Cancer Control Research7R01CA117840-06
Gould, Michael K.
EVALUATION OF COMPUTERIZED DECISION SUPPORT FOR MANAGING LUNG NODULES
DESCRIPTION (provided by applicant): In previous research, members of our group developed ALCHEMIST, a computerized decision support system that generates patient- and site-specific recommendations for patient management based on decision- analytic models. Subsequently, we developed ALCHEMIST-SPN, a web-based decision support system for managing patients with solitary pulmonary nodules (SPN), by linking a rigorously developed decision analytic model with the existing ALCHEMIST architecture for computer-based decision support. Concurrently, we performed a statistical validation of the decision model, and showed that decision support with ALCHEMIST-SPN has the potential to improve quality-adjusted survival, reduce cost, and reduce the frequency of unnecessary surgery in patients with benign nodules and inappropriate "watchful waiting" in patients with malignant nodules. Management of lung nodules is a very attractive target for computer-based decision support; it is a common problem, multiple possible management strategies exist, and the optimal management approach depends on pre-test probability, the risk of surgical complications, patient preferences, and other factors. In this second revision of our competing renewal application, we seek funding to continue our systematic evaluation of ALCHEMIST-SPN. Specifically, we have revised our aims and methods based on helpful feedback from reviewers of our previous applications. The revised proposal includes three specific aims: (1) to conduct a rigorous process of usability testing and peer review, to enable design and refinement of a decision support system that is highly responsive to the needs of potential users; (2) to determine whether ALCHEMIST-SPN has the potential to improve outcomes in patients with lung nodules under idealized laboratory conditions, by conducting a randomized, controlled trial that compares management of clinical vignettes by physicians following exposure to either ALCHEMIST-SPN or a static flow-chart algorithm; (3) to implement ALCHEMIST-SPN in two health care settings, by integrating the decision support system into the clinical workflow of SPN management, and to conduct a "before and after" implementation study of ALCHEMIST-SPN in these settings to determine whether clinicians will use the decision support system and adhere to its recommendations. By accomplishing these aims, we will not only evaluate the performance of a highly innovative computerized decision support system for managing lung nodules, but also improve the system's performance based on feedback from both experts and practicing clinicians in laboratory and clinical settings. Ultimately, we believe that our approach to computerized decision support with ALCHEMIST-SPN has the potential to improve management in patients with lung nodules, and may be broadly applicable to other challenging problems in thoracic oncology and clinical medicine. PUBLIC HEALTH RELEVANCE This research aims to evaluate a novel, computer-based system for improving management in patients with solitary pulmonary nodules (or "spots on the lung"), which sometimes represent a curable form of lung cancer. By evaluating and eventually implementing the system, we aim to reduce the frequency of unnecessary surgery in patients with non-cancerous lung nodules, and reduce the number of delayed surgeries that may lead to missed opportunities for cure in patients with cancerous nodules.