Barrier analysis for daily fruit and vegetable consumption among low income consumers in Hanoi, Vietnam.

Barrier analysis is completed with at least 45 doers of the behavior of interest and 45 non-doers who do not practice the behaviour. In this case, the behaviour of interest was the consumption of an adequate quantity and diversity of fruits and vegetables by low income adults in the focal neighbourhoods. Doers and non-doers were selected from the list of 300 lower-income households in the focal neighborhoods using screening questions. Respondents were asked to recount the fruits and vegetables they consumed in the past 24 hours and whether the amount they consumed was typical for them or not. Trained enumerators from NIN converted the reported serving sizes into an approximate number of grams using standardized conversion factors. Based on the answers to the screening questions, the respondents were classified as ‘doers’, ‘non-doers’ or ‘do not interview’. Doers met fruit target (200 g fruit) and vegetable target (200g vegetable from at least two MDDW groups) and normally consumed the same amount of fruits and vegetables or more. Nondoers did not meet the fruit target, did not meet the vegetable target and normally consume the same amount or less. We used a quota sampling approach in which households were contacted in randomized order. The gender to be interviewed for each household was randomly assigned so that we did not include men and women from the same household in the sample. The respondents were contacted by telephone by representatives of the district health centres and the screening questions were completed over the phone if the respondent was comfortable. In other cases, based on the preference of the respondents, the enumerators made an appointment to visit their household to complete the screening questions in person. Those who were classified as doers and non-doers were invited to complete the barrier analysis survey in person at the local health centre and an appointment was arranged with several options for times ranging from the early morning to early evening. The estimated self-reported portion size of doers was 348 g of fruit and 297 g of vegetables from two groups in the past 24 hours. By contrast, the estimated self-reported portion size of non-doers was 101 g of fruit and 128 g of vegetables in the past 24 hours.