Abstract
Raw fish consumption in restaurants, for example, Sashimi style, is popular worldwide. In Vietnam, raw fish dishes are also traditionally prepared and consumed in private households. However, the habits of eating raw or otherwise inadequately cooked fish can be associated with risks of acquiring fishborne zoonotic trematode (FZT) infection. The present study was done in a fish-farming community in Nam Dinh, Vietnam, to obtain information about habits of eating raw fish dishes and risks for human FZT infection. Discussions were held in different groups divided by gender and age on raw-fish-eating behavior. A total of 180 household members were interviewed and their stool samples analyzed to identify risk factors of FZT infection. There was awareness about the risk of liver fluke infections from eating raw fish. However, many older people accepted these risks and continued eating raw fish, as they know effective drug treatment is available. Raw fish dishes are consumed at social gatherings from shared plates and dipping sauces using the same chop sticks. This is likely to pose risks of crosscontamination with FZT metacercariae to different food items as indicated by the finding that 25.8% of household members that stated not to have eaten raw fish were infected. In total, 32.2% fish farm household members were infected with FZT. The odds of FZT infection was 2.3 times higher (p = 0.013) for those eating raw fish than for those who did not eat raw fish. Among the people eating raw fish, those eating raw fish in restaurants had 3.6 times higher odds of FZT infection (p = 0.009) than people eating raw fish at home. A successful program to control FZT must be based on in-depth knowledge on the social and anthropological determinants of people's raw-fish-eating behavior and hygiene practices as well as production of FZT-free fish for human consumption.
Introduction
One million Vietnamese are estimated infected with FZT, in particular in the Red River Delta, where FZT is endemic (Kino et al., 1998; De et al., 2003; Dung et al., 2007; Thach et al., 2008). To reduce the risk of FZT infection in humans, the Vietnamese government has together with international organizations implemented numerous campaigns in northern Vietnam to advise people stop eating raw fish. Despite these efforts the FZT prevalence in humans remains high (Kino et al., 1998; De et al., 2003; Dung et al., 2007; Thach et al., 2008). The limited success of the campaigns to obtain a sustainable reduction in FZT prevalence may be because such campaigns do not seem to build upon insights into peoples practices of preparing raw fish dishes as wells as their knowledge and attitudes toward such dishes and the risk of FZT infection. The present study was therefore carried out to obtain information about habits of preparing and eating raw fish dishes and risks for FZT infection among people in fish-farming communities in Nam Dinh province, northern Vietnam.
Materials and Methods
Study sites and selection of households
Four communes in Nghia Hung district, Nam Dinh province, in the Red River Delta of northern Vietnam, were selected. Here, fish farms are mainly small-scale and family-based, and the aquaculture produce generates income together with other activities such as rice culture and live stock rearing.
Group discussions were held in Nghia Hong and Nghia Son communes, Nghia Hung district, to obtain information on people's habits of eating raw fish. The information obtained was subsequently used to formulate a questionnaire used in a quantitative risk factor study on human FZT infection (see below). In each commune, individuals were divided into four groups consisting of older men (≥30 years), younger men (<30 years), older women (≥30 years), and younger women (<30 years), all coming from households having a fish pond. The division into these groups allowed individuals irrespective of age and gender to freely speak about their knowledge, attitudes, and practices of eating raw fish. Each group discussion (7–12 persons) lasted 1–2 h and was recorded digitally and by taking notes. Oral consent for such recording, which was anonymous, was obtained from all individuals before discussion.
Household members in Nghia Lac and Nghia Phu communes, Nghia Hung district, were selected for individual interviews and for stool sample analyses to identify and assess risk factors of FZT infection with focus on habits of eating raw fish. Households with a fish pond were randomly selected from a list available at the local health clinic and were part of an ongoing longitudinal study on prevention and control of FZT transmission in aquaculture (Phan et al., 2010). Households with no fish ponds and farms that participated in a previous study on human FZT prevalence (Dung et al., 2007) were excluded, because infected individuals were provided praziquantel to treat the FZT infection. A total of 104 households with fish ponds were randomly selected from the adjusted lists of households.
