• Download Summary

PCORI funded the development of PCORnet^®, the National Patient-Centered Clinical Research Network, to make research faster, easier, and less costly to conduct. PCORnet is made up of Partner Networks of healthcare systems, patients and communities, and health plans that harness the power of large amounts of health data.

PCORI supports brief, descriptive projects to assess the feasibility of conducting research using data gathered and shared securely through PCORnet. This project is one of several designed to test the network while addressing priorities identified by PCORI and its stakeholders.

What was the project about?

Patient-reported outcome measures, or PROMs, are surveys that ask patients how they feel and what activities they can do. PCORnet created a shared database system that includes data from patients’ electronic health records, or EHRs. In this project, the team helped three medical centers start collecting PROMs about heart failure. The team uploaded the data to PCORnet to test ways of using the data to improve patient care.

What were the results?

Could the project team help collect PROMs from patients? The team successfully collected PROMs from patients at each medical center and uploaded that data into the PCORnet Common Data Model (CDM).

How did patients and doctors feel about using the PROMs? Patients felt that PROMs helped them talk about their symptoms with their doctors. Doctors felt that PROMs helped them provide better care for patients.

How were patients who filled out PROMs different from patients who didn’t fill out these surveys? Patients who completed PROMs were younger, but they did not differ in gender, race, or ethnicity. Patients who didn’t complete PROMs were sicker and had more medical problems.

Which patients had more symptoms of heart failure? Patients with more symptoms of heart failure were more likely to be women and have other health problems, like obesity and depression.

Who was in the project?

The project team gathered PROM data from 1,054 patients from three medical centers in the United States between 2019 and 2020. The team also gathered data on patient characteristics from patients who did not complete PROMs during the same time.

What did the project team do?

First, the project team helped the three medical centers collect PROM data from patients using patients’ EHRs. The team then uploaded the data to PCORnet. The team also interviewed 10 patients with heart failure and 8 doctors at the medical centers to learn how they felt about using PROMs. Using the data, the team looked at differences between patients who filled out PROMs and those who didn’t. The team also looked at the characteristics of patients who had more symptoms of heart failure.

What were the limits of the project?

The project team didn’t collect as much data as they hoped to due to the short timeframe of the project. The results may differ if the team had data from more patients.

More data will be available as patients fill out PROMs over time for future projects to use.

How can people use the results?

Researchers can use the results when planning studies that collect and use PCORnet data.

PCORI funded the development of PCORnet^®, the National Patient-Centered Clinical Research Network, to make research faster, easier, and less costly to conduct. PCORnet is made up of Partner Networks of healthcare systems, patients and communities, and health plans that harness the power of large amounts of health data.

PCORI supports brief, descriptive projects to assess the feasibility of conducting research using data gathered and shared securely through PCORnet. This project is one of several designed to test the network while addressing priorities identified by PCORI and its stakeholders.

Objective

To incorporate patient-reported outcome measures (PROMs) collected during routine clinical care for patients with heart failure into the PCORnet Common Data Model (CDM) and conduct descriptive analytics of the relationship between these data and other patient characteristics

Project Design

The project team had three aims:

• Aim 1. Implement data collection for PROMs through the electronic health records (EHRs) of three academic health centers across the United States using the PCORnet CDM and interview patients and clinicians about their experiences using these PROM data during clinical encounters.
• Aim 2. Use a PCORnet query from the three medical centers to compare demographic and clinical characteristics of patients who completed PROMs with patients who did not.
• Aim 3. Using data from the PCORnet query, evaluate the association of the Kansas City Cardiomyopathy Questionnaire (KCCQ-12) scores with patient comorbidities.

PROMs are surveys that ask patients how they feel and what activities they can do. The team integrated three PROMs into the EHRs of three academic health centers: the KCCQ-12, which focuses on heart failure symptoms; the Patient-Reported Outcomes Measurement Information System (PROMIS®) Global Health scale, which focuses on general health status; and the Patient Health Questionnaire (PHQ-2), which focuses on depression. The team incorporated PROMs into the PCORnet infrastructure using the PCORnet CDM, which organizes data into a standard structure for use by researchers. The team then interviewed a group of 10 patients and 8 clinicians to understand their experiences in using PROM data as part of clinical care.

Next, the project team ran a PCORnet query of PROM data collected from 1,054 adult patients at the three medical centers between 2019 and 2020. The team compared the characteristics of patients who completed the three PROMs with 3,126 patients who did not. The team also examined the association between patients’ PROM scores and patient characteristics, including patient demographics and clinical comorbidities.

Results

Aim 1. The team successfully implemented PROMs at each academic health center and uploaded PROM data to PCORnet. The team assigned PROMs to be completed by 4,180 patients at the three centers, and 1,054 patients completed the questionnaires. Based on the qualitative interviews, patients felt that PROMs helped them communicate their symptoms to their doctors, and doctors found the information from PROMs to be useful for clinical care.

Aim 2. The team successfully ran a PCORnet query and compared patients who completed PROMs with patients who did not. At one of the three medical centers, patients who completed PROMs were significantly younger (p<0.05) than patients who did not complete PROMs; no other statistically significant differences in demographics were seen. Patients who did not complete PROMs had a higher burden of clinical comorbidities, including ischemic heart disease, valvular heart disease, peripheral vascular disease, renal failure, and depression.

Aim 3. The team successfully ran a PCORnet query and found that a significantly larger proportion of female patients had severe heart failure symptoms, defined as a KCCQ-12 score less than 25. Patients with KCCQ-12 scores less than 25 also had a higher burden of clinical comorbidities, including chronic obstructive pulmonary disease, obesity, and depression. Lastly, patients with KCCQ-12 scores less than 25 had significantly lower general health status as defined by the PROMIS Global Health scale and significantly higher depressive symptoms as defined by the PHQ-2.

Limitations

The timeframe of the project and data latency related to the PCORnet query limited the number of patients with PROM data, which precluded more advanced quantitative analysis such as linear and logistic regression.

Conclusions and Relevance

The project team successfully integrated three PROMs into EHRs at three academic health centers and queried these data using the PCORnet CDM. Patients who did not complete PROMs were generally more ill than those that did, raising the importance of optimizing PROM completion rates. The team found that the PCORnet CDM was a useful tool for extracting and analyzing PROM data.

Future Needs

Future projects could use PROM data from more patients, allowing for an analysis of the association between PROM data and outcomes such as hospitalizations and mortality.