PolySafe

Name: PolySafe Tagline: Protecting patients from polypharmacy

Solution in brief: PolySafe is a clinical decision-support engine that automatically applies STOPP/START and deprescribing guidelines to identify inappropriate medications and recommend safer alternatives, assisting clinicians reduce polypharmacy and minimise adverse drug reaction at scale.

Background and Problem Statement: Polypharmacy—the concurrent use of five or more medications—is a rapidly growing global health challenge, driven by ageing populations and the increasing prevalence of chronic diseases. It poses one of the most significant medication-related risks in contemporary healthcare. The impact is substantial and well-documented. Patients exposed to polypharmacy experience elevated risks of hospitalisation, mortality, adverse drug reactions, drug-drug interactions, falls, cognitive decline, and medication errors. These risks escalate with the number of medications prescribed. Polypharmacy also imposes a considerable economic burden. Medication-related harm contributes to a notable proportion of hospital admissions—many of which are preventable—resulting in higher healthcare costs, extended hospital stays, and increased pressure on systems already constrained by workforce and resource limitations. The issue is inherently systemic. Patients with multiple chronic conditions frequently receive prescriptions from various providers, often without comprehensive visibility into the full medication regimen. This fragmented care heightens the likelihood of inappropriate prescribing, duplication, and harmful interactions. With populations ageing, the scale of polypharmacy is projected to expand further. By 2050, the global population aged 60 and over is expected to double, potentially exposing millions more individuals to preventable medication-related harm absent effective interventions. Polypharmacy represents a critical, scalable opportunity for innovation. Enhanced medication management can reduce hospitalisations, avert avoidable harm, decrease healthcare expenditure, and meaningfully improve patient outcomes—positioning it as a high-priority area for digital health, artificial intelligence, and clinical decision-support solutions.

How common is polypharmacy in Ireland? In Ireland, polypharmacy is particularly prevalent among older adults, reflecting trends in ageing and multimorbidity. • Among community-dwelling adults aged over 50, approximately one in five (around 19–20%) regularly take five or more medications. • This rises to 31% in those aged 65 and over, and 37% in those aged 75 and over (based on data from The Irish Longitudinal Study on Ageing – TILDA). • In primary care settings, polypharmacy rates among older adults (≥65 years) have increased significantly over time; for example, from 17.8% in 1997 to 60.4% in 2012 in repeated cross-sectional analyses. • Recent studies (including applications of STOPP/START criteria) indicate that potentially inappropriate prescribing often co-occurs with polypharmacy, affecting around 31% of community-dwelling older adults in some samples. • Excessive polypharmacy (10 or more medications) is also notable, particularly in older groups, with historical increases from low single digits to over 20% in some cohorts aged 65+. • Prevalence increases sharply with age, mirroring global patterns: higher rates are observed in those with greater morbidity, lower educational attainment, and medical card eligibility. Key Irish statistics on risks and outcomes Individuals on polypharmacy in Ireland face heightened risks, consistent with international evidence: • Polypharmacy is associated with increased adverse drug reactions (ADRs), with cumulative incidence rates of around 27% over extended follow-up in some older cohorts. • It contributes to higher healthcare utilisation: those aged 65+ with polypharmacy (about one-third of this group) account for over half of hospital inpatient and outpatient visits in this age bracket. • Broader evidence links polypharmacy to elevated risks of hospitalisation and mortality, though specific Irish hazard ratios align with global ranges (e.g., increased all-cause mortality and admission risks in polypharmacy-exposed older adults). • Medication-related issues, including potentially inappropriate prescribing, remain a driver of preventable harm, with opportunities for deprescribing and optimisation via tools like STOPP/START.

Solution in detail: We are developing an innovative clinical decision-support algorithm to assist medical care providers—such as general practitioners, geriatricians, pharmacists, and nurses—in efficiently identifying Potentially Inappropriate Medications (PIMs) and Potential Prescribing Omissions (PPOs) in older adults (≥65 years) with polypharmacy and/or multimorbidity. The core goal is to automate the detection and flagging of medications that are potentially harmful, redundant, or no longer aligned with current evidence-based benefit-risk profiles, while simultaneously proposing safer, more appropriate therapeutic alternatives that minimize side effects, reduce anticholinergic/sedative burden, avoid therapeutic duplication, prescription cascades, and overall decrease the contribution to polypharmacy. This approach aligns closely with established European standards: • STOPP/START criteria version 3 (2023) — the primary explicit tool for screening PIMs (133 criteria to avoid/stop/reduce) and PPOs (57 criteria to consider initiating/optimizing). Recent studies (2024–2025) confirm v3 detects ~17–74% more issues than v2, with enhanced sensitivity for common classes like NSAIDs, anticholinergics, PPIs, and cardiovascular drugs, though some criteria have variable evidence strength. • EuGMS (European Geriatric Medicine Society) advice — including position papers on polypharmacy, fall-risk-increasing drugs (e.g., STOPPFall integration), and pharmacology SIG recommendations emphasizing structured reviews, shared decision-making, and deprescribing in frail/multimorbid patients. • Deprescribing.org evidence-based resources — drug-class-specific algorithms (e.g., updated statin deprescribing algorithm as of September 2025, PPIs, benzodiazepines, antipsychotics, antihyperglycemics) providing practical tapering schedules, monitoring plans, patient information, and rationale for discontinuation when harms outweigh benefits (e.g., in frailty, limited life expectancy, or stable chronic conditions). By automating this process, the algorithm eliminates the need for pharmacists or clinicians to manually review every single medication in a patient's record—a time-intensive task that currently takes 10–20 minutes per patient and is often unsustainable in busy primary care, outpatient, or hospital settings. Instead, the algo would: • Ingest structured electronic health record (EHR) data (e.g., medication list, diagnoses, age, renal/liver function, frailty indicators). • Apply rule-based logic derived from STOPP/START v3 to flag PIMs/PPOs in real-time or batch mode. • Cross-reference with deprescribing.org algorithms for prioritized deprescribing candidates and safe replacement options (e.g., switching from a long-term high-dose PPI to on-demand use or H2-blocker if appropriate; deprescribing statins in primary prevention with advanced frailty). • Suggest evidence-based alternatives that are less likely to cause adverse effects (e.g., fewer falls, cognitive impairment, drug interactions) and do not exacerbate polypharmacy (e.g., avoiding therapeutic duplication or low-value additions). • Incorporate EuGMS principles for patient-centered elements, such as prompts for shared decision-making discussions, consideration of goals of care, life expectancy, and monitoring post-change (e.g., withdrawal symptoms, rebound effects, or re-emergence of symptoms). This tool aims to scale high-quality medication optimization across healthcare systems, reduce adverse drug events, improve adherence and quality of life, lower healthcare costs associated with polypharmacy harms, and support deprescribing without requiring specialist input for every case—while always deferring final decisions to clinical judgment and patient preferences. In essence, it's shifting from labor-intensive, manual polypharmacy reviews to intelligent, automated, guideline-embedded support that empowers providers to focus on complex cases, patient conversations, and implementation, ultimately making safer prescribing more accessible and routine in European geriatric and primary care practice. If integrated into EHRs or clinical software, it could represent a significant step toward digital health solutions for medication appropriateness in aging populations.

Guidelines exist—but clinicians cannot manually apply hundreds of rules in seconds.