Brain Safe: Consumer Intervention to Reduce Exposure to Drugs Linked to Alzheimer's Disease

To calculate the Total Standardized Daily Dose (TSDD), we first calculated the Standardized Daily Dose (SDD) for each anticholinergic medication with an ACB Score of 2 or 3. SDD was calculated by multiplying the strength by the units per dose and frequency per day and number of refills. The product was divided by the minimum effective daily dose (where available) or the minimum effective geriatric dose to standardize the quantity across multiple classes of medications. The SDD for all anticholinergics with an ACB score 2 or 3 was then summed for each participant for each 6-month period, and divided by the number of days in the relevant six month period to arrive at a TSDD.

Range: The minimum possible value is 0 (optimal outcome; indicating zero exposure to ACB 2 or 3 medications). There is no fixed theoretical maximum score(higher value more exposure), as the upper limit is mathematically determined by the total volume and combination of medications a participant is prescribed.

Construct: Measures natural log-transformed average daily exposure to medications with an Anticholinergic Cognitive Burden (ACB) score of 2 or 3 to normalize data distribution.How Computed: Medication Standardized Daily Dose (SDD) = (strength × units/dose × frequency/day × refills) / minimum effective daily or geriatric dose. The SDD for all eligible drugs is summed over 6 months, divided by the number of days in that period to find raw TSDD, and natural log-transformed ($\\ln(x+1)$ or similar) for the final score.Scale Range: Minimum value is 0 (representing 0 raw TSDD, or zero exposure). There is no fixed theoretical maximum, as it is mathematically determined by the volume of medications prescribed.Interpretation: A value of 0 is the optimal outcome (no exposure). Higher values represent greater exposure (worse outcome).

An overall cognitive score evaluating three domains: memory/new learning, executive function, and processing speed. Calculated by taking the mean of standardized Z-scores from the following specific tests: Hopkins Verbal Learning Test-Revised (HVLT-R) total and delayed recall; Trail Making Test Parts A and B; Phonetic Fluency; Semantic Fluency (adjusted); and Symbol Digit Modalities Test (correct).

A Z-score of 0 represents the baseline study population mean.

Standardized scores theoretically range from -3.0 to +3.0. Higher positive Z-scores (standard deviations above the mean) represent better cognitive performance (better outcome). Lower negative Z-scores represent worse cognitive performance (worse outcome). No clinical threshold is defined. Z-scores were derived using baseline means, with timed tests (e.g., Trails) inverted prior to averaging so higher always means better performance.

Computer-based assessment of Choice Reaction Time (CRT). It is used to evaluate executive function, attention, and psychomotor speed by measuring the time elapsed between the presentation of one of multiple possible stimuli and the participant's correct response.

The final score represents the average response time (in milliseconds/seconds) across all valid test trials.

The minimum value is 0. There is no fixed theoretical maximum, as the upper limit is bounded only by the participant's maximum delay.

Lower values represent faster reaction times, indicating better executive functioning and psychomotor speed (a better outcome). Higher values indicate slower reaction times (a worse outcome).

The Health Utilities Index Mark 3 (HUI3) is a self-reported, multi-attribute health status classification system used to measure overall health-related quality of life and functional capacity.

The score is calculated using responses across eight dimensions of health (vision, hearing, speech, ambulation, dexterity, emotion, cognition, and pain). These subscales are combined using a mathematically weighted scoring algorithm to compute a single total utility index score.

The total utility score ranges from a minimum of -0.36 (representing a health state considered worse than death) to a maximum of 1.00 (representing perfect health). A score of 0.00 represents death.

Higher values represent a better health-related quality of life (a better outcome). Lower values represent worse health status (a worse outcome).

Hopkins Verbal Learning Test (HVLT) Total Recall. It is a paper-based list learning and recall assessment used to evaluate verbal memory and new learning capacity.

Participants are read a list of 12 words and asked to freely recall as many as possible across three consecutive learning trials. The total recall score is calculated by summing the number of correctly recalled words across all three trials.

The total recall score ranges from a minimum of 0 to a maximum of 36.

Higher values represent a greater number of words correctly recalled, indicating better memory and verbal learning function (a better outcome). Lower scores indicate worse memory function (a worse outcome).