Group discussions, formulation of questionnaire, and interview
Group discussions were conducted at the local health clinic and included the principal investigator and three colleagues who helped taking notes. Local health and aquaculture extension officers were also present in the groups, but did not take part in the discussions. Topics introduced and discussed were eating raw fish (yes, no), species, and origin of fish eaten raw; how the quality of fish was assessed; where raw fish dishes were consumed; utensils, ingredients, and methods used to prepare raw fish dishes; people involved in preparing the raw fish dishes; type of person (including age and gender) eating raw fish; season and frequency of consumption of raw fish dishes; and knowledge on FZT, including sources of information, risk of infection, and drug treatment (Table 1).
Questionnaires used for assessing risks for human infection were formulated based on the information provided in the group discussions and then pilot tested through interviews of five staff at the Research Institute of Aquaculture No. 1 in Bac Ninh and five persons in each of the communes of Nghia Lac and Nghia Phu who did not participate in the interview later on. Main topics covered by the questionnaire were identical to the topics discussed in the groups. The use of questionnaires to investigate the habit of eating raw freshwater fish has previously been found to be a rapid, cost-effective, and operationally feasible tool for the identification of individuals at risk for FZT (Yajima et al., 2009).
Stool sampling and FZT identification
All household members interviewed were invited to provide a stool sample for examination of FZT. The examination was conducted by well-trained and experienced staff of the National Institute of Malariology, Parasitology, and Entomology (NIMPE) in Hanoi and carried out at the local communal health clinics according to previously published procedures (Dung et al., 2007). Briefly, a total of 180 stool samples were analyzed within 24–48 h of collection. Two Kato-Katz smears were prepared from each stool sample and analyzed for FZT eggs by using a standard kit provided to NIMPE by the World Health Organization. Fecal slides were examined by light microscopy.
Data analysis
Quantitative data collected for the risk factor study were analyzed using the Statistical Analysis System (SAS®, version 9.1) software. The descriptive analysis was performed by means of frequencies of FZT infected household members stratified by commune, gender, age, and eating raw fish.
Initially, an analysis was performed to evaluate the effect of commune, age groups (<30; ≥ 30 year old), gender, and eating raw fish (yes, no) on the risk of human FZT infection. Human infection status (positive, negative) was the outcome. The second analysis included only people who reported eating raw fish, and the outcome was their FZT infection status. In total, 91 people reported that they were eating raw fish, but only 86 people responded to all the questions on habits of eating raw fish and these were therefore included in the data analysis. Factors studied included place of eating raw fish, frequency of eating raw fish, fish species consumed, fish size, origin of fish, and utensils used to prepare raw fish dish. If people responded “yes” to both eating raw fish in restaurant and at home, then they were included in the group “eating raw fish at restaurant.” If people responded that they had eaten raw fish dishes prepared using wild-caught fish species, then the fish species were grouped as “not cultured species.”
In both analyses, logistic analysis was used with household nested in commune as a random effect. Backward elimination was applied to obtain the resulting model, including significant effects. Confounding between excluded potential risk factors with identified risk factors were tested by using 2 × 2 tables and adding the factors to the resulting model one by one for testing level of significance. Odds ratios were calculated only for significant effects. A p-value of <0.05 was considered significant.
The qualitative data obtained from the group discussion by recording and as written notes were validated by comparison of the recorded information and written notes. Only information that was relevant for the topic of discussion was collected and summarized. Important and illustrative statements by people during group discussion were translated into English and quoted.
Ethics consideration
Before the study was carried out, households were invited to participate in a group meeting where information on the aim, objectives of the project, the benefit, risks, and rights of project participants were discussed. Participating households provided written consent about their willingness to participate in the project by delivering a signed form. Ethics approval to collect stool samples from households' members was obtained from the NIMPE, Vietnam. All people tested positive for helminth infection were treated for free of charge. The project also provided free of charge stool test for anyone who wished that in the communes. The results from the stool test were informed in confidence to people who provided stools.
