Integration of administrative claims and clinical data to advance real-world evidence generation in oncology.

Abstract

Objective: The objective of this study is to evaluate capabilities of the Generating Evidence Excellence (GEx) research environment using Non-Small Cell Lung Cancer (NSCLC) as a use-case.

Methods: Patients with incident NSCLC were identified in GEx from 01/01/2015 to 12/31/2019. Two GEx data containers were used: (1) claims from the Healthcare Integrated Research Database (HIRD) linked to clinical data from a Cancer Care Quality Program (CCQP) (HIRD+CCQP); (2) claims data from the HIRD and a second large commercially insured claims dataset (CD2) linked to electronic medical records from IntrinsiQ Specialty Solutions (IQSS) (HIRD+CD2 + IQSS). Patient demographics and all-cause mortality were contrasted with the American Community Survey and the United States Cancer Statistics (USCS) public use database. Descriptive analyses were completed via SAS Enterprise Guide 7.12 and SEER*stat.

Results: The study sample comprised of 6,233 (HIRD+CCQP) and 1,176 (HIRD+ CD2 + IQSS) patients with NSCLC. Both GEx data containers had similar patient demographics with a mean age 63 years and 71% White, while 70% of USCS patients were ≥65 years and 85% White. Most patients resided in the South/Midwest regions of the US (74% (HIRD+CCQP) and 86% (HIRD+CD2 + IQSS). Patients in GEx over-represented the working-age group, i.e. 65% were between 18 and 65 years old. Stage IV disease was frequently recorded at diagnosis (59.4% HIRD+CCQP, 44.0% HIRD+CD2 + IQSS, 46.1% USCS). All-cause mortality rate was 38.8 (HIRD+CCQP) and 27.3 (HIRD+CD2 + IQSS) per 100 person-years.

Conclusions: The analysis of NSCLC as a use-case demonstrated that GEx is a valuable data ecosystem with a generalizable working-age patient population and comprehensive data to facilitate timely real-world evidence generation in oncology.