Maternal willingness to pay for infant and young child nutrition counseling services in Vietnam

Background Alive & Thrive Vietnam, a 6-year initiative (2009–2014), has developed and incorporated elements of social franchising into government health services to provide high-quality nutrition counseling services to improve infant and young child feeding practices. One element of franchising that has not yet been implemented is fee for service, which is a potential financing mechanism for sustaining services in the long run. Objective This research aims to estimate maternal willingness to pay (WTP) for nutrition counseling services and to examine potential factors associated with their WTP. Design and methods Data were drawn from an impact evaluation survey of 2,511 women with a child <2 years old from four provinces in Vietnam. An iterative bidding technique was employed to explore individual WTP. The first bid was defined as VND 20,000 (~US$ 1), which was approximately the level of the actual service cost. Depending on the participant response, the bid increased or decreased. Finally, the respondents were asked about the highest price they would be willing to pay for the service. Results Overall, 92.6% of clients reported a need for nutrition counseling services for children <2 years. The WTP rates at bid levels of VND 5,000, 10,000, 20,000, 40,000, and 100,000 were 95.2, 94.4, 90.7, 68.9, and 33.4%, respectively. The mean and median of the maximum WTP were VND 58,500 and 50,000, respectively. In multiple regression models, WTP rates were higher among younger women, the Kinh majority group, and better educated and wealthier women. Conclusion A high demand for nutrition counseling coupled with a WTP by almost all segments of society would potentially cover costs of delivery for nutrition counseling services in Vietnam.

T he benefits of breastfeeding to infant growth and development (1), as well as long-term health outcomes (2), are well recognized. Despite its known benefits, breastfeeding practice remains sub-optimal in developing countries (3), including Vietnam (4). Breastfeeding promotion and complementary feeding education have been highlighted in the recent Lancet series on maternal and child undernutrition as a cost-effective approach to reduce the burden of child undernutrition (5).
Alive & Thrive (A&T) is an initiative funded by the Bill & Melinda Gates Foundation aiming to reduce undernutrition and death caused by sub-optimal infant and young child feeding (IYCF) practices in Bangladesh, Ethiopia, and Vietnam. In Vietnam, A&T in partnership with the National Institute of Nutrition (NIN) has incorporated elements of franchising and established the ''Mat Troi Be Tho'' (MTBT) model (or Little Sun in English) to shape and support the demand for IYCF services. The MTBT franchise model has two main purposes: 1) to standardize and monitor services to ensure that counseling on IYCF is uniform and of good quality; and 2) to build on the existing healthcare infrastructure and decentralized services promoted by the Government of Vietnam to ensure reach, utilization, and sustainability. The model also focuses on good quality and relevant training of health workers to build the capacity of those who are in positions to encourage and support mothers to practice optimal IYCF (6). Findings from a process evaluation of the A&T program showed that the nutrition counseling service based on the IYCF franchise model has significantly improved the mother's breastfeedingrelated practices and that mothers report satisfaction with the service they receive at the franchise (7).
Collecting fees for services is a critical element of a franchise model for three main reasons. First, worldwide experience shows that consumers tend to value services more when they pay for them; second, having a mechanism for generating revenues is an important part of attracting franchisees and incentivizing high performance; and finally, long-term sustainability of the franchise will depend on its ability to generate revenue (e.g. through the eventual establishment of franchise dues that are based on a percentage of revenues generated through client fees) that enable it to recover operating costs and pay for franchise advertising and promotions. In Vietnam, the MTBT franchises are nested in public sector health facilities. Under current laws, the Government of Vietnam levies no fees for health services (examining and treating diseases) provided to children B6 years of age through the public sector. Hence, MTBT franchises are reluctant to apply fees for service for franchise nutrition counseling service because they are not accustomed to these charging fees for such services, particularly for children. To date, nutrition counseling services have been provided for free. In the long run, however, the sustainability of the model would rely on establishment of a suitable financing mechanism such as user fees or inclusion of health insurance.
User fees would be heavily dependent on willingness to pay (WTP) of clients for the service. The WTP concept generally refers to the economic value of a good to a person or a household under given conditions (8Á12). Information on WTP can be useful for designers and planners in assessing economic viability of projects, setting affordable fee, evaluating policy alternatives, assessing financial sustainability, and designing socially equitable subsidies. This article aims to estimate maternal WTP for nutrition counseling services in Vietnam and to examine the factors associated with it.