Results
Information from group discussions
A summary of the information provided on the topics discussed is presented in Table 1. Both older and young people explained that they heard about the risk of being infected with liver flukes when eating raw fish from information provided by the government through television and posters placed in the community. However, very few people had heard about risk of infection with intestinal flukes. Older people of both gender often provided remarks like: “Don't worry, there are drugs to treat the liver flukes” (Đù'ng lo, có thuốc chũ'a sán lá gan mà), although some expressed that, “We do not allow our children to eat raw fish.” Several older men even said like a joke that they “would rather die because of the fluke than not having their desire for raw fish fulfilled” (Thà chết vì sán còn ho'n là chết vì thèm). Some older men and women stated that older people had a stronger and more resistant body compared with young people that could cope with FZT infection. Several older women expressed: “Very tasty sauce served with raw fish dishes, I like to eat herbs dipped into the sauce” (Nu'ó'c chắm cá gỏi ngon lắm, cú' thích lắy rau tho'm chắm vó'i nu'ó'c chắm ăn). Some boys said that they would like to “try to experience (the taste of raw fish)” (Thú' cho biết).
Common fish species used to prepare raw fish dish were silver carp (Hypophthalmichthys molitrix), grass carp (Ctenopharyngodon idellus), rohu (Labeo rohita), and mrigal (Cirrhinus mrigala). Only live fish were used to prepare the dishes. In contrast to other food preparation activities, only men were reported to clean the fish using wooden chop boards, where the fillets would be in close contact with other fish parts like skin, head, fins, and the like. Fillets were soaked twice in a mild saline (1%–3% NaCl) solution. Slices of raw fish were mixed with rice powder and eaten with herbs and sauce and always accompanied by intake of strong alcohol, for example, rice vodka. Mostly adult men eat raw fish, whereas the women reported that they mainly eat herbs dipped into the sauce. The children also had meals together with those eating raw fish, but the children consumed other types of food.
Risks for human FZT infection
Results of the descriptive analysis on human FZT infection are presented in Table 2. A total of 58/180 (32.2%) household members who provided stool samples were infected with FZT. It was noted that 23/89 (25.8%) people infected with FZT stated that they had never eaten raw fish. Among these, 73.9% were >20 years of age and 69.6% were women. The youngest person infected was a 14-year-old boy.
Total number of people providing a stool sample.
Number of people infected with FZT.
Number of people having stool examined.
Identified risk factor.
FZT, fishborne zoonotic trematode.
Odds for FZT infection in household members eating raw fish were 2.3 times significantly higher (p = 0.013; 95% CI: 1.2–4.4) than in household members not eating raw fish. Age was found to be confounder with eating raw fish since >80% of people <30 years of age stated that they did not eat raw fish, whereas about 70% of people >30 years of age reported to eat such dishes.
Table 3 shows the descriptive analysis of potential risk factors for FZT infection in family members of fish farms eating raw fish. Among those who ate raw fish, people eating raw fish in restaurants had 3.6 times significantly higher (p = 0.009; 95% CI: 1.4–9.0) odds to get an FZT infection than people eating raw fish at home.
Number of people eating raw fish responded to the question.
Number of people eating raw fish who are positive with FZT.
Number of people eating raw fish who had stool examined.
Identified risk factor.
Discussion
The study demonstrated that 38.5% of household members eating raw fish were infected with FZT. Surprisingly, 25.8% of household members stating not to eat raw fish were also found infected with FZT. Thus, FZT infect the people not only through consumption of raw fish dishes, but also through other unknown sources and transmission routes. FZT metacercariae could have contaminated the hands of people or utensils utilized during preparation and consumption of the raw fish dishes. The tradition of sharing several dishes from common plates when having meals will increase the risk of transmitting FZT metacercariae and contaminate other types of food, such as by sharing the sauce, or people using the same chop sticks when picking raw fish. The contamination with metacercariae could also occur during preparation of the different types of fish dishes. However, it may also be that some people incorrectly stated not to eat raw fish or that they were infected through unknown sources and types of foods. Further studies are needed to explain how people not eating raw fish are getting infected with FZT, including studies on adherence of metacercariae to utensils and other surface types.
