Implementation of the Short-Form 36 (SF-36) in childhood cancer survivors: An analysis on measurement properties across 5 European countries

PurposeThe Short Form-36 (SF-36) is widely used in many research contexts and cultures to assess health-related quality of life (HRQOL). We investigated the measurement properties of the SF-36 in a large cohort study among childhood cancer survivors living in 5 European countries. MethodsThe PanCareLIFE project includes adult survivors of childhood cancer living in the Czech Republic, France, Germany, The Netherlands, and Switzerland. We invited 19,268 survivors aged >18 years, and 10,077 (53%) returned questionnaires. Of these, 9,871 (98%) were included in the analyses. We assessed HRQOL with the SF-36 version 1 (V1) or version 2 (V2) and investigated its performance. We analysed data completeness, floor and ceiling effects, item-internal consistency, item-discriminant validity, reliability, and scaling assumptions focusing on country-and version-specific differences. ResultsData completeness was high but differed between countries (90% in the Czech Republic to 96% in France). Floor effects were negligible, but ceiling effects in 5 out of 8 SF-36 scales were present (50-70%). V2 had slightly fewer ceiling effects than V1 in 2 scales. Item-internal consistency, item-discriminant validity, and reliability were good. Scaling assumptions were met, with 4 scales representing strong mental health content, and the other 4 representing strong physical health content. ConclusionThe SF-36 (V1 and V2) provided robust measurement properties among childhood cancer survivors, and differences in study design across the 5 European countries did not affect measurement properties. Researchers need to take into account that ceiling effects exist which might limit sensitivity among participants who fared particularly well. PLAIN ENGLISH SUMMARYThis study investigated how well the Short Form-36 (SF-36), a commonly used tool for assessing health-related quality of life, performs in childhood cancer survivors from five different European countries. We looked at the amount of survivors answering to all of the SF-36 questions, how many indicated the most positive or the most negative answer category, whether the questions were consistent within the different aspects of the questionnaire and how consistent and dependable it measured quality of life in the survivors. We found that overall, the SF-36 measured health-related quality of life well in all the countries. This is reassuring for researchers considering its use in studies involving multiple centers across Europe.