Socio-Economic Disparities in Attitude and Preference for Menu Labels among Vietnamese Restaurant Customers

Calories and nutrition labeling on restaurant menus are powerful policy interventions to reduce the burden of obesity epidemic. However, the success of this policy requires an assurance of equal benefits among customers with different characteristics, especially people at a higher risk of poor health outcomes and eating habits. This study examined the sociodemographic disparities in the attitude and preference for calories and nutrition labeling on menus among customers in various food facilities. A cross-sectional study was conducted with 1746 customers of food facilities in Hanoi, Vietnam, who were recruited by using a multistage sampling method. Socio-economic characteristics, attitudes regarding the necessity and preferences for calories, and nutrition labeling on menus were analyzed. Multivariate logistic regression was employed to determine the associated factors with attitudes and preferences. Results show that most of the sample understood the necessity to have calories and nutrition labeling (59.8%), and 71.8% preferred to have calories and nutrition labeling. People who often visited food facilities (Odd Ratio (OR) = 1.36; 95% confident interval (CI) = 1.06–1.74) and had higher education and were more likely to understand the necessity of calories and nutrition labeling. Factors such as being homemakers, often going to dine-in restaurants, and perceiving that labeling was unnecessary were negatively associated with preferences for calories and nutrition labeling. The results of this study encourage policymakers to implement calories and nutrition labeling in the future. Health education interventions to improve knowledge and attitude as regards calories and nutrition labeling on menus are important, particularly for males, less-educated individuals, and high-income people.


Introduction
Calories and nutrition labeling have been proposed as a cost-effective policy intervention against obesity and other malnutrition epidemics globally [1]. Informing nutrition contents (e.g., calories, nutrients, fat, etc.) to customers empowers them to purchase healthy food and have balanced diets [2,3]. This is especially important in settings where an increasing number of people are routinely eating in restaurants instead of at home. Food at these facilities has higher calories and poorer nutrients, and is often served in large portions, which may lead to overconsumption [4,5]. Previous studies indicated that calories and nutrition labeling on menus or menu boards at restaurants promoted customers' food choices, increased their perceptions, and reduced their calories intake [6]. These positive effects could result in decreasing the burden of the obesity epidemic [7]. Thus, regulations requiring calories and nutrition labeling on menus have been implemented officially in some states of the U.S.A. and is of concern in other places such as the United Kingdom, China, and several Asian countries [8][9][10]. Global research has shown widespread interest from customers in seeing calories and nutrition labeling on menus or menu boards at restaurants, with about 50-70% of customers preferring to have and use calorie information on the menus [11][12][13].
Despite the rapid increase of obesity rates across all population groups, a remarkably higher burden was observed among people who were young, women, belonged to minority groups and had low income [14,15], leading to socio-economic disparities. These disparities are more likely to increase if public health policies cannot engage all demographic segments equally. Therefore, in order to become an effective public health tool, calories and nutrition labeling policies should assure equal benefits for consumers with different sociodemographic characteristics, particularly people who are at a higher risk of poor health and eating habits. Some prior studies found that female clients were more likely to use calorie information on the menus, while there were some mixed results according to age, education, and income groups [16][17][18][19].
The prevalence of overweight and obesity in Vietnam has been rising proportionately with the economic growth, especially in urban areas. Two national surveys indicated that the rate of individuals with overweight and obesity approximately doubled from 3.7% in 2000 to approximately 7% in 2005 [20]. A study in Ho Chi Minh city-a Vietnamese metropolis-in 2015 found that 24% males and 19% females were overweight and obesity [20]. Importantly, the occurrence of overweight and obesity is increasingly observed in Vietnamese preschool children and adolescents, which may be due to the expansion of fast-food restaurants, sedentary lifestyles, as well as the academic burden [21,22]. In 2010, the Vietnam National Assembly enacted the Law on Food Safety following the CodeX Alimentarius (a joint United Nations and World Health Organization Commission) guideline, requiring nutrition labeling on the pack of food products [8,23]. However, the law does not provide regulations for calories and nutrition labeling on the menus of food facilities [23]. Therefore, it is hard to find nutrition labels on the menus of Vietnamese restaurants.
Given the dearth of information about calories and nutrition labeling on the menus in Vietnam, this study examined the sociodemographic disparities in the attitudes and preferences regarding calories and nutrition labeling on the menus among customers in various food facilities. The result will be expected to partly contribute to developing nutritional strategies for alleviating the overweight and obesity epidemic in Vietnam.

