Polypharmacy in people aged 65 and over

Polypharmacy single map PHO analysis Consumer summary (531 KB, PDF) 

Polypharmacy refers to the concurrent use of multiple medicines by a person. It can be beneficial (‘appropriate polypharmacy’) or harmful (‘problematic polypharmacy’).[1]

  • Appropriate polypharmacy has been described as ‘prescribing for an individual for complex conditions or for multiple conditions in circumstances where medicines use has been optimised and where the medicines are prescribed according to best evidence’.[1]
  • Problematic polypharmacy is the prescribing of many medicines inappropriately (five or more medicines is often quoted) or adding an inappropriate medicine to an existing regime.[2]

The goal of this Atlas domain is to identify whether there is wide variation in rates that may highlight areas for further local investigation.

Domain update 2021

Updated with 2019 data

A total of 25,000 people aged 65 years and over received the ‘triple whammy’ in 2019.

The triple whammy is the combination of an angiotensin converting enzyme (ACE) inhibitor/angiotensin receptor blocker (ARB), a diuretic and a non-steroidal anti-inflammatory drug (NSAID). Medsafe notes an increased risk of acute kidney injury with this combination, especially in people with risk factors for renal failure and in older adults.[3] The combination should be avoided if possible.

  • Of those aged 65 years and over, 3.2 percent were dispensed the triple whammy in the same 90-day period. 
  • This equates to 25,000 people in 2019. 
  • Rates were significantly higher in younger Māori and Pacific peoples than those in the Asian or European/other groups. Rates reduced significantly with age for all populations.
  • This indicator does not include those who bought an NSAID over the counter or had a prescription in a previous period. 
  • It is unknown whether people actually took this combination. 

Table 1: Age group and ethnic group of those dispensed the triple whammy in a quarter, Aotearoa New Zealand, 2019

Ethnicity Age
65–74 75–84 85 and over
% count % count % count






Pacific peoples

























Note: Ninety-five percent confidence intervals are shown in brackets.

A total of 3,400 older people were dispensed the combination of a benzodiazepine or zopiclone and a strong opioid following a public hospital event.

The combination of a benzodiazepine or zopiclone and a strong opioid carries an increased risk of over-sedation. The Federal Drug Agency in the USA issued an alert in 2017 warning against prescribing this combination due to the risk of serious side effects, including slowed or difficult breathing and death. It recommends using the combination only in patients for whom alternative treatment options are inadequate. The UK government has issued a similar drug safety update.

  • We have corrected a technical error in which dispensing of both medicines on the same day was not accounted for correctly in previous publications of this Atlas domain. This means the number of people in 2017 dispensed both a strong opioid and a benzodiazepine/zopiclone on the same day has changed from 800,000 to 15,000. Rates have changed significantly as a result.
  • Around 3,400 older people were dispensed the combination of a benzodiazepine or zopiclone and a strong opioid following a public hospital event in 2019.
  • This included only those who had not had a dispensing for benzodiazepine or zopiclone in the previous three months, suggesting the combination was started in hospital.
  • Rates are significantly higher for Māori, Pacific and Asian people than for European/other.
  • Rates decreased with age, from 26 percent of those aged 65–74 years to 22 percent in the 85 and over age group.
  • The opioids Atlas domain contains data on a similar indicator for people aged 20 years and over.

Table 2: People aged 65 years and over receiving a benzodiazepine or zopiclone and a strong opioid following a public hospital event, by age group, Aotearoa New Zealand, 2019 

  Age (%)
65–74 75–84 85 and over Total
Dispensed a benzodiazepine/zopiclone and a strong opioid following a public hospital event




(22.9 – 24.4)

Note: Ninety-five percent confidence intervals are shown in brackets.

Polypharmacy is associated with:

  • people not taking medicines as prescribed
  • significant costs to patients and health services
  • poor health outcomes, such as adverse drug events, drug interactions, admissions to hospital and death.

The frequency of adverse drug events increases with the number of medicines taken: from 13 percent with two medicines and 58 percent with five medicines to 82 percent with seven or more.[4]

Polypharmacy is more likely to be appropriate in the robust 'young elderly' while problematic polypharmacy is more likely to occur in the frail 'old elderly'. Hence a focus on the 85 years and over age group may be most appropriate, where the doses used may be as important as the number of medicines.

Overall, the rate of use of long-term medicines has reduced since 2012 but rates continue to be high for Pacific peoples and in those aged 85 and over.

