Development of an Osteogenesis Imperfecta (OI) specific Quality of Life measure

Osteogenesis Imperfecta (OI) is a hereditary disorder effecting approximately 1 in 20,000 births. Symptoms include; low bone mass, recurrent fractures, varying degrees of short stature and deformity. There is currently no disease specific quality of life (QoL) measure for children with OI. This study uses a mixed methods approach to develop a QoL measure for the paediatric OI population. Patient reported outcome measure development is an iterative process, moving back and forth between concept elicitation, questionnaire development, pre-testing and psychometric analysis. In order to encourage a balance between good content validity, alongside promoting a robust, reliable and responsive measure, the methods chosen involved several stages: • Literature review to ensure no suitable QoL measure already existed and to begin eliciting themes. • Interview and focus groups with the target population to uncover relevant concepts, develop a conceptual framework and subsequently validate themes. • Questionnaire development; transforming themes into items, using the children’s’ language to ensure high content validity and acceptability. • Pre-testing the instrument alongside a sample of the OI population, making revisions as required. • Psychometric evaluation to assess validity, reliability and responsiveness of the questionnaire, informing potential item elimination and revision of the measure. Interviews and focus groups with the target population uncovered six main themes when describing QoL in children with OI; being safe and careful, reduced function, pain, fear, independence and isolation. These themes and related sub themes informed the development of the conceptual framework, which alongside the children’s own thematic based quotes, was used to develop the OIQoL. Pre-testing of the OIQoL highlighted logistical issues and understanding, which lead to revisions of the initial version. The final version underwent field testing; concerns around construct validity and internal consistency reliability highlighted the need to undertake further psychometric techniques on a larger cohort prior to item elimination