Polypharmacy and inappropriate prescribing

Polypharmacy and deprescribing in Australian general practice

Polypharmacy — five or more regular medications — affects around 40% of community-dwelling Australians aged 65 and over, and approximately 70% in residential aged care. Adverse drug events cause around 20% of unplanned hospital admissions in older adults.

Deprescribing is the planned, supervised reduction or cessation of medications no longer providing net benefit. STOPP/START criteria and shared decision-making guide the process.

High-risk drug classes in older adults include benzodiazepines, anticholinergics, antipsychotics in dementia, long-term proton-pump inhibitors, opioids, and aspirin for primary prevention without established heart disease.

Polypharmacy — conventionally five or more regular medications — is the norm rather than the exception for older Australians. Around 40% of community-dwelling adults aged 65 and over take five or more medications; in residential aged care facilities, that proportion rises to approximately 70%. According to RACGP polypharmacy resources, adverse drug events account for around 20% of unplanned hospital admissions in older adults — a largely preventable burden.

Deprescribing is the planned, supervised reduction or cessation of one or more medications where the harm outweighs the benefit for that individual at their current stage of life. It is not medication nihilism — many older Australians appropriately require multiple well-indicated medications. The clinical task is to distinguish appropriate from inappropriate polypharmacy and to act on the distinction.

This article covers definitions, high-risk drug classes, clinical assessment tools, the deprescribing process, drug-specific protocols, Australian operational pathways, and populations requiring particular attention.

A. Core clinical — the AU general-practice framework

Definitions

Polypharmacy is conventionally defined as five or more regular medications, though some definitions use four or six. Hyperpolypharmacy refers to ten or more. Inappropriate polypharmacy describes medications without a current clear indication, drugs with significant drug-drug or drug-disease interactions, therapeutic duplications, or situations where harm clearly exceeds benefit for that individual.

Deprescribing is the planned, supervised tapering or cessation of medications to reduce harm and improve outcomes. It is a normal part of good prescribing across the life course, not a failure of care.

High-risk drug classes for older adults

Benzodiazepines and Z-drugs (zolpidem, zopiclone): The drug class carrying the highest deprescribing priority in older adults. Risks include falls and hip fractures, delirium, cognitive impairment, paradoxical agitation, dependence, traffic accidents, and complex sleep behaviours. The TGA has issued Boxed Warnings for Z-drugs regarding sleep-driving, somnambulism, and related complex behaviours. CBT for insomnia (CBT-I) is first-line for insomnia at any age and is accessible via a Mental Health Care Plan. All Z-drugs and most benzodiazepines are monitored under SafeScript/RTPM.

Anticholinergics: Anticholinergic burden accumulates across many drug classes — tricyclic antidepressants, oxybutynin, hyoscine, diphenhydramine, promethazine, and several antipsychotics all contribute. Consequences include cognitive impairment, falls, urinary retention, constipation, and worsening glaucoma. The Anticholinergic Burden (ACB) Scale quantifies this cumulative load; a score of 3 or more is associated with meaningfully increased dementia and falls risk.

Antipsychotics in dementia: Used for behavioural and psychological symptoms of dementia (BPSD), antipsychotics carry increased mortality and stroke risk (FDA Black Box Warning; Therapeutic Guidelines are concordant). Use in dementia with Lewy bodies is particularly hazardous. Non-pharmacological interventions are first-line for BPSD. If antipsychotics are necessary, risperidone is the preferred PBS-listed agent; supervised taper should be attempted after 3 months of stable behaviour.

Opioids: Falls, sedation, constipation, cognitive impairment, dependence, respiratory depression, and opioid-induced hyperalgesia. The Faculty of Pain Medicine, ANZCA identifies limited long-term evidence for opioids in chronic non-cancer pain, with harms predominating. All opioids are monitored under SafeScript/RTPM.

