Section 1: Introduction — When Did Anyone Last Review This List?

A 74-year-old woman attends her primary care clinic in Port of Spain. Her medication list runs to eleven items:

Amlodipine 10 mg daily. Lisinopril 20 mg daily. Hydrochlorothiazide 25 mg daily. Metformin 1000 mg twice daily. Glibenclamide 5 mg twice daily. Omeprazole 20 mg daily. Diazepam 5 mg at night. Amitriptyline 25 mg at night. Aspirin 81 mg daily. Ferrous sulphate 200 mg twice daily. Promethazine 25 mg as needed.

Three of these were started during a hospital admission five years ago — the omeprazole, the diazepam, and the iron. The omeprazole was for stress ulcer prophylaxis during the admission. The diazepam was for acute anxiety related to the hospitalisation. The iron was for a mild anaemia that resolved within two months. All three were continued at discharge. None were reviewed.

The amitriptyline was started by a doctor who has since retired. The indication is unclear — possibly insomnia, possibly neuropathic pain. Nobody has asked whether it is still needed.

The promethazine was added because the patient complained of poor sleep. It is treating the insomnia that the diazepam was also supposed to be treating — except that the diazepam stopped working years ago and nobody adjusted the plan.

The aspirin was started for "heart protection" at a time when primary prevention with aspirin was standard practice. Guidelines have since changed. The bleeding risk in a 74-year-old now outweighs the cardiovascular benefit.

The glibenclamide is maintaining her HbA1c at 5.8% — well below target. She is overtreated. She has had three hypoglycaemic episodes in the past year, each attributed to "not eating enough" rather than to the medication.

This patient does not need eleven medications. She needs someone to look at the list and ask: which of these should she still be taking?

1.1 Why deprescribing is harder than prescribing

Prescribing a new medication is a positive act. It addresses a complaint. It demonstrates action. It takes thirty seconds to write.

Stopping a medication is a negative act. It requires reviewing the original indication, assessing whether it still applies, considering the risks of withdrawal, communicating the decision to the patient, and accepting responsibility if the patient's condition changes after the drug is stopped. It takes time, confidence, and a willingness to question a previous decision — sometimes someone else's decision.

In Caribbean clinical practice, where consultations are short, medication lists are long, and patients may interpret stopping a drug as "the doctor giving up," deprescribing meets resistance at every level: the clinician, the patient, and the system.

This report is designed to make it easier. Twenty drugs. Clear criteria for when to stop. Taper schedules where needed. Communication scripts for the conversation. A printable checklist for the clinic wall.

1.2 What deprescribing is — and what it isn't

Deprescribing is the planned, supervised process of reducing or stopping a medication that is no longer needed, no longer effective, or causing more harm than benefit. It is not:

• Abruptly stopping medications without a plan
• Withholding necessary treatment to save money
• Telling patients their previous doctor was wrong
• A one-time event — it is an ongoing process at every medication review

The goal is not to reduce the pill count for its own sake. The goal is to ensure that every medication on a patient's list has a current indication, is at the right dose for the patient's current physiology, and is still the best available option.

Section 2: The Five Rules of Deprescribing

Before reviewing the twenty specific drugs in Section 3, these five questions apply to any medication on any patient's list.

Rule 1: Is there still an indication?

The original reason for the prescription may no longer exist. The stress ulcer healed. The anxiety resolved. The anaemia corrected. The infection cleared. But the drug continued because nobody actively stopped it.

Caribbean Practice Note: In settings where patients move between public and private facilities without shared records, the original indication may be lost entirely. The drug appears on the list with no documented reason. When this happens, the safest approach is to trace the indication before assuming the drug is still needed.

Rule 2: Is the drug still the right choice?

Guidelines change. A drug that was first-line five years ago may no longer be recommended. Aspirin for primary cardiovascular prevention was standard practice until recently — it is no longer recommended for most patients over 60. Methyldopa for hypertension was appropriate during pregnancy — continuing it post-partum when safer alternatives exist is not.

Rule 3: Is the dose still appropriate?

Renal function declines with age. Body weight changes. Hepatic metabolism slows. A dose that was appropriate at 60 may not be appropriate at 75. This is not just about renal dose adjustment (covered in Report 2) — it is about whether the therapeutic target itself has changed. Blood pressure targets in an 80-year-old are not the same as in a 50-year-old. HbA1c targets in a frail elderly patient are not the same as in a newly diagnosed 45-year-old.