  • On average, 31 percent of people aged 65 years and over received five or more long-term medicines (five or more of the same chemicals dispensed over two consecutive quarters). There was a 2.3 percent reduction between 2012 and 2019. These findings are in the context of an ageing population and an international trend towards increased prescribing over time.[1]
  • Pacific peoples were more likely to receive five or more long-term medicines than any other ethnic group. It is particularly worrying that the rate has increased sharply for Pacific peoples receiving 11 or more unique long-term medicines, from 5.5 percent in 2016 to 6.5 percent in 2019. This increase is more rapid for Pacific peoples than for any other ethnicity.
  • Also of concern is that the rate of long-term medicines dispensed is significantly higher with older age, from 23 percent of those aged 65–74 years to 52 percent of those aged 85 years and over.
  • Around 4 percent of older people received 11 or more long-term medicines. The rate increased sharply with age, with people aged 85 years and over being 2.5 times more likely to receive 11 or more medicines than those aged 65–74 years. 

Table 3: People aged 65 years and over receiving five or more long-term medicines, by age, Aotearoa New Zealand, 2019

  Age (%)
65–74 75–84 85 and over
Dispensed 5 or more long-term medicines 23.1 39.1 51.6
5–7 long-term medicines 14.6 22.8 28.7
8–10 long-term medicines 5.9 11.2 16.3
11 or more long-term medicines 2.7 5.1 6.7

Table 4: People aged 65 years and over receiving five or more long-term medicines, by ethnic group, Aotearoa New Zealand, 2019

  Long term medicines dispensed
Ethnic group (%) 5 or more 5–7 8–10 11 or more
Māori 33.1 18.1 10.2 4.8
Pacific peoples 46.4 25.2 14.6 6.5
Asian 27.0 15.4 7.8 3.9
European/other 30.9 18.8 8.4 3.7

Māori and Pacific peoples received more medicines at a younger age compared with those identifying as Asian or European/other.

Table 5 shows the percentage and count by age group and ethnic group of those dispensed five or more long-term medicines. 

Table 5: People dispensed five or more long-term medicines, by age group and ethnic group, Aotearoa New Zealand, 2019

Ethnic group Age group (years)
65–74 75–84 85 and over
% Count % Count % Count
Māori 29.3 10736 41.3 5694 41.7 1251
Pacific peoples 43.5 6217 53.7 3160 45.2 633
Asian 21.7 7344 35.9 4922 46.0 1432
European/other 21.8 78,676 38.8 78599 52.3 42043

Antipsychotic and benzodiazepine or zopiclone dispensing increases significantly with age.

In older people, certain classes of medicines carry a substantially higher risk of adverse effects. Two examples presented in this Atlas domain are antipsychotics and a benzodiazepine or zopiclone. Common adverse effects include impaired functional ability, agitation, confusion, blurred vision, urinary retention, constipation, postural hypotension and falls. These increase if both classes of medicine are taken together. This indicator cannot assess inappropriate use of these medicines; however, high rates of prescribing may indicate misuse or overuse.

  • The use of psychotropic agents increased with age; of people aged 85 years and over, on average 7.1 percent received an antipsychotic and 18.7 percent received a benzodiazepine or zopiclone.
  • The rates for a combination of antipsychotics and a benzodiazepine or zopiclone were low but increased with age, from 0.8 percent of those aged 65–74 years up to 3.4 percent of those aged 85 years and over. 
  • Rates of antipsychotic use have increased slightly since 2012 whereas benzodiazepine or zopiclone use has reduced significantly.
  • Rates of benzodiazepine or zopiclone use are significantly higher for European/other than any other ethnic group, at all ages.

Table 6: People dispensed an antipsychotic, by age group and ethnic group, Aotearoa New Zealand, 2019

Ethnic group Age group (years) (%)
65–74 75–84 85 and over Total
Māori 2.2 3.6 5.9 2.8
Pacific peoples 1.9 3.6 6.3 2.6
Asian 1.2 1.9 5.1 1.7
European/other 2.0 3.2 7.2 3.0
Total 2.0 3.1 7.1 2.9

Table 7: People dispensed a benzodiazepine or zopiclone, by age group and ethnic group, Aotearoa New Zealand, 2019

Ethnic group Age group (years) (%) 
65–74  75–84  85 and over  Total
 Māori  4.7  6.5  8.4  5.4
 Pacific peoples  2.5  4.6  6.3  3.3
 Asian  5.5  9.6  14.4  7.2
 European/other  8.8  12.7  19.5  11.4
 Total  8.0  12.0  18.7  10.4

Psychotropic medication increases the risk of falling [5] and there is evidence that reducing psychotropic medication can result in no or limited worsening of key outcomes such as sleep quality or behavioural problems.[6] The falls Atlas domain shows that, in 2016, 25 percent of people aged 85 years and over made at least one ACC claim for a fall; hip fractures following a fall were most common in that age group, at a rate of 21.8 per 1,000.