Long-term proton pump inhibitors (PPIs): Without a current clear indication, long-term PPI use is associated with B12 deficiency, hypomagnesaemia, fracture risk, Clostridioides difficile, and possible CKD progression. Many patients continue PPIs beyond the period of need. Step-down therapy is practical and well tolerated with alginate (Gaviscon) cover for rebound acid in the transition period.

NSAIDs: Gastrointestinal bleeding (especially combined with anticoagulants, antiplatelets, or corticosteroids), acute kidney injury, hypertension exacerbation, and heart failure exacerbation. Topical NSAIDs carry a substantially more favourable systemic risk profile for focal musculoskeletal pain.

Sulfonylureas in older adults: Gliclazide, glimepiride, and glibenclamide carry hypoglycaemia risk — particularly when HbA1c targets are set too tight in frail elderly individuals. Switching to safer agents (metformin if eGFR permits, DPP-4 inhibitors, SGLT-2 inhibitors, or GLP-1 receptor agonists) reduces this risk substantially.

Aspirin for primary prevention in older adults without established heart disease: The ASPREE trial (McNeil et al., NEJM 2018) — a large Australian and US RCT in healthy adults aged 70 and over (65 and over for Aboriginal and Torres Strait Islander peoples) — found increased bleeding without cardiovascular benefit. Aspirin for primary prevention in this population should be reviewed and ceased where the indication is absent. Secondary prevention in those with established ischaemic heart disease or ischaemic stroke is a separate clinical question with net benefit.

Statins in life-limiting illness: A 2015 RCT by Kutner and colleagues (JAMA Internal Medicine) found statin cessation in patients with a life expectancy under one year was safe, reduced pill burden, and did not worsen quality of life. Ceasing statins in palliative or life-limited individuals is supported by trial data, reduces medication burden, and does not increase mortality risk.

B. Assessment tools and the deprescribing process

Tools

STOPP/START criteria (O’Mahony et al., Age and Ageing 2023) — the 2023 update provides explicit lists: STOPP (Screening Tool of Older Persons’ Prescriptions) identifies potentially inappropriate medications by class and clinical context; START (Screening Tool to Alert doctors to Right Treatment) identifies guideline-supported medications that are frequently under-prescribed. Both lists are openly available and practical to apply in a structured review.

Beers Criteria (American Geriatrics Society 2023) — a US-derived list, updated every 3 years; useful reference, interpreted in the Australian prescribing context.

Anticholinergic Burden (ACB) Scaleavailable online as a calculator — quantifies cumulative anticholinergic load across all medications; a total ACB score of 3 or more is associated with increased dementia risk and falls.

Drug Burden Index (DBI) — weighted index incorporating both sedative and anticholinergic components; correlates with falls, cognitive decline, and functional impairment.

deprescribing.org — provides evidence-based algorithms for common drug classes including PPIs, benzodiazepines, antipsychotics, antihyperglycaemic agents, and cholinesterase inhibitors, as well as patient-facing handouts.

Choosing Wisely Australia — lists specific prescribing decisions with low-value indications that clinicians and patients may reconsider.

The deprescribing process — five steps

1. Comprehensive medication reconciliation: Request all prescribed, over-the-counter, complementary, and recreational substances. The “brown bag review” — asking the person to bring every medication and supplement to the appointment — is the most practical method. Reconcile against pharmacy dispensing records. For each drug, confirm the current indication, dose, duration, efficacy, side effects, and interactions.

2. Identify deprescribing targets: Apply STOPP/Beers criteria explicitly, calculate the ACB/DBI, and use clinical judgement regarding indication, current benefit, frailty, and life expectancy. Which drug provides the least current benefit? Which carries the greatest current harm? Which belongs to the highest-risk class for this individual?

3. Prioritise: Deprescribe one medication at a time where practical. Prioritise the drug with the worst harm-to-benefit ratio for that person. Factor in patient priorities — which medication the person finds most burdensome or distressing.

