Radiologists, Reduce Variability in Care

Health care is increasingly complex and demanding, but the real problem is that the delivery of care is highly variable, Giles Boland, MD, professor of radiology at Massachusetts General Hospital, said at RSNA 2015.

Institutions have different practices and policies, which can create confusion for patients, but even within the same institution, and the same department, policies conflict.  The problem, Boland said, is that the incentives that drive high quality and safety are not really aligned with the fee for service model, which is relatively unrestricted and, at the moment, not tied to quality or safety.

This, of course, is mandated to change in the next few years.

“The Institute of Medicine recently looked at the diagnostic process and found that when errors occur, it’s due to failure of engagement, failure of information gathering, failure in information integration, failure to establish an explanation of the health problem, and failure to communicate to patients this explanation,” Boland said. “This is the business we’re in.”

“This is a pretty damning statement form the Institute of Medicine,” he said. “Scientific insights are poorly managed, evidence is poorly used, and the experience is poorly captured, which leads to fragmented systems and variable practices.”

Boland noted that even in his subspecialty, you can ask 10 radiologists what to do about an incidental finding and probably get five or six different answers.

In order to adapt to new payment models, radiologists need to identify and how where they create value and contribute to the outcome of the patient.[[{"type":"media","view_mode":"media_crop","fid":"46171","attributes":{"alt":"Giles Boland, MD","class":"media-image media-image-right","id":"media_crop_9997543368409","media_crop_h":"0","media_crop_image_style":"-1","media_crop_instance":"5347","media_crop_rotate":"0","media_crop_scale_h":"0","media_crop_scale_w":"0","media_crop_w":"0","media_crop_x":"0","media_crop_y":"0","style":"height: 227px; width: 170px; border-width: 0px; border-style: solid; margin: 1px; float: right;","title":"Giles Boland, MD","typeof":"foaf:Image"}}]]

“The metrics that we measure really mean nothing to patients or other caregivers in the system,” Boland said. “We need to ask ourselves, ‘what business are we in as radiologists?’”

Radiologists are in the information business, Boland said. “We are here to aggregate all of the biomedical data, not just images, but data in the electronic medical records.”

Currently, radiologists have to go mine for relevant data in the electronic medical records. Better integration of the EMR and imaging will ultimately lead to a better report and more appropriate communication to referrers, he said.

Reporting language is another obstacle in delivering value in radiology as radiologists often use variable language to mean the same thing. Words like “suspicious” or “concerning” are frustrating to referrers, Boland said.

“Sometimes, the key findings are unclear, we use a fraction of the available information that resides in the medical health record,” Boland said. “Our recommendations are variable and we communicate in different ways.”

Boland also noted that there is no national mandate for the reporting of critical findings. While colleges and professional organizations publish guidelines, each hospital reports critical findings differently and usually depending on what their task force comes up with, he said.

Actionable reports will deliver better value because it will affect outcomes, Boland said.

To measure value, Boland also points to the report, the radiologist’s primary product. A simple question to ask is, “Did the referring provider find the information useful?”

“In almost all cases [in my organization], I have no idea [if the information was useful]”, Boland said. “I get no feedback.”

Boland suggested reports come equipped with functionality for the referring provider to disclose if the information in the report was useful.

Another measurable metric radiologists can use to prove their value is imaging’s effect on length of stay. This, especially, is a metric that hospital executives will find valuable.

“An organization might have 5.12 days average length of stay,” Boland said. “If we can image that patient earlier, expeditiously, and report on them correctly, what’s our contribution to reducing that length of stay? Let’s quantify it.”

Metrics should be meaningful, transparent, truthful, reflect real valuable outcomes for patients, and easily understood by stakeholders, Boland said.

“We don’t actually know how we are going to get paid moving forward, but we have to think about these things seriously,” he said.