Case study: Drexel University improves cohort selection for HIV and Hepatitis C

Situation

Drexel wanted to reduce the amount of manual chart review needed to identify patients with specific disease comorbidities. Manual review was noted as taking 20 person months to find 700 patients with HIV and Hepatitis C due to the complexities of patient narratives and referrals.

Solution

Drexel used Linguamatics I2E queries to quickly identify patient groups from both structured and unstructured text, alleviating manual effort. I2E’s iterative development and graphical user interface enabled queries to be quickly developed and assessed.

Success

Using I2E, searching for disease codes and terminology took one person less than half a day to identify 1,100 patients with HIV and Hepatitis C. Identifying new disease cohorts is easy and can be quickly iterated to select new groups for study.