Impact of diagnostic workup location on time to treatment for patients with non-small cell lung cancer seen in the emergency department.

Abstract

e13518 Background: Most non-small cell lung cancer (NSCLC) diagnoses begin in the emergency department (ED), despite ED diagnoses being associated with higher mortality compared with those made outside the ED. Patients with new NSCLC are often admitted from the ED for complications or to expedite workup, as earlier treatment correlates with better survival. However, hospital stays are resource-intensive and may not reduce time to treatment initiation (TTI). We compared characteristics and TTI for patients with an ED diagnosis of NSCLC who underwent inpatient versus outpatient workups. Methods: We identified 5443 patients at Kaiser Permanente Northern California, a large integrated health system, with new NSCLC diagnosed within 30 days of an ED visit from 1/1/2018 to 12/31/2021. We randomly sampled 600 for chart review and excluded patients who died within 30 days or lacked pathologic confirmation. Using Cox proportional hazards, we adjusted for age, sex, race, neighborhood deprivation, comorbidities, and stage to assess the impact of workup location on TTI for patients receiving cancer-directed treatment within 30 days. Results: Of 480 patients included, 286 (60%) underwent inpatient workups and 194 (40%) had outpatient workups. In unadjusted analyses, hospitalized patients were more likely to have stage IV disease (69% vs. 53%, p< 0.001) and to receive treatment within 30 days (40% vs. 29%, p = 0.005), but less likely to be treated at 1 year (74% vs. 85%, p = 0.005). After adjustment, there was no difference in TTI between groups for all patients or patients with stage IV disease treated within 30 days. Conclusions: Patients admitted after an ED diagnosis of NSCLC had similar adjusted TTI to those who underwent outpatient workups. Our findings suggest that for patients who may receive early treatment and do not require hospitalization, outpatient NSCLC workup does not delay treatment initiation. Identifying patients appropriate for outpatient workup could help preserve hospital resources without impeding timely treatment. Characteristics and outcomes by workup location after ED diagnosis of NSCLC. Univariate analysis Total 1 n = 480 Hospital 1 n = 286 Outpatient 1 n = 194 p-value 2 Age, mean (SD) 71.6 (11.3) 71.2 (11.8) 72.2 (10.7) 0.2 Male sex 205 (43) 119 (42) 86 (44) 0.6 Neighborhood Deprivation Index (NPI): 4th quartile (most deprived) 115 (24) 67 (23) 48 (35) 0.7 Charlson Comorbidity Index (CCI) ≥5 137 (29) 90 (31) 47 (24) 0.085 Stage IV disease 299 (62) 196 (69) 103 (53) <0.001 Multivariate analysis Hazard Ratio 95% CI p-value 3 Any treatment at 30 days 0.90 0.72, 1.14 0.377 Any treatment at 30 days, stage IV only 0.89 0.66, 1.19 0.428 1 n (%), unless otherwise stated. 2 Chi-square test (categorical variables) or Wilcoxon rank-sum test (continuous variables), unadjusted. 3 Cox proportional hazard model comparing outpatient to hospital; adjusted for age, sex, race, NPI, CCI, and stage.