruque Q, Haque QF, Shekhar HU, Begum S, 2010. Institutionalization of Healthy Street Food System in Bangladesh: A Pilot Study with Three Wards of Dhaka City Corporation as a Model. Dhaka, Bangladesh: National Food Policy Capacity Strengthening Programme (NFPCSP) Available at: http://www.nfpcsp.org/agridrupal/sites/default/files/pR_7_of_04_Final_Techncial_Report_-_Approved.pdf. Accessed February 26, 2015. [Google Scholar]
- 7.Abdussalam M, Kaferstein FK, 1993. Safety of street foods. World Health Forum 14: 191–194. [PubMed] [Google Scholar]
- 8.Todd EC, Greig JD, Bartleson CA, Michaels BS, 2008. Outbreaks where food workers have been implicated in the spread of foodborne disease. Part 5. Sources of contamination and pathogen excretion from infected persons. J Food Prot 71: 2582–2595. [DOI] [PubMed] [Google Scholar]
- 9.Vollaard AM, Ali S, van Asten HA, Ismid IS, Widjaja S, Visser LG, Surjadi C, van Dissel JT, 2004. Risk factors for transmission of foodborne illness in restaurants and street vendors in Jakarta, Indonesia. Epidemiol Infect 132: 863–872. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Rane S, 2011. Street vended food in developing world: hazard analyses. Indian J Microbiol 51: 100–106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Nizame FA, Nasreen S, Halder AK, Arman S, Winch PJ, Unicomb L, Luby SP, 2015. Observed practices and perceived advantages of different hand cleansing agents in rural Bangladesh: ash, soil, and soap. Am J Trop Med Hyg 92: 1111–1116. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Nizame FA, Unicomb L, Sanghvi T, Roy S, Nuruzzaman M, Ghosh PK, Winch PJ, Luby SP, 2013. Handwashing before food preparation and child feeding: a missed opportunity for hygiene promotion. Am J Trop Med Hyg 89: 1179–1185. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Luby SP, Halder AK, Tronchet C, Akhter S, Bhuiya A, Johnston RB, 2009. Household characteristics associated with handwashing with soap in rural Bangladesh. Am J Trop Med Hyg 81: 882–887. [DOI] [PubMed] [Google Scholar]
- 14.Manes MR, Kuganantham P, Jagadeesan M, Laxmidevi M, Dworkin MS, 2016. A step towards improving food safety in India: determining baseline knowledge and behaviors among restaurant food handlers in Chennai. J Environ Health 78: 18–25. [PubMed] [Google Scholar]
- 15.Onyeneho SN, Hedberg CW, 2013. An assessment of food safety needs of restaurants in Owerri, Imo state, Nigeria. Int J Environ Res Public Health 10: 3296–3309. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Bryan FL, Bartleson CA, Christopherson N, 1981. Hazard analyses, in reference to Bacillus cereus, of boiled and fried rice in Cantonese-style restaurants. J Food Prot 44: 500–512. [DOI] [PubMed] [Google Scholar]
- 17.Barro N, Bello AR, Savadogo A, Ouattara CAT, Iiboudo AJ, Traoré AS, 2006. Hygienic status assessment of dish washing waters, utensils, hands and pieces of money from street food processing sites in Ouagadougou (Burkina Faso). Afr J Biotechnol 5: 1107–1112. [Google Scholar]
- 18.WHO/UNICEF. Joint Monitoring Programme (JMP) for Water Supply and Sanitation Improved Unimproved Water Sources Sanitation Facilities. Available at: http://www.wssinfo.org/definitions-methods/watsan-categories/. Accessed November 30, 2015.
- 19.Dreibelbis R, Winch PJ, Leontsini E, Hulland KR, Ram PK, Unicomb L, Luby SP, 2013. The integrated behavioural model for water, sanitation, and hygiene: a systematic review of behavioural models and a framework for designing and evaluating behaviour change interventions in infrastructure-restricted settings. BMC Public Health 13: 1015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Bangladesh Bureau of Statistics , 2011. 2011 Population & Housing Census: Preliminary Results. Dhaka, Bangladesh: Bangladesh Bureau of Statistics. [Google Scholar]
- 21.NIPROT , 2016. Bangladesh Urban Health Survey 2014. Dhaka, Bangladesh: National Institute of Population Research and Training (NIPORT), Associates for Community and Population Research (ACPR), and ICF International. [Google Scholar]
- 22.Government of the People’s of Republic of Bangladesh MoLG, Rural Development and Cooperatives, Local Government Division, 2012 National Hygiene Promotion Strategy for Water Supply and Sanitation Sector in Bangladesh 2012. Available at: https://itn.buet.ac.bd/publications/sector-documents/documents/nhps.pdf. Accessed January 9, 2016.