Rule 4: Is it causing a prescribing cascade?

Is another drug on the list treating this drug's side effect? If so, removing the first drug may allow the second to be removed as well. The prescribing cascades documented in Report 1 Section 7 are deprescribing opportunities in reverse: identify the cascade, remove the trigger drug, and the secondary drug becomes unnecessary.

Rule 5: Have the patient's goals of care changed?

A statin in a 55-year-old with high cardiovascular risk is clearly appropriate. A statin in an 85-year-old with advanced dementia and a life expectancy measured in months raises a different question. Preventive medications — those prescribed to reduce a risk that will materialise in 5–10 years — may no longer be appropriate when the patient's life expectancy or quality of life has changed. This is a conversation, not a formula.

Section 3: The Twenty

PPIs — The Biggest Opportunity

1. Omeprazole

Why it was started: Gastro-oesophageal reflux, peptic ulcer, gastroprotection during an NSAID course, stress ulcer prophylaxis in hospital, dyspepsia.

Why it should be reviewed: The Beers Criteria recommends avoiding PPI use beyond 8 weeks unless a specific high-risk indication is present (erosive oesophagitis, Barrett's oesophagus, chronic NSAID use with gastroprotection, or chronic corticosteroid use). Long-term PPI use is associated with C. difficile infection, hypomagnesaemia, osteoporotic fractures, vitamin B12 deficiency, and community-acquired pneumonia. Most patients on long-term PPIs were started for a short-term indication that resolved months or years ago.

How to stop: Step down to half dose for 2–4 weeks, then to alternate days for 2 weeks, then stop. Rebound acid hypersecretion can cause temporary symptom recurrence — counsel the patient that this is withdrawal, not the return of the original condition. Offer antacids (calcium carbonate) for breakthrough symptoms during the taper.

What to use instead: If the patient has ongoing dyspepsia without a high-risk indication, as-needed antacids or lifestyle modification (meal timing, weight management, avoiding late eating). If gastroprotection is still needed (chronic NSAID use), continue the PPI with documented justification.

Caribbean Practice Note: Omeprazole is the most overprescribed drug in Caribbean primary care. It is started in hospital, continued at discharge, and never reviewed. A clinic-wide PPI review — identifying every patient on a PPI for more than 8 weeks and checking whether a high-risk indication still exists — is one of the highest-yield deprescribing interventions available.

2. Pantoprazole / Lansoprazole

Same profile as omeprazole. Different brand, same problem. The taper approach is identical. If the patient was switched from omeprazole to pantoprazole or lansoprazole at some point (often for formulary reasons), they may now have two PPI prescriptions in their history — a duplication that is itself a signal for review.

Benzodiazepines

3. Diazepam

Why it was started: Anxiety, insomnia, muscle spasm, alcohol withdrawal, procedural sedation.

Why it should be reviewed: Diazepam is a long-acting benzodiazepine with active metabolites (nordiazepam, oxazepam) that can accumulate over days in older adults. Long-term use is associated with cognitive impairment, delirium, falls, hip fractures, motor vehicle accidents, and physical dependence. Tolerance to the anxiolytic and hypnotic effects develops within weeks, meaning the drug is no longer doing what it was prescribed for — but stopping it produces withdrawal symptoms that are mistaken for the return of the original condition.

How to stop: Never stop abruptly — benzodiazepine withdrawal can cause seizures. Reduce by 10–25% of the current dose every 2–4 weeks. If the patient is on a high dose, the initial reductions can be larger; slow down as the dose gets lower. The final reduction (from a very low dose to zero) is often the hardest — extend the taper interval to 4–6 weeks for the last step. Consider switching to an equivalent dose of a shorter-acting benzodiazepine (oxazepam) for the taper if the patient finds the long half-life fluctuations of diazepam difficult.

What to use instead: For anxiety: SSRI (sertraline, escitalopram) — takes 2–4 weeks to reach full effect, so start before the benzodiazepine taper is complete. For insomnia: sleep hygiene counselling, melatonin, low-dose trazodone. For muscle spasm: non-pharmacological approaches, short-course low-dose baclofen (with renal check).

Caribbean Practice Note: Diazepam is on the Jamaica VEN list and the OECS EML. It is also one of the most commonly continued medications in Caribbean practice. Patients may have been on diazepam for years — sometimes decades. The taper requires patience, clear communication, and follow-up. Telling a patient who has taken diazepam for ten years to "just stop" is clinically dangerous and practically impossible. Plan for a 3–6 month taper.

