Society to Improve Diagnosis in Medicine Funds 17 Projects in Seed Grant Program

Maryland-based MedStar Health will launch and evaluate a new sepsis-specific clinical decision support tool that uses a validated sepsis staging model to provide earlier and more accurate characterization of sepsis patients. MedStar is just one of 17 organizations awarded a grant by the Society to Improve Diagnosis in Medicine (SIDM) as part of its Diagnostic Quality Improvement (DxQI) Seed Grant Program to improve the quality, accuracy, and timeliness of diagnoses.

Inaccurate and delayed diagnoses are the most common, catastrophic, and costly of medical errors. It is estimated that 12 million adults in the United States experience a diagnostic error every year in outpatient settings alone, and diagnosis failures result in as many as 80,000 premature deaths every year in U.S. hospitals.

This funding round of $3 million offers initial awards are up to $50,000 for 17 organizations testing interventions to improve the quality, accuracy, and timeliness of diagnoses. Most of the grantees selected in this first round will focus on developing interventions to reduce diagnostic errors in three specific disease categories—cancers, vascular events, and infections—which account for most inaccurate or delayed diagnoses that result in serious harm or death. Several grantees are developing improvement projects that address racial and gender disparities in diagnosis.

Ultimately, the program will help clinicians, hospitals and health systems take specific steps to reduce diagnostic errors by building an evidence base of effective interventions. Once tested, SIDM will create a clearinghouse of the interventions and be a catalyst for the adoption of effective practices by other organizations across the country.

“There are very few proven real-world solutions to minimize the harm diagnostic errors can cause,” said Gerard Castro, Ph.D., M.P.H., director of quality improvement at the Evanston, Ill.-based SIDM, in a statement. “The DxQI grants awarded through the program are a good first step in learning what kinds of best practices we can build into the diagnostic process to ensure that patients receive a diagnosis that is accurate, determined in a timely manner, and effectively communicated. ”

The DxQI grants align with SIDM’s support for an inclusive approach to reducing diagnostic errors where patients and frontline health professionals engage in developing and testing approaches to improve diagnostic quality and safety.

The principal investigator on MedStar’s project is Kristen Miller, Dr.P.H., M.S.P.H., scientific director of MedStar Health National Center for Human Factors in Healthcare and an associate professor of emergency medicine at Georgetown University School of Medicine. “We are eager to work towards improving diagnosis of sepsis,” she said in a statement. “Diagnostic error is a significant problem that the healthcare community needs to address, specifically diagnosis of sepsis which continues to increase annually despite increased awareness. We are proud to be part of a cohort developing specific strategies to improve the accuracy and timeliness of diagnosis. This is just one of many continuing efforts to improve patient care at MedStar Health.”

The 17 grantees awarded in this round are:

• Advocate Aurora Health is initiating an infection screening program for patients in outpatient clinics to diagnose infection and reduce sepsis.

• Atrium Health is evaluating the effectiveness of cognitive machine learning to identify the likelihood that a patient had COPD even if there is no documented medical history of COPD in their health record.

• The Atrium Health Levine Children’s Hospital is improving the accuracy of the referral process for underserved pediatric patients from primary care clinics to a rheumatology specialist in an effort to reduce delayed diagnoses.

• Beth Israel Deaconess Medical Center is collaborating with the BCH Spanish Collaborative to develop a patient-centered framework and online tool that makes it easier to “coproduce” diagnoses and share visit notes with patients with limited English proficiency.

• Brigham Health/Brigham and Women’s Hospital is aiming to reduce missed and delayed diagnosis of cancer by improving the communication loop for patients at-risk for cancer who have abnormal test results.

• The Johns Hopkins University School of Medicine is training physicians in bedside diagnosis through a symptom-specific virtual patient curriculum with direct performance feedback to improve diagnostic reasoning skills.

• Kaiser Permanente East Bay is applying a quality improvement workflow in the emergency department (ED) for patients identified as high risk for diagnostic errors due to factors such as cognitive impairment, psychiatric illness, or social issues.

• Maine Medical Center is finding ways to ensure that patients at-risk for a liver disease complication receive the right diagnostic testing and treatment as soon as possible when they come to the hospital.

• McMaster University is implementing an intervention to optimize testing for pulmonary embolism in the ED and reduce inappropriate use of computed tomography scans while maintaining or improving patient safety.

• MedStar Health is launching a sepsis-specific clinical decision support tool to improve the diagnosis of sepsis by leveraging electronic health record data to provide earlier data on potential sepsis patients.

• Northwell Health aims to improve cardiovascular disease screening for cancer survivors by utilizing a clinical decision support tool integrated into electronic health records.

• Northwestern Memorial HealthCare is implementing a survey to determine how frequently diagnoses change when patients transition from the ED to hospital medicine to identify diagnostic errors.

• Tufts Medical Center and Floating Hospital for Children aims to improve the timeliness of tests and diagnosis of cancers by enhancing an electronic alert that notifies a clinician when a patient fails to complete a recommended and scheduled radiology test.

• The University of Michigan and Hurley Medical Center are developing criteria for electronic health records alerts to improve pediatric ED patient vital sign reevaluation to improve identification of children at risk of sepsis or serious illness.

• The University of Texas Health Science Center at San Antonio & University Health System is evaluating the use of EmBrACE (Emergent Breast Abnormality Care and Evaluation) as a pathway designed to decrease hospital admissions and time to diagnosis for unfunded and underserved ED patients with a breast abnormality.

• The University of Pittsburgh School of Medicine’s multicenter quality improvement project focuses on improving newborn healthcare delivery systems to improve diagnosis for early-onset sepsis and reduce the use of antibiotics.

• The Veterans Education and Research Association of Michigan and VA Ann Arbor Healthcare System are creating an electronic tool to improve the delivery of abnormal pulmonary nodule results captured incidentally in the ED to the patient’s primary care physicians to coordinate future imaging and biopsy.

The DxQI grantees will identify and implement small, practical tests of change, then begin to build evidence supporting interventions that could reduce harm due to diagnostic error if replicated and spread across other hospitals and health systems. SIDM will release a report annually throughout the three-year program about interventions that were effective and those that require further research and funding.

The goal is to share the learnings with the broader healthcare community in the months and years to come.