https://immattersacp.org/weekly/archives/2014/11/18/2.htm

Communication scores vary significantly within practices, especially low-performing ones

Patients' scoring of physician communication varied much more by physician than by practice, raising concerns about assessment of performance on a practice level, according to a new study.


Patients' scoring of physician communication varied much more by physician than by practice, raising concerns about assessment of performance on a practice level, according to a new study.

Researchers in England looked at surveys returned by more than 7,000 patients treated at 25 general practices by 105 physicians. The surveys asked patients about access, waiting times, opening hours, and continuity and interpersonal aspects of care, as well demographic information. Results were published by The BMJ on Nov. 11.

When the communication scores were specifically examined, the study found that 6.4% of the variation was due to differences between physicians, compared to 1.8% of variation from practices, after the researchers controlled for differences in age, sex, ethnicity, and health status. The study also suggested that practices that performed worse overall on the surveys had more variation between physicians, while “higher-performing practices usually contain only higher-performing doctors,” the authors wrote. They concluded that aggregating communication survey scores from a practice could mask variation in physician performance.

Given that the variation seemed to be greater in lower-performing practices, it's possible that practice-level assessments could be effectively used as a screen to identify practices in need of individual-level assessments, the study authors suggested. Another potentially useful finding of the study was that a reliable score for an individual physician could be collected with 46 patient ratings, the authors also noted.

An accompanying editorial praised the research for its insight into how to measure and improve quality of primary care but called the conclusions about differences in variation between high- and low-performing practices contentious, “since the paper does not report whether any of the practices or doctors were statistically significantly above or below average on their performance.” It's also still uncertain how to respond to physicians who show low performance on such surveys, the editorial said.