Barriers to survivorship care in the Consortium for New England Childhood Cancer Survivors (CONNECCS).

10035 Background: Although long-term adverse consequences of childhood cancer treatment may be mitigated by screening, prevention, and interventions, many survivors do not take advantage of survivorship care. We hypothesized that patients who are at risk for poor compliance with long-term follow-up are identifiable at diagnosis. Methods: To identify factors associated with poor follow-up compliance, 7 CONNECCS institutions evaluated a childhood acute lymphoblastic leukemia (ALL) survivor cohort diagnosed 1996-99. Data collected included: diagnosis year, age, race, ethnicity, gender, insurance, distance from center, CNS disease, and risk classification. Primary endpoints were compliance with 5 and 10-year follow-up. Differences in compliance were tested using chi-squared or t-tests. Logistic regression (including institution as a clustering variable) was used to calculate adjusted odds ratios (OR). Results: At diagnosis, the 358 ALL patients were: female (47%), age= 6.5 + 4.6 years, white/non-Hispanic (84%), black non-Hispanic (7%), high-risk (52%), CNS involvement (10%), privately insured (68%). Private insurance (OR 4.0; 95% CI 2.1-7.8) significantly increased the odds of 5-year compliance. Compliance with 10-year follow-up increased with private insurance (OR 3.3; 95% CI 1.4-8.1) but decreased with CNS disease (OR 0.36; 95% CI 0.31- 0.42) and with years of age (OR 0.93; 95% CI 0.88- 0.96). Conclusions: We evaluated predictors of long-term follow-up based on disease/demographic characteristics at diagnosis to identify cohorts in need of early interventions. In this regional cohort, patients from lower socioeconomic background (without private insurance) at diagnosis were less likely to participate in long-term follow-up care at 5 and 10 years from diagnosis. Older survivors and those with CNS disease were less likely to be in follow-up at 10 years. Future studies should investigate reasons why follow-up compliance is affected by 1) private insurance at diagnosis, 2) older age, and 3) CNS disease. Remediable causes might include: understanding of risk, adolescence/young adult transitions, and healthcare access.