Does patient-centered health care mean never saying "no"?

Many commentators on U.S. health care, including me, have written that our existing models of care are far from patient-centered. In a Health Affairs piece written last May, pediatrician and Institute for Healthcare Improvement president Donald Berwick, took this one step further by arguing that instead of revolving around the needs and wants of patients, health systems are generally designed to meet the needs of clinicians. He relates an episode during which a close friend who was having chest pain requested that he accompany her to the cardiac catheterization lab for emotional support and to help explain the procedure's results afterwards. The nurse and cardiologist both rejected his friend's request, giving no explanation other than "it's just not possible." (And I remember, with dismay, hearing these exact words from a physician when I wanted to accompany my newborn daughter to the hospital nursery for her first bath.)

After tracing the recent history of "patient-centeredness" as an aim of various groups devoted to improving health care quality, Berwick proposes what he calls an "extremist" definition of patient-centered care:

The experience (to the extent the informed, individual patient desires it) of transparency, individualization, recognition, respect, dignity, and choice in all matters, without exception, related to one's person, circumstances, and relationships in health care.

To me, this definition sounds hardly revolutionary, and more like common sense. The sticking point for most clinicians, of course, is "without exception." Does this mean that clinicians should be required to fulfill patients' all-too-common requests for tests and procedures that have inconclusive clinical benefit or are likely to lead to harm? Well, in many cases, such as antibiotics for colds or imaging for acute low back pain, most of us already do, despite knowledge of potential adverse consequences. One might assert that these seemingly irrational requests occur because the health professional hasn't done enough to explain the risks and benefits of various options, or more importantly, understand what those risks and benefits mean from the patient's perspective.

This topic leads me a study about getting to "no" published in today's Archives of Internal Medicine, in which "standardized patients" (trained actors posing as patients) visited the offices of internists and family physicians, and, after complaining of vague symptoms of fatigue, requested a specific medication for depression that they said they had learned about in an advertisement. In the majority of these encounters (111 out of 199), doctors simply fulfilled the request, even though the information that the actors gave clearly did not meet criteria for the diagnosis of depression. Of the 88 who did not, 31 either gave no reason for saying no or deflected the request by ordering (unnecessary) tests or prescribing a sleep aid. In the remaining 53 encounters, the doctor attempted to further explore the content of the request or suggested a referral to a mental health counselor to accomplish the same end.

Although the study's authors did not explicitly endorse a particular approach to saying "no," it seems to me that the last one, which one might call the "patient-centered" strategy, is the most likely to benefit the physician-patient relationship and improve health care in the long run. I'm not saying that this negotiation is easy, and it certainly isn't the sort of thing that is routinely taught in medical school or residency. It's only what primary care physicians do, day in and day out - and like many other aspects of the current practice model, we need to do it better.