e18138 Background: Patients diagnosed with leukemia and lymphoma typically present with nonspecific symptoms, making a timely diagnosis difficult. Little is known about factors associated with delays in diagnosis. We hypothesized that age, minority race/ethnicity, and low income are associated with greater time to diagnosis. Methods: Using the OptumLabs Data Warehouse, which includes claims data for privately insured enrollees in a large US health plan, we identified 17,536 pediatric (0-14 y), adolescent (15-21 y), and young adult (22-39 y) patients diagnosed with acute leukemia or lymphoma between 2001-17. Using this retrospective cohort, potential cancer-related symptoms occurring up to 6 months pre-diagnosis were identified. Delay was defined as > 3 months from symptom onset to diagnosis. Contingency table analysis with chi-squared tests and unconditional logistic regression were used to estimate the association between sociodemographic factors and delays in diagnosis. Results: Seventy-eight percent of patients had a diagnosis of a cancer-related symptom in the 6 months prior to diagnosis. The most common presenting symptoms were lymphadenopathy, fever, and cytopenias. The median days to diagnosis was longer in young adults (93) than children (86) or adolescents (81) (p = < 0.0001). For pediatric v. AYA patients, median days to diagnosis differed for those with constitutional symptoms (18 v. 37, p = < 0.001), infectious symptoms (93 v. 74, p = < 0.001), and cytopenias (11 v. 22, p = < 0.001). Multivariable analysis identified younger age, female sex, and low household income to be significantly associated with delays in diagnosis (table below). Conclusions: In this large cohort of privately insured patients, adolescents had the shortest time to diagnosis. We saw no disparities by race/ethnicity or education but observed that low income ( < $40K) and female patients had greater odds of delays in diagnosis. [Table: see text]

e21515 Background: While cancer is the leading cause of non-accidental death in children and AYAs, factors associated with delays in diagnosis in young patients with cancer are poorly understood; we sought to fill this knowledge gap. Methods: Using the OptumLabs Data Warehouse’s claims data for commercially insured enrollees in a large US health plan—we identified pediatric [0-14 years (y)] and AYA (15–39 y) patients diagnosed with soft tissue sarcomas (STS), bone tumors (BT) and germ cell tumors (GCT) during 2001–17 and continuously enrolled 6 months prior to diagnosis. Time to diagnosis was calculated as days between first medical encounter associated with a possible cancer symptom and diagnosis date. Median times from first symptom to diagnosis were compared using Wilcoxon Rank Sum test. Multivariable logistic regression identified sociodemographic and clinical factors associated with longer time ( > 3 months) from symptom to diagnosis. Results: Of the 11,395 patients, 86% presented to medical care with symptoms prior to diagnosis [STS: 2,228 (89%); BT: 1,565 (87%); GCT: 5,904 (84%)]. The most common symptoms were pain and swelling. STS had the longest median days to diagnosis (92), followed by BT (91) and GCT (49). There was a significant difference (p < 0.001) in median days to diagnosis by age for BT (0–14y: 69; 15–21y: 77; 22–39y: 105) and GCT (0–14y: 96; 15–21y: 34; 22–39y: 49), but not for STS. Patients in households with ≥ a college degree (OR 1.96, 95% CI 1.06–3.64, vs < high school) and seeing a specialist (excluding oncologists) (OR 2.54, CI 2.03–3.19, vs primary care) at first symptom presentation was associated with a longer delay, while older age (22–39y: OR 0.77, CI 0.63–0.94, vs 0-14y) and male sex (OR 0.58, CI 0.51–0.66) were associated with a shorter delay in diagnosis. Conclusions: This study demonstrates that, in a commercially insured population, time to diagnosis varies by cancer type and is impacted by clinical and sociodemographic factors. Shorter time to diagnosis may represent delays in presenting to medical care or more acute presentations of symptoms, therefore patient-reported symptoms and barriers to care data should be collected to better define strategies to reduce delays in diagnosis.