Covert Cerebrovascular Disease Detected by Artificial Intelligence (C2D2AI): A Platform for Pragmatic Evidence Generation for Stroke and Dementia Prevention - Project Summary Incidental imaging findings are common; an estimated 15-30% of all diagnostic imaging and 20-40% of CT examinations contain at least one incidental finding. Comprised of both covert brain infarct (CBI) and white matter disease (WMD), covert cerebrovascular disease (CCD) is a common incidental finding among adults over 50 and is a risk factor for stroke and dementia. Our epidemiological work, leveraging a natural language processing (NLP)-enhanced real-world cohort, indicates that incidentally discovered-(id-)CCD patients are at two to three-fold the risk for future stroke and dementia, consistent with prior findings. Health systems have rushed to comply with the 21st Century Cures Act Information Blocking Rule (hereafter “Rule”) which requires immediate disclosure of id-CCD through release of radiology reports, often without the benefit of interpretation by referring physicians. Prior to the Rule, only about 10% of patients were informed of CCD incidental findings. Early reports suggest that incidental findings have contributed to heightened patient distress and increases in workload for clinicians faced with more portal messages, referral requests, and calls to address incidental findings. Yet, existing bioethical guidance related to disclosure of incidental findings is largely silent on this issue, responsive instead to concerns raised by genetics/genomics and clinical research literatures, which present unique considerations that are not directly transferrable to other routine care. Lack of guidance regarding how to disclose incidental findings and support patients who discover new diagnoses absent patient tailored education about the condition, risks, and potential treatment options underscores an important gap in capacity pertaining to ethically disclosing incidental findings in routine care and developing supports for patients and clinicians for patients with id-CCD, with generalizable findings for other patient populations. Our preliminary findings underscore unmet needs and increasing demand for development of evidence-based disclosure statements and educational materials to support patients and clinicians. This study addresses ethical tensions for clinicians determining how to balance ethical obligations for transparency and patient autonomy with obligations to protect patients from unnecessary distress. Our studies will complement the Parent Grant and build capacity to improve practice guidelines and education for disclosure of id-CCD. We will use data about patient and clinician preferences for disclosure, education, and engagement, to develop patient-centered disclosure statements and an education toolkit to improve quality care for older adults with id- CCD and their providers by clarifying how to best ethically disclose incidental findings, support patient autonomy and engagement, and promote appropriate health care use.