Study Design
Participants were 1746 customers in fast-food restaurants, dine-in restaurants, street food restaurants, and other food facilities (such as cafeterias, street food vendors, etc.). They were recruited for a cross-sectional survey which was conducted in Hanoi from October to November 2015. Hanoi is the capital of Vietnam, having 577 communes clustered within 30 districts. According to the General Statistics Office in 2016, the population in Hanoi was young given that 52.2% of people were 15 years old or above. Most of the residents were female (51.0%) and living in urban areas (53.6%) [24]. In this study, the eligible criteria included: (1) aged ≥15 years old; (2) using food services in selected food facilities; and (3) provided informed consent to participate in this study.
We performed a multistage sampling method to recruit respondents. First, among 29 districts of Hanoi, we randomly selected 176 communes. Then, in each commune, we listed all food facilities that were registered with local authorities, and randomly picked ten facilities. Finally, the data collectors visited these facilities and recruited the third customer after them. A total of 1760 clients were invited to participate in the study, and data of 1746 customers were used for analysis (99.2%). We excluded data from 14 clients because they decided to withdraw during the interview.

Measures and Instruments
We constructed a structured questionnaire and piloted it with 20 consumers to validate the tools. After revision, the questionnaire was used by the data collection teams who were Master of Public Health students at Hanoi Medical University. These students were trained to collect the data consistently and ensure the quality of data. Respondents were interviewed face-to-face within 15-20 min.
The questionnaire included socioeconomic characteristics (age, gender, education attainment, marital status, living location, employment, monthly household income); self-reported height and weight; attitudes and preferences for calories and nutrition labeling on the menus in the restaurants.
For the attitudes and preferences, we asked respondents to report whether they frequently visited food facilities for food services, preferable types of food facilities (fast-food, dine-in restaurant, street food, or others), criteria for ordering food (name, nutrition, introduction, price or others), attitudes regarding the necessity of calories and nutrition labeling (with five-point Likert scale from 'very unnecessary' to 'very necessary'). People were categorized into the 'necessary' group if they selected 'very necessary' or 'necessary'; otherwise, they were belonged to 'not necessary' group. We also asked them about the preferences for having calories and nutrition labeling ('yes/no').

Statistical Analysis
We analyzed the data using STATA software version 12.0 (StataCorp. LP, College Station, TX, USA). p-value < 0.05 was used for identifying the statistical significance. We used a multivariate logistic regression to identify the associated factors with "Attitudes regarding the necessity of calories and nutrition labeling" (necessary/not necessary) and "Prefer to have calories and nutrition labeling" (yes/no). These models were combined with a forward stepwise selection strategy to produce the reduced models.

Ethics Approval
The study protocol was reviewed and approved by the IRB of the Hanoi Health Department (code: 06/CCATVSTPHN). We obtained the written informed consents from participants. Their data were only used for research and kept confidentially.

Mean SD
Monthly household income (million VND) 5.2 5.7 Table 2 presents that 68.6% clients reported that they frequently visited food facilities. Street food restaurants were the preferable facility of 43.9% customers, following by the dine-in restaurants (42.2%) and fast food restaurants (41.2%). Name and nutrition of food were the two favorable criteria when ordering food (with 48.6% and 47.6%, respectively), followed by the introductory statement and price of food (with 43.1% and 21.5%, correspondingly). Most of the sample felt that it was necessary or very necessary to label nutrition on the menus (59.8%), and 71.8% preferred to have food label on the menus.  Table 4 shows that most of female customers perceived that menu labeling was necessary (63.8%) and preferred menu labels (74.9%). These rates were significantly higher than those in males (53.4% and 66.7%, respectively). People belonged to the age group ≥60 years (48.7%), being separated/divorced/widowed (40.0%), attaining < high school education (40.2%), and being blue-collar workers (45.9%) had the lowest percentages compared to other groups in having positive attitudes regarding menu labels. These tendencies were also observed in preferring to have menu labels. These differences were statistically significant (p < 0.05). Meanwhile, we did not find any statistically significant differences among income groups, living locations, BMI categories, and often food facilities visit (p > 0.05).   Table 4 shows that people who were male (OR = 0.54; 95% CI = 0.43-0.68), and had higher income were less likely to perceive that calories and nutrition labeling was necessary. Otherwise, often visiting food facility (OR = 1.38; 95% CI = 1.08-1.77) and having higher education had positive associations with feeling the necessity of calories and nutrition labeling. In addition, people who were homemakers/others, who often going to dine-in restaurants (OR = 0.38; 95% CI = 0.18-0.80) were less likely to prefer to have calories and nutrition labeling. Meanwhile, having positive attitudes with menu labeling was significantly associated with preferring to have menu labels (OR = 32.62; 95% CI = 21.96-48.46). Body mass index or overweight/obesity was not associated with attitudes and preferences for menu labels.