4. Shared decision-making: Discuss goals of care, preferences, frailty, and life expectancy with the person and, where appropriate, their family or carer. Explain the rationale, expected benefits, possible withdrawal effects, the monitoring plan, and what to do if symptoms recur. Document the discussion and the agreement reached.

5. Planned taper, monitoring, and re-engagement: Gradual tapering is required for benzodiazepines, opioids, gabapentinoids, antidepressants (particularly paroxetine, venlafaxine, sertraline), beta-blockers, long-term corticosteroids, and PPIs (for rebound acid). Abrupt cessation is acceptable for statins, vitamins, most over-the-counter agents, and PPIs with alginate cover. Monitor for withdrawal symptoms, disease recurrence, and functional or cognitive change. Restarting a medication if symptoms return is a valid clinical outcome — failed deprescribing attempts are expected and should not discourage future efforts.

C. Specific deprescribing protocols by drug class

Benzodiazepines and Z-drugs: Taper by approximately 10% of the current dose every 2–4 weeks; for long-standing use, schedules spanning months are often necessary. Converting short-acting agents to diazepam equivalents provides smoother titration. CBT-I is first-line for the underlying insomnia. SafeScript/RTPM monitoring applies across all Australian jurisdictions with active programs. Counsel about withdrawal symptoms: anxiety, insomnia, autonomic arousal, and rarely seizures with abrupt cessation.

Antipsychotics for BPSD: Attempt a supervised taper at 3 months of stable behavioural control, with non-pharmacological alternatives in place. DBMAS (Dementia Behaviour Management Advisory Service) provides specialist support for complex or refractory presentations.

Long-term PPIs without ongoing indication: Step-down example: omeprazole 20 mg daily → 10 mg daily → alternate days → on-demand use. Rebound acid secretion for 2–4 weeks is common and expected; alginate cover (Gaviscon) provides symptom relief and avoids misinterpreting rebound as GORD relapse. Continue PPI for Barrett’s oesophagus, severe oesophagitis, Zollinger-Ellison syndrome, and co-administration with antiplatelet agents in high-risk individuals.

Statins in life-limited illness: Discuss cessation when life expectancy is estimated at less than 12 months or goals shift to comfort-focused care. Evidence supports safety without quality of life detriment.

Aspirin for primary prevention: Review and cease based on ASPREE 2018 evidence. Continuation is appropriate in established secondary prevention (ischaemic heart disease, ischaemic stroke, peripheral arterial disease).

Bisphosphonates: A drug holiday after 5 years of oral bisphosphonate use (or 3 years of annual zoledronate) is appropriate for individuals at low fracture risk, to reduce the risk of atypical femoral fractures and osteonecrosis of the jaw. Reassess bone mineral density and fracture risk every 1–3 years; reinitiate therapy if risk increases.

Anticholinergics: Switch to an alternative drug class wherever possible — for example, oxybutynin to mirabegron for overactive bladder; tricyclic antidepressant to SSRI for depression; choosing antipsychotics with lower anticholinergic burden profiles.

Antidepressants: Taper over at least 4–8 weeks after long-term use; some agents (paroxetine, venlafaxine, sertraline) require months due to discontinuation syndrome — dizziness, brain zaps, gastrointestinal upset, and mood disturbance. Reassess the ongoing indication at regular intervals.

Diabetes medications in frail older adults: Relax HbA1c target — for example, to 8% in frail or life-limited individuals. Cease or switch sulfonylurea to metformin (if eGFR ≥45), DPP-4 inhibitor, SGLT-2 inhibitor (with renal and cardiovascular indication criteria), or GLP-1 receptor agonist. Deintensify insulin regimens when hypoglycaemia is a recurring problem.

