Reducing Risk of Dementia Through Deprescribing (R2D2)

A cluster-randomized controlled trial (RCT) called "Reducing Risk of Dementia through Deprescribing" (R2D2) to evaluate the impact of a deprescribing intervention on important cognitive and safety outcomes.

The R2D2 study will test whether the adverse cognitive effects of anticholinergic medications are reversible by implementing a pharmacist-based deprescribing intervention for older adults within primary care practices. Two groups will be recruited: providers (physicians and advanced practice providers including nurse practitioners), and patients. Primary care providers of those prescribed eligible anticholinergic medications will be recruited for participation in the study, and their patients who also meet eligible criteria will be subsequently approached and recruited. Participants will be randomized to one of two groups: the deprescribing intervention group or usual care; the intervention group will receive a pharmacist-based deprescribing intervention, while the usual care group will receive care as usually provided by their primary and/or specialty care providers. The intervention and follow-up data collection will occur over 24 months in order to test the long-term impact of the intervention on the planned clinical outcomes. Study outcomes include cognition (primary) and safety (secondary) through validated self-reported scales.

This cluster-randomized trial will randomize at the level of physicians. Physicians agreeing to participate in the trial will be randomized to intervention or usual care in blocks of two or four. Physician randomization status will determine participants' study group. Physicians randomized to usual care will not have access to the intervention. Study outcomes will be collected directly from participants; no outcomes will be collected from physicians.

The principle investigator and all outcome assessor will be blinded to the arm assignment of the subjects. No access to unblinded data will be provided to blinded staff.

The active intervention group (ACT) will receive a pharmacist-based deprescribing intervention focused only on targeted anticholinergic medications. The intervention pharmacist will serve as the central source of communication between participants, providers, and (as needed) dispensing pharmacy to coordinate the deprescribing process. The study pharmacists will navigate a shared-decision model between the physicians and participant in order to personalize the selection of appropriate alternatives and switch/titration schedules. Importantly, the pharmacist will supervise the titration and deprescribing plan and communicate with both participants and physicians throughout the study.

Those in the usual care group will not have access to the study intervention, but will receive a one-time information packet through the mail reviewing risks of polypharmacy, but no information specific to anticholinergic medications. They will receive clinical care as usually provided by their primary care or specialty care physicians.

Cognitive composite scores will be collected from participants at baseline, 6, 12, 18 and 24 months and changes in the composite scores over time will be compared between the intervention and usual care groups. An overall cognitive composite score including measures of information processing speed, memory, and executive function will be conducted at each time point from the average of each measure's z-score, constructed by subtracting the mean baseline scores and dividing by the baseline standard deviation. The z-score transformation of the cognitive composite score will have a mean of 0 and standard deviation of 1 at baseline, with higher scores representing improvement in cognition.

Participant self-reported, 4-item scales evaluating depression, anxiety, pain, and insomnia will be collected for each participant at each time point. Each PROMIS measure raw score can be converted to a T-score where 50 represents the general population norm for that symptom and each 10-point deviation represents one standard deviation (SD) from the population norm. Changes in the T-scores over time will be compared between the intervention and usual care groups.

Participant self-reported measure of health-related quality of life, evaluating domains including vision, hearing, speech, ambulation, dexterity, emotion, cognitive function and pain. These attributes produce a single score on a standardized utility measure with individual health domain scores ranging from 0.00 (maximum impairment) to 1.00 (no impairment) and the multi-attribute (HUI) scores ranging from 0.36 to 1.00 with anchors 0.00 = dead and 1.00 = perfect health. HUI scores will be collected at each outcome assessment and change in HUI scores over time will be compared between the intervention and usual care groups.

Inclusion Criteria for Participants: Age 65 and older; At least one office visit to their primary care physician within the previous 12 months; Use of a target anticholinergic medication within the last two weeks OR medical record evidence of exposure to target anticholinergic medications at or above a cognitive risk threshold in the prior 12 months Able to communicate in English; Access to a telephone Exclusion Criteria for Participants: Permanent resident of an extended care facility (nursing home) Diagnosis of schizophrenia, bipolar disorder, or schizoaffective disorder defined by International Classification of Diseases (ICD) version 9/10 codes Diagnosis of Alzheimer's Disease or Related Dementia as determined by (a), (b), or (c) below: ICD-9/10 codes, or Current use of a medication for Alzheimer's Disease or a Related Dementia, or A pattern of responses to the Functional Activities Questionnaire (FAQ) that indicate dementia (i.e., ≥ 3 FAQ items are scored at "requires assistance," or if ≥ 1 FAQ item is scored at "dependent").

Data collected through this trial will be made available to either advance scientific research in a way that is allowed by the federal regulations that protect research subjects, or for the purpose of auditing or program evaluation by the government or funding agency. Data sharing for scientific purposes is essential for further translation of research results into knowledge, products, and procedures to improve human health. To protect participants' rights and confidentiality, protected health information will be limited to the minimum necessary for authorized oversight before the data are shared.The final dataset will include demographic, clinical, and limited genetic data. The final subject-level, de-identified dataset will be made available to qualified individuals within the scientific community.

The final data set will be made available no later than 9 months of the database lock or at the time of on-line publication of the primary results, whichever comes first.

We anticipated access to the study data will be facilitated in collaboration with the Global Alzheimer's Association Interactive Network (GAAIN). As a GAAIN Data Partner, data from this study will be shared in aggregate through the GAAIN portal (gaain.org). GAAIN provides a global infrastructure for cooperative research by linking data repositories that have collected information from participants who are at risk for or have been diagnosed with Alzheimer's disease. The GAAIN network allows the investigator to retain ownership of the data collected in each study, and can detach the data at any time. Only de-identified data is shared with GAAIN or any other research entity.