Eating raw fish is a well-known pathway for FZT infection in humans (Keiser and Utzinger, 2005, 2009; Lun et al., 2005). The high prevalence of FZT found among members of fish-farming households (32.2%) corroborates findings from other human prevalence surveys conducted in the same area (Kino et al., 1998; De et al., 2003; Dung et al., 2007; Thach et al., 2008). In northeast Thailand, a 10-year control program (1984–1994) that tried to change the raw-fish-eating habits showed that consumption of raw fish was reduced from 14% to 7%, whereas occasional consumption of raw fish still remained as high as 42% (Jongsuksuntigul and Imsomboon, 1997). Similar experiences have been obtained in Nam Dinh province, where FZT infection remains high despite several efforts and campaigns by the government to inform people about the risk of infection when eating raw fish. As indicated in the present and other studies, the habits of eating raw fish is influenced by several factors, including cultural traditions, religion, ethnicity, age, and gender (Azim et al., 2008). For a better understanding of the epidemiology of FZT and determinants of eating raw fish, in-depth qualitative social–anthropological studies are needed. To generate more knowledge on people's awareness, attitudes, and practices of eating raw fish, and also on how they perceive raw fish and associated health risks, such information is essential to develop effective educational campaigns.
Consumption of raw fish fishes in restaurant, for example, Sashimi style, is popular worldwide, including Vietnam, mainly because of the economic development. Hence FZT infections are now seen across all socioeconomic classes (RITM, 2006). In the present study, people who had eaten raw fish dishes at restaurants had a higher risk of FZT infection than those eating raw fish only at home (OR = 3.6). Often, the awareness of the importance of proper hygiene conditions when preparing food seems higher in private homes compared with the awareness of owners and kitchen staff of rural restaurants like in Nam Dinh province. This may explain the higher risk for FZT infection seen when eating raw fish at restaurants. Adequate hygiene and prevention of crosscontamination during raw fish preparation is important to reduce risk of FZT infection as illustrated by the findings of Chi et al. (2009), where the prevalence (6.1%) of FZT in fish meat was significantly lower than the prevalence (34.9%) found in the other parts, such as head, gills, fins, skin, and muscle tissue from the tail.
The results from the group discussions showed that people in private homes always washed their chopping boards and knifes well because of fishy smell. They kept the fish fillet meat separated from the other parts to preserve the taste and flavor of the meat. These practices are likely to reduce the contamination risk of FZT. In the restaurants, the staff has to prepare many fish and types of dishes within short time, and therefore the risk of crosscontamination, for example, from chopping boards, may be higher. In Vietnam, live fish are traditionally used to prepare raw fish dishes, and it is uncertain if fish or meat fillets that have been frozen can be accepted or affordable by restaurant owners and consumers. It is well documented that freezing of fish will inactivate the FZT metacercariae (FDA, 1998). Thus, freezing of fish to be used for raw dishes is mandatory in many developed countries.
Other variables such as “using separate chop board for preparing the fish,” “origin of fish,” and “fish species used to prepare raw dish” had the differences in FZT prevalences of >15% but did not show significant effects due to limitations in sample size. Further studies are needed to investigate the importance of these factors to the FZT infection in human.
In the present study, people <30 years old were found to have a lower risk of FZT infection than older people. However, we also showed that age was confounded with eating raw fish. Thus, older people consumed raw fish dishes more often than younger people. It should also be noted that the life span of FZT in the human host will have a direct influence of how long an infected person will excrete eggs.
Conclusion
It was shown that 38.5% of household members eating raw fish were infected with FZT, but also that 25.8% of household members stating not to eat raw fish were infected. Eating raw fish at restaurants was associated with high risk for FZT infection. Thus, FZT infect people not only through direct consumption of raw fish dishes, but also through other unknown sources and because of crosscontamination with FZT metacercariae via hands and utensils during preparation and consumption of the raw fish dishes. With the increasing importance of aquaculture, FZT transmission should be controlled in cultured fish as an essential part of any FZT prevention strategy.
Footnotes
Acknowledgments
The authors would like to thank staff of NIMPE and CEDMA for their involvement in stool analysis and questionnaire interviews. The authors thank the people from Nghia Hong, Nghia Son, Nghia Lac, and Nghia Phu communes, Nghia hung district, Nam Dinh province, for their participation in the study. The Danish International Development Assistance (Danida) is gratefully acknowledged for its financial support through the project entitled “Fish Borne Zoonotic Parasites in Vietnam” (FIBOZOPA;
Disclosure Statement
No competing financial interests exist.