- 23.Keller H; NSP, IPHN , 2003. Life in the Chars in bangladesh, Nutritional Surveillance Project Bulletin no. 14. Dhaka, Bangladesh: Helen Keller; Available at: http://socialprotection.gov.bd/wp-content/uploads/2017/06/Nutrition-Livelihoods-in-Chars.pdf. [Google Scholar]
- 24.Guest G, Bunce A, Johnson L, 2006. How many interviews are enough?: an experiment with data saturation and variability. Field Methods 18: 59–82. [Google Scholar]
- 25.Rabe-Hesketh SA, 2006. Multilevel modelling of complex survey data.J R Statist. Soc 169: 805–827. [Google Scholar]
- 26.United Nations, 2018. Sustainable Developement Goal 6. Available at: https://sustainabledevelopment.un.org/content/documents/19901SDG6_SR2018_web_3.pdf. Accessed June 30, 2018. [Google Scholar]
- 27.WHO/UNICEF , 2017. WASH in the 2030 Agenda. New global Indicators for Drinking Water, Sanitation and Hygiene. Geneva, Switzerland: World Health Organization and the United Nations Children’s Fund (UNICEF). [Google Scholar]
- 28.WHO , 2016. Monitoring WASH in Health Care Facilities. Geneva, Switzerland: World Health Organization. [Google Scholar]
- 29.Huda TM, Unicomb L, Johnston RB, Halder AK, Yushuf Sharker MA, Luby SP, 2012. Interim evaluation of a large scale sanitation, hygiene and water improvement programme on childhood diarrhea and respiratory disease in rural Bangladesh. Soc Sci Med 75: 604–611. [DOI] [PubMed] [Google Scholar]
- 30.Story M, Neumark-Sztainer D, French S, 2002. Individual and environmental influences on adolescent eating behaviors. J Am Diet Assoc 102: S40–S51. [DOI] [PubMed] [Google Scholar]
- 31.Gielen AC, Sleet D, 2003. Application of behavior-change theories and methods to injury prevention. Epidemiol Rev 25: 65–76. [DOI] [PubMed] [Google Scholar]
- 32.Mensah P, Yeboah-Manu D, Darko KO-, Ablordey A, 2002. Street foods in Accra, Ghana: how safe are they? Bull World Health Organ 80: 546–553. [PMC free article] [PubMed] [Google Scholar]
- 33.Nizame FA, Leontsini E, Luby SP, Nuruzzaman M, Parveen S, Winch PJ, Ram PK, Unicomb L, 2016. Hygiene practices during food preparation in rural Bangladesh: opportunities to improve the impact of handwashing interventions. Am J Trop Med Hyg 95: 288–297. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Amin N, Pickering AJ, Ram PK, Unicomb L, Najnin N, Homaira N, Ashraf S, Abedin J, Islam MS, Luby SP, 2014. Microbiological evaluation of the efficacy of soapy water to clean hands: a randomized, non-inferiority field trial. Am J Trop Med Hyg 91: 415–423. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.BBS-UNICEF , 2010. Multiple Indicator Cluster Survey Bangladesh 2009. Dhaka, Bangladesh: Bangladesh Bureau of Statistics (BBS)-United Nation’s Children Fund (UNICEF). [Google Scholar]
- 36.Ercumen A, Naser AM, Unicomb L, Arnold BF, Colford JM, Jr., Luby SP, 2015. Effects of source- versus household contamination of tubewell water on child diarrhea in rural Bangladesh: a randomized controlled trial. PLoS One 10: e0121907. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Angulo FJ, et al. 1997. A community waterborne outbreak of salmonellosis and the effectiveness of a boil water order. Am J Public Health 87: 580–584. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.PAHO , 1994. PAHO (Pan American health organization) Drinking Water Supply. Health Conditions in the Americas. Washington, DC: Scientific Publications; vol. 54, 274–277. [Google Scholar]
- 39.Mankee A, et al. 2003. Bacteriological quality of “doubles” sold by street vendors in Trinidad and the attitudes, knowledge and perceptions of the public about its consumption and health risk. Food Microbiol 20: 631–639. [Google Scholar]
- 40.Cousens S, Kanki B, Toure S, Diallo I, Curtis V, 1996. Reactivity and repeatability of hygiene behaviour: structured observations from Burkina Faso. Soc Sci Med 43: 1299–1308. [DOI] [PubMed] [Google Scholar]
- 41.Islam MS, Hasan MK, Khan SI, 1993. Growth and survival of Shigella flexneri in common Bangladeshi foods under various conditions of time and temperature. Appl Environ Microbiol 59: 652–654. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Parvez SM, Kwong L, Rahman MJ, Ercumen A, Pickering AJ, Ghosh PK, Rahman MZ, Das KK, Luby SP, Unicomb L, 2017. Escherichia coli contamination of child complementary foods and association with domestic hygiene in rural Bangladesh. Trop Med Int Health 22: 547–557. [DOI] [PubMed] [Google Scholar]
- 43.Alam M-U, et al. 2017. A behavior change intervention to improve shared toilet maintenance and cleanliness in urban slums of Dhaka: a randomized control trial. Trop Med Int Health 22: 1000–1011. [DOI] [PubMed] [Google Scholar]
- 44.Ashraf S, et al. 2017. Nonrandomized trial of feasibility and acceptability of strategies for promotion of soapy water as a handwashing agent in rural Bangladesh. Am J Trop Med Hyg 96: 421–429. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Draeger CL, Akutsu R, de Oliveira KES, da Silva ICR, Botelho RBA, Zandonadi RP, 2019. Unhygienic practices of health professionals in Brazilian public hospital restaurants: an alert to promote new policies and hygiene practices in the hospitals. Int J Environ Res Public Health 16: E1224. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.