Empirical Application of the Sentinel EHR and Claims Data Partner Network to Enhance ARIA Sufficient Inferential Requests and Atypical Descriptive Requests (Use Case Package 2)

Health insurance claims data currently form the backbone of the Sentinel Distributed Database (SDD), owing to their complete capture of outpatient pharmacy dispensing records, medical encounters, and hospitalizations during well-defined periods of health plan enrollment. Analyses conducted in the SDD as part of Sentinel’s Active Risk Identification and Analysis (ARIA) system provide vital information to the FDA to aid in regulatory decision making. However, in many circumstances uncertainties remain at the conclusion of an ARIA analysis, often attributed to lack of data availability in insurance claims for important variables for addressing the regulatory question.

The Sentinel Innovation Center (IC) has organized a Development Network to enable access to electronic health records (EHRs) and linked claims data including information from unstructured and structured EHRs. Timely access to granular EHR data can be valuable for conducting supporting analysis to strengthen the validity of ARIA analyses conducted in the SDD using health insurance claims data alone. Access to granular free-text data from EHRs is also important to address atypical regulatory questions.

This project will develop empirical case studies in the following 5 aims to address targeted use cases in the IC Development Network:

(1) Rapid confounder balance evaluation of factors unmeasured in Sentinel claims data

(2) Correcting claims analyses for unmeasured confounding using subset calibration tools

(3) Real-time validation of code-based algorithms

(4) Identifying use of cannabis-derived products (CDP) from free-text notes 

(5) Expand on a principled quantitative bias analysis (QBA) at the design stage that could allow for better understanding of the uncertainties associated with potential unmeasured confounding

This Sentinel Innovation Center (IC) project is led by Mass General Brigham, Kaiser Permanente Washington Health Research Institute and Vanderbilt University Medical Center, in collaboration with Harvard Pilgrim Health Care Institute and Harvard Medical School.