Convergent and discriminant validity of quality of life measures used in burn populations

Abstract

Introduction The primary goal of this study was to investigate convergent validity, i.e. overlapping constructs, of the Burn Specific Health Scale-Brief (BSHS-B), the Short Form-36 items Health Survey (SF-36) and the European Quality Of Life Five Dimensions (EQ-5D) within the International Classification of Functioning Disability and Health (ICF) framework. A secondary goal was to examine the discriminant validity of the questionnaires according to burn severity (no surgery versus 1 or more surgeries). Methods A prospective multi-centre study in adult patients with burns was conducted. At the 9 months assessment, two generic questionnaires, i.e. the SF-36 and EQ-5D, and the BSHS-B were completed. Pearson correlations were used to evaluate convergent validity. Linear discriminant analysis was used to evaluate discriminant validity. Results At 9 months post-burn data from 184 persons were available of which 131 (71%) were male, mean TBSA burned was 11.8% (SD = 10.2). Sixty five (34%) patients did not need surgery, 128 (66%) patients required one or more surgeries. Higher convergence was shown between the generic SF-36 and the condition specific BSHS-B whereas the EQ-5D showed lower convergence with the BSHS-B especially in the domain Activity. The generic scales discriminated across all scales whereas not all BSHS-B scales were able to differentiate problem levels across burn severity groups. Conclusion This study demonstrates that the ICF is useful to classify scales in order to identify overlapping areas as well as to uncover gaps in relation to patient reported outcomes. Both the SF-36 and EQ-5D showed the ability to distinguish levels of functioning across burn severity groups. As the BSHS-B performed less well and relevant domains of functioning were not addressed, there is room for improvement and modification of this condition specific questionnaire to better capture burn patients’ functioning.