More Accurate, Timely Medical Diagnoses Require Overhaul, Report from National Academy of Medicine States

Patient care will greatly improved if more attention is paid to getting the diagnosis right and getting it faster, according to a report released today by the National Academy of Medicine.

But ensuring correct, timely diagnoses will take across-the-board reforms that target the way doctors are trained, the way healthcare rewards those who spend time with patients and families, and the way the legal system handles information about mistakes and near-misses.

Today’s report, “Improving Diagnosis in Healthcare,” published by the National Academies Press, is a successor to the groundbreaking report, “To Err is Human: Building a Safer Health System,” which 15 years ago launched the process that today has made measuring healthcare quality a centerpiece of reform efforts. (See “Building Upon the Strong Foundation of National Healthcare Quality,” in this month’s issue of The American Journal of Managed Care.)

However, the report from the National Academy of Medicine, formerly the Institute of Medicine, asserts that for all the progress made in improving healthcare quality, the essential step in the chain—assessing and diagnosing what is wrong with the patient—has received less than adequate attention. The 21-member Committee on Diagnostic Error in Health Care examined the diagnostic process in light of the following:

• At least 5% of US adults experience diagnostic error each year in outpatient settings, although the report acknowledges the lack of data on the scope of errors.
• Postmortem exams show diagnostic errors contribute to roughly 10% of patient deaths.
• Diagnostic errors are responsible for 6% to 7% of hospital adverse events.
• By far, diagnostic errors account for the largest number of paid medical malpractice claims; errors are twice as likely to result in deaths compared with other claims.

“Getting the right diagnosis is a key aspect of health care—it provides an explanation of a patient’s health problem and informs subsequent health care decisions,” the report states. “Improving the diagnostic process is not only possible but it also represents a moral, professional, and public health imperative. Achieving that goal will require a significant re-envisioning of the diagnostic process and a widespread commitment to change among healthcare professionals, health care organizations, patients and their families, researchers and policymakers.”

Notably, the committee defined a diagnostic error as both the “failure to establish an accurate and timely explanation of the patient’s health problem,” as well as the “failure to communicate that explanation to the patient.” The report’s emphasis on doing more to listen to patients and families (or other caregivers) elevates the patient as a partner in the medical delivery process.

Christine K. Cassel, MD, president and CEO of the National Quality Forum, was a member of the committee that developed the report, just as she took part in the original “To Err Is Human” effort. As an internist, Cassel said she has felt since that time that a report focusing on the diagnostic process is “very much needed.”

“We have measures for diabetes, or heart disease, or hypertension,” she said. “We don’t have measures for what to do when the patient says, ‘I feel dizzy,’ or ‘I feel fatigued.’ And yet, that’s a really big part of healthcare.”

Cassel said patients have to be “empowered” to take part in the process, and the healthcare team has to be given enough time to get a complete history. “Often doctors are so rushed, and the system overloaded with so many external pressures, that the doctor or nurse doesn’t have time to ask, ‘Do you remember anything else?’” But sometimes, Cassel explained, it’s that last question that could change a doctor’s opinion about what is the root cause of a symptom.

Managed care plans will be heartened by the report’s attention to “overdiagnosis,” in which a condition is identified and treated even though it is unlikely to affect a patient’s health and well-being. Overdiagnosis represents “a challenge to healthcare quality,” even though it is not an error, and may only be seen in hindsight on a populationwide basis.

The report outlined 8 distinct goals, with recommendations for each, to achieve improved and timelier diagnoses:

More effective teamwork. Greater collaboration is essential, both within care teams and between primary care providers and specialists. The report calls for more involvement with pathologists, radiologists and other diagnostic specialists. A section of the report addresses the growing importance of molecular diagnostics, and how the increased use and sophistication of these tools demands better attention to this step in the care chain.

More must be done to educate patients and families about the diagnostic process, “to create environments in which patients and families are comfortable.” Patients and families must feel safe to share information and speak up about errors and near misses.

