Prescription bagIntroduction
Declining benefits of prescribing in an elderly population
Increasing risks of prescribing in an elderly population
Proton pump inhibitors (PPIs)
Z-hypnotic medications
Antipsychotic medication
Diabetic medication
Calcium supplementation and bisphosphonates



Elderly patients with dementia who are residents in care homes are frequently prescribed multiple medications. The benefits of each must be weighed against the risk of harm through adverse drug reactions. The benefits of preventative medication in an elderly population are known to reduce as their associated risk of harm increases with time. Polypharmacy is also an independent risk factor for negative physical health consequences having been associated with falls, worsening cognitive impairment, hospitalisation and mortality. Unnecessary polypharmacy also incurs a financial cost and the additional time required on medication rounds diminishes opportunities for more positive interaction with residents.

This review article summarises the rationale for reducing general levels of drug prescribing and lists specific medications whose risk profile becomes less favourable in this cohort.


Declining benefits of prescribing in an elderly population

Firstly, robust trial data regarding efficacy and risk of harm of medications in this cohort is often not available due to the infrequent participation of this group in clinical trials.

Furthermore, medications prescribed to prevent the emergence of disease typically have time to exert their effect in a younger population. However, for an older, frailer population there may be less time for this benefit to be realised.

A 2011 BUPA Report1 revealed that the median length of stay for residents in BUPA residential and nursing homes in the UK was 462 days. In the subset of residents with dementia, the median was 43 days less. In other words, half of all residents with dementia had died approximately 13 months after arrival. Life expectancy was even shorter if analysis was confined to nursing home residents.

Whilst this short life expectancy will vary according to the specific frailty of the person in question, there is an argument that the philosophy should be to move towards more palliative, though not end of life care.


Increasing risks of prescribing in an elderly population

A study in Australia revealed that one in four elderly persons in the community were hospitalised for medication related problems over five years.2

The risk of adverse events also increases dramatically with the number of medications prescribed and has been estimated at 82% for concurrent prescriptions of seven drugs or more,3 highly applicable to patient groups in residential homes who are reported to be prescribed on average seven medications.4 Polypharmacy is an independent risk factor for falls when combined with a medication known to also increase this risk5 and has been associated with an increased risk of cognitive impairment.6

Adverse effects may be misinterpreted as symptoms in new diseases and be treated with medication thus exacerbating the problem. These may be direct side effects from one medication or due to an interaction eg. selective seretonin reuptake inhibitor (SSRI) and non-steroidal anti-inflammatory drugs (NSAIDs) leading to gastro-intestinal haemorrhage.

There are also direct and indirect cost implications for unnecessary prescribed medication in a healthcare system under increasing financial pressure.



In this subgroup of patients, total cholesterol has been shown to fall naturally without the need for pharmaceutical intervention. It is hypothesised that this is due to the catabolic nature of chronic disease.7

Additionally, most statins require administration late at night for maximal effectiveness. If residents are woken up for administration of medication, this disruption to the sleep cycle is likely to be harmful.


  • Benefits of preventative medications limited by poor life expectancy
  • Financial cost
  • Reducing time-consuming drug administration, freeing time for care
  • Increased risk of drug interactions
  • Taking medication may be unpleasant for the patient
  • Polypharmacy has strong associations with:
    • Falls
    • Cognitive Impairment
    • Hospitalisation
    • Mortality



Addressing hypertension has well established cardiovascular benefits and uncontrolled hypertension is a risk factor for Alzheimer’s and vascular dementia.8 However, more advanced dementia itself causes autonomic dysregulation resulting in a subsequent fall in blood pressure. Orthostatic hypotension and persistently low blood pressure is reported to be found in 39–52% of patients with organic dementia.9

Orthostatic hypotension may increase the risk of falls by up to 50%12 and recent research would suggest that it confers a 15% increase in the long- term risk of dementia including Alzheimer’s disease.10

The progressive nature of dementia therefore requires careful monitoring of the potential development of autonomic dysregulation and reconsideration of the merits of continuing anti-hypertensive prescription in this group.


Proton pump inhibitors (PPIs)

Indications for PPIs include treatment of gastro-oesophageal reflux disorder and prophylaxis of gastrointestinal bleeding associated with antiplatelet therapy.

Whilst this may be appropriate, PPIs are also known to double the incidence of Clostridium difficile diarrhoea amongst hospital in-patients.11 Furthermore, amongst patients who contract Clostridium difficile, those on acid suppression therapies were 2.4 times more likely to have complications and 4.7 times more likely to die,12 an effect likely to be even stronger in the elderly and in the high risk population residing in care homes.