Discussion
The current study highlighted the positive attitudes and high demand for calories and nutrition labeling by consumers in food facilities in a Vietnamese urban setting. We also explored the existing socio-demographic disparities in the attitudes and preferences for calories and nutrition labeling in food facilities, which can potentially be used to develop interventions tailored for different groups of customers in the future.
In this study, we found that a high proportion of respondents understood the necessity of calories and nutrition labeling and preferred to have calories and nutrition labeling in the food facilities. These results were consistent with other findings, which demonstrated that calories and nutrition labeling had a widespread support from the public [11][12][13]25]. Importantly, the nutrition of food was the second most important information that the clients used for ordering food, and people who often visited food facilities were more likely to perceive the necessity for menu labels. Indeed, we found two-thirds of customers visited food facilities frequently, which might put them at higher risk of obesity. The literature suggested that people who have meals outside the home more than five times per week were more likely to be obese [26]. These results were very critical that our sample were aware of the importance of menu labeling intervention in protecting their health and preventing overweight/obesity. Therefore, adopting the low-cost tool such as posting calories information on menus should be considered to inform people about healthy food choice.
Our analysis indicated that there were disparities in the attitudes and preferences for calories and nutrition labeling in certain socio-demographic characteristics such as gender and education. Men were less likely to perceive the necessity of calories and nutrition labeling, which was similar to other studies [11,19,25]. Otherwise, women preferred restaurants having caloric information on the menus because this information could help them to choose the lower-calorie dishes and control their diets [11,25]. People who were well-educated were also observed to have a favorable response to calories and nutrition labeling in food facilities. This may be explained by the fact that people with higher education had a higher likelihood to have healthy behaviors (doing physical exercise, eating a healthy diet, not smoking or drinking alcohol, etc.) or seek health information frequently to have better health outcomes [27].
Nonetheless, although income was not associated with the attitudes regarding menu labeling in the univariate analysis, in the multivariate model, income was negatively related to the attitudes. This finding was different from previous studies, which found that wealthier people had more interested in calories and nutrition labeling [19]. The reason for this phenomenon was not clear.
In fact, we observed that lower-income individuals were more likely to choose fast-food restaurants, while higher-income people were more likely to visit dine-in restaurants. In addition, people often selecting dine-in restaurants-albeit not statistically significant and not included in the final model-were less likely to prefer calories and nutrition labeling. We supposed that they believed the food in dine-in restaurants had more balanced and healthier nutrition compared to fast-food or street food restaurants, which made them feel that calories and nutrition labeling was not necessary [28].
The study findings suggest several implications. First, policymakers should consider implementing interventions requiring calories and nutrition labeling not only in fast-food restaurants but also in dine-in restaurants and other food facilities. Second, educational interventions about the importance of a healthy diet and the necessity of calories and nutrition labeling should be provided, particularly for male and high-income individuals, in order to encourage them to control their calories and nutrient intake. This, in turn, will help to control the obesity epidemic that is increasing in Vietnam. Finally, a study to examine the barriers and facilitators of implementing calories and nutrition labeling from the providers' perspective should be conducted to provide a comprehensive view of the feasibility of this intervention.
This study has strengths in a large sample size with various types of restaurants selected. Nonetheless, several limitations should be pointed out. First, this study was conducted only in Hanoi, a metropolitan area of Vietnam. Moreover, the socio-demographic characteristics of respondents in this study were slightly different compared to these characteristics of the general population in Hanoi. Therefore, the result had a limited generalizability that might not apply to other settings. Second, the cross-sectional design does not allow us to identify the causal relations between attitudes and preferences and its associated factors. Finally, there are some features that we did not take into account in this study, such as the effective approach to communicate caloric and nutritional information. In addition, we could not test the effect of calories and nutrition labeling on the reduction of energy and nutrients consumed. Further studies should be conducted to fill these gaps.

Conclusions
In conclusion, the positive attitudes and preferences for calories and nutrition labeling by customers found in this study should inform actions to implement this intervention in the future. Educational interventions to improve knowledge and attitudes regarding calories and nutrition labeling are important, particularly for male, less educated individuals, and high-income people. Further research is needed to examine the opinions of food sellers and the most effective way for calories and nutrition labeling.