Deprescribing in Internal Medicine: A Practical Approach to Rational Medication Discontinuation

DR Neeraj Manikath , claude.ai

Abstract

Deprescribing—the systematic process of tapering or stopping medications that may no longer be beneficial or may be causing harm—has emerged as a critical competency in modern internal medicine. While evidence-based guidelines extensively address when to initiate medications, guidance on discontinuation remains comparatively limited. This review synthesizes current evidence on deprescribing strategies, highlights common clinical scenarios, and addresses prevalent misconceptions that contribute to therapeutic inertia and polypharmacy-related harm.

Introduction

Polypharmacy, traditionally defined as the concurrent use of five or more medications, affects approximately 40% of older adults and is associated with increased risks of adverse drug events, falls, cognitive impairment, and mortality. The average hospitalized patient receives 10 to 15 medications at discharge, yet studies suggest that 20% to 50% of these medications may be inappropriate or unnecessary. Despite this, deprescribing remains underutilized in clinical practice, constrained by prescribing inertia, fear of disease recurrence, fragmented care, and insufficient training in medication discontinuation.

The fundamental principle underlying deprescribing is that medication appropriateness exists along a continuum that changes with patient age, comorbidities, functional status, and goals of care. What was appropriate at diagnosis may become inappropriate years later. This review provides a structured framework for identifying deprescribing opportunities and implementing discontinuation safely.

The Deprescribing Framework

When to Consider Deprescribing

Deprescribing should be considered in several clinical contexts. The first involves medications lacking current indication, such as proton pump inhibitors (PPIs) initiated during acute hospitalization and continued indefinitely despite resolution of the precipitating condition. Studies demonstrate that up to 70% of chronic PPI use lacks appropriate indication, exposing patients to risks of Clostridium difficile infection, pneumonia, and osteoporotic fractures without corresponding benefit.

Limited life expectancy represents another crucial scenario. When prognosis is measured in months rather than years, medications aimed at preventing long-term complications become futile and potentially harmful. Statins require approximately two years of treatment to demonstrate cardiovascular benefit in primary prevention. Continuing such medications in patients with advanced dementia or metastatic cancer prioritizes theoretical future benefit over current quality of life, often resulting in unnecessary pill burden and adverse effects.

The harm-benefit balance may shift unfavorably when adverse effects outweigh therapeutic benefits. Anticholinergic medications prescribed for urinary incontinence may worsen cognitive function in elderly patients. Antihypertensive medications may cause postural hypotension and falls while blood pressure naturally declines with advanced age and frailty. Recognition of these evolving risk-benefit profiles is essential.

Preventive medications in frail elderly patients warrant particular scrutiny. Bisphosphonates for osteoporosis prevention in an 85-year-old with limited mobility and multiple comorbidities may pose greater harm through esophageal irritation and atypical fractures than benefit from fracture prevention. The number needed to treat increases substantially with age and comorbidity burden, while the number needed to harm often decreases.

Pearl: The "Start Low, Go Slow, But Also Stop" Principle

While geriatric pharmacotherapy emphasizes cautious initiation, equally important is the adage to periodically reassess and discontinue. Every medication should justify its continued prescription at each encounter. The question should not be "Is there a reason to stop this medication?" but rather "Is there still a compelling reason to continue it?"

Evidence-Based Deprescribing Strategies by Drug Class

Proton Pump Inhibitors

PPIs represent one of the most commonly overprescribed medication classes. Appropriate indications include documented erosive esophagitis, Barrett's esophagus, and gastric ulcer prophylaxis in patients receiving chronic NSAID therapy with additional risk factors. The majority of long-term PPI use, however, lacks documentation of appropriate indication.

Deprescribing approaches include abrupt discontinuation for patients using PPIs for more than eight weeks without appropriate indication, or gradual tapering for those concerned about rebound acid hypersecretion. A stepped approach involves reducing to the lowest effective dose, transitioning to histamine-2 receptor antagonists, then stopping entirely with as-needed antacid use. Studies demonstrate successful PPI discontinuation in 60% to 80% of patients without symptom recurrence.

Oyster: Rebound Acid Hypersecretion

A common barrier to PPI deprescribing is rebound acid hypersecretion, which occurs in approximately 40% of patients following abrupt discontinuation after prolonged use. This physiologic phenomenon results from compensatory gastrin elevation during PPI therapy. Symptoms typically emerge within two weeks of discontinuation and resolve within four weeks. Educating patients about this temporary phenomenon and providing as-needed antacids improves deprescribing success rates. This is not GERD recurrence but rather a self-limited withdrawal phenomenon.

