Alert fatigue and poor workflow integration in electronic health record systems are familiar problems to Sara Chokshi, DrPH, MSPH, Assistant Director of the Healthcare Innovation Bridging Research, Informatics & Design (HiBRID) Lab and her fellow researchers at NYU Langone, who study how digital tools impact clinical practice and how digital innovations can be better integrated to improve care delivery.
In a recently published feasibility study, Chokshi outlined the work of a team led by principal investigators Devin Mann, MD, MS and Andrea B. Troxel, ScD to create a more user-friendly EHR-embedded clinical decision support module that uses behavioral economics (BE) nudges to increase physician adherence to Choosing Wisely guidelines targeting overtreatment of diabetes among older adults.
“A lot of our work focuses on improving the provider experience with the EHR because it is, for lack of a better word, so noisy,” said Chokshi, who works closely with Mann and others in an informatics research group at NYU. “Just as we get too many alerts on our phones and we begin to ignore them, the same thing happens with the EHR.”
The research includes a recently completed six-month pilot study of its “BE-” module at five primary care clinics in their faculty group practices at NYU Langone serving the target population, patients age 76 and older. The Choosing Wisely guidelines, developed by the American Geriatrics Society, recommend less intensive control of blood sugar levels for older adults based on life expectancy, and the use of metformin as the preferred diabetes medication.
The team recently completed the pilot phase of their study, which includes the development and pilot of the BE-EHR module and is awaiting word on funding to conduct a clinical trial of the module.
Besides being text heavy, many traditional alerts in EHRs can be difficult to read and poorly designed, said Chokshi, adding that the research team used key informant interviews and workflow assessments and other design thinking strategies to create the clinical decision support (CDS) module.
The CDS module offers a variety of behavioral economic “nudges” within the EHR including: suggesting metformin when a physician tries to prescribe another diabetes medication; asking for a justification if a physician orders a medication other than metformin; providing defaults in diabetes management order sets that suggest metformin; listing metformin in medication preference lists, and using anchoring by updating hemoglobin A1c lab reports for eligible patients to show less aggressive treatment goals. Additional nudges were developed outside of the EHR, such as an email leveraging peer comparisons or embedded .gif files or animations.
The alerts don’t stop physician action or interrupt the workflow, said Chokshi, but instead prompt providers to consider the Choosing Wisely recommendation. “There are many good reasons the best practice alerts are there, for quality, safety and better care. But is there a way to make them better without adding to the cognitive load?” she said, referring to the volume of information directed toward physicians in patient encounters each day.
The study states: “Integrating behavioral economics strategies and electronic health records using various CDS tools is a novel approach to improving guideline adherence that also seeks to minimize negative impacts on clinical workflow and cognitive load.”
Chokshi said they launched nudges outside of the EHR to simply generate greater awareness about the Choosing Wisely campaign and its more than 600 recommendations from specialty societies. The use of behavioral economics is also helping the researchers determine what affects physician decision-making about how to best treat their patients’ medical conditions.
“Physician behavior is notoriously difficult to change,” Chokshi said. “So the other impetus (in this research) is to uncover what things motivate physicians when it comes to care decisions.”
- A Case Study: Reducing Routine Oxygen Tests
- Achieving Change at Scale
- Overuse and Vulnerable Patients