Trial Attrition

Wave 5 response rates were lower among boys than girls. Wave 5 participants had lower delinquent behaviour scores than non-participants, but did not differ in emotional health, psychosomatic problems, or alcohol and drug use. Dropouts differed from completers in feedback use, motivation source, relevance rating, and overall assessment, but not in prior Delphi experience, information seeking, or consensus expectations. The final analysed sample was severely underpowered, with about 10% power instead of the planned 90%. Attrition was interpreted as influenced by both methodological design and contextual constraints. Wave 5 participation dropped sharply to 949 adolescents, or 50.4%. Dropouts did not attribute withdrawal to insufficient subject-matter expertise. Questionnaire complexity, unclear wording, and difficult feedback can increase cognitive burden and dropout risk. Illness, scheduling conflicts, and email delivery problems were treated as contextual contributors to dropout beyond researchers' control.