THE LIVING WITH SICKLE CELL QUESTIONNAIRE: PSYCHOMETRIC PROPERTIES OF A NEW DISEASE- SPECIFIC SELF-REPORT MEASURE OF HEALTH-RELATED QUALITY OF LIFE FOR PEDIATRIC SICKLE CELL DISEASE

Health-related quality of life (HRQoL) remains a relatively new area of investigation in pediatric sickle cell disease (SCD).The purpose of this study was to determine the reliability, validity, and sensitivity to change of the Living with Sickle Cell Questionnaire (LSCQ), a new disease-specific HRQoL self-report instrument for children and adolescents with SCD. The LSCQ was constructed using a ‘clinimetric’ approach, which assigned items identified as important by patients to one of four traditional domains of HRQoL: Somatic Sensation, Cognitive/Psychological Functioning, Social Functioning, and Physical Functioning. Self-report versions of the PedsQL, a generic HRQoL questionnaire for children and adolescents, were co-administered in counterbalanced fashion to help establish questionnaire properties. One hundred children and adolescents with SCD (mean age: 11.7 years) participated in the study at two time points approximately two weeks apart. At the second visit, patients were asked to estimate change in their overall health since the first administration. Parents provided demographic information. High internal consistency (Cronbach’s alpha > .70) was found for each LSCQ domain at both timepoints. The LSCQ instrument’s sensitivity to patient-reported change in health between testing timepoints was only weakly demonstrated. Child demographic and disease-related factors predicted LSCQ Total Score and PedsQL About My Feelings and About School scores at Time 2 only. Confirmatory factor analyses performed at Time 1 and Time 2 showed high intercorrelations among items belonging to the same parent domain for the four-factor traditional HRQoL model, but indicated poor model fit at both time points. Follow-up Principal Component analyses for the LSCQ computed at Time 1 and Time 2 yielded four-factor solutions that accounted for 68% and 71% of the variance explained, respectively, but did not confirm the hypothesized four-domain generic HRQoL model. Rather, the four factors organized around themes of treatment, impact of fatigue, and pain and mobility, with some variation at each time point. High item loadings on the first factor suggest that a unidimensional structure might underlie HRQoL as measured by the LSCQ. Hypothesized associations between LSCQ and the PedsQL domains did not emerge as predicted. Findings are discussed in the context of disease-specific HRQoL conceptualization, instrument construction and child response style.Ph.D