Abstract
OBJECTIVE: Cancer survivors report deficits in social functioning even years after completing treatment. Commonly used measures of social functioning provide incomplete understanding of survivors' social behavior. This study describes social activities of survivors and evaluates the psychometric properties of the Social Activity Log (SAL) in a cohort of long-term survivors of hematopoietic stem cell transplantation (HSCT) for cancer. METHODS: One hundred and two (5-20 year) survivors completed the SAL, Short-Form-36 Health Survey (SF-36), and other patient-reported outcomes. Principal components analysis determined the factor structure of the SAL along with correlations and regressions to establish validity. RESULTS: Principal component analysis yielded three factors in the SAL: 'non-contact events' (e.g. telephone calls), 'regular events' (e.g. played cards), and 'special events' (e.g. concerts), which explained 59% of the total variance. The SAL possessed good internal consistency (Cronbach's alpha=0.82). SF-36 social function and SAL were moderately correlated (r=0.31). In linear regressions, physical function and depression explained 16% of the variance in the SAL (P<0.001), while physical function, depression, and fatigue predicted 55% of the variance in SF-36 social function (P<0.001). CONCLUSIONS: Results support the use of the SAL as a measure of social activity in cancer survivors who received HSCT. Although the SAL is designed to measure social behaviors, SF-36 social function assesses subjective experience and is more strongly associated with depression and fatigue. The SAL appears to be a promising tool to understand the behavioral social deficits reported by long-term survivors of cancer.