Effective communication about possible risks and benefits of medical interventions to patients has been an integral part of medical care since the doctrine of informed consent was adopted. In recent years, it has become increasingly more important due to changes in the accepted model of the doctor-patient relationship that promote more active involvement of patients in decisions about their care, the rise of evidence-based medical practice, and the increasing emphasis placed on preventive measures that are intended to reduce future health risks. Effective communication about the possible outcomes of medical interventions is also an essential component of methods for improving the translation of biomedical research findings into practice. Despite its importance, there are large gaps in our knowledge of how to effectively convey information about uncertain future events to patients and practitioners.
The specific aims of this study are: 1) To provide initial estimates differences among currently recommended numeric and graphic formats in their ability to convey information to patients about possible outcomes that vary across a representative range of clinically important likelihoods and consequences;2) To provide initial estimates of interactions between the effectiveness of the risk communication formats and patient characteristics including: numeracy, literacy, educational level, socio- economic status, age, and racial/ethnic background;and 3) To develop an analytic framework and to gather data about key parameters for use in planning and conducting a follow-up R01-level study that will thoroughly address these issues. The patient population will consist of demographically-stratified sample of 360 patients at least 18 years of age who are patients at three outpatient practices in Rochester New York. An incomplete factorial design with two five-level within-subject factors will be used. One factor will be five currently recommended risk communication formats: vertical bar charts, grouped icon displays, risk scales, frequency diagrams, and numeric frequency information. The other factor will be five outcomes of a disease screening program: risk of disease, risk of dying from disease, risk of a serious screening side effect, risk of dying from a screening side effect, and risk of a false positive screening test result. Each subject will be randomly assigned to respond to risk information in five of the possible 25 combinations such that they are presented with each risk outcome and each risk format in some combination. To facilitate the design of a future comprehensive R01-level study, data from this study will be used to gain initial estimates of risk format and subject variables on the differences among the five risk presentation formats on the following outcomes: a) comprehension, b) usefulness for decision making, c) hazard-response consistency, d) uniformity, e) patient evaluation, and f) direction of communication errors. We will also determine if there are associations between these outcomes and either the nature of the risk or individual patient characteristics including demographics, education, literacy, and numeracy.
Effective communication about possible risks and benefits of medical interventions is essential for ensuring the delivery of consistently high quality health care. Despite its importance, there are large gaps in our knowledge of how to effectively convey information about uncertain future outcomes of health care decisions to patients and practitioners. This project will begin to address these gaps by: a) comparing how well five currently recommended risk communication formats convey risk information to patients across a range of clinically important probabilities and outcomes, and b) determining whether patient characteristics such as demographics, numeracy, and literacy affect the effectiveness of health risk communications.
| Dolan, James G; Qian, Feng; Veazie, Peter J (2012) How well do commonly used data presentation formats support comparative effectiveness evaluations? Med Decis Making 32:840-50 |
