A key challenge for the field of health services research is translating research findings into health care innovations and then scaling them for widespread national use. This process has been disappointingly slow for high-priority patient safety topics, including diagnostic errors. The Leapfrog Group Recognizing Excellence in Diagnosis initiative aims to address this challenge. The problems related to diagnostic error are clear: every year, over 250,000 hospital inpatients will experience harm from a diagnostic error. Diagnostic errors are the leading cause of medical malpractice claims, and account for as many as 17 percent of all hospital adverse events. Indeed, in 2015, the National Academies of Medicine (NAM) found that “most people will experience at least one diagnostic error in their lifetime, sometimes with devastating consequences.”
The NAM report concluded that the health care system has “largely unappreciated” the harm from diagnostic errors and anticipated that the problem of diagnostic errors would only get worse as patient care gets increasingly complex. Often hospitals don’t appreciate the magnitude of the problem, and even those that do aren’t sure where to start to tackle errors in diagnosis. In 2019, Leapfrog partnered with researchers from Johns Hopkins University, Baylor College of Medicine, and the Society to Improve Diagnosis in Medicine to address this research translation issue.
With funding from the Gordon and Betty Moore Foundation, and under the guidance of a national advisory group chaired by Drs. Mark L. Graber and Hardeep Singh, we identified and published a set of evidence-based practices that hospitals should adopt to reduce harm to patients from diagnostic errors. Based on AHRQ’s definition of a diagnostic safety event, diagnostic errors include a delayed, wrong, or missed diagnosis or a diagnosis not communicated to the patient.
Our first step was to identify all the possible practices that we could recommend to hospitals. We conducted a literature review, and carefully combed through the seminal publications from the Agency for Healthcare Research and Quality, the National Quality Forum, NAM, and others. To supplement published materials, we conducted in-depth interviews with leading practitioners in risk management, laboratory and radiology, health information technology, and clinical medicine. This search identified well over 100 interventions that have been suggested or shown to improve diagnosis in practice. These interventions were identified at the organizational and system leadership level, as well as in the diagnostic process itself. Key themes included address the hospital’s culture around diagnosis and error reporting, allocating the financial, technological, and staffing resources that clinicians need to succeed in their work, engaging patients as partners in their diagnosis and care plan, and facilitating communication of diagnostic information in transitions between settings of care.
We then leveraged the expertise of the national advisory group, including patient and payor stakeholders and experts in implementation science and the various diagnostic specialties. This group met with the core team to rank these recommendations using three prioritization criteria:
- Strength of evidence: practices endorsed by experts, or shown to have improved practice
- Likely to be adopted: practices that are actionable and practical
- Likely to have impact: practices most likely to improve diagnostic processes and outcomes
Twenty-nine practices were ultimately recommended, and this set was reviewed again by the Advisory Group for specificity and clarity. A final report, published in July 2022, presents these twenty-nine practices along with a summary of the evidence base for each one, and implementation examples. Some key practices include helping patients and their caregivers communicate complete and accurate information (including providing qualified interpreters), promoting effective teamwork including the entire care team, and communicating clear instructions to patients and continuing care clinicians to look for test results pending at discharge or clinical changes.
Two key lessons emerged from this research translation process. First, it is critical to facilitate unstructured conversations with multidisciplinary experts with a variety of perspectives. The free flow of ideas and comments provides the basis to find commonalities across the viewpoints of different stakeholders. Through this process, we added several practices that would otherwise not have been recognized, such as facilitating patient reporting of diagnostic errors or giving patients pathways to escalate care when needed. Second, as with most interventions in health care, the possible scope of topics to consider is enormous. By focusing on Leapfrog’s areas of measurement expertise in hospital safety and quality and prioritizing those with the highest potential impact on reducing harm to patients, we finalized a manageable list of key practices for hospitals to implement.
The report is only the first step. To achieve real, measurable progress, we need transparency and accountability. In 2023, we will publish a report reviewing the results of a pilot test of the implementation of these practices, based on a survey of 115 hospitals. The next step is to integrate these practices into the Leapfrog Hospital Survey, and, after a trial period, publicly report on hospitals’ progress on implementation starting in 2025.
We believe our process – multistakeholder input, iterative piloting and testing, and leveraging public accountability - can serve as a model to help drive positive change in patient safety and reducing harm from diagnostic error.
The Leapfrog Group is an AcademyHealth organizational member. Organizational Members are a critical link in AcademyHealth’s ability to effectively advocate for the field, and support the future field of health services researchers. Organizational Members gain visibility among AcademyHealth membership, enjoy unique networking opportunities, and benefit from event discounts. Click here to learn more.