Assessment of Sample Frame Validity After Pilot Recruitment for a Population Based Study of Infant Food Allergy

RATIONALE: HealthNuts is a cross-sectional population-based study aiming to measure the prevalence of and risk factors for food allergy in 12 month old infants. Using data from the pilot study we assessed (1) whether the sampling frame is reflective of a representative population sample and (2) whether responders and non-responders differ significantly. METHODS: The HealthNuts study sampling frame designed comprises infants presenting for vaccination at immunisation clinics in 29 of the 31 local government areas in and around Melbourne. Regarding Aim 1, we compared the demographic features of participating families with those of the Australian Bureau of Statistics major statistical regions. Regarding Aim 2, we administered a non-responder survey to 109 parents who chose not to participate in the study. RESULTS: A high participation rate (74%) was obtained for the pilot study of 419 infants. Percentages of individuals in high income groups were higher in areas with higher SES. Rates of English only spoken in homes were also found to rank similarly to ABS statistics. Non-responders and responders were similar in sex ratio (55.7% versus 50.0% female, χ2, p = 0.32) and amount of food allergy in the immediate family (13.5% versus 13.4%, χ2, p = 0.96). However, child exposure to peanut differed (47.4% in non-responders versus 27.4%, χ2, p < 0.05). CONCLUSIONS: Pilot recruitment for the HealthNuts study demonstrated that demographics of participants reflected the population from which they were sampled. In comparing responders with non-responders, family history of allergy was not influential on study participation but prior consumption and tolerating peanuts was.