Table 1 Summary of methodology of case-control studies
Reference (country)Case selectionControl selectionExposure measurementConfounding factors consideredSample sizeAge (years)
Stene et al (2000)21 (Norway, Vest-Agder)National Childhood Diabetes Register (NCDR)Random selection from official population registerQuestionnaireAge, sex, duration of exclusive breastfeeding, maternal education, maternal use of vitamin D supplements in pregnancy78 cases, 980 controlsCases: mean 11.1, SD 3.6; controls: mean 8.5, SD 4.7
EURODIAB20
    AustriaNCDRRandom sample from schoolsQuestionnaireCombined results adjusted for age, sex, low birth weight, short duration of breast feeding, maternal age and study centre95 cases, 346 controlsNot specified
    BulgariaNCDRRandom sample from schools and polyclinicsInterview126 cases, 430 controlsNot specified
    LatviaNCDRPopulation registerInterview109 cases, 293 controlsNot specified
    LithuaniaNCDRRandom sample from polyclinicsQuestionnaire111 cases, 250 controlsNot specified
    LuxembourgNCDRRandom sample from preschools and schoolsInterview55 cases, 176 controlsNot specified
    RomaniaNCDRRandom sample from health service registerInterview82 cases, 276 controlsNot specified
    N. IrelandNCDRGeneral practitioner registerQuestionnaire168 cases, 419 controlsNot specified
Stene et al (2003),22 (Norway)NCDRRandom selection from populationQuestionnaireAge, sex, breastfeeding, maternal age, education and smoking during pregnancy, time of weaning, number of siblings, family history of type 1 diabetes545 cases, 1668 controlsCases: mean 10.9, SD 3.4; controls: mean 9.3, SD 4.1
Tenconi et al26 (Italy, Pavia province)NCDRRandom selection from non-diabetic hospitalised patientsInterviewAge, sex, residency, family history of type 1 diabetes, drugs taken during pregnancy, type of delivery, early feeding, neonatal and most common childhood diseases, history of surgical operations, severe infections159 cases, 318 controlsCases: mean 23.24, SD 9.36; controls: mean 23.38, SD 9.64