PPIs have been associated with chronic kidney disease and an increased risk of dementia though a cause and effect relationship has not yet been established.13 Careful consideration of the benefits of PPIs therefore needs to be weighed against the risks listed above.


  • Dizziness/light headedness on standing
  • Progression in dementia
  • Unexplained falls
  • Transient loss of consciousness
  • Polypharmacy
    • Diuretics and anti-hypertensives
    • Alpha blockers eg. tamsulosin/doxazosin


Z-hypnotic medications

Sleep disturbance in dementia is both common and disabling, leading to daytime hyper-somnolence, night-time wandering predisposing to falls and potential disruption to other residents in care homes. Z-hypnotic medications are commonly prescribed to address this. It is thought that 12–16% of patients with dementia in Scotland are prescribed a hypnotic or anxiolytic,14 though in care homes this figure is likely to be an underestimate with studied international rates varying from 22% to 47%.15 Their main benefit is felt to be a potential reduction of sleep latency, but a large meta-analysis has shown that The Number Needed to Treat (NNT) in order to achieve a 25 minute improvement was 13. The same study estimated that two out of every 13 patients prescribed hypnotics would experience an adverse effect.16 These included an increased risk of falls, poor coordination and the development of tolerance and subsequent withdrawal which can occur as early as one week into nightly administration.

Orthostatic hypotension may increase the risk of falls by up to 50%

As a result of this NICE guidelines suggest that hypnotic prescription should be limited and reserved for severe insomnia which is impacting on quality of life and only after an unsuccessful trial of non-pharmacological methods and optimal sleep hygiene.17

Following optimisation of non-pharmacological methods, an alternative choice to hypnotic medication is low dose trazodone (50mg nocte), which has some evidence for improving sleep in those with Alzheimer’s Dementia according to a Cochrane Review18 and may be particularly helpful in suspected comorbid depression.


  • Increase day time stimulation (music, games, books)
  • Minimise daytime napping
  • Encourage bedtime routine (i.e. consistent times for going to bed and getting up)
  • Reduce intake of stimulants/caffeine six hours prior to bedtime
  • Regular toileting before bedtime
  • Encourage relaxation 2–3 hours before bedtime
  • Encourage peace and quiet at bedtime minimising disruption by staff or other residents overnight


Antipsychotic medication

The risks of anti-psychotic medication for elderly patients with dementia are increasingly well understood. There is an increased risk of stroke (approximately 2.5% per year compared with an average baseline of approximately 1%). There is also an increased risk of death with a study estimating that following 46 weeks of use, those on quetiapine were twice as likely to have died. There is also evidence to suggest that antipsychotics are ineffective for wandering, shouting, screaming, verbal agitation, insomnia, sexual disinhibition or any other type of non-physically aggressive symptom.

Antipsychotics are indicated in those with psychosis (most commonly persecutory delusions in Alzheimer’s Dementia) and physical aggression when the need is great and if other methods, such as treating pain, are unsuccessful. However, even in these situations they have limited efficacy with NNTs estimated as five for risperidone, 10 for olanzapine and 30 for quetiapine. Risperidone is the only antipsychotic licenced for treatment of behavioural problems in people with dementia and this extends for only six weeks.

Alternative non-pharmacological approaches are preferable (see Table 4). If absolutely necessary, when prescribed, a time limited prescription is highly recommended. Approximately 50% of behavioural symptoms dissipate by six months due to further progression of dementia. A stop date allows a review of medication efficacy and ensures that the symptoms have not remitted naturally.


Diabetic medication

Diabetes mellitus is a common comorbidity in care homes. The standard target for medical intervention is tight glycaemic control to minimise risks of short term complications, eg. hyperglycaemia and longer-term complications, eg. end-organ damage. Elderly, cognitively impaired patients are at high risk of hypoglycaemia. Causes for this may include impaired counter-regulatory responses and an intrinsic increasing lack of hypoglycaemic awareness. Furthermore, weight often decreases as dementia progresses, reducing insulin resistance often without a corresponding decrease in the dose of prescribed medications.19, 20

Adverse effects from hypoglycaemia are known to include falls with associated fracture and injury, cardiovascular events, transient cognitive impairment, seizure, coma and death. Importantly, hypoglycaemia in those with dementia has been shown to have a bidirectional relationship, with hypoglycaemia worsening cognitive impairment and progression of dementia increasing the likelihood of hypoglycaemia potentially setting up a vicious cycle.21

An approach whereby quality of life is prioritised over the risk of long-term complications and where medications prone to causing hypoglycaemia are avoided or reviewed in the context of weight and response is therefore recommended.


