Scitech  Diagnosis gone digital

Electronic medical assistants helpful but not always effective

A lmost every field has adopted digital technology, and medicine is no exception. However, the transformation of health informatics in the past decade has not simply been a change in tools of the trade, but a change in the very way knowledge is acquired and applied.

As a discipline that brings together health care and information science, health informatics is involved in setting up resources like search engines that doctors can use to retrieve clinical data. These tools can be grouped into two categories – information retrieval systems (IRS) and clinical decision support systems (CDSS).

Pierre Pluye, a physician and associate professor in the Department of Family Medicine at McGill University, investigates these electronic resources. He explained how IRS provide a way to filter the staggering amount of available information down to only the most relevant.

“There are 57 million abstracts on Medline [an online biomedical database]. Physicians do not have time [to read every single one]… because basically you would have to read 24 hours a day, seven days a week just to keep updated,” said Pluye.

CDSS differ from IRS in that they provide patient-specific information. Clinicians can use calculator-type programs that look at a patient’s history to determine their likelihood of contracting diseases or experiencing medical complications.

Still, the existence of such readily available information can create tension in the clinical setting. The digital diagnoses sometimes conflict with the clinician’s assessment and experience. But Pluye points out that physicians are under no obligation to incorporate new information into their practice.

“All physicians know tools like Medline…[but] a study by researcher Marc Berg found that often physicians will turn the tool off when they disagree with it,” said Pluye.

At other times, recommendations provided by digital support systems clash with peer or hospital policies.

“You have to talk to your peers, your group, consult with them, ask, ‘Do we want to do this?’” Pluye said.

While some studies have indicated that decision support systems improve practitioner performance, these tools have yet to be adopted on a larger scale, and clinicians still seem skeptical. A 2005 report in the Journal of the American Medical Association stated that effects of these decision-making systems on medical outcomes are still inconclusive.

Pluye makes certain to note, however, that clinicians are much more likely to use technology to look up comparable medical research rather than to base a clinical decision on the results.

“There is something appealing about the myth of automatic medicine. The patient enters the room, you enter the information, and results appear,” Pluye said. “It’s not like that at all. There are very few decision support systems…. In daily practice, people use the information retrieval systems.”

Recently, artificial intelligence technologies have been developed to aggregate relevant data faster, and to learn from patient histories and diagnoses to improve future results. How the field of medicine and the role of physicians will accommodate and adapt to this new technology remains to be seen.

With more patients using online information sources, and physicians using electronic knowledge resources, the dynamic of physician-patient relationships has changed. Pluye, however, saw no threat to the future of the profession. Instead, he pointed to the stronger communication between patients and physicians that will follow new technological developments.

“Do you prefer a clinician that supposedly knows everything? Do you prefer to meet and share negotiations with a clinician?” Pluye said. “Whatever the machine, or the book, or your brain says, the issue is compliance. With such tools you can have better communication with patients.”