Project Summary

This implementation project is complete.

PCORI implementation projects promote the use of findings from PCORI-funded studies in real-world healthcare and other settings. These projects build toward broad use of evidence to inform healthcare decisions.

This PCORI-funded implementation project used a risk prediction tool—shown to identify patients at high risk for developing diabetes—to help patients and their doctors make decisions about preventive treatment.

One in three adults in the United States are at risk for diabetes. Effective treatments are available to prevent diabetes, including medicine and lifestyle changes. But not all patients require treatment. Identifying which patients would benefit most from treatment may help patients and clinicians, such as doctors or nurses, prevent diabetes among patients at greatest risk.

What was the goal of this implementation project?

In a completed PCORI-funded pilot project, researchers looked at data from a landmark clinical trial and other studies to see how well treatment prevented diabetes in different groups of patients. They found that patients at highest risk for diabetes benefited most from preventive treatment. Patients at lower risk got much less or no benefit. Researchers used these findings to create a prediction tool that estimates patients’ risk for diabetes over three years as well as potential benefits from treatment. Patients and clinicians can use this information to help decide together on ways to prevent diabetes, either metformin or a lifestyle program called the Diabetes Prevention Program, or DPP.

This project worked with two health systems, Premier Medical Associates in Pennsylvania, and Mercy in Missouri, to make the prediction tool available at 52 primary care clinics.

What did this project do?

The project team adapted the prediction tool to work with different electronic health record, or EHR, systems. At Premier, a calculator drew from data in the Allscripts EHR to estimate risk. The calculator did not require staff to enter any data. At Mercy, a SMART on FHIR app linked to Epic EHR data to provide the risk estimates.

Clinicians could access the risk prediction information during the patient visit and produce a report. The report included information on each patient’s risk for developing diabetes and how well treatment was likely to work to reduce risk for each patient.

To support the use of the prediction tool, the project team:

  • Developed resources on how to use the tool, including a site-specific short video
  • Trained clinicians on how to use the tool and discuss results with patients
  • Provided sites with technical assistance
  • Identified and trained a site champion to promote the use of the tool, fix issues, and work with clinicians who weren’t using the tool to understand their concerns
  • Provided feedback reports to clinician champions on clinicians’ use of the tool

What was the impact of this project?

During the project, a total of 96 clinicians used the prediction tool with more than 2,500 patients.

At Premier, clinicians used the tool with 79% of their patients with prediabetes over a 31-month period. The prediction tool identified about half of these patients as having high risk for diabetes. Prescriptions for metformin increased fourfold among these patients. Clinicians also referred 490 patients at high risk to the DPP. Of these, 124 patients followed up; 64 enrolled and met the program’s targeted level of weight loss.

Clinicians at Premier reported that the tool was useful in helping to predict patient risk of diabetes. Patients reported feeling more confident in making decisions once they understood their risk of diabetes. The availability of the prediction tool also facilitated screening for diabetes at Premier. With the increased screening that took place during the project, 148 patients learned they had diabetes.

At Mercy, clinicians piloted the prediction tool with manual data entry. During the pilot, clinicians used it with 58% of their patients with prediabetes. Referral to the DPP increased from 0% to 13% among patients with prediabetes. Metformin prescriptions increased from 3% to 17%.

Completion of the project using the EHR at Mercy was halted by the COVID-19 pandemic. However, the project team did test the EHR-based approach. During testing, clinicians accessed the method with nearly 1,000 patients. The SMART on FHIR app used at Mercy is now available for use with Epic or other EHR systems.

More about this implementation project:

Stakeholders Involved in This Project

  • AMGA (American Medical Group Association)
  • Premier Medical Associates/Allegheny Health Network (Monroeville, PA)
  • Mercy (St. Louis, MO)
  • Interopion (Salt Lake City, UT)

Publicly Accessible Project Materials

For more information about these materials, please contact the Project Team at [email protected].

The project team developed these materials, which may be available for free or require a fee to access. Please note that the materials do not necessarily represent the views of PCORI and that PCORI cannot guarantee their accuracy or reliability.

Project Achievements

  • Adapted diabetes risk prediction tool to work with real world data

  • Created SMART on FHIR app version of the tool
  • Successfully integrated the prediction method into EHR systems at two health systems
  • Demonstrated the feasibility of using the prediction method in clinical practice at 52 primary care clinics
  • An initial estimate conducted by an independent auditor at Premier suggested cost savings based on lowering diabetes rates over the three-year project.

Implementation Strategies

  • Adapted the prediction method to work with sites’ existing EHRs and workflows, including integrating a clinical decision support tool into two different EHR systems (Allscripts and Epic)
  • Provided sites with tools to support implementation, including manuals, EHR-specific guidance, and posters
  • Developed and used online training, including educational webinars and videos
  • Trained clinicians on how to use the prediction model and discuss and interpret results
  • Used a phased implementation approach
  • Identified and prepared physician champions at sites
  • Conducted site visits
  • Provided technical assistance to sites, including practice facilitation
  • Provided sites with audit and feedback reports
  • Partnered with national stakeholder organizations to promote implementation

Evaluation Measures

To document implementation:

  • Proportion of all patients in system identified as having prediabetes with whom the risk prediction tool was used during a primary care visit
  • Provider uptake of the tool
  • Knowledge transfer to and satisfaction among providers by email survey
  • Knowledge transfer to and satisfaction among patients by mail survey

To assess healthcare and health outcomes:

  • Referral rates to the Diabetes Prevention Program or metformin prescriptions among patients with prediabetes

More to Explore...

PCORI Stories

Tailoring Study Results to an Individual Patient
This study shows that data from large clinical studies can provide not just the average effect of a treatment, as most studies now do, but indicate which patients are likely to benefit—or not. Principal Investigator David Kent’s team is now working with the American Medical Group Association to spread the risk model to 50 clinics in two major health systems.

Videos

Implementing a Diabetes Risk Assessment Tool into Clinical Practice
Hear researchers describe their PCORI-funded project, which used electronic health records to develop and test a prediction method that allows doctors to see information on a patient’s risk of developing diabetes. They are now working to implement the prediction method tool in practice through a PCORI Dissemination and Implementation Award.

Project Information

David Kent, MD, MS
Tufts Medical Center
$749,999
Improving Diabetes Prevention with Benefit-Based Tailored Treatment: Disseminating Patient-Centered Estimates of Benefit

Key Dates

February 2022
2017

Study Registration Information

Related PCORI-Funded Pilot Project

This implementation project focuses on putting findings into practice from this completed PCORI-funded pilot project: Predicting Who Will Respond Best to Medical Treatments

Related Dissemination and Implementation Project

Putting a Diabetes Risk Prediction Tool into Practice to Support Shared Clinical Decision Making

Tags

Project Status
Health Conditions Health Conditions These are the broad terms we use to categorize our funded research studies; specific diseases or conditions are included within the appropriate larger category. Note: not all of our funded projects focus on a single disease or condition; some touch on multiple diseases or conditions, research methods, or broader health system interventions. Such projects won’t be listed by a primary disease/condition and so won’t appear if you use this filter tool to find them. View Glossary
State State The state where the project originates, or where the primary institution or organization is located. View Glossary
Last updated: December 7, 2022