  • Assessing and treating any co-existing pain linked with agitation
  • Identifying and avoiding antecedent factors i.e. Behavioural Analysis
  • Consider alternatives:
    • Memantine can reduce psychotic and agitation symptoms
    • Carbamazepine has some evidence for intermittent aggression
    • If antipsychotic medication required:
      • Low dose prescribing with a stop-date to allow for reassessment
      • Recommend three months duration
      • Ideally stop date on a Monday allowing residential home to call the prescriber if symptoms recur



Benzodiazepines are frequently prescribed for insomnia in the elderly, but also for behavioural symptoms associated with dementia for which there is no real evidence base. Use for longer periods of time is now known to be fraught with risk with an increased likelihood of further cognitive deterioration, delirium, falls, fractures and even death.22, 23 They may induce paradoxical agitation leading to dose increases or further polypharmacy. Once established, benzodiazepines are also potentially challenging to withdraw. They have been identified as the medication class most in need of specific deprescribing advice by healthcare professionals.24 Hence they should be introduced only with extreme caution and withdrawn as soon as possible.


  • Reframing the issue as care optimisation
  • Maximising time efficiency through
    • Targeting high risk patient groups
    • Targeting high risk medications
  • Accessing deprescribing guides and regimens
  • Multidisciplinary meetings focusing on deprescribing
  • Providing personalised information regarding risks and benefits to families or patients where applicable


Calcium supplementation and bisphosphonates

Bisphosphonates are commonly prescribed to elderly patients in order to reduce the risk of osteoporotic fractures. The optimum duration of treatment has been controversial with many patients prescribed bisphosphonates indefinitely. A recent FDA report comparing multiple studies reported that those discontinuing bisphosphonates had no difference in fracture rate compared to those on active treatment and that three to five years of treatment was enough to confer this benefit. The report concluded that there was no significant advantage in continuing therapy beyond five years.25 Additionally whilst those in residential care and with dementia are much more likely to fall, a subset will have such markedly impaired mobility as to be bed bound and thus be at a reduced risk for fracture.

There are also risks in continuing long-term bisphosphonate therapy. A systematic review and meta-analysis published in 2013 showed that bisphosphonates were associated with an increased risk of subtronchanteric femoral fractures, femoral shaft fractures and atypical femoral fractures with adjusted RR of 1.7 [95% CI 1.22–2.37].

In summary, emerging evidence demands careful consideration of the benefits of longer term bisphosphonate therapy especially in the group with very limited mobility with a lower baseline risk of fracture.



Even having identified medications whose risk appears to outweigh benefit in absolute terms it can be challenging to initiate a withdrawal of the drug for multiple reasons. There may be concerns regarding reversing the prescription decision of another clinician, changing medications in a patient who appears to tolerate the current regime, fear of precipitating relapse or withdrawal, a sense that care is being ‘downgraded’ and limited time to initiate withdrawal.26

However, these barriers are not insurmountable. Rather than deprescribing being seen as a ‘downgrade’, it should be viewed as an optimisation of care and an attempt to reduce the likelihood of adverse drug events. With limited resources and time, patients at the highest risk of adverse events could be targeted. Furthermore, within this group, withdrawal of medications with the highest risk could be prioritised.27 There are specific discontinuation guides available to provide a framework and assist with withdrawing medications. Finally, where possible, multi-disciplinary meetings with deprescribing advice from pharmacologists have been shown to be helpful.28



Elderly care home residents with dementia are very likely to be prescribed multiple medications yet this is the group most likely to experience adverse effects as a direct result. This population has a limited life expectancy and the benefit, particularly of preventative medications, decreases with time. This article has focused on medications demanding particular scrutiny. There is a temptation to see discontinuing medications as a demotion in care, where in fact a re-evaluation of their effectiveness plays a crucial role in maximising quality of life and preventing iatrogenic harm.

It can therefore be argued that for the frail elderly population in care homes, the philosophy should be symptom based treatment (though not end of life care).


Dr J Blackman, Consultant Psychiatrist, Victoria Centre, 53 Downs Way, Swindon
Dr S Manchip, Consultant Psychiatrist, Victoria Centre, 53 Downs Way, Swindon
Conflict of interest: none declared



1. Forder J, Fernandez JL. Length of stay in care homes. Report commissioned by Bupa Care Services, PSSRU Discussion Paper 2769. 2011. Canterbury: PSSRU

2. Kalisch LM, Caughey GE, Barratt JD, et al. Prevalence of preventable medication-related hospitalizations in Australia: an opportunity to reduce harm. Int J Qual Health Care 2012; 24: 239–49

3. Atkin PA, Veltch PC, Veitch EM, Ogle SJ. The epidemiology of serious adverse drug reactions among the elderly. Drugs Aging 1999; 14: 141–52

4. Elliott RA, Thomson A. Assessment of a nursing home medication review service provided by hospital-based clinical pharmacists. Aust J Hosp Pharm 1998; 29: 255–60

5. Ziere G, Dieleman JP, Hofman A, et al. Polypharmacy and falls in the middle age and elderly population. Br J Clin Pharmacol 2006; 61(2): 218–23

6. Niikawa H, Okamura T, Ito K, et al. Association between polypharmacy and cognitive impairment in an elderly Japanese population residing in an urban community. Geriatr Gerontol Int 2016 doi: 10.1111/ggi.12862.

