Relationship-centered Care and the Patient-Physician Relationship

I recently observed a third-year medicine resident interview a 77-year-old established clinic patient with diabetes, hypertension, and mild degenerative joint disease, all well controlled. For the first 19 minutes, of a visit scheduled for 20, the patient focused on reconfirming that her symptoms, unchanged for months, were related to her diagnoses. She also talked animatedly about her granddaughter's upcoming wedding and the dress she had bequeathed her. As the resident prepared to leave, the patient said, “Oh by the way, Dr. Green, could you give me prescriptions for Valium, Lomotil, and Oscal?” With virtually no time left for discussion, the resident grudgingly wrote scripts for all three medications.

After the encounter I asked the resident her opinion of the visit. She said, “I can't believe she waited until the very last minute to bring up her real concerns. I really like this patient but I'm really unhappy with how this visit went.” Interviewed as part of a practice improvement project, the patient said, “I love Dr. Green. She's a good listener and really cares about me as a person. She also trusts me to tell her what medications I need and doesn't waste time telling me how to take them.”

This scenario illustrates at least 4 factors that can make physician-patient relationships challenging to engage in and research. They are: 1) communication preferences; 2) visit expectations; 3) communication and relationship; and 4) satisfaction. Several articles in this issue of the Journal of General Internal Medicine explore these issues.

1. Communication preferences: A growing literature supports the assertion that patient-centered interviewing is associated with positive care outcomes.^1–4 These findings have led some, including the Institute of Medicine, to call for adoption of a patient-centered approach to all patients.⁵ As tantalizing as these results are, a fundamental question remains: do all patients prefer and benefit from patient-centered care?

   Swenson et al.'s article addresses this question.⁶ Using video vignettes of simulated discussions about complementary and alternative medicine (CAM), the authors found that while a majority of patients preferred a patient-centered style, more than 30% did not. Despite differences by age, education, use of CAM, and physician interview style, patient preferences varied across the board. The take-home lesson from this study is clear: one size does not fit all. Using a strictly patient-centered approach may leave as many as one third of patients preferring a different approach. Physician flexibility is the skill necessary to address variation in patient preferences.

2. Visit expectations: It is clear that the patient and resident above had very different ideas and expectations of one another. A recent study found that a fifth of all patients had visit expectations that were unmet and 9% never got to mention their expectations at all.⁷ Arthur Kleinman, an anthropologist and physician, has coined the term “explanatory model” to describe differing cultural expectations and beliefs physicians and patients have about one another.⁸

   One might expect unmet expectations and patient satisfaction to be inversely related. Everyone has a story about a patient who didn't get what they wanted, “fired” their physician, and never came back. This assumption makes the findings by Peck et al. of special interest.⁹ The authors found that patient satisfaction was extremely high irrespective of whether expectations were or were not met. One possible explanation is that satisfaction is too general to measure the effect(s) of unmet expectations. Another more intriguing possibility is that expectations for tests and treatments are only one component of what defines satisfying relationships. Additional research will be necessary to better understand how expectations and satisfaction relate.

3. Communication and relationship: The Accreditation Council for Graduate Medical Education (ACGME) now requires that all trainees demonstrate competence in 6 areas of practice, including communication and relationship building.¹⁰ Other high-stakes testing organizations such as the American Board of Internal Medicine (ABIM), American Board of Medical Specialties (ABMS), and the Educational Commission for Foreign Medical Graduates (ECFMG) have adopted and are testing for these competencies. In the past, good bedside manner was seen as a desirable, but essentially unmodifiable, physician communication trait. Health services researchers have now demonstrated that communication skills associated with improved outcomes such as adherence can be taught, learned, and put into practice.^4,11–13

   Schneider et al. evaluated 7 dimensions of relationship quality and patient adherence to antiretroviral therapy.¹⁴ Six of the 7 measures of relationship quality were significantly associated with adherence. Two (sexual behaviors and adherence dialog) were condition specific and involve communication skills that have been shown in other contexts to be transferable.^15,16 In short, physicians can learn to improve relationship quality and with it their chances of improving patient adherence.

4. Satisfaction: Satisfaction with medical care has been a focus for research for at least 3 decades.¹⁷ Within the patient-centered care paradigm satisfaction has frequently been taken to mean patient satisfaction. In fact, some large health care organizations base clinician salaries and bonuses on patient satisfaction ratings. Such ratings may be misleading, however, as the scenario above illustrates. Focusing exclusively on the patient, who was highly satisfied, misses the fact that the physician was equally dissatisfied. Including physician perspectives in rating satisfaction expands the measurement focus from patient centered to relationship centered, an approach advocated in several recent studies.^18–23

   Zandbelt et al.'s study illustrates the benefits of a relationship-centered approach.²⁴ Using 5 visit-specific measures, they found that physicians were generally less satisfied than their patients, and that the correlation between measures of overall satisfaction were modest. That patients were more satisfied is not surprising given the ceiling effect found in many patient satisfaction studies. What is of real interest here is that patients and physicians appear to define satisfaction using different relationship qualities. For patients, it's physician gender and self-efficacy in communication; for physicians, it's higher education, better mental health, dominant language proficiency, and limited information needs. This study underscores the importance of understanding the different perspectives sufferers and healers adopt in judging each other and the care process, which in turn reveal the wonders and sheer complexity of human social relations.

The Journal of General Internal Medicine has been a leader in publishing research on communication and relationship. This issue continues that tradition with articles that explore the conceptual and empirical bases of patient-centered care. Progress in science is never measured in absolutes, and is rarely linear. As we contemplate the pros and cons of this new form of care, it is worth recalling Plato's observations about physician-patient relationships from over 3,000 years ago.

“Slaves, to speak generally, are treated by slaves…. A physician of this kind never gives a servant an account of his complaint nor asks for any; he gives him some empirical injunction with the finished air of a dictator and then is off in hot haste to the next ailing servant…. The free practitioner, who, for the most part, attends free men, treats their diseases by going into things thoroughly from the beginning in a scientific way…. He does not give his prescriptions until he has won the patient's support, and when he has done so, he steadily aims at producing complete restoration to health by persuading the sufferer into compliance.”²⁵