4. Alprazolam

Same class, shorter acting, higher potency. Alprazolam produces more intense withdrawal symptoms than diazepam because of its short half-life and high receptor affinity. Taper more slowly — reduce by 10% every 2–4 weeks. Consider switching to an equivalent diazepam dose first (alprazolam 0.5 mg ≈ diazepam 10 mg), then tapering the diazepam.

5. Lorazepam

Why it persists: Lorazepam is frequently started in hospital (for agitation, insomnia, or procedural sedation) and continued at discharge without a stop date. The patient goes home on a "temporary" benzodiazepine that becomes permanent because nobody reviews it. Same taper approach as diazepam — reduce by 10–25% every 2–4 weeks.

Antidepressants Past Their Indication

6. Amitriptyline (low-dose)

Why it was started: Insomnia, neuropathic pain, migraine prophylaxis, "nerves."

Why it should be reviewed: Low-dose amitriptyline (10–25 mg at night) is one of the most commonly prescribed drugs in Caribbean primary care for indications that may no longer apply. If the insomnia has resolved, if the pain has settled, or if the patient no longer remembers why they were started on it, the drug is contributing anticholinergic burden (ACB Score 3 — see Report 1) without clinical benefit.

How to stop: Reduce by 25% every 2–4 weeks. At low doses (10–25 mg), a single step-down to half dose for 2 weeks then stop is usually sufficient. Watch for discontinuation symptoms (insomnia rebound, headache, nausea) — these are withdrawal, not relapse.

What to use instead: For insomnia: sleep hygiene, melatonin. For neuropathic pain: gabapentin (with renal adjustment per Report 2). For migraine prophylaxis: propranolol.

7. Paroxetine

Why it persists: Paroxetine has the most severe discontinuation syndrome of any SSRI — dizziness, electric shock sensations ("brain zaps"), nausea, insomnia, irritability, and flu-like symptoms. These symptoms are frequently mistaken for relapse of the original depression, leading clinicians to restart the drug. The result is a patient who cannot stop paroxetine because every attempt produces withdrawal that looks like the disease.

How to stop: Taper extremely slowly — reduce by 10% every 4–6 weeks. Liquid formulation allows precise dose reductions if available. If the patient is on 20 mg, consider: 20 → 15 → 10 → 7.5 → 5 → 2.5 → stop, with each step lasting 4–6 weeks. Counsel the patient explicitly that withdrawal symptoms are temporary and do not mean the depression is returning.

Antihypertensives in Overtreated Elderly

8. The Third or Fourth Antihypertensive

Why it should be reviewed: Blood pressure targets in older adults have been debated extensively. In a frail 80-year-old, a systolic BP of 150 mmHg may be acceptable — and pushing it below 130 with three or four drugs may cause more harm (falls from orthostatic hypotension, dizziness, acute kidney injury) than the cardiovascular benefit it provides. If a patient is on three or more antihypertensives and has postural symptoms, the answer may be to remove a drug rather than add another.

How to stop: Remove one agent at a time (usually the most recently added or the one most likely to cause side effects). Monitor blood pressure weekly for 4 weeks. If BP remains within an acceptable range for the patient's age and frailty, the drug was unnecessary.

9. Methyldopa Post-Partum

Why it persists: Methyldopa is the first-line antihypertensive in pregnancy. After delivery, many women remain on methyldopa because nobody switches them to a more appropriate long-term agent. Methyldopa causes sedation, depression, and orthostatic hypotension — all avoidable by switching to amlodipine, lisinopril, or another standard antihypertensive post-partum.

How to stop: Switch directly to an alternative antihypertensive. No taper needed — methyldopa can be stopped and the new agent started on the same day. Monitor blood pressure daily for one week.

Antidiabetics in Overtreated Elderly

10. Glibenclamide With HbA1c Below 6.5%

Why it should be reviewed: An HbA1c below 6.5% in an older adult on a sulfonylurea is not good control — it is overtreatment. The patient is at risk of hypoglycaemia, which in older adults presents as confusion, falls, or apparent stroke. The ACCORD trial demonstrated that aggressive glucose lowering in older adults with type 2 diabetes increased mortality. An HbA1c of 7.0–7.5% is an appropriate target in most older adults; 7.5–8.0% in frail elderly patients.