7. Solomon A, Kareholt I, Ngamdu T, et al. Serum cholesterol changes after midlife and late-life cognition. Neurology 2007: 751–56

8. Zhenchao G, Viitanen M, Fratiglioni L, Winblad B. Low blood pressure and dementia in elderly people: The Kungsholmen Project. BMJ 1996; 312 [Available from :]

9. Passant U, Warkentin S, Gustafson L, et al. Orthostatic hypotension and low blood pressure in organic dementia: A study of prevalence and related clinical characteristics. International Journal of Geriatric Psychiatry 1997; 12(3): 395–403

10. Wolters FJ, Mattace-Raso FUS, Koudstaal PJ, et al. Orthostatic hypotension and the long-term risk of dementia: a population-based study. PLoS Med 2016 [published online 11 October, doi:10.1371/journal.pmed.1002143]

11. Dial S, Alrasadi K, Manoukian C, et al. Risk of Clostridium difficile diarrhoea among hospital inpatients prescribed proton pump inhibitors: cohort and case-control studies. CMAJ 2004; 171(1): 33–38

12. Morrison RH, Hall NS, Said M, et al. Risk Factors Associated With Complications and Mortality in Patients With Clostridium difficile Infection. Clinical Infectious Diseases 2011; 53(12): 1173–78

13. Lazarus B, Chen Y, Wilson FP, et al. Proton Pump Inhibitor Use and the Risk of Chronic Kidney Disease. JAMA Intern Med 2016; 176(2): 238–46

14. Guthrie B. Time trends in psychoactive prescribing to people with dementia in Scotland 2001-2011. Published 2012. University of Dundee, NHS Fife

15. Connelly P. Hypnotic Use In People With Dementia – Are Our Eyes Open To The Problem? Old Age Psychiatrist 2013 (57)

16. Glass J, et al. Sedative hypnotics in older people with insomnia: meta-analysis of risks and benefits. BMJ 2005; 331: 1169

17. Guidance on the use of zaleplon, zolpidem and zopiclone for the short-term management of insomnia. Technology appraisal guidance [TA77]. NICE April 2004

18. Mcleeary, et al. Cochrane Library – Pharmacotherapies for sleep disturbance in Alzheimer’s Disease Online 21/3/2014

19. Holstein A, Plaschke A, Egberts EH. Clinical characterisation of severe hypoglycaemia – a prospective population-based study. Exp Clin Endocrinol Diabetes 2003; 111: 364–69

20. Boucai L, Southern WN, Zonszein J. Hypoglycaemia associated mortality is not drug-associated but linked to comorbidities. Am J Med 2011; 124: 1028–35

21. Yaffe K, Falvey CM, Hamilton N, et al. Association between hypoglycaemia and dementia in a biracial cohort of older adults with diabetes mellitus. JAMA Intern Med 2013; 173: 1300-6

22. Kripke, D. Chronic hypnotic use: deadly risks, doubtful benefit. Sleep Medicine Reviews; 4(1): 5–20

23. Weich S, Pearce HL, Croft P, et al. Effect of anxiolytic and hypnotic drug prescriptions on mortality hazards: retrospective cohort study. BMJ 2014; 348: g1996

24. Farrell B, Tsang C, Raman-Wilms L, et al. What Are Priorities for Deprescribing for Elderly Patients? Capturing the Voice of Practitioners: A Modified Delphi Process. PLoS ONE 2015; 10(4)

25. Background Document for Meeting of Advisory Committee for Reproductive Health Drugs and Drug Safety and Risk Management Advisory Committee. FDA September 2011 (

26. Scott I, Anderson K, Freeman C, Stowasser D. First do no harm : A real need to deprescribe in older patients. MJA 2014; 201(7)

27. Best Practice Advocacy Centre New Zealand. A practical guide to stopping medicines in older people. Best Practice Journal 2010(27). [Accessed Oct 2016]

28. Bregnhal L, Thirstrup S, Kristensen MB, et al. Combined intervention programme reduces inappropriate prescribing in elderly patients exposed to polypharmacy in primary care. Eur J Clin Pharmacol 2009; 65: 199–207