Most experts advocate for early detection of cognitive impairment (CI) so that patients and caregivers can be prepared for making difficult decisions and to improve quality of life, but studies show that screening alone isn't sufficient to change clinician actions related to early detection. Using predictive modelling developed with machine learning methods and sophisticated clinical decision support (CDS) tools, it is possible to identify patients at elevated risk for CI and make it much easier for primary care to engage and support patients and caregivers in meaningful care planning. The goal of this study is to implement and evaluate a low-cost, highly scalable CI-CDS system integrated within the electronic health record that has high potential to improve early CI detection and care and translate massive public and private sector investments in health informatics into tangible health benefits for large numbers of people.

The prevalence of Alzheimer's disease (AD) and AD-related dementias (ADRD) is expected to triple by 2050, contributing to decreased quality of life, increased medical care utilization, and additional burden on an already stressed primary care system. Many clinicians lack confidence to assess, diagnose and manage cognitive impairment (CI), and more than 50% of patients with CI are undiagnosed. Unfortunately, studies show that even in settings with high rates of standardized CI screening, very few patients who screen positive have documentation of any clinician follow-up action. To address these important problems, a machine learning model (called MC-PLUS) was developed and validated using electronic health record (EHR) data to identify patients at elevated risk of a future dementia diagnosis (AD/ADRD). A web-based and EHR-integrated CI clinical decision support (CI-CDS) system was also developed and validated to engage patients and clinicians in conversation about elevated dementia risk, and to give clinicians the confidence and tools they need to diagnose and manage CI. Both MC-PLUS and the CI-CDS system have been added into an existing web-based CDS platform that has high use rates and primary care clinician satisfaction, and is already seamlessly integrated within the EHR. This CDS platform improves outcomes for patients with chronic diseases such as diabetes and high cardiovascular risk as shown in published studies. After systematically validating the CI-CDS system with expert champions and conducting a pilot test at three primary care clinics, a pragmatic, clinic-randomized, controlled clinical trial is now being implemented in 38 primary care clinics randomized to receive CI-CDS or usual care (UC). All primary care visits that take place at the randomized clinics after the CI-CDS system is implemented will be screened for intervention eligibility. Patients will be accrued into the study on the date of their first visit during the accrual period that meets all intervention eligibility criteria and followed for the duration of the observation period. Primary care clinicians in the intervention clinics will be encouraged but not required to use the CI-CDS with eligible patients, so that the decision to use or not use CI-CDS at a given clinical encounter is up to the clinician. The CI-CDS user interface will provide updated clinical recommendations at primary care encounters for patients with elevated CI risk or with a CI diagnosis. The interface will enable the user to administer diagnostic screening exams (e.g., Montreal Cognitive Assessment (MoCA), Patient Health Questionnaire (PHQ-9)), place quick orders (e.g., referrals, procedures, lab assessments, medications), accurately diagnose CI, provide patient education materials (e.g., diagnoses, legal documents, community resources), and manage CI (e.g., visualize trends in screening exams, lab values, medications). If successful, the CI-CDS system will improve rates of new CI diagnosis and narrow existing sociodemographic disparities in adults with elevated dementia risk at index visit in CI-CDS compared to UC clinics. The impact of the intervention on care management and care plans will be evaluated using EHR data and chart audits. Change in clinician confidence in CI detection and care management will be evaluated in CI-CDS compared to UC clinics. Determinants of clinician actions in response to the CDS system will be assessed using behavior change theory and technology acceptance constructs, and phone surveys of patient and caregiver dyads will be conducted to evaluate intervention effects on feelings of preparedness for decision making and distress. The CI-CDS system is immediately scalable to large numbers of patients through the existing non-commercialized CDS platform already in use for millions of patients in care systems spanning 14 states. The CDS system implemented as described could maximize return on massive investments that have been made in EHR systems, and provide a prototype to rapidly and consistently translate evolving evidence-based CI guidelines into personalized CI care and guidance within primary care.

Inclusion Criteria:

• Primary care office visit at a randomized clinic during the accrual period, AND

• Patient is age 65 or over, AND

• Patient has no CI diagnosis documented in the EHR prior to the visit, AND

• Patient has:

  • Any abnormal score on a comprehensive cognitive assessment (MoCA, MMSE or SLUMS) in the prior 18 months, OR
  • a MiniCog score of <3 in the prior 18 months and there is no evidence of subsequent comprehensive cognitive assessment (MoCA, MMSE, SLUMS), OR
  • No cognitive testing in the last 18 months and a risk of a dementia diagnosis in the next 3 years >=15% as calculated by the MC-PLUS algorithm developed in the R61 phase

AND

-First visit during the accrual period at which all prior inclusion criteria are met

Exclusion Criteria:

• Patient has received active parenteral chemotherapy within the last year
• Patient has stage 4 or equivalent cancer diagnosis
• Patient is enrolled in hospice care or palliative care program