The GABRIEL Advanced Surveys: study design, participation and evaluation of bias

Summary Genuneit J, Büchele G, Waser M, Kovacs K, Debinska A, Boznanski A, Strunz‐Lehner C, Horak E, Cullinan P, Heederik D, Braun‐Fahrländer C, von Mutius E, the GABRIELA Study Group. The GABRIEL Advanced Surveys: study design, participation and evaluation of bias. Paediatric and Perinatal Epidemiology 2011; 25: 436–447. Exposure to farming environments has been shown to protect substantially against asthma and atopic disease across Europe and in other parts of the world. The GABRIEL Advanced Surveys (GABRIELA) were conducted to determine factors in farming environments which are fundamental to protecting against asthma and atopic disease. The GABRIEL Advanced Surveys have a multi‐phase stratified design. In a first‐screening phase, a comprehensive population‐based survey was conducted to assess the prevalence of exposure to farming environments and of asthma and atopic diseases (n = 103 219). The second phase was designed to ascertain detailed exposure to farming environments and to collect biomaterial and environmental samples in a stratified random sample of phase 1 participants (n = 15 255). A third phase was carried out in a further stratified sample only in Bavaria, southern Germany, aiming at in‐depth respiratory disease and exposure assessment including extensive environmental sampling (n = 895). Participation rates in phase 1 were around 60% but only about half of the participating study population consented to further study modules in phase 2. We found that consenting behaviour was related to familial allergies, high parental education, wheeze, doctor diagnosed asthma and rhinoconjunctivitis, and to a lesser extent to exposure to farming environments. The association of exposure to farm environments with asthma or rhinoconjunctivitis was not biased by participation or consenting behaviour. The GABRIEL Advanced Surveys are one of the largest studies to shed light on the protective ‘farm effect’ on asthma and atopic disease. Bias with regard to the main study question was able to be ruled out by representativeness and high participation rates in phases 2 and 3. The GABRIEL Advanced Surveys have created extensive collections of questionnaire data, biomaterial and environmental samples promising new insights into this area of research.