Medical Imaging Decision And Support

Overview

Rationale: Diagnostic medical imaging is essential in health care and guides patient management decisions. Rapid advances in imaging technology have enabled diagnosis at an early disease stage. Nevertheless, striving for early diagnosis leads to substantial overuse of diagnostic tests (too many tests are performed) and overdiagnosis (diagnosed abnormalities that are unlikely to affect a patient's health or well-being), with dire consequences to patient outcomes and health care costs. Computerized decision support systems (CDSS), such as the European Society of Radiology (ESR) iGuide, have been developed to guide referring physicians at the point of care. Objective: The overall objective of this project is to promote the appropriate, meaningful, value-based and personalized use of medical imaging. The specific objectives of this study are to evaluate the trends over time with implementation of a CDSS to guide imaging referrals (ESR iGuide) in key outcomes related to the appropriate use of diagnostic imaging tests, and to compare these trends between departments randomized between implementation of the active intervention (with decision support) and the control condition (no decision support). Study design: The study will be performed as a multi-center cluster randomized trial with departments acting as clusters combined with a before-after-revert design. Four hospitals with each 8 participating departments for a total of 32 clusters will be included in the study. All departments will start with the control condition. Subsequently, stratified per specialty (surgical vs non-surgical), half of the departments - chosen at random - will switch to the active intervention while the other half in the control condition. Study population: The study population consists of patients of eligible participating departments for whom a diagnostic imaging procedure is considered, and for whom a diagnostic imaging request is entered in a computerized order entry in such a form that it is registered by the CDSS. Intervention: The control condition will consist of computerized order entry with structured data entry of the clinical indication and tracking of the imaging exams requested. The CDSS is implemented but without decision support and physicians are blinded to these referral guidelines. Without their knowledge their imaging requests are categorized as appropriate, under certain conditions appropriate, or inappropriate. They receive no feedback to their requests. In the active intervention the CDSS is implemented with decision support. Physicians are made aware of the referral guidelines and appropriate exams are suggested. Their imaging requests are categorized as appropriate, under certain conditions appropriate, or inappropriate and feedback is provided. Physicians are able to change their request in response to the feedback. In the revert condition decision support is removed to study the sustainable educational effect of temporary use of the system. Main study parameters/endpoints: The primary outcome is the proportion of inappropriate diagnostic imaging exams requested per cluster (department). Inappropriateness is determined by the decision support system and is based on the ESR imaging referral guidelines. Secondary outcomes are the overall absolute number of imaging exams ordered (total and by exam type), estimated medical radiation from diagnostic imaging in investigated groups, and estimated medical costs for diagnostic imaging from the healthcare system perspective in investigated groups. Amendment November 2023: At Johannes Gutenberg University Mainz, the CDSS could not be integrated into the hospital's internal computerized order entry system as required for the study. Therefore, departments from this hospital were not eligible for inclusion. Therefore, and 24 instead of 32 departments were from three locations (Augsburg, Kiel, Lübeck) were randomized accordingly. In a qualitative description of the process of integrating the CDSS in the local hospitals, Mainz will still be included.