Prevalence of food allergy in Vietnam: comparison of web-based with traditional paper-based survey

Background: Web-based surveys (WBS) are increasingly applied in epidemiological studies as an appealing alternative to traditional survey methods. Rapid data collection, reduced expenditure and ease of access to large populations are some of the clear advantages of online surveys. However, WBS are still subject to limitations in terms of sample size, response rate and other additional biases compared to traditional survey methods. In the present study, we seek to validate data on food allergy (FA) in two independent sample populations collected from a WBS, and compare it to a paper-based survey (PBS). Methods: Data collected from two survey modes were compared by hypothesis testing for independent sample population. The WBS included 1185 respondents, while the PBS included 9039 respondents. Results: Overall, the data from the WBS were comparable to the PBS conducted over the same period of time in Vietnamese adults. There were no effects of different survey modes on the lifetime prevalence of doctor-diagnosed FA (5.7%; P = 0.7795, beta = 0.05) and IgE-mediated FA (5.8%; P = 0.9590, beta = 0.05). Both surveys showed the dominance of seafood allergy in this population (up to 2.6%), followed by beef allergy. Close correlation was seen in the patterns of FAs and different clinical symptoms. The contribution of family history of allergic diseases and place of residence to FA were confirmed in both surveys. Conclusions: The consistency of the WBS results with the PBS indicates a promising application of online surveys as an economic and validated model for future epidemiological studies, specifically in developing countries.