Choosing doctors, choosing patients

The September 30, 2008 New York Times has a new feature section on “Decoding your Health”, which includes, among other articles, “Searching for Clarify: A Primer on Medical Studies” by Gina Kolata, an outstanding science writer. It also includes “You Can Find Dr. Right, With Some Effort” by Roni Caryn Rabin, which offers sound advice about how patients might go about selecting physicians.

The problem of choosing a physician is not a trivial one.

Patients have relatively little guidance to go on besides:

  • Who do my friends see?
  • Where is this physician located?
  • Is this physician covered by my health care plan, if any?
  • What is his/her specialty?
  • What are his/her office hours?
  • What is his/her name? (which is often informative about the physician’s gender and sometimes other demographics)

The article discusses some expanded strategies:

  • Rating systems (by consumers, by payers, by peers)
  • Hospital affiliations
  • ABMS certification
  • Interviewing the physician

The article concludes by quoting Dr. Samantha Collier, the chief medical officer for HealthGrades, who points out physicians “need to be a good match for you.” She’s talking about having a comfortable relationship, but the idea of patient/physician fit is one that I’ve been interested in for some time.

In 2006, colleagues and I wrote an a paper on the development of a measure of patient and physician fit. The process involved surveying both physicians and patients at my University, and asking about preferences for the physician’s behavior around six dimensions of interaction: considering non-medical aspects of the patient’s life, knowledge of herbal medicine, degree of physician decision making, providing information to the patient, considering patient’s religion, and treating the problem as the patient perceives it. We then looked to see how well patients and their physicians fit, whether fit was related to satisfaction, and whether there was room for improvement in fit.

The first interesting result was that both patients and physicians had a wide variety of preferences. This means that no physician could be a perfect fit to every patient.

The second interesting result was that fit on some dimensions was associated with satisfaction. Among patients seeing internists, fit on degree of physician decision making led to greater satisfaction; among patients seeing family physicians, fit on providing information led to greater satisfaction. It wasn’t that patients whose doctors provided more information (or shared decisions more) were more satisfied — it was patients whose doctors had similar preferences for these dimensions, whatever those preferences were. For example, patients who didn’t want to be given a lot of information were more satisfied when working with physicians who didn’t prefer to give a lot of information. Similar findings have been reported by Edward Krupat at Harvard in his work on the Patient-Practitioner Orientation Scale (PPOS).

The third interesting result was that patients and physicians who fit best may not automatically find each other. We found that in many cases, patients were not seeing the physician in their practice group who would have been their best fit.

Although we didn’t investigate the question directly, I suspect that fit only rarely and only slightly improves over time in the patient-physician relationship. Find the right physician for a patient is where the big improvements happen, and if better fit leads to more satisfying patient/physician relationships — and better relationships lead to better continuity, adherence, and care — we ought to be trying to provide patients and physicians with the guidance necessary to make those choices.

Taking fit seriously provides an important role for the physician as well. These dimensions of clinical behavior are often only implicitly observed or inferred by patients because many patients and physicians don’t discuss them explicitly. Physicians can help by letting patients know where they stand on such issues as providing information, and decision making. Physicians who practice in groups might consider developing an intake survey for patients that would enable them to place the patient with the best-fitting physician first.


  1. I wonder how patient-physician fit changes over time becasue of changes in a patient’s condition. Is a physician who is a good fit for me now when I am health still going to be a good fit for me as I develop chronic diseases? As my needs change in terms of health care perhaps the physician who best fits also changes.

    A related question to how should patients select a physician is how should a physician select a consulting physician? As a physician I find it difficult to judge the expertise of the consultant. When I send a patient to another physician it is because I realize I need help because I lack the required expertise. I may feel the consultant is doing too much or too little, but am I in a good position to make a valid judgment?

    Certainly I can assess how well the consultant treats me as a referring physician. Do they send me a report in a timely fashion? Do they treat me with respect? I can also ask my patient whether they liked the consultant. I used to think these were surrogates for quality but then within a few years two consultants I used got into trouble with our state medical board over issues of competence. I was surprised by this and very distressed by this because I had thought I was making good selections. If I am unable to judge the competence of a fellow physician it seems not very likely that a patient can effectively judge the competence of their personal physician.

Leave a Reply

Your email address will not be published.