Project Summary

One of PCORI’s goals is to improve the methods that researchers use for patient-centered outcomes research. PCORI funds methods projects like this one to better understand and advance the use of research methods that improve the strength and quality of comparative effectiveness research.

This research project is in progress. PCORI will post the research findings on this page within 90 days after the results are final.

What is the project about?

Patients may not always take medicines as directed. If patients take some but not all of their medicine or stop taking their medicine too soon, the medicine may not work to treat a health problem.

Researchers want to learn why and how patients take medicines in order to find ways to support patients in taking their medicines as directed. One way to study this problem is by looking at doctors’ notes in electronic health records, or EHRs. But researchers can’t easily analyze these notes with current methods.

In this study, the research team is creating a method for converting doctors’ EHR notes into a more usable format for research. The new method uses natural language processing, or NLP. In NLP, computer programs interpret and code written language.

How can this project help improve research methods?

Results from this study may help make doctors’ EHR notes more useful for looking at how patients take medicine.

What is the research team doing?

The research team is developing NLP methods to capture information from doctors’ notes about how patients take medicine. First, the team is working with patients to learn about how patients take medicine. For example, patients are helping the team understand what information patients usually share or don’t share with doctors about taking their medicine. Then the team is developing an NLP computer program to identify patterns in the information about how patients take medicine in doctors’ notes. Finally, the team is testing the NLP program using existing clinical data on diabetes and depression treatment. The team wants to see how well the NLP program works compared with existing ways of measuring how patients take medicine.

Research methods at a glance

Design Elements Description
Goal
  1. Develop a patient-centered model of medicine adherence using EHR text
  2. Develop an NLP computer program to identify patterns in medicine adherence data from EHR text
  3. Evaluate the NLP program using existing diabetes and depression treatment adherence data 
Approach Qualitative methods; natural language processing; machine learning; modeling

Project Information

Kirk Roberts, PhD
The University of Texas Health Science Center at Houston
$954,189
NLP for Medication Adherence: Complex Semantics and Negation

Key Dates

November 2018
March 2023
2018

Study Registration Information

Tags

State State The state where the project originates, or where the primary institution or organization is located. View Glossary
Last updated: March 4, 2022