Benzodiazepines and Sedative-Hypnotics

Chronic benzodiazepine use for insomnia or anxiety in older adults exemplifies medications where harms frequently exceed benefits. Long-term use is associated with cognitive impairment, falls, hip fractures, and paradoxical worsening of anxiety. Despite these risks, benzodiazepines are often continued for years or decades.

Successful deprescribing requires gradual tapering, typically reducing the dose by 25% every two weeks for short-acting agents or every month for long-acting agents. For patients taking benzodiazepines for more than one year, even slower tapers over six months may be necessary. Converting short-acting agents to equivalent doses of long-acting benzodiazepines before tapering can reduce withdrawal symptoms.

Cognitive behavioral therapy for insomnia demonstrates superior long-term outcomes compared to pharmacotherapy and should be offered concurrently with deprescribing efforts. Patient education about tolerance, dependence, and the limited efficacy of chronic benzodiazepine therapy enhances motivation for discontinuation.

Hack: The Therapeutic Letter

For patients resistant to deprescribing, consider the therapeutic letter approach. Provide written information explaining why the medication is being reconsidered, potential harms of continuation, and benefits of discontinuation. This allows patients time to process the information and discuss concerns with family members, often resulting in greater acceptance of deprescribing recommendations.

Antihyperglycemic Medications

Tight glycemic control in elderly patients with diabetes, particularly those with multiple comorbidities or functional impairment, increases hypoglycemia risk without reducing macrovascular complications. Guidelines now recommend relaxing HbA1c targets to 8.0% to 8.5% in patients with limited life expectancy or significant comorbidity burden.

Deprescribing sulfonylureas should be prioritized given their high hypoglycemia risk. Insulin regimens can often be simplified, eliminating mealtime bolus doses while maintaining basal insulin for patients with erratic oral intake. Metformin, with its favorable safety profile and potential pleiotropic benefits, may be reasonably continued unless contraindicated by renal dysfunction or intolerable gastrointestinal side effects.

Antihypertensive Medications

Blood pressure treatment targets have shifted toward individualization based on frailty, falls risk, and life expectancy. The SPRINT trial established benefits of intensive blood pressure control to less than 120 mmHg systolic in ambulatory older adults without diabetes or prior stroke. However, these results should not be extrapolated to frail, institutionalized, or functionally impaired patients.

For frail elderly patients experiencing orthostatic hypotension, falls, or syncope, liberalizing blood pressure targets to 140 to 150 mmHg systolic is reasonable. Deprescribing should prioritize medications with the greatest adverse effect burden, such as alpha-blockers and centrally acting agents, while maintaining ACE inhibitors or ARBs that provide additional renal and cardiac protection.

Fallacy: "Never Stop a Statin"

A pervasive misconception is that statin therapy, once initiated, should never be discontinued. While statins provide clear mortality benefit in secondary prevention for patients with established cardiovascular disease, their role in primary prevention in elderly patients with limited life expectancy is questionable. The PROSPER trial demonstrated that statins initiated after age 70 for primary prevention do not reduce all-cause mortality despite reducing cardiovascular events.

In patients with advanced dementia, terminal illness, or severe frailty, statin discontinuation does not increase short-term cardiovascular events and improves quality of life by reducing pill burden. The decision to continue or discontinue should be individualized based on life expectancy, goals of care, and patient preferences.

Osteoporosis Medications

Bisphosphonates and other osteoporosis medications merit reconsideration in patients with declining functional status, severe frailty, or terminal illness. Fracture prevention requires years of treatment to demonstrate benefit, making these medications poor choices when life expectancy is limited.

For patients who have received five years of bisphosphonate therapy with adequate response (stable or improved bone mineral density, no fractures), a drug holiday may be appropriate for those at moderate risk. High-risk patients with prior fragility fractures or severe osteoporosis should generally continue therapy. Denosumab requires special consideration, as discontinuation results in rapid bone loss and rebound fracture risk; transitioning to bisphosphonate therapy before stopping is recommended.

Barriers to Deprescribing and Strategies to Overcome Them

Clinical Inertia

Therapeutic inertia, the failure to discontinue medications despite changing clinical circumstances, represents a significant barrier. Contributing factors include time constraints, fear of adverse outcomes, and lack of training in deprescribing. Implementing systematic medication reviews at transitions of care, particularly hospital discharge and nursing home admission, creates structured opportunities for reassessment.