Education and training. Physicians must be better trained in clinical reasoning and their ability to communicate with patients, families, and other healthcare professionals, their appropriate use of tests, and the application of results on future decisions. The report notes, for example, that the development of sophisticated tests has perhaps de-emphasized the importance of the physical exam; the development of the “Stanford 25” is an attempt to return to the days of the focus on the hands-on exam. The report recommends that accrediting agencies ensure that all healthcare professionals maintain competencies in making good diagnoses.

Use of health information technology (IT). Health IT, while useful, can limit decision-making and the barriers it poses must be identified. Special attention must be paid to cases where health IT is not well-integrated into the workflow and where there is difficulty sharing patient information. The report identifies the problem that health IT vendors may pose in limiting information about the risks of using technology, and calls for sharing more information about user experiences.

In a specific recommendation, the report calls on the Office of the National Coordinator of Health IT to set standards for vendors for interoperability to support “effective, efficient, and structured flow of patient information across settings to facilitate the diagnostic process by 2018.” It further calls on the HHS Secretary to require vendors to “routinely submit products for independent evaluation and notify users about potential adverse effects on the diagnostic process related to the use of their products.”

Learning from mistakes, near misses. The report finds that few healthcare groups have process to identify and learn from errors and near misses and seeks a greater use of “postmortem examination research,” similar to the Choosing Wisely initiative that has reduced the use of unnecessary medical tests and procedures.

Specifically, the report calls on Medicare and accrediting organizations to require such steps as conditions of participation, and to create practices to provide “systematic feedback on diagnostic performance.” HHS should set aside funds to do postmortem audits on representative samples of patient deaths.

A better work culture. Creating a less fragmented, nonpunitive work environment is crucial to reducing diagnostic errors, the report states. Healthcare organizations that adopt a culture “in which the diagnostic process occurs to support the work and activities of patients, their families, and health care professionals” will be better off.

Addressing the legal environment. Recommendations here will be difficult to implement, because so many are beyond the control of healthcare systems themselves. While the report calls for better reporting of errors and near misses in an environment “without the threat of legal discovery or disciplinary action,” so mistakes can be studied; in many cases this may require changes to state laws. Separately, the report notes that the current medical liability system “sets up barriers to improvements in quality and patient safety.”

The report calls for the potential use of alternate dispute mechanisms and demonstration projects, such as special administrative health courts, to resolve claims, but recognizes there are powerful interests that have a stake in maintaining the “the current tort-based system.”

“Medical liability reform has been on our radar for years,” said Marcus Rayner, president of the New Jersey Civil Justice Institute, whose group advocates for liability reform. “This report shows it is not just a cost issue as some claim, it is a quality of care issue. Hopefully lawmakers will see this and realize it is time to make some changes so our legal system is not reducing access to care, or the quality of care.”

Better payment models. At least one of the goals may be on its way to fulfillment: the report calls for payment structures that reward collaboration and consultation on cases, rather than procedures and tests. Payment for so-called “evaluation and management” (E&M) services have been less emphasized under the old fee-for-service model, but this may be changing. The report notes that half of Medicare payments are to be based on alternate models by 2018.

The report specifically calls for creating current procedural terminology codes and coverage for E&M services to encourage collaboration and consultation with pathologists, radiologists, and other diagnostic specialists who have been underutilized in the past.

More research on diagnostic process, errors. The report finds that much more research is needed on how errors occur, how frequent they are, and how they can be avoided. While the report gave conservative estimates on the frequency of errors, it states frankly: “The committee recognized that, perhaps not surprisingly, the available research estimates were not adequate to extrapolate a specific estimate or range of the incidence of diagnostic errors in clinical practice today.”

In addition to funding from federal agencies, including the Veterans Administration, the report suggests some funds could come from property and liability insurers, who pay medical liability claims.

Research is needed not only on how diagnostic errors occur, but also into measurement science, so that hospitals and healthcare systems can track errors. Cassel noted the report’s findings that there remains the same dearth of data on diagnostic mistakes that she observed when she took part in “Too Err is Human” 15 years ago.

Cassel noted that the original report brought about a culture of change and showed that healthcare quality could improve in an atmosphere that was nonpunitive. That same philosophy is needed to improve diagnostics. “We need an openness,” she said. If a person reports a problem or makes a mistake, “that person should get some feedback. It should be understood that the only way we learn from those mistakes is if we get feedback.”