Clinical versus statistical significance: interpreting P values and confidence intervals related to measures of association to guide decision making

Abstract

Pharmacists need to apply outcomes from studies to reduce risk and improve patient care. Interpretation of outcomes is based on a variety of assessment tools, such as P values and confidence intervals (CIs). P values determine statistical significance of data, while CIs suggest the degree of clinical application. Many health care providers might not have the skill set required to carefully examine and interpret statistical results and then are required to assume that the researchers of the study correctly interpreted and presented the statistical results. The reluctance to examine statistical data often reflects a misconception that concepts such as P values and CIs are difficult to understand, while in reality, both can be interpreted once basic definitions and applications are understood. Measures of association such as number needed to treat can serve as effective tools for quantifying important parameters that ultimately affect patient care. A basic understanding of how to interpret and apply P values and CIs enhances one's ability to effectively assess the validity of results from the literature. An informed reader, armed with tools for critical analysis, is in the best position to evaluate studies and thereby discern which information is applicable to a specific patient care decision.