Patient and Family Concerns

Patients often resist deprescribing due to fear of symptom recurrence or disease progression. Effective communication strategies include shared decision-making, emphasizing quality of life over disease-oriented outcomes, and offering trial discontinuation with close monitoring. Framing deprescribing as "taking control of your health" rather than "giving up on treatment" improves acceptance.

Pearl: The Deprescribing Conversation

Structure deprescribing discussions using the following framework: acknowledge the reason the medication was started, explain how circumstances have changed, describe potential benefits of stopping, address concerns about discontinuation, and propose a trial period with reassessment. For example, "We started this sleeping pill two years ago when you were having trouble sleeping after your surgery. Now that's resolved, continuing this medication puts you at risk for falls and confusion. I'd like to gradually reduce the dose over the next few months and see how you do. If your sleep worsens, we can always reconsider."

Practical Implementation

The Systematic Medication Review

Implement a structured approach to medication review at each clinical encounter. Document the indication for each medication, confirm ongoing appropriateness, assess for adverse effects, and evaluate patient adherence and understanding. For patients taking more than ten medications, prioritize review of medications with the highest risk-benefit ratio using validated tools such as the Beers Criteria, STOPP/START criteria, or the Medication Appropriateness Index.

Monitoring After Deprescribing

Establish monitoring plans for medication discontinuation. For benzodiazepines, schedule follow-up within two weeks to assess withdrawal symptoms. For antihypertensives, recheck blood pressure within one to two weeks. For PPIs, counsel patients about potential rebound symptoms and provide rescue antacids. Clear monitoring protocols reduce anxiety about deprescribing and facilitate early identification of adverse consequences.

Oyster: The Deprescribing Cascade

Just as prescribing cascades occur when new medications are prescribed to treat side effects of existing medications, deprescribing cascades can occur when stopping one medication allows discontinuation of others. For instance, discontinuing NSAIDs may allow PPI discontinuation. Stopping diuretics may alleviate orthostatic hypotension, permitting liberalization of blood pressure medications. Recognizing these opportunities amplifies deprescribing benefits.

Conclusion

Deprescribing represents an essential clinical skill that balances evidence-based medicine with patient-centered care. As our patient populations age and accumulate comorbidities, the ability to thoughtfully discontinue medications becomes as important as knowing when to initiate them. By challenging therapeutic inertia, educating patients and families, and implementing systematic medication review processes, clinicians can reduce polypharmacy-related harm while improving patient quality of life. The question we should ask is not simply "What medications should I prescribe?" but equally "What medications should I stop?"

References

1. Page AT, Potter K, Clifford R, et al. Deprescribing in older people. Maturitas. 2016;91:115-134.

2. Scott IA, Hilmer SN, Reeve E, et al. Reducing inappropriate polypharmacy: the process of deprescribing. JAMA Intern Med. 2015;175(5):827-834.

3. Reeve E, Gnjidic D, Long J, Hilmer S. A systematic review of the emerging definition of 'deprescribing' with network analysis: implications for future research and clinical practice. Br J Clin Pharmacol. 2015;80(6):1254-1268.

4. American Geriatrics Society Beers Criteria Update Expert Panel. American Geriatrics Society 2023 updated AGS Beers Criteria for potentially inappropriate medication use in older adults. J Am Geriatr Soc. 2023;71(7):2052-2081.

5. Farrell B, Pottie K, Thompson W, et al. Deprescribing proton pump inhibitors: Evidence-based clinical practice guideline. Can Fam Physician. 2017;63(5):354-364.

6. Tannenbaum C, Martin P, Tamblyn R, et al. Reduction of inappropriate benzodiazepine prescriptions among older adults through direct patient education. JAMA Intern Med. 2014;174(6):890-898.

7. Wright RM, Sloane R, Pieper CF, et al. Underuse of indicated medications among physically frail older US veterans at the time of hospital discharge. J Am Geriatr Soc. 2009;57(9):1587-1592.

8. Sjöberg C, Ohlsson H, Wallerstedt SM. Association between statin discontinuation and mortality in patients aged 75 years and over: a Swedish register-based cohort study. Eur J Clin Pharmacol. 2018;74(11):1493-1500.

9. Iyer S, Naganathan V, McLachlan AJ, Le Couteur DG. Medication withdrawal trials in people aged 65 years and older. Drugs Aging. 2008;25(12):1021-1031.

10. Thompson W, Farrell B. Deprescribing: what is it and what does the evidence tell us? Can J Hosp Pharm. 2013;66(3):201-202.