Enrollment experiences in a pediatric longitudinal observational study: The Environmental Determinants of Diabetes in the Young (TEDDY) study

Abstract

Objective: Our objective was to identify characteristics of infants and their families who were enrolled, refused to enroll, or were excluded from The Environmental Determinants of Diabetes in the Young (TEDDY) study.

Method: 16,435 infants screened at birth and identified as at increased genetic risk for type 1 diabetes (T1DM) were placed into one of three categories: enrolled, excluded, or refused to enroll. Enrollment, exclusion and refusal rates were compared across countries and between infants from the general population (GP) and infants with a first degree T1DM relative (FDR). A multivariate logistic model was used to identify factors associated with TEDDY enrollment.

Results: TEDDY enrollment, exclusion, and refusal rates differed by country and by GP/FDR status but reasons for refusal to enroll were similar across countries and GP/FDR populations. Sweden had the highest enrollment rate, US had the highest exclusion rate, and Finland had the highest refusal rate. FDR infants were more likely to enroll than GP infants. Inability to re-contact the family was the most common reason for exclusion. Primary reasons for refusal to enroll included protocol factors (e.g. blood draws) or family factors (e.g., too busy). Study enrollment was associated with FDR status, European country of origin, older maternal age, a singleton birth, and having another child in TEDDY.

Conclusions: Findings highlight the importance of country specific estimates for enrollment targets in longitudinal pediatric studies and suggest that enrollment estimates should be lowered when the study involves the general population, painful procedures, or makes multiple demands on families.

Figure 1

Percentage of excluded, refused and enrolled shown for all HLA eligible infants and by general population or first degree relative status and by TEDDY country.