How to stop: Reduce glibenclamide dose by 50%. Recheck HbA1c in 3 months. If HbA1c remains below 7.5%, reduce further or stop. If the patient is on both glibenclamide and metformin, stop the glibenclamide first — metformin alone may be sufficient.

11. Inpatient Insulin Sliding Scale Continued After Discharge

Why it persists: Sliding-scale insulin (regular insulin dosed according to pre-meal blood glucose) is an inpatient tool for managing hyperglycaemia during acute illness. It is not intended for outpatient use. Yet patients are sometimes discharged with a sliding scale because nobody converted them to a standard outpatient regimen. The result is reactive, unpredictable glucose management that increases hypoglycaemia risk and confuses patients.

How to stop: Convert to a basal insulin regimen (NPH or glargine) based on total daily sliding-scale insulin requirements during the last 48 hours of admission. Or, if oral agents were held during admission, restart them and stop insulin entirely if the patient's pre-admission control was adequate.

Antibiotics

12. Prophylactic Antibiotics Continued Beyond Indication

Why it persists: Antibiotic prophylaxis for recurrent UTIs (nitrofurantoin, trimethoprim), surgical site infection prevention, or rheumatic fever prevention (benzathine penicillin) has specific durations. When these durations are not documented, the antibiotic continues indefinitely. Long-term antibiotic use promotes resistance, disrupts gut microbiota, and — in the case of nitrofurantoin — carries a risk of pulmonary fibrosis and peripheral neuropathy with prolonged use.

How to stop: Review the original indication and check whether the recommended duration has been met. If it has, stop. If uncertain, consult the relevant guideline (UTI prophylaxis: typically 6–12 months then reassess; surgical prophylaxis: 24 hours post-procedure; rheumatic fever: duration depends on severity and age).

13. Metronidazole Continued After Course Completion

Why it persists: Metronidazole is prescribed for a defined course (typically 7–14 days) for conditions like bacterial vaginosis, C. difficile infection, or amoebic dysentery. Occasionally the prescription is renewed "just in case" or because the patient requests it. Long-term metronidazole use causes peripheral neuropathy (potentially irreversible) and CNS toxicity.

How to stop: Stop. No taper needed. If the patient's symptoms have resolved, the course is complete.

GI Drugs

14. Metoclopramide Continued Long-Term

Why it should be reviewed: Metoclopramide carries a risk of tardive dyskinesia — involuntary, repetitive movements of the face and tongue — that increases with duration of use and cumulative dose. The FDA recommends limiting use to 12 weeks. In Caribbean practice, metoclopramide is frequently continued for months or years for gastroparesis or chronic nausea without reassessment.

How to stop: Stop or taper over 1–2 weeks. If the patient needs ongoing prokinetic therapy, domperidone (where available) has lower CNS penetration and lower tardive dyskinesia risk, though it carries its own cardiac risk (QT prolongation at high doses).

15. Laxatives Continued After Opioid Stopped

Why it persists: Opioid-induced constipation requires prophylactic laxative use (Report 1, Cascade 7). When the opioid is stopped, the laxative should be stopped too. But the laxative often continues because nobody links the two prescriptions. The patient remains on a stimulant laxative indefinitely, causing abdominal cramping, electrolyte disturbances, and — with long-term stimulant laxative use — possible colonic dysmotility.

How to stop: When the opioid is stopped, stop the laxative. If the patient has developed habitual laxative dependence, taper by switching from a stimulant laxative to an osmotic laxative (macrogol/PEG), then reduce the osmotic laxative over 2–4 weeks.

Cardiovascular

16. Aspirin for Primary Prevention

Why it should be reviewed: Until recently, low-dose aspirin (75–100 mg daily) was widely recommended for primary cardiovascular prevention in patients with risk factors. The ASPREE, ARRIVE, and ASCEND trials (2018) changed this: in patients without established cardiovascular disease, the bleeding risk of aspirin outweighs the cardiovascular benefit, particularly in adults over 60. Current guidelines recommend against aspirin for primary prevention in most patients over 60.

How to stop: Stop. No taper needed. Continue aspirin if the patient has established cardiovascular disease (secondary prevention) — the indication is different and the benefit-risk balance favours continuation.