What questions might the data prompt?

  • How do similar district health boards compare? 
  • Why are greater proportions of European/other people aged 65 years and over dispensed a benzodiazepine or zopiclone than Māori or Pacific peoples?
  • Why is there a difference between the North and South Islands in the use of antipsychotics and a benzodiazepine or zopiclone?
  • What role might secondary clinicians have in influencing prescribing patterns in their community?

More information on medication review and managing polypharmacy

Recommended reading

Corbett A, Burns A, Ballard C. 2014. Don’t use antipsychotics routinely to treat agitation and aggression in people with dementia. BMJ 349. DOI: 10.1136/bmj.g6420. URL: https://www.bmj.com/content/349/bmj.g6420 (accessed April 2021). 

Gnjidic D, Tinetti M, Allore HG. 2017. Assessing medication burden and polypharmacy: finding the perfect measure. Expert review of clinical pharmacology 10(4): 345–7. https://doi.org/10.1080/17512433.2017.1301206.

Guthrie B, Makubate B, Hernandez-Santiago V, et al. 2015. The rising tide of polypharmacy and drug-drug interactions: population database analysis 1995–2010. BMC Med 13: 74. https://doi.org/10.1186/s12916-015-0322-7.

Jackson G, Gerard C, Minko N, et al. 2014. Variation in benzodiazepine and antipsychotic use in people aged 65 years and over in New Zealand. NZMJ 127(1396): 67–78.

Consumer resources from the Health Quality & Safety Commission

Let’s plan for better care– a health literacy initiative to help consumers prepare well for their visit to the GP or other primary care health professional. www.hqsc.govt.nz/our-programmes/partners-in-care/publications-and-resources/publication/2424 external link

Let’s plan pharmacy week factsheets and poster – resources from 2015 for helping people learn more about their medicines by talking to their pharmacist, or taking away information. The aim is to improve health literacy and reduce harm from high-risk medicines. www.open.hqsc.govt.nz/patient-safety-week/lets-plan/lets-plan-pharmacy-week

Primary care guide to working with consumers – www.hqsc.govt.nz/our-programmes/partners-in-care/publications-and-resources/publication/3777

Three steps to better health literacy – a guide for health care professionals. (Please note this resource is currently being updated and will be replaced) www.hqsc.govt.nz/our-programmes/partners-in-care/publications-and-resources/publication/2046


For analysis, people were assigned to the district health board (DHB) where they normally live (DHB of domicile).

Data for this Atlas domain was drawn from the Pharmaceutical Collection, which contains claim and payment information from community pharmacists for subsidised dispensing.

The data presented does not allow for analysis of patients’ condition or the effectiveness of dose provided. This means it was not possible to assess the appropriateness or otherwise of prescribing. Instead, proxy markers were employed, beginning with a simple count of the number of long-term medicines taken by older people. Unsubsidised or over-the-counter medicines are not included. Data also does not indicate whether people took the medicine.

Download the methodology (246 KB, PDF)



  1. Duerden M, Avery T, Payne R. 2013. Polypharmacy and medicines optimisation. Making it safe and sound. London: The King's Fund. URL: www.kingsfund.org.uk/publications/polypharmacy-and-medicines-optimisation (accessed August 2015).
  2. Aronson J. 2004. In defence of polypharmacy. British Journal of Clinical Pharmacology 57(2): 119–20.
  3. Medsafe: www.medsafe.govt.nz/profs/PUArticles/June2013NSAIDS.htm.
  4. Patterson S, et al. Interventions to improve the appropriate use of polypharmacy for older people. Cochrane Database Syst Rev 2012, Issue 5, Art. No. CD008165.
  5. Health Quality & Safety Commission. 2014. Topic 8: Medicines: balancing intended benefits and increased falls risks. Wellington: Health Quality & Safety Commission. URL: www.hqsc.govt.nz/our-programmes/reducing-harm-from-falls/publications-and-resources/publication/2879/ (accessed October 2017).
  6. Hill KD, Wee R. 2012. Psychotropic drug-induced falls in older people: a review of interventions aimed at reducing the problem. Drugs Aging 29(1): 15–30. URL: www.ncbi.nlm.nih.gov/pubmed/22191720 (accessed August 2015).

Last updated 25/06/2021