One of the ways Picker Institute supports patient-centered care is by recognizing significant contributions to achieving patient-centered care nationwide. “Conversations with Leaders in the Field of Patient-Centered Care” is a regular feature that highlights these contributions. This Conversation is with Dr. Charles Burger, a strong proponent of the problem-knowledge coupler principle.
A Conversation with Dr. Charles Burger
“The problem-knowledge coupler principle is simple: Gather a large number of variables (medical history findings, physical exam findings, laboratory data) and use a computer to sort them into all the diagnostic or treatment possibilities for a patient’s unique clinical situation.”
Charles Burger, MD
“The Use of Problem-Knowledge
Couplers in a Primary Care Practice”
In 1968, Dr. Lawrence Weed, a physician researcher, posited that the rapid growth of medical knowledge was creating a demand for new ways of providing information in support of evidence-based medical practice. Following his own passion for a disciplined approach to medical record documentation to optimize the care provided to each individual patient, he developed the problem-knowledge coupler.
Problem-knowledge couplers comprise a sophisticated, evidence-based diagnostic and treatment decision support tool, embedded in technology, to manage condition and disease, reduce life-style risk and provide primary care. Couplers match unique patient information and unique characteristics with an extensive medical database to provide guidance tailored to unique individuals.
Couplers are developed through a collaboration among clinicians, informaticians and librarians. They recognize that functionality must be predicated upon combining unique patient information, gleaned through relevant structured question sets, with the appropriate knowledge found in the world’s peer-reviewed medical literature. Two pilot studies indicate that couplers can meet the gold standards of decision making within both a primary care and a specialty practice.
Issues remain about how to best integrate problem-knowledge couplers into clinical practice and whether large-scale outcomes research will support the findings of pilot studies. However, problem-knowledge couplers represent a promising approach that might portend a new model for healthcare delivery in the next millennium.
Dr. Charles Burger, the medical director of Evergreen Woods Primary Care in Bangor, Maine, has been working with computerized problem-knowledge couplers since 1984. Click here to read his article about their use in a primary-care practice, which appeared in the Spring 2010 (Kaiser) Permanente Journal.
Dr. Weed proposed his problem-knowledge coupler theory in 1968 with his article on “Medical Records That Guide and Teach.” More than 40 years later, the idea seems to be a novelty still, yet people like you, who have incorporated it into practice, have nothing but praise for it. Why is it taking so long for it to catch on?
This is by far the most complicated issue here. Physicians who are concerned that it is not “experiential” enough have probably never even used the tool. However, more and more often, thoughtful physicians are recognizing that the job we are trying to do is cognitively impossible due to the complexity of the information that we have to deal with, and that they’re unable to keep in mind either generally or specifically all the elements of the cases they’re dealing with.
As with all tools, one must be trained to use it. That takes time and commitment. And as with all learning, one goes from novice to master over a period of time, and that can be a significant barrier.
One must follow the path I outlined in the article—reorganizing your practice around the use of new software—and that is a difficult process. It takes a commitment to certain principles to push through on that.
Finally, I am sure that many physicians are offended by the thought that they need a tool for their brain!
Dr. Weed was the president of PKC (Problem-Knowledge Coupling) Corp. in South Burlington, Vt. Do you think there is any onus attached to the fact that the theory has been commercialized?
It’s sad that Dr. Weed is no longer the president—he was forced out a couple of years ago, and I think this has been a disaster. But I do agree that the commercial product limits success to a certain level. I know it is the commercial aspect that dismays some of the leading names in healthcare and that many professionals support the idea but are dissuaded from actively moving it forward.
Early on, Dr. Weed tried to interest the Library of Medicine, as well as the NEJM, in taking over the project, but nothing happened. It should be managed as though it were an open-source software company.
Forty-two years of technology have brought us a long way. Do you think early reluctance to buy into the proposition had anything to do with the amount of work it must have required, which must have been staggering, before it was fully computerized?
There has been a considerable evolution of the product, but I am not sure that the lack of technology has had anything to do with the reluctance to embrace the science. Look how slowly physicians have adapted to the EMR!
Is there any proof that PKC results in better outcomes more often? Have those measurements been made?
There have not been enough good studies to say one way or the other. Does the science enable doctors to be more thorough? Yes! Does the patient get better information? Yes! Do I feel that the reliability of the system improves by using them? Yes! Outcomes are a system function, and no single tool can improve them by itself , any more than the medical record can. The problem is that we do not have a system in medicine that has feedback loops so that we can get wiser over time. Here’s a schematic outline of what that process might look like.
In his new book, Thinking, Fast and Slow, Nobel-Prize winner Daniel Kahneman points out the inherent biases that are built into our decision- making process. These same biases are at work when physicians make patient-care decisions. Couplers provide at least some protection against these biases by forcing us to collect all the relevant information on patient problems up front before considering options. We know, for instance, that on average providers make an initial diagnosis within 30 seconds of talking with a patient. From that time on they look for information that confirms that judgment and ignore that which refutes it (called confirmation bias).
Have there been attempts to integrate this science into medical curricula? If so, where have they succeeded?
Sadly, no, though I am trying very hard to put the program in the hands of medical students.
Where can PKC go from here?
Without tools like these, patients take a risk every time they visit their healthcare provider, and I think it is inevitable that this very important technological development will become as vital a part of healthcare as MRIs and CT scans. In my own practice, we have successfully integrated a sophisticated clinical support system into our busy primary-care practice with no loss of productivity. We have standardized inputs at the front end (itself a quality gain),with the variations occurring in the outputs (options) generated by each unique patient situation. We have minimized the chances that the rare or unusual case will be missed, and we are able to provide detailed, current information for the patient. We have shown that it can be done. The only question is whether the profession is willing to minimize the limitations of the human mind to deal with complex data through the use of new tools, and I think that with good, strong leadership that can be accomplished.
Dr. Weed’s new book, Medicine in Denial, published in March 2011, offers a very clear blueprint for using PKC to build a healthcare system that would serve as a hope for the future rather than what it is now, a looming liability.