Deprescribing in Elderly Patients: A Practical Framework

This post is for consultant pharmacists and clinical pharmacy directors who lead medication regimen reviews (MRR) in skilled nursing / long-term care (SNF) and other post-acute settings. You will get a structured, evidence-informed framework for deprescribing in elderly patients — one you apply with your own clinical judgment, not a set of directives. Nothing here is patient-specific clinical, dosing, or legal advice.

What deprescribing means — and what it does not

Deprescribing is the planned, supervised reduction or discontinuation of a medication when its potential harms outweigh its benefits for a given resident. It is a positive clinical act, not simply "stopping drugs." Done well, it is proactive, patient-centered, and fully documented.

Older adults are especially vulnerable to polypharmacy, cumulative anticholinergic burden (ACB), and drug interactions. Age-related changes in pharmacokinetics and pharmacodynamics can amplify adverse effects that a younger patient would tolerate. The consultant pharmacist's role is to surface these risks and frame options for the prescriber and interdisciplinary team.

Ground your framework in recognized criteria

A defensible deprescribing program rests on published, periodically updated criteria rather than intuition. The tools most consultant pharmacists rely on include:

• AGS Beers Criteria (2023 update) — flags medications considered potentially inappropriate in older adults, including drugs to use with caution and select drug-disease interactions. See the American Geriatrics Society for the current version.
• STOPP/START version 2 — Screening Tool of Older Persons' Prescriptions (STOPP) identifies potentially inappropriate prescribing; Screening Tool to Alert to Right Treatment (START) identifies potential prescribing omissions.
• Anticholinergic burden (ACB) scales — quantify cumulative anticholinergic exposure, which is associated with cognitive and functional risk in older adults.

These criteria are guides, not mandates. They flag candidates for review; the licensed pharmacist and prescriber decide what is appropriate for the individual. Because each tool is revised over time, always confirm the current edition from its publishing authority before relying on a specific entry.

A five-step deprescribing framework

The following sequence organizes the work. Adapt it to your facility's policies and the interdisciplinary team's workflow.

1. Reconcile the full regimen. Build a complete, current medication list — prescription, over-the-counter, and as-needed. You cannot deprescribe what you cannot see. Duplicate therapy and orphaned orders often surface here.
2. Screen against criteria. Run the regimen against Beers 2023, STOPP/START v2, and an ACB scale. Note potentially inappropriate medications, prescribing omissions, and cumulative anticholinergic load, along with the clinical context that may justify or contraindicate each.
3. Prioritize by risk and goals of care. Not every flag warrants action, and not all at once. Weigh the drug's likelihood of harm, the resident's goals of care and life expectancy, monitoring burden, and the risk of withdrawal or rebound. High-risk combinations and drugs with poor benefit-to-harm profiles rise to the top.
4. Recommend, with rationale and monitoring. Frame each recommendation for the prescriber: what to consider changing, why, how to monitor for withdrawal effects or return of the underlying condition, and what to watch for. The consultant pharmacist proposes; the prescriber decides. Tapering, where indicated, is individualized — never a fixed schedule applied blindly.
5. Document and follow up. Record the recommendation, the clinical reasoning, the prescriber's response, and the plan for reassessment. Deprescribing is iterative; a drug not appropriate to stop today may be revisited at the next review.

Common deprescribing targets in older adults

Certain drug classes recur as deprescribing candidates because published criteria and geriatric literature repeatedly flag their risk-benefit balance in older adults. You evaluate each in context rather than treating any class as automatically inappropriate:

• Sedative-hypnotics and long-acting benzodiazepines, given fall and cognitive risk.
• Antipsychotics used without a supported indication — an area of intense CMS survey focus in SNFs, tied to gradual dose reduction (GDR) expectations.
• Strongly anticholinergic agents that drive up ACB.
• Proton pump inhibitors continued without a documented ongoing indication.
• Medications requiring intensive monitoring where the monitoring is impractical or the benefit is marginal for the resident's goals of care.

In the SNF setting, deprescribing intersects directly with regulatory expectations around unnecessary medications, GDR for psychotropics, and the consultant pharmacist's monthly MRR. Confirm current federal requirements and F-tag definitions at the source rather than from memory — the Centers for Medicare & Medicaid Services publishes the governing guidance, and it is updated periodically. See cms.gov for current State Operations Manual guidance.

Build a defensible audit trail

A deprescribing recommendation is only as strong as its documentation. Surveyors, medical directors, and your own quality program need to see the clinical reasoning, the criteria applied, and the outcome. Capture:

• Which criterion or scale triggered the review, and the specific clinical rationale.
• The recommendation as communicated to the prescriber.
• The prescriber's decision — accepted, modified, or declined with reason.
• The monitoring and reassessment plan.

An append-only record that ties each recommendation to a consultant's e-signature turns deprescribing from an informal suggestion into a defensible clinical act. That is the difference between a note and an audit trail.

Practical cautions

Deprescribing carries its own risks. Abrupt discontinuation of certain agents can provoke withdrawal or rebound. Some "potentially inappropriate" medications are entirely appropriate for a specific resident. And criteria conflict at times — a drug flagged by one tool may be indicated under another. This is exactly why the framework centers the pharmacist's judgment and the prescriber's authority, not an algorithm's verdict.

Standardizing the screening step is what frees your clinical time for the judgment step. When flag detection is consistent and comprehensive, you spend your attention where it matters: prioritizing, reasoning, and communicating.

How WeConsultRx helps

The WeConsultRx clinical engine codifies 49 rules across Beers 2023, STOPP/START v2, and anticholinergic burden, so every regimen you review is screened consistently against recognized geriatric criteria — and every flag arrives with the context you need to exercise judgment quickly. Recommendations, prescriber responses, and your e-signature flow into an append-only audit trail built for SNF survey readiness. Explore the platform, see the SNF solution, or request a demo to see the framework in practice.