Data sources
Data for this study were collected within the context of a larger evaluation of A&T's interventions in Vietnam (13). The end line round was carried out between June and August 2014 in 40 communes from four provinces (Thai Nguyen, Thanh Hoa, Quang Ngai, and Vinh Long), spanning the northern, central, and southern regions of Vietnam and are geographically representative of the 15 provinces in which A&T operates.
Sample size, sampling, and study subjects The sample size was estimated based on the expected change in rates of stunting and different IYCF indicators as a result of the A&T intervention, 80% power to detect those differences, and an alpha level of significance of 0.05. Cluster sampling (based on communes) was applied to select a total sample size of 5,800 children aged 0Á59.9  months (14). The study subjects for this study were a subset of 2,511 women who had at least one child aged 0Á24 months.

Data collection
The contingent valuation method (CVM) was applied for data collection in our study. CVM is a survey-based economic practice that asks individuals how much they are willing to pay for a particular good or service (9,15,16). In this approach, interviewees were given different bids before being asked about their maximum WTP for nutrition counseling services.
An iterative bidding technique was used to explore individual WTP that involved a scenario to describe the IYCF counseling service (Box 1) followed by a sequence of dichotomous choice questions (i.e. yes or no to the bid offered in each question) and a final open-ended question about the mothers' maximum WTP (Fig. 1). Mothers were offered the service at a starting price of VND 20,000 (about US$ 1.00), which was approximately the actual cost of the service (17). For a mother who accepted the price, the bid was raised to 40,000. If she again accepted this amount, the bid was raised to 100,000. At this level, the mother was asked about her maximum WTP either if she accepted or rejected the 100,000 bid. In a corresponding way, if the mother rejected the starting bid of 20,000, the bid was lowered to 10,000. A mother who accepted this bid was asked about her maximum WTP. But for a mother, who also rejected 10,000, the bid was lowered to 5,000 and then the mother was asked about her maximum WTP. Details of bidding questions are presented in Box 2.
In addition, we collected a number of variables that capture different domains of household socioeconomic characteristics such as the ownership of house and land, housing quality (e.g. house construction materials), and access to services (water, electricity, gas, and sanitation services). Household assets were measured by asking a number of questions on the possession of different types of durable goods, productive assets, small animals, and livestock. Wealth index quintile was derived by Principal Component Analysis (PCA) with 17 main types of durable assets and 7 housing conditions. PCA is the 'reduction dimensions' statistic technique that allows us to investigate the linear combination of assets in the household through variance maximization, in order to reflect the wealth index of households (18). Other information collected included mother's characteristics (age, education, and ethnicity).

Data management and analysis
Data were entered into EpiData management software. Statistical analyses were carried out using Stata 12. Descriptive analysis was applied to report background characteristics of study participants. A logistic regression model was performed to estimate the probability of being willing to pay for the nutrition service at the bid of VND 20,000 based on socioeconomic status of the mothers. Log-linear regression model was applied to identify socioeconomic correlates that would influence the maternal WTP amount (as the data on WTP amount are not normally distributed). A significant level of 0.05 was used.

Ethical considerations
Ethical approval was obtained from the Institutional Review Board of the Institute of the Social and Medical Studies in Hanoi and from the International Food Policy Research Institute, Washington, DC. Informed consent forms were obtained from all subjects before participating in the study.

