Development and testing of clinical decision support tools to optimize the process of lung cancer screening in primary care settings.

554 Background: Primary care providers (PCPs) routinely address health screening during clinical encounters and are ideally positioned to promote awareness among patients who would benefit from lung cancer screening (LCS). Integrating clinical decision support (CDS) tools for LCS into the electronic health record (EHR) within primary care settings can potentially serve as a strategy for the implementation of LCS. This study aimed to (1) identify PCPs’ recommendations for LCS CDS tools and (2) develop LCS CDS tools and test their usability in primary care settings. Methods: User-centered design principles were used to gather information from 23 PCPs about their recommendations for LCS CDS tools. Qualitative research methods, including individual interviews and group meetings, were used to collect data on the desired components of LCS CDS tools that would be useful in primary care. Braun and Clarke’s thematic analysis was used to analyze the qualitative data. Member checking was conducted to validate the accuracy of the recommendations. A set of EHR-based CDS tools was developed based on the recommendations. The usability of the CDS tools was tested using a think-aloud methodology with five PCPs and five patients. Iterative changes were made until no further feedback was provided by the PCPs and patients who reviewed the CDS tools. Results: Four components for LCS CDS tools were identified as important: (1) streamline the assessment process to increase the accuracy of tobacco use data collection, (2) provide shared decision making educational resources for patients before the visit and integrate shared decision making tools into the best practice alerts for use by PCPs, (3) provide non-interruptive best practice advisory alerts that were easy to see within the EHR, and (4) streamline the process for ordering CT scans and enhance communication across departments about results and future testing. The CDS tools that were developed included: (1) a patient-reported tobacco use survey, (2) order set for tobacco treatment, (3) best practice alerts to notify PCPs about LCS eligibility based on the accuracy of information available from the tobacco use survey and the EHR, (4) shared decision-making educational materials, and (5) LCS CT order sets. PCPs provided positive feedback about the CDS tools except for the order set for tobacco treatment. PCPs indicated that the order set for tobacco treatment was difficult to use since it disrupted their workflow. Patients provided positive feedback about the educational resources and stressed the importance of having tobacco use questions framed in a way that minimized stigma associated with tobacco use. Conclusions: PCPs recommended four components for CDS tools to facilitate LCS, with a focus on strategies to streamline clinical workflows. Future testing of the CDS tools is needed to determine whether they facilitate implementation of LCS.