Clinician-patient communication measures: drilling down into assumptions, approaches, and analyses

To critically examine properties of clinician-patient communication measures and offer suggestions for selecting measures appropriate to the purposes of research or clinical practice assessment.

We analyzed different types of communication measures by focusing on their ontological properties. We describe their relative advantages and disadvantages with respect to different types of research questions.

Communication measures vary along dimensions of reporter (observer vs. participant), focus of measurement (behavior, meaning, or quality), target, and timing. Observer coded measures of communication behavior function well as dependent variables (e.g., evaluating communication skill interventions, examining variability related to gender or race), but are less effective as predictors of perceptions and health outcomes. Measures of participants' judgments (e.g., what the communication means or how well it was done) capture patients' or clinicians' experiences (e.g., satisfaction) and can be useful for predicting outcomes, especially in longitudinal designs.

In the absence of a theoretically coherent set of measures that could be used across research programs and applied setting, users should take steps to select measures with properties that are optimally matched to specific questions.

Quality assessments of clinician-patient communication should take into account the timing of the assessment and use measures that drill down into specific aspects of patient experience to mitigate ceiling effects.