Results
Demographic and socioeconomic characteristics of the respondents are summarized in Table 1. Average age was 27.7 (95.14), with the majority (58.7%) belonging to Box 1. The scenario to describe the IYCF counseling service Good morning/afternoon. I am ________ from the Hanoi Medical School. We are inviting you to participate in the nutrition counseling session aiming to improve infant and young child practices (IYCF) for mothers with children B24 months old. This IYCF counseling session will take place in the 'Mat Troi Be Tho' counseling room in our commune health center (CHC). In each session, the trained CHC doctors or nurse will introduce one specific topic related to breastfeeding and complementary feeding practices. They will also listen to any questions or concern you may have, then advice you the best way to feed your child or support you to try the optimal practice. The CHC doctor/nurse will use the counseling card books and other materials to help you understand and remember the information. You will also be provided with a motherÁbaby book and leaflets where you can find many useful information for feeding your child and follow the growth of your child. The IYCF counseling session will be offered for pregnant women in the third trimester and mothers with children less than 24 months. You may already participated in this IYCF counseling session, which have been provided for free so far because we got funded from the Alive & Thrive project. However, in the future, this service will no longer get funded and we will need to collect a fee from participants. We would like to ask if you are interested in this IYCF counseling session, and if yes, how much you are willing to pay for the service based on our questions below. 25Á34 year age group. Almost all of the women identified themselves as the Kinh (majority group) and 13.7% belong to ethnic minority groups. More than 90% of mothers had completed secondary school or higher, and their main occupation was farming (Â40%).
Overall, 92.6% of clients reported a need for nutrition counseling services, indicative of potential demand (results not shown). The rates of WTP for counseling services at different price levels are presented in Fig. 2. More than 90% of mothers report WTP for these services at a price of VND 20,000 or less, two thirds are willing to pay VND 40,000, and only a third are willing to pay 100,000 VND. Overall, the mean and median of the maximum WTP were VND 58,400 and 50,000, respectively (7.4% of women who did not need the nutrition counseling had WTP of zero).
Bivariate analyses showed that the WTP rates for counseling services at price levels of VND 20,000 were significantly associated with mother's age, education, occupation, and wealth index ( Table 2). These results were confirmed in a multiple logistic regression model except for mother's occupation. Younger mothers (18Á24 years) are more likely to be willing to pay for nutrition counseling services (OR: 2.26; p00.001), as are mothers belonging to the Kinh ethnic group compared to other minority ethnic groups (OR: 1.5; p00.04), and mothers with greater than a primary education (OR Â1.7, pB0.05). Additionally, those in the higher socioeconomic quintile are also more likely to be willing to pay for the nutrition counseling services. Similar results were found for other bid levels (results now shown).
Results from a multiple log-linear regression model analyzing socioeconomic correlates of the maximum WTP amount revealed three significant determinant factors: younger mothers and mothers belonging to higher wealth index quintiles were willing to pay a higher amount as compared to those in the lowest quintile (Table 3). In addition, mothers working as a salaried government employee and small traders/self-employed are willing to pay a higher amount compared to farmers. Child age is neither associated with WTP rate nor WTP amount.

