RSNA 2017: Understanding error, potential improvements in the diagnostic process

Twitter icon
Facebook icon
LinkedIn icon
e-mail icon
Google icon
 - RSNA

According to the 2015 report from the Institute of Medicine "Improving Diagnosis in Health Care," every person with access to healthcare will experience at least one diagnostic error in her lifetime. The diagnostic process for clinicians can be complex and requires collaboration between patients, clinicians and healthcare providers to provide the best treatment plan. There remains room for improvement in reducing diagnostic error, as discussed during a Nov. 28 session at RSNA 2017 in Chicago.  

Presenters Timothy Mosher, MD, Danny Kim, MD, and Jeffrey Myers, MD, led the dynamic session, "Understanding Error and Improvement in Diagnosis," which focused on three objectives to reducing diagnostic error:  

  • Understanding sources of error in the diagnostic imaging pathway.
  • Understanding the barriers and impediments to overcoming them. 
  • Knowing the role of team, culture and technology in minimizing error, improving communication and ultimately providing better care.  

"Diagnostic error is the failure to do two things: one is establishing accurate and understandable explanation for the patient's health problems, and, the second, to communicate that explanation to the patient," said Mosher, a professor of radiology at Penn State University.  

According to recent research, roughly 17 percent of medical errors are attributed to diagnostic errors. Alluding to Robert Wachter's article, "Why diagnostic errors don't get any respect – and what can be done about them," Mosher listed six challenges when reducing errors in clinical practice: 

  • Diverted attention and resources. 
  • Diagnosis in the outpatient setting with a fragmented delivery system. 
  • System and cognitive causes.
  • Wide intervals between error and adverse outcome.
  • Business models lacking in reducing error. 
  • Lack of accountability.  

The diagnostic process, according to Mosher, should be focused around the patient, who is "the one person involved in this highly fragmented system." 

The complex nature of the diagnostic process involves the accuracy of the diagnosis, the creation and ventilation of the patient's case, communication between physicians and patients, follow-up with the patient and treatment options.  

Mosher simplified the process into three distinct phases:

  • The pre-analytic phase relies on system-based solutions and occurs before the patient engages with the health system. This includes applying public health applications to actions taken because the patient may not recognize the need to get healthcare or failure to have access to resources or expertise.
  • The post-analytic phase also relies on system-based solutions. However, it involves a failure to communicate a diagnosis back to the patient due to IT infrastructure and tracking software.
  • The analytic phase is when traditional diagnostic process failure points occur. For example, gathering data and information from an improper test or inaccurate report could lead to errors.   

Along with diagnostic errors, failure modes in diagnosis include no-fault errors made by clinicians, system errors and, most notably, cognitive errors.  

"In order to see where these failure points occur, we're really going to have to look at diagnostic thinking," Mosher said.  

Cognitive errors are difficult to reduce and can stem from knowledge deficits or perception errors. Perception errors are the most common type of diagnostic error made in radiology, which are data driven and can either be from a recognition, search or synthesis error or an over-reliance or overconfidence in heuristics. Mosher concluded that radiologists need to be monitoring, reflecting and calibrating heuristics to lessen diagnostic error.  

Continuing the discussion, Danny Kim, MD, director of quality in the radiology department at NYU Langone Medical Center, evaluated the impediments and barriers to the diagnostic process.  

Technology, medical liability, reimbursement, and organizational culture are the four main barriers to improving the diagnostic process, according to Kim. Additional barriers include clinicians' unwillingness to disclose diagnostic errors to patients, knowledge limitations, mixed messages from senior management about what to disclose, and lacking = confidence in admitting error.  

In terms of reimbursement, the U.S. fee-for-service (FFS) healthcare payment model favors quantity over quality. There is higher reimbursement for procedures and diagnostic testing than for evaluation and management services, according to Kim. Additionally, documentation guidelines created to facilitate clinical reasoning are repetitive and extensive, making it harder for physicians to find treatment information for patients, Kim stated.  

Kim explained barriers that prohibit change in an organizational culture include system inertia and insufficient involvement of management.  

Kim urged radiologists must strive toward a "just culture," which includes learning from errors and accepting accountability.

Jeffrey Myers, MD, professor of pulmonary and thoracic pathology at the University of Michigan, discussed the importance of Patient and Family Centered Care (PFCC).  

"Patient and family centered care is an approach to the planning, delivery and evolution of healthcare that is grounded in mutually beneficial partnerships among providers, patients and families," Myers said. "It redefines the relationships in healthcare by placing an emphasis on collaboration with people of all ages, at all levels of care and in all health care settings."  

Myers explained that PFCC is working with patients.  

"We have forgotten why we're here. Patients and families are at the center of what we do," said Myers to the attendees when discussing the University of Michigan's "Michigan Innovative Personalized Patient-centered Pathology" program, more commonly referred to as MiP3, established in 2014.  

Goals of the MiP3 program include increasing patient access to understandable information, optimizing technology to bring better and faster information to the patient, and implementing multi-disciplinary initiatives. Myers said that although PFCC isn't common among radiology practices and most clinical practices yet, switching to it is worth the effort in the long term.   

"Patient family-centered care is not some touchy-feely thing to make us feel better. It's a hard-nose strategy to do hard things in a way that makes sense," he said.