The Value of Assessing and Addressing Communication Skills

Research published in the late 1960s by Korsch et al^{1 - 2} is widely considered the foundation for contemporary inquiry into the patient-physician relationship. In a diverse set of studies since then, effective communication has been linked with increases in patient and physician satisfaction, better adherence to treatment plans, more appropriate medical decisions, better health outcomes, and fewer malpractice claims.^{3 - 6} Recent research has provided evidence-based guidance about specific aspects of the patient-physician interaction, such as greetings and self-disclosure.^{7 - 8} In addition, surveys continue to indicate that physicians are the preferred source of health information,⁹ highlighting the importance of ensuring effective patient education and counseling.

Education and accreditation initiatives have evolved along with the research trajectory.¹⁰ Indeed, the focus on communication now extends throughout the continuum of medical education and practice: US and Canadian medical schools must teach and assess communication skills to maintain accreditation, and interpersonal and communication skills are considered a core area of competency for both residents and practicing physicians.^{11 - 13} The United States Medical Licensing Examination (USMLE) has included a clinical skills assessment since June 2004.¹⁴ Known as USMLE Step 2 CS, this examination focuses on interactions with standardized patients and is administered at 5 regional test centers in the United States.

As noted in the article by Tamblyn and colleagues¹⁵ in this issue of JAMA, communication features prominently in USMLE Step 2 CS as well as the clinical skills examination administered by the Medical Council of Canada (MCC). Tamblyn et al conducted a cohort study of 3424 physicians licensed to practice in Ontario and/or Quebec who took the MCC clinical skills examination between 1993 and 1996. These physicians were followed up through 2005 to determine whether communication scores on the MCC examination were related to the rate of patient complaints regarding communication or quality of care. The researchers found that patient-physician communication scores varied widely for the study physicians and that a 2–standard deviation decrease in communication score was associated with about 1 additional retained complaint per 100 practice-years.

This is not a huge difference, but it takes on added weight when combined with the finding that communication scores in the bottom quartile were clearly associated with patient complaints. It appears that low scores on the communication component of a standardized clinical skills examination indicate higher risk for patient complaints to medical regulatory authorities for physicians practicing in Canada. It is not clear if low-scoring physicians generated particular complaint types or if physicians in the top 3 quartiles generated a different pattern of complaints. In any case, this longitudinal study suggests that identifying and appropriately addressing communication deficits among physicians may have considerable downstream benefit. The Canadian experience may not directly translate to USMLE Step 2 CS, which has a very high pass rate among first-time examinees, although it does raise the question of how best to deal with candidates who fail.

Another important question is whether collecting data at the local level also may have predictive value. The logic of the study by Tamblyn et al can be extended backward to encompass assessments made during the medical school years. For instance, Papadakis et al¹⁶ reported a strong positive correlation between a record of unprofessional behavior during medical school and subsequent disciplinary action by a state medical board. The strongest associations were with behaviors described as “irresponsible” and with poor responsiveness to feedback. Poor academic performance was also associated with disciplinary action, but less so. Widespread recognition of the potential utility of this work has led the Federation of State Medical Boards to make available to each US medical school complete, school-specific data on disciplinary actions taken by any state medical board concerning that school's medical graduates (Tim R. Knettler, MBA, vice president of Member Resource Centers and Services/interim chief financial officer, Federation of State Medical Boards, written communication, July 18, 2007).

The communication skills and behaviors assessed by the MCC clinical skills examination are in many ways distinct from the deficiencies in professional behavior studied by Papadakis et al, but both studies demonstrate the ability to predict difficulties in clinical practice based on assessment of noncognitive performance. However, ensuring the reliability of these assessments is essential. In terms of clinical skills examinations, reliability can be enhanced by increasing the number of cases and by enhancing the quality of training of standardized patients. It is difficult to completely train tendencies toward leniency or stringency out of standardized patients, and these tendencies can factor into decisions about whether students pass the examination. Accordingly, some medical educators are beginning to use statistical analyses supported by Item Response Theory to estimate “true” scores and better calibrate assessments.^{17 - 18}

Assuming that problems can be identified in a reliable manner, remediation becomes an issue for physicians-in-training and physicians-in-practice. There is a wide variety of approaches to teaching communication skills. Many medical schools use a combination of discussion, experience with simulated patients, and videotape review to provide a solid foundation for basic communication skills; some schools follow up with focused attention on advanced skills such as breaking bad news. Workshops for residents and practicing physicians abound, ranging in duration from a few hours to a few days. While the effect of these workshops is variable, it appears that even brief workshops can result in positive change if they focus on particular communication behaviors rather than try to cover too much ground at once.¹⁹

An open issue is whether training in communication skills per se is the most appropriate remediation resource for candidates who perform poorly on a standardized clinical skills examination. It is reasonable to assume that examinees would be trying to do their best during an examination that is explicitly linked to licensure. As noted by Tamblyn et al,¹⁵ “Examples of patient-physician communication that would receive a low score include condescending, offensive, or judgmental behaviors, or ignoring patient responses during the encounter.” These are not nuanced communication behaviors. Thus, it may be worth encouraging direct conversations with examinees about the attitudes and emotions that generated, for example, disrespectful communication during the encounter in question. It is unlikely that a clinical skills examination would be the first instantiation of such behavior.

This observation underscores the importance of addressing professional skills and perspectives early and often during medical education.²⁰ In terms of communication, initiatives could include more systematically assessing interpersonal skills during the admissions process, better connecting clerkship and residency experiences to earlier training in communication skills, and ensuring that clinical skills assessments include a communication component. The momentum is building. More medical schools are adopting a competency-based approach that features interpersonal and communication skills, paralleling the framework that is in place for residency programs and for maintenance of certification. At the same time, clinical skills laboratories and simulated-patient expertise are now expanding within US medical schools and rapidly spreading to graduate and continuing medical education. The finding that clinical skills examination scores predict future patient complaints is an important step toward establishing the value of efforts to improve both medical education and patient care.