Methods/Survey Development. Sixty-four survivors between the ages of 18-35 years and 21 years of age or younger when diagnosed with cancer responded to newspaper advertisements and flyers posted in four pediatric cancer centers announcing a “quality of life” study. Volunteers participated in semi-structured interviews lasting 30-120 minutes and responded to one question – “How has having had cancer affected your life?” while an interviewer continuously probed for effects in physical, psychological, social, and spiritual/religious domains. All interviews were transcribed and subject to analyses that involved organizing data into meaningful sub-categories from which to develop a set of candidate survey items. These analytic procedures involved open coding and constant comparative analysis, resulting in 86 mutually exclusive codes from which investigators derived 250 candidate items. These items were then subject to expert review (13 clinicians and researchers; 17 cancer survivors/patient advocates) and cognitive testing (n=13 survivors) for purposes of item revision and reduction.
Results/Psychometric Evaluation. Ultimately, 82 candidate items represented respondents' perceptions of cancer's impact upon various aspects of their lives, and were administered in a comprehensive multi-site quality of life survey. Based on responses from 519 eligible respondents, exploratory factor analyses yielded eight factors/subscales composed of 45 items describing distinct constructs: Life Outlook, Body/Health, Talking with Parents, Identity, Cognition, Health Care, Socializing and Finances. Intraclass correlation coefficients for test-retest reliability ranged from 0.62-0.91 across the subscales, and Cronbach's alpha as measure of internal consistency reliability ranged from 0.70-0.86. Assessment of concurrent validity involved examination of correlations between the IOC-YA subscales and standardized measures of psychological well-being and QOL, and resulted in statistically significant directional associations consistent with a priori hypotheses. Examination of IOC-YA subscales across select demographic and health status variables also derived convergent and discriminant results consistent with a priori hypotheses.
Conclusion & Implications. This report represents the initiation of an effort to identify and define specific areas of concern for young adult survivors of childhood cancer. Its product – a field tested and psychometrically sound instrument – will enable investigators to assess aspects of cancer's impact that have yet to be well established and measured empirically. The IOC-YA also has the potential to assist clinicians in identifying salient survivorship issues in clinical and community settings.