Discussion
Little is known about WTP for preventive nutrition counseling services in Vietnam, or elsewhere. Several studies report WTP of specific nutrition products (e.g. lipid-based nutrient supplements) or nutrition counseling services for patients (e.g. obesity, chronic diseases) but not for receipt of preventive nutrition service for infants and young children (19Á21). The results from this study provide evidence for instituting user fees as potential financing mechanisms for ensuring sustainability of the model. These results also indicate that nutrition counseling services are highly valued by clients as evidenced from the high reported need and WTP. Consequently, the results call for incorporating preventive nutrition services for children within the health system in Vietnam. We found that 90.7% of mothers in the study areas were interested in and willing to pay for preventive nutrition counseling services at the price of VND 20,000 (ÂUS$ 1), which was about the level of the actual service cost (22). This high WTP rate reflects the fact that the nutrition counseling service was greatly appreciated by the women. Results from the process evaluation of the A&T program reported changes in quality of care at the franchises and also showed high levels of client satisfaction with the counseling skills of health providers, usefulness of the advice received, facility appearance, and counseling materials (23).
The mean of maximum WTP amount was approximately VND 58,400 (about US$ 2.7), which is approximately the cost of other routine medical checkups and  examinations at commune health centers (24). In addition, the median of maximum WTP of VND 50,000 implies the reasonably high consensus among children's caregivers on the financial value of the nutrition counseling service.
Our findings also suggest that higher education or better economic status were significantly associated with higher rate of WTP for the nutrition service. Economic status was also proven to be an independently significant correlate of the WTP amount. These findings are not novel, and are expected, but show a high degree of validity of our WTP data. The findings are in line with our hypothesis that the more people understand the importance of the nutrition counseling, the more likely they are willing to pay for the service. Further, poorer households (with lower income) should have lower demand for all goods in general, including nutrition service due to income constraint. Our findings are in line with a study in Korea, showing the younger, richer, and more-educated participants were significantly more likely to be interested in nutrition topics and willing to participate in nutrition program (20). The lower WTP among worse-off households could result in a lower chance of having access to improved preventive nutrition counseling services by this vulnerable group. This suggests that the Government of Vietnam should provide special support to people with low education and/or who belong to a worse-off group, in order to facilitate access to this important service.
We hypothesized that demands for infant and young child nutrition counseling services would be varied by Although user fees are not currently being implemented in many MTBT communes, there is a great interest in exploring this mechanism to ensure the sustainability of the model. In Vietnam, there are three provisos attached to the user fee mechanism including the removal of legal restrictions on charging for health services through the public sector; WTP by women and other caregivers for counseling services; and, finally, consideration of the costs and benefits of introducing a user fee more generally. Evidence from this study addresses the core question on WTP by mothers and caregivers. Additionally, data from the study can be used to inform further economic analysis (e.g. cost-benefit, cost-effectiveness, and cost-utility) and policy dialogue on charging user fees for preventive nutrition services for children under six in Vietnam.
We need to note some methodological constraint associated with the use of the CVMs in this study. Eliciting consumer preferences through in-person interviews was not an easy task. Several potential biases might be introduced because of the way the questions were framed, the contingent valuation scenarios, the elicitation method used, and the survey method that was conducted.
To overcome these challenges, we conducted several field visits in order to develop appropriate contingent valuation scenarios and questions. We also implemented a number of cognitive interviews to make sure that the contingent valuation scenarios and questions were easy to understand among the local people. Careful training of enumerators and field-testing also helped to ensure validity and reliability of the study findings. We used the bidding game technique because it has been the most widely used in developing countries (25Á27). The bidding game technique is more market realistic than a single open-ended question (28) and more reliable than a single dichotomous choice question (11). A disadvantage of the bidding game is the threat of startingpoint bias, where the respondent's final WTP value is not independent of the first bid prompted by the interviewer (25,29). The starting-point bias is known as 'an anchoring effect' (30), which occurs when the first bid influences the WTP amount as the respondent may consider it as a 'normal' value. We set up the starting point based on the actual cost of the service (22) as well on the experience from our pilot study. Our pilot showed that the starting bids of VND 5,000, VND 10,000, and VND 20,000 were accepted by 94, 92, and 91% of the respondents.
Our findings were also possibly affected by other biases such as 1) information bias, which occurs when the WTP depends on who does the interview, what information about the counseling service is provided, and what other information the respondents have about the service. We selected interviewers with good research experiences and did careful training based on the standard and clear scenario about the counseling service to minimize these biases; 2) strategic bias, which occurs when a respondent purposely states a WTP higher or lower than the true level. In this study, one respondent may think that it is better with a low estimated WTP in order to have the services provided for free also in the future. There may even be respondents stating a zero WTP as a protest against the idea of having user fees pay for the services. Another respondent may think that a high estimated WTP is good for securing the future of the services and therefore overestimate her true WTP, etc. We consider the risk of a strategic bias where women would overstate their true WTP to be unlikely because the scenario clearly explained to the mothers that fees may be used to fund these services in the future. So the mothers were aware of the fact that, in the future, they might have to pay these amounts. A strategic bias where mothers would underestimate their true WTP would, to the extent that it exists, mean an underestimation of the elicited WTP in this study. Since the elicited WTP is high relative to the cost of provision, the risk of this bias does not present a large problem for this study.

Conclusions
There is a high demand for nutrition counseling services and a WTP by almost all segments of society for these services, which would potentially cover cost of delivery. Findings from our study provide not only evidence for establishing user fees as a potential financing mechanism, but also demonstrate that preventive nutrition counseling services are highly valued. We recommend that decision makers in Vietnam consider incorporating preventive nutrition counseling services into routine health systems.