Background: Patients undergo millions of noninvasive diagnostic imaging tests, including computed tomography, magnetic resonance imaging, and nuclear medicine, in the United States annually. Based on Appropriate Use Criteria (AUC) developed by professional medical societies, 20-30% of these tests may be inappropriate, i.e., the potential benefit-to-risk ratio is not maximized. Limited data on appropriateness of imaging within the Veterans Health Administration (VHA) suggests that inappropriate use is similar to that outside the VHA. A variety of methods have been employed to encourage more appropriate use of noninvasive imaging. One of the most popular is an electronic decision support tool (DST). These software products are added to patient care workflows to capture meaningful AUC data and reduce unnecessary testing. A recent VHA Evidence Synthesis Program (ESP) report concluded that little is known about the implementation of DSTs and possible negative ramifications of their use. Further, the cost to the VHA of noninvasive imaging testing, which has not been calculated, is needed to determine the burden of inappropriate tests. Specific Aims: In the course of this award, we will achieve the following specific aims: (1) assess potential barriers and facilitators to successful adoption of a DST, (2) determine the effectiveness of a DST using a stepped implementation process, and (3) estimate the fiscal effectiveness of a DST in the VA health system by developing a micro-costing model for delivery of noninvasive imaging tests. Significance: Through these investigations, we will glean valuable insight into how to adjust processes to improve the ordering of millions of costly tests. If effective at reducing unnecessary tests, the results from this proposal will ultimately improve Veterans' access to care and reduce radiation risks. Innovation: Study of DSTs has focused primarily on effectiveness, with little attention paid to the implementation process and possible adverse effects such as provider dissatisfaction and inefficiency. Our investigations will break new ground in studying these aspects of DSTs. The use of qualitative and micro- costing methods as we propose is innovative given their underuse for clinically oriented research. Methods: We will apply the Practical Robust Implementation and Sustainability Model (PRISM) and Theoretical Domains Framework (TDF) to the specific aims for this award. In Specific Aim 1, we will use qualitative and quantitative methods to identify possible barriers and facilitators to DST implementation through interviews and surveys of providers who order noninvasive tests. In Specific Aim 2, implementation of a DST for noninvasive imaging will be assessed from the rate of testing, the proportion of testing that is inappropriate, and feedback from providers in three discrete steps. First, providers blinded to AUC for imaging tests will order tests as usual. Second, at the point of ordering, they will receive immediate feedback on appropriateness of the test and be given the option to change the order. Third, if the provider's decision is not consistent with AUC, the provider will be required to give a justification for the test order. These steps will help us understand the implementation process for DSTs and to address gaps in the literature raised by the recent ESP. In Specific Aim 3, the cost of noninvasive imaging tests will be calculated using information from administrative databases, time-and-motion observation of care delivery, and cost diaries from Veterans. Expected Results: We anticipate that through these investigations, we will learn and be able to share important lessons about implementation of a DST for noninvasive imaging. The results will be the basis for investigator initiated research grants on further spread of DSTs within the VA system and adaptation to other imaging/testing modalities such as echocardiography.

Noninvasive imaging tests, including computed tomography (CT), magnetic resonance imaging (MRI), and nuclear medicine are performed millions of times every year in the United State. An estimated 20-30% of these tests do not benefit the patient and are considered inappropriate. Most of these tests involve radiation, which is imparts a small risk of future cancer. Inappropriate tests may delay or limit access to patients with appropriate indications for testing. Reducing inappropriate testing could reduce costs and make funds available for other Veteran services. In this award, we propose a series of investigations to better understand why inappropriate tests are ordered, what the cost is to the VA system, and implement a system to reduce inappropriate tests. Veterans will directly benefit from this research through reduction in wait times and avoiding harm from radiation; indirectly the VA health system will benefit from cost avoidance.