D. Australian operations

MBS-funded medication reviews

MBS Online provides several funded pathways for medication review:

  • Home Medicines Review (HMR — item 900) — an accredited community pharmacist visits the person’s home, conducts comprehensive review, and provides written recommendations to the GP; PBS-funded, frequently under-utilised
  • Residential Medication Management Review (RMMR — item 903) — pharmacist-led review for residential aged care residents; funded at entry and at least annually
  • MedsCheck (item 90035) and Diabetes MedsCheck (item 90034) — pharmacy-based review without a home visit
  • 75+ Health Assessment (item 705) — structured health assessment including a medication review component
  • ATSI Health Assessment (item 715) — includes medication review
  • GP Chronic Condition Management Plan (GPCCMP — items 965/967) — replaced 721/723/732 from 1 July 2025; applicable for polypharmacy in the context of chronic disease
  • Standard consultations (items 23, 36, 44) — for medication review within a general practice consultation
  • MyMedicare registration (health.gov.au) — voluntary registration with a named general practice for continuity of chronic disease care, supporting ongoing polypharmacy management

SafeScript and Real-Time Prescription Monitoring (RTPM)

SafeScript is active across Victoria, NSW, South Australia, Tasmania, Western Australia, and Queensland. It provides real-time visibility of monitored substances for prescribers and pharmacists: all Schedule 8 medications (opioids, benzodiazepines, ketamine, methylphenidate), and selected Schedule 4D drugs including Z-drugs, certain benzodiazepines, and gabapentinoids (pregabalin, gabapentin). Checking SafeScript before prescribing a monitored substance is both clinically important and legally required in jurisdictions where it is active.

Resources for GPs and patients

E. Special populations

Residential aged care (RACF) residents. Polypharmacy prevalence is highest here. RMMR (item 903) is funded at admission and annually. DBMAS provides specialist support for antipsychotic reduction in BPSD. Choosing Wisely Australia has specific recommendations targeting RACF overprescribing. Falls risk, delirium risk, and goals of care all inform the deprescribing priority list.

Frail older adults. Frailty assessment using the Clinical Frailty Scale should inform blood pressure targets, HbA1c targets, and opioid thresholds — applying standard disease targets uniformly in frailty causes harm. The RACGP aged care clinical guidelines provide frailty-informed prescribing frameworks.

Aboriginal and Torres Strait Islander peoples. High chronic disease burden and high polypharmacy prevalence, combined with health literacy considerations and the cultural context of medications, require collaborative and culturally responsive medication review. Item 715 (ATSI Health Assessment) provides a funded entry point for structured review.

People with dementia. Informed consent for shared decision-making may require engagement with a substitute decision-maker or reference to an advance care directive. Anticholinergic burden, antipsychotics, and opioids each warrant explicit attention in dementia at every medication review.

People approaching end of life. Goals of care shift substantially; preventive medications — statins, antihypertensives, bisphosphonates, aspirin — often offer no meaningful benefit within the person’s remaining life expectancy and add pill burden. Proactive, compassionate deprescribing conversations are valued by patients and families in this context.

When to escalate

Refer to a geriatrician or clinical pharmacist specialist for:

  • Complex polypharmacy in frail elderly patients where straightforward review is insufficient
  • High anticholinergic burden with significant cognitive or functional decline
  • Multiple drug-drug or drug-disease interactions requiring systematic pharmacological analysis
  • BPSD where antipsychotic deprescribing requires specialist DBMAS support
  • Complex end-of-life prescribing and medico-legal considerations in RACF settings

Comprehensive Geriatric Assessment is appropriate for older adults with polypharmacy combined with frailty, falls, delirium, or progressive functional decline.

What this article is and is not

This is general health information drawn from current Australian prescribing guidelines — Therapeutic Guidelines (eTG), Australian Medicines Handbook (AMH), RACGP polypharmacy resources, NPS MedicineWise, and Choosing Wisely Australia. It is not personal medical advice and does not create a doctor–patient relationship. Medication review decisions are made together with the treating GP and relevant specialists.

For consumer-facing resources: NPS MedicineWise, HealthDirect, and deprescribing.org patient resources.