Caribbean Practice Note: Aspirin for primary prevention is deeply embedded in Caribbean prescribing culture. Many patients were started on "baby aspirin" years ago and believe it is protective. The conversation requires explaining that the evidence has changed — not that the original decision was wrong, but that the science has moved on. "We used to think aspirin helped prevent heart attacks in people who hadn't had one. The latest research shows it doesn't, and it increases your risk of bleeding. So we're going to stop it."

17. Statin in a Patient Over 80 With Limited Life Expectancy

Why it should be reviewed: Statins reduce cardiovascular events over a 5–10 year horizon. In a patient with a life expectancy shorter than that — due to advanced dementia, terminal illness, or severe frailty — the preventive benefit will not be realised. Continuing the statin adds pill burden, costs, potential side effects (myalgia, which can reduce mobility and quality of life), and drug interactions without measurable benefit.

How to stop: Stop. No taper needed. Statins have no withdrawal syndrome. Monitor lipids only if the clinical context changes.

Caribbean Practice Note: This is the most difficult deprescribing conversation in the series because it touches on prognosis and mortality. The approach is not "you don't need this anymore" but "this medication works by preventing something that takes years to develop. At this stage, it's adding pills without adding benefit. Let's focus your medications on what's helping you right now."

CNS Drugs

18. Haloperidol Continued After Acute Psychosis Resolved

Why it persists: Haloperidol is frequently started in the emergency department or on the ward for acute agitation or psychosis. It is effective acutely. But it is sometimes continued at discharge — or continued for months — without a clear plan for reassessment. Long-term haloperidol carries risks of tardive dyskinesia, extrapyramidal symptoms, QT prolongation, and metabolic effects.

How to stop: If the acute episode has resolved and the patient is not on haloperidol for a chronic psychiatric condition (schizophrenia, bipolar disorder), reduce the dose by 25–50% every 2–4 weeks with psychiatric monitoring. If the patient was started on haloperidol for delirium in hospital and the delirium has resolved, stop.

19. Promethazine Continued as a Sleep Aid

Why it persists: Promethazine is used as a sedative, often added when a patient complains of poor sleep. It is an ACB Score 3 drug (Report 1) — high anticholinergic burden, risk of confusion, excessive sedation, and extrapyramidal effects. It is also frequently co-prescribed with a benzodiazepine that was itself prescribed for the same complaint, creating a duplication of sedative therapy.

How to stop: Stop. No taper needed. Address the underlying sleep problem: sleep hygiene counselling, review other medications that may be causing insomnia (SSRIs, corticosteroids, caffeine-containing drugs), and consider melatonin if a medication is needed.

Other

20. Iron Supplements Continued After Anaemia Resolved

Why it persists: Iron supplements (ferrous sulphate, ferrous fumarate) are prescribed for iron-deficiency anaemia. Once the haemoglobin has normalised and iron stores are replete, the supplement should be stopped. But it often continues indefinitely — causing constipation, nausea, black stools (which can mask GI bleeding), and reduced absorption of other medications (levothyroxine, tetracyclines, fluoroquinolones).

How to stop: Check ferritin and haemoglobin. If both are normal, stop. Recheck in 3 months to confirm stability. If the patient was prescribed iron for a specific cause of blood loss (menstruation, GI bleeding) that has since resolved, there is no reason to continue.

Section 4: How to Have the Conversation

4.1 The three barriers

Deprescribing fails when the conversation fails. Three barriers appear consistently:

"My doctor put me on this." The authority barrier. Patients trust the prescribing decision and may interpret stopping a drug as an insult to the original prescriber — or as a sign that the new clinician doesn't know what they're doing.

Script: "Dr [name] made the right decision at the time. This medication was the best option when it was started. But your situation has changed — [your kidney function is different / the condition it was treating has resolved / we now have better options]. Updating your medications is part of good care, the same way Dr [name] would update them if they were still seeing you."

"But I've been taking it for years." The familiarity barrier. The drug feels like a safety net. Stopping it feels risky.

Script: "I understand — it's been part of your routine for a long time. That's actually part of the reason I want to review it. Medications should be doing something useful for you right now, not just sitting on the list because they've always been there. Let's try reducing it gradually and see how you feel. If anything changes, we can always adjust."

"What if I get worse?" The fear barrier. The patient worries that stopping a drug will cause the original condition to return.