Hopkins Verbal Learning Test (HVLT) Delayed Recall. It is a paper-based assessment used to evaluate delayed verbal memory retention and recall capacity.

After a standard delay interval (typically 20 to 25 minutes) following the initial learning trials, participants are asked to freely recall the original list of 12 words. The delayed recall score is the total number of words correctly recalled from memory.

The score ranges from a minimum of 0 to a maximum of 12.

Higher values represent a greater number of words retained and recalled after the delay, indicating better memory retention (a better outcome). Lower scores indicate worse memory retention (a worse outcome).

Number Correct Line Segments. This is a performance-based metric derived from the Trail Making Test (TMT) used to assess visual search, scanning, processing speed, and executive function, particularly when a participant is unable to complete the full timed test.

The score is calculated by counting the total number of correct line segments a participant successfully draws connecting the sequences of encircled numbers (or numbers and letters) within the allotted test parameters.

The score ranges from a minimum of 0 to a maximum of 25.

Higher values represent a greater number of correctly drawn segments, indicating better cognitive processing speed and executive functioning (a better outcome). Lower values indicate worse cognitive performance (a worse outcome).

Number Correct Line Segments (TMT Part B). This is a performance-based metric derived from the Trail Making Test Part B used to assess executive function, cognitive flexibility, and visual-motor tracking. It is particularly useful for quantifying performance when a participant is unable to complete the full timed test.

The score is calculated by counting the total number of correct line segments a participant successfully draws connecting the 25 encircled numbers and letters in an alternating sequence (1, A, 2, B, 3, C...) within the allotted test parameters.

The score ranges from a minimum of 0 to a maximum of 25.

Higher values represent a greater number of correctly drawn segments, indicating better cognitive flexibility and executive functioning (a better outcome). Lower values indicate worse cognitive performance (a worse outcome).

Digit-Symbol Substitution Test (DSST), utilizing the Wechsler Adult Intelligence Scale-IV (WAIS-IV) Coding subtest. It is a paper-based assessment used to evaluate processing speed, visual-motor coordination, and sustained attention.

Participants are provided a key pairing numbers (1 through 9) with simple geometric symbols. They are given a strict time limit (120 seconds) to draw the correct corresponding symbols beneath a series of randomized numbers. The final score is the total number of correctly drawn symbols within the time limit.

The score ranges from a minimum of 0 to a maximum of 135 (the maximum possible items on the WAIS-IV Coding form).

Higher values represent a greater number of correctly matched symbols, indicating faster processing speed and better cognitive performance (a better outcome). Lower scores indicate slower processing speed (a worse outcome).

Medication Perceptions Questionnaire. It is a 9-item survey developed in-house (adapted from Health Belief Model constructs and the revised Patients' Attitudes Towards Deprescribing Questionnaire) to assess participants' beliefs and attitudes regarding their medications and deprescribing.

Participants rate items on a 5-point Likert agreement scale. The final score is calculated as a subscale average of the answered items (which allows for missing data if a participant skipped a question).

The subscale average score ranges from a minimum of 1 to a maximum of 5.

Higher scores indicate more disagreement (less agreement) with the survey statements. Lower scores indicate stronger agreement. The clinical favorability (better/worse outcome) depends on the specific construct of the subscale statement.

The Self-Reported Deprescribing Behavior Survey is an in-house, 4-item questionnaire designed to assess a participant's self-reported actions and behaviors related to stopping or reducing their medications (deprescribing).

Item Scoring: Participants rate each of the 4 items individually using a 5-point Likert scale (1 = Strongly Agree, 5 = Strongly Disagree).

Score Generation: The scores of the 4 individual items are mathematically averaged to calculate a single overall composite score.

Scale Range: The overall composite average score ranges from a minimum of 1 to a maximum of 5.

Outcome Direction: Lower overall average values (closer to 1) indicate stronger agreement with active deprescribing behaviors, which represents a better clinical outcome. Higher overall average values (closer to 5) indicate greater disagreement, which represents a worse clinical outcome.

Proportion of months out of the total study time the participant logged into the application at least once.