Sources cited

  1. RACGP — Polypharmacy and deprescribing resources
  2. Therapeutic Guidelines (eTG)
  3. Australian Medicines Handbook (AMH) Aged Care Companion
  4. STOPP/START criteria 2023 — O’Mahony et al., Age and Ageing
  5. Beers Criteria — American Geriatrics Society 2023
  6. deprescribing.org
  7. Choosing Wisely Australia
  8. NPS MedicineWise
  9. ASPREE Trial — McNeil et al., NEJM 2018
  10. Kutner JS et al. — Statin cessation — JAMA Internal Medicine 2015
  11. Anticholinergic Burden calculator (ACB)
  12. SafeScript Victoria / RTPM
  13. MBS Online — medication review items
  14. Faculty of Pain Medicine, ANZCA
  15. TGA
  16. HealthDirect

Frequently asked questions

  • What is the difference between polypharmacy and inappropriate polypharmacy?

    Polypharmacy is a numerical descriptor — five or more regular medications — and is not inherently problematic. Many older adults with multiple chronic conditions appropriately require five or more well-indicated medications. Inappropriate polypharmacy describes the subset where medications lack a current clear indication, where drug interactions are present, where duplications exist, or where harm outweighs benefit for that individual at their stage of life. The clinical task is to identify inappropriate polypharmacy through systematic review and deprescribe selectively, not to reduce medication numbers as an end in itself.

  • Which medications are most dangerous for older adults?

    Benzodiazepines and Z-drugs (zolpidem, zopiclone) cause falls, hip fractures, delirium, cognitive impairment, and dependence in older adults. Anticholinergic medications — including some antidepressants, bladder agents, antihistamines, and antipsychotics — accumulate burden across multiple drug classes, impairing cognition and bladder function. Antipsychotics in dementia carry increased mortality and stroke risk. Opioids increase falls and cognitive impairment. NSAIDs worsen kidney function and blood pressure. Sulfonylureas cause hypoglycaemia in those with tight glucose targets. Aspirin for primary prevention without established heart disease was shown to cause net harm in healthy Australians over 70 in the ASPREE trial.

  • How does a doctor safely stop a long-term medication?

    Safely stopping a long-term medication requires a planned approach: confirm the original indication is still current, discuss the proposed change with the person and their family or carer, plan a taper schedule where required, and arrange monitoring and clear re-engagement criteria if symptoms return. Gradual tapering is essential for benzodiazepines, opioids, antidepressants, and proton pump inhibitors (to manage rebound acid). The key message is that restarting a medication if symptoms return is a valid outcome — failed deprescribing is not a reason to avoid future attempts, and it does not cause harm to try.

  • What are the main tools GPs use to identify medication problems in older adults?

    The main clinical tools are the STOPP/START criteria (Screening Tool of Older Persons' Prescriptions and Screening Tool to Alert doctors to Right Treatment), updated in 2023. STOPP identifies potentially inappropriate medications; START identifies evidence-based treatments that may be missing. The Anticholinergic Burden (ACB) Scale quantifies cumulative cognitive and functional risk across all medications with anticholinergic properties. The Drug Burden Index incorporates both sedative and anticholinergic components. The deprescribing.org website provides practical algorithms for the most commonly reviewed drug classes including proton pump inhibitors, benzodiazepines, and antipsychotics.

  • Are Home Medicines Reviews covered under Medicare?

    Yes. The Home Medicines Review (HMR, MBS item 900) is a Medicare-funded service in which an accredited community pharmacist visits the patient at home, conducts a comprehensive review of all medications, and provides recommendations to the GP. It is especially valuable for older adults living independently with complex medication regimens. For residents in aged care facilities, the Residential Medication Management Review (RMMR, MBS item 903) provides an equivalent service. Both are frequently under-utilised despite the clear benefit in reducing adverse drug events and unplanned hospitalisations.

Source quality

Sources grouped by evidence tier. AU primary tier first; international where AU is silent or lagging; named-author reconstruction where guidelines have not yet caught up. How tiers work.