Script: "That's a fair concern. Here's what we'll do: we'll reduce it slowly, not stop it all at once. I'll see you in [2–4 weeks] to check how you're doing. If you feel worse, we can go back to the current dose. But in many cases, patients feel better with fewer medications — less drowsiness, fewer side effects, less to remember."

4.2 The trial reduction

The most effective deprescribing tool is the trial reduction: "Let's try reducing it for two weeks and see how you feel." This reframes the decision from permanent to provisional, reduces patient anxiety, and gives the clinician a data point. If the patient does well on the lower dose, the next reduction follows. If not, the dose goes back up — no harm done.

4.3 When NOT to deprescribe abruptly

Some medications must never be stopped abruptly:

• Benzodiazepines — seizure risk (taper over weeks to months)
• Antiepileptics — seizure risk (taper under specialist guidance)
• Corticosteroids (long-term) — adrenal crisis (taper over weeks)
• Beta-blockers — rebound tachycardia and hypertension (taper over 1–2 weeks)
• Opioids (long-term) — withdrawal syndrome (taper over weeks to months)
• Clonidine — rebound hypertensive crisis (taper over 1 week)
• SSRIs / SNRIs — discontinuation syndrome (taper over 2–4 weeks; paroxetine and venlafaxine need slower tapers)

Section 5: The Deprescribing Checklist

Print this page and post it in your clinic.

One rule: Before the next repeat prescription, ask: does this patient still need every drug on this list?

Section 6: About ElesRx

ElesRx identifies deprescribing opportunities automatically. When a clinician enters a medication list, the system flags:

• Beers Criteria drugs in patients aged 65 and older — drugs that warrant review or safer alternatives
• Prescribing cascades — where a second drug is treating the side effect of the first, and removing the first may allow the second to be stopped
• Therapeutic duplications — where two drugs from the same class are prescribed concurrently, one of which may be unnecessary
• Long-term PPI use without a documented high-risk indication

Every deprescribing opportunity in this report is a flag that ElesRx can generate in real time for every patient.

elesrx.com — free tier available. Full access $9.99/month.

ElesRx is a product of PIPPS Smart Apps, a division of J.C. Epiphany Limited (Jamaica, est. 1998).

Section 7: Methodology and References

7.1 Data sources

Deprescribing recommendations in this report are drawn from published deprescribing guidelines, the ElesRx clinical database, and the sources listed below. Drug-specific Beers Criteria flags reference the AGS 2023 update (Report 1, Section 4). Prescribing cascade data references the ElesRx prescribing_cascades table (Report 1, Section 7).

7.2 Limitations

This report profiles twenty drugs. They were selected for their prevalence in Caribbean practice and the frequency with which they are continued without review. The list is not exhaustive. Many other medications are candidates for deprescribing in individual patients — the five rules in Section 2 apply broadly.

Deprescribing decisions must be individualised. This report provides general guidance, not patient-specific recommendations.

7.3 Author and conflict of interest disclosure

This report was authored by Juliet Duncan, BPharm, founder of J.C. Epiphany Limited and developer of ElesRx. The author has a commercial interest in ElesRx. This report is published freely as a contribution to Caribbean clinical education. No external funding was received.

7.4 Citation

Duncan J. The Deprescribing Report: 20 Medications Caribbean Clinicians Should Be Stopping. ElesRx Clinical Reports, Report 4. Published 2026 at elesrx.com/reports/deprescribing-report/. J.C. Epiphany Limited, Jamaica.

References

1. 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. doi:10.1111/jgs.18372

2. Scott IA, Hilmer SN, Reeve E, et al. Reducing inappropriate polypharmacy: the process of deprescribing. JAMA Intern Med. 2015;175(5):827–834. doi:10.1001/jamainternmed.2015.0324

3. Reeve E, Shakib S, Hendrix I, Roberts MS, Wiese MD. Review of deprescribing processes and development of an evidence-based, patient-centred deprescribing process. Br J Clin Pharmacol. 2014;78(4):738–747. doi:10.1111/bcp.12386

4. McNeil JJ, Wolfe R, Woods RL, et al. Effect of aspirin on cardiovascular events and bleeding in the healthy elderly (ASPREE). N Engl J Med. 2018;379(16):1509–1518. doi:10.1056/NEJMoa1805819

5. ACCORD Study Group. Effects of intensive glucose lowering in type 2 diabetes. N Engl J Med. 2008;358(24):2545–2559. doi:10.1056/NEJMoa0802743

6. 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. PMC5429265