The Science of Quality Improvement

Abstract

Scientific rigor should be consistently applied to quality improvement (QI) research to ensure that healthcare interventions improve quality and patient safety before widespread implementation. This article provides an overview of the various study designs that can be used for QI research depending on the stage of investigation, scope of the QI intervention, constraints on the researchers and intervention being studied, and evidence needed to support widespread implementation. The most commonly used designs in QI studies are quasi-experimental designs. Randomized controlled trials and cluster randomized trials are typically reserved for large-scale research projects evaluating the effectiveness of QI interventions that may be implemented broadly, have more than a minimal impact on patients, or are costly. Systematic reviews of QI studies will play an important role in providing overviews of evidence supporting particular QI interventions or methods of achieving change. We also review the general requirements for developing quality measures for reimbursement, public reporting, and pay-for-performance initiatives. A critical part of the testing process for quality measures includes assessment of feasibility, reliability, validity, and unintended consequences. Finally, publication and critical appraisal of QI work is discussed as an essential component to generating evidence supporting QI initiatives in radiology.

Keywords: Quality improvement; metrics; pay-for-performance; public reporting; research methods.