Polypharmacy Risk Score

The foundational layer of the score is a simple count of concurrent active medications. Polypharmacy is typically defined by thresholds: minor (2-4 drugs), moderate (5-9 drugs), and hyperpolypharmacy (≥10 drugs). The score algorithm parses the Best Possible Medication History (BPMH) and active orders to count distinct active ingredients, excluding temporary or PRN medications based on configurable rules. This raw count serves as the base coefficient before clinical weighting is applied.

The score applies a multiplicative weight based on the cumulative Anticholinergic Cognitive Burden (ACB). Each medication is assigned an ACB score (1: mild, 2-3: strong) from validated scales. The engine sums these values and applies a risk multiplier. A high ACB score is strongly correlated with cognitive impairment, falls, and delirium in geriatric populations. The system automatically maps RxNorm codes to ACB scales to calculate this burden without manual pharmacist input.

A parallel weighting factor derived from the Sedative Load Model, which classifies medications as hypnotics, anxiolytics, or sedating antihistamines. The engine identifies drugs with central nervous system depressant effects and assigns a sedation rank. This load is cross-referenced against the Beers Criteria to flag potentially inappropriate medications. A high sedative load dramatically increases the risk score due to the multiplicative risk of respiratory depression and falls when multiple sedatives are combined.

The risk score dynamically adjusts based on the patient's estimated Glomerular Filtration Rate (eGFR). The engine evaluates each active medication against renal dose adjustment guidelines from drug monographs. Medications requiring dose reduction or contraindicated at the patient's current eGFR level receive an elevated risk coefficient. This component ensures the score reflects not just the number of drugs, but the pharmacokinetic appropriateness of the regimen for the individual patient's organ function.

The score incorporates the density and severity of Prospective Drug-Drug Interactions (PDDIs) within the regimen. The engine analyzes all active medication pairs against a curated interaction database, classifying each as contraindicated, major, moderate, or minor. The risk score increases proportionally to the number of major and contraindicated interactions. This component prevents the score from treating a regimen of 10 non-interacting drugs the same as a regimen of 5 drugs with multiple severe interaction pairs.

A binary flag that applies a global risk multiplier for patients aged 65 and older. When activated, the engine cross-references the entire medication list against the Beers Criteria and the STOPP/START criteria. Each potentially inappropriate medication (PIM) identified adds a penalty to the score. This component accounts for the altered pharmacodynamics and pharmacokinetics of aging, where standard dosing assumptions no longer hold and the risk of adverse outcomes is inherently elevated.

Validated scoring systems that quantify the cumulative cognitive and physical impact of medications with anticholinergic properties. These scales assign a weight (e.g., 0-3) to each drug based on its receptor affinity and clinical effect.

• Anticholinergic Cognitive Burden (ACB) Scale: The most widely used tool, summing scores to predict delirium and cognitive decline.
• Anticholinergic Drug Scale (ADS): Focuses on serum anticholinergic activity.
• Clinical Relevance: A high cumulative burden score is strongly associated with increased risk of falls, dementia, and all-cause mortality in geriatric populations.

A pharmacodynamic risk model that calculates the cumulative sedative effect of a patient's medication regimen, independent of anticholinergic activity. This model is critical for distinguishing between different mechanistic pathways of adverse drug reactions.

• Mechanism: Weights drugs based on their ability to depress the central nervous system (e.g., benzodiazepines, antipsychotics, opioids).
• Outcome: High sedative load is a primary predictor of functional impairment, motor vehicle accidents, and nocturnal falls.
• Integration: Often combined with anticholinergic scores for a holistic Drug Burden Index (DBI).

An implicit, judgment-based tool used to evaluate the overall appropriateness of each medication in a polypharmacy regimen. Unlike explicit criteria, MAI requires clinical expertise to assess ten distinct elements of prescribing quality.

• 10 Domains: Includes indication, effectiveness, dosage correctness, practical directions, drug-drug interactions, drug-disease interactions, duplication, duration, and cost.
• Scoring: Each domain is rated as 'A' (appropriate), 'B' (marginally appropriate), or 'C' (inappropriate).
• Correlation: Higher MAI scores directly correlate with increased hospitalization rates and polypharmacy risk scores.

The computational simulation of drug absorption, distribution, metabolism, and excretion (ADME) in aging physiology. This modeling underpins the precision of polypharmacy risk scores by accounting for age-related organ decline.

• Key Variables: Estimates glomerular filtration rate (eGFR), reduced hepatic blood flow, and increased body fat percentage.
• Drug-Drug Interaction (DDI) Prediction: Uses physiologically based pharmacokinetic (PBPK) models to simulate how multiple substrates compete for the same CYP450 enzyme pathways.
• Output: Generates a pharmacokinetic risk score that predicts toxic accumulation before it occurs clinically.

Explicit, evidence-based screening tools designed to identify potentially inappropriate prescribing in older adults. These criteria serve as a deterministic rules engine that feeds into the polypharmacy risk score.

• STOPP (Screening Tool of Older Persons' Prescriptions): Identifies drugs that should be stopped due to risk-benefit imbalance (e.g., duplicate therapy, wrong duration).
• START (Screening Tool to Alert to Right Treatment): Identifies medications that should be started for common geriatric conditions but are currently omitted.
• Automation: AI systems parse these criteria against structured EHR data to generate a real-time prescribing appropriateness score.

A dose-normalized pharmacological risk metric that measures the total exposure to medications with anticholinergic and sedative properties. The DBI is the mathematical backbone of many modern polypharmacy risk scores.

• Formula: DBI = Σ [D / (D + δ)], where D is the daily dose and δ is the minimum effective dose.
• Clinical Validation: A higher DBI is linearly associated with poorer physical function, slower gait speed, and reduced grip strength in longitudinal studies.
• Automation: AI models extract dose and frequency data from unstructured notes to calculate the DBI dynamically without manual pharmacist input.