Emergency department diagnosis and guideline concordant care in gastric cancer.

Abstract

300 Background: Roughly half of new cases of gastric cancer are diagnosed in the Emergency Department (ED-dx). Patients with ED-dx have poorer clinical outcomes than those who are not ED-dx. However, the reasons for this difference are unclear. We aim to explore whether ED-dx is associated with receipt of Guideline-Concordant Therapy (GCT) for patients with gastric cancer in a large New Jersey-based health system. Methods: Gastric cancer patients, identified via single institution cancer registry, were included in a retrospective chart review from 2016-2022. Univariable and multivariable analyses were performed to evaluate the association between sociodemographic factors and utilization of GCT as defined by the NCCN by cancer stage. Logistic regression was used to estimate covariate adjusted odds ratios for the association of ED-dx with GCT. Results: A total of 331 patients were included, of which 154 (46.5%) were ED-dx and 69 (20.8%) did not receive GCT. Among patients who did not receive GCT, 60.9% were ED-dx, compared to 42.8% of those who got GCT (p=0.007). Despite adjusting for demographic factors and comorbidities, ED-dx patients had a higher probability of not receiving treatment (OR 2.13 [1.19-3.83]). However, this association was no longer significant when socioeconomic conditions (OR 1.69, [0.91-3.15]) and clinicopathologic variables (OR 1.27, [0.65-2.50]) were also accounted for. Factors most strongly associated with not receiving GCT were age >75 years (OR 7.40 [2.46-22.26]), Charlson comorbidity index > 3 (4.22 [1.61-11.04]), having Medicaid or no health insurance (3.49 [1.36-8.98]), living in areas with moderate compared to low area deprivation index (3.32 [1.37-8.04]), having stage IV disease (2.33 [1.06-5.15]) and having gastric cancer histology other than adenocarcinoma (OR 2.63 [1.11-6.23]). Conclusions: Lower rates of oncologic treatment compared to those diagnosed in other settings are largely attributable to differences in socioeconomic and clinicopathologic variables, not whether patients were ED-dx. Further investigation is warranted to understand the poor outcomes observed among ED-dx gastric cancer patients with a particular focus on modifiable or early identifiable factors to improve outcomes. Multivariable logistic regression for the association of being diagnosed in the emergency department and receiving GCT. Model 1 Model 2 Model 3 OR (95%CI) OR (95%CI) OR (95%CI) Diagnosis at Emergency Department 2.13 (1.19-3.83) 1.69 (0.91-3.15) 1.27 (0.65-2.50) Model 1: Adjusted for age, sex, and ethnicity, Charlson comorbidity index. Model 2: Adjusted for age, sex, ethnicity, Charlson comorbidity index, area deprivation index, and insurance. Model 3: Adjusted for age, sex, ethnicity, Charlson comorbidity index, area deprivation index, insurance, symptoms, stage, and gastric cancer histology type.