The Survivor Unmet Needs Survey (SUNS) for haematological cancer survivors: A cross-sectional study assessing the relevance and psychometric properties

Abstract

Background: Relevant and psychometrically sound needs assessment tools are necessary for accurate assessment of haematological cancer survivors unmet needs. No previous study has developed nor psychometrically evaluated a comprehensive needs assessment tool for use with population-based samples of haematological cancer survivors. This study aimed to assess the validity and reliability of the Survivor Unmet Needs Survey (SUNS) with haematological cancer survivors. Methods. The relevance, content and face validity of the SUNS to haematological cancer survivors was assessed using qualitative interviews. Psychometric evaluation was conducted using data collected from haematological cancer survivors, aged 18-80 years at recruitment and recruited from four Australian cancer registries. Construct, convergent and discriminant validity; internal reliability and floor and ceiling effects were assessed. A second survey was completed by a sub-sample of survivors recruited from two of the four registries to assess test-retest reliability. Results: Results from 17 qualitative interviews confirmed the relevance, face and content validity of the original items of the SUNS for use with haematological cancer survivors. Overall, 1,957 eligible haematological cancer survivors were contacted by the cancer registries. Of these 1,280 were sent a survey, and 715 returned a survey (37% of eligible survivors contacted and 56% of survivors sent a survey). A total of 529 survivors completed all 89 items of the SUNS and were included in the exploratory factor analysis. Exploratory factor analysis supported the original five-factor structure of the SUNS. Evidence for convergent validity was established, with all five domains of the SUNS illustrating a moderate positive correlation with all three subscales of the Depression Anxiety and Stress Scale (DASS-21). All Cronbach's alpha values were above 0.9 and all corrected item-total correlations were acceptable (>0.2). Criteria for discriminant validity was not met, with only 10 of the 15 (67%) a-priori hypotheses supported. Test-retest reliability was acceptable for 40 of the 89 items (45%) and for three of the five domains. Significant floor effects were evident for all five domains. Conclusions: The SUNS demonstrates evidence for multiple features of validity and reliability as a measure of unmet needs for haematological cancer survivors. However, evidence supporting some psychometric properties was limited. © 2014 Hall et al.; licensee BioMed Central Ltd.