Article Text

Frailty, polypharmacy and deprescribing


Multimorbidity and associated polypharmacy present a significant and increasing challenge to patients, carers and healthcare professionals.1,2 While it is recognised that polypharmacy can be beneficial, there is considerable potential for harm, particularly through drug interactions, adverse drug events and non-adherence.1 Such harms are amplified in people who are frail and who may require interventions to be tailored to their individual needs rather than strictly following guidance designed to manage single diseases. It is important to develop an approach that allows patients to make informed decisions and prioritise medicines for continuation or discontinuation, in order to maximise benefit and minimise harm.1

The term ‘deprescribing’ has been suggested in recognition that the skills utilised in stopping medicines need to be as sophisticated as those used when initiating drug treatment.3 Key to deprescribing, as with all medical interventions, is the active participation of the patient to ensure that their preferences and choices are taken into account. Particular care is needed when end-of-life considerations apply, so that treatment is optimised and the burden of taking medicines is minimised.4 Although evidence is sparse, this article provides some practical observations on deprescribing.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

About frailty

Frailty is recognised as a distinct health state, usually related to the ageing process, in which multiple body systems gradually lose their in-built homeostatic or repair mechanisms.5 Drug handling in older people is altered and the metabolism and excretion of medicines is decreased. As a result, older people are more sensitive to medicines and this effect increases with frailty. In addition, people who are frail are at particular risk of adverse outcomes, such as worsening of physical and mental wellbeing, after an apparently minor event (e.g. urinary or respiratory tract infections, or the introduction of new medication). Frail individuals are also at greater risk of injury through falls, resulting in institutional care, hospitalisation or death.5,6 Research suggests that there is an association between a patient's level of frailty, the inappropriateness of their medication and the likelihood of developing adverse drug reactions.7 While frailty is strongly correlated with age, not all older people are frail; around 10% of people aged over 65 years have frailty, rising to 25–50% of those aged over 85 years.5,6

The British Geriatric Society has published guidance that highlights several measures that can be used to identify frailty.6 These include: walking or gait speed (taking more than 5 seconds to walk four metres); the ‘get up and go’ test (taking more than 10 seconds to stand up and walk three metres from a chair and sit back down again); assessment of polypharmacy (taking five or more medicines); and, the PRISMA 7 questionnaire (see Box 1).8 Furthermore, the guidance recommends conducting a personalised medication review for older people with frailty, taking into account the number and type of medicines.

Box 1

PRISMA 7 questions8

  • Are you more than 85 years old?

  • Are you male?

  • In general do you have any health problems that require you to limit your activities?

  • Do you need someone to help you on a regular basis?

  • In general, do you have any health problems that require you to stay at home?

  • If you need help, can you count on someone close to you?

  • Do you regularly use a stick, walker or wheelchair to get about?

A score of ≥3 positive responses suggests the need for further clinical review.

Many drugs that are particularly problematic for frailty (e.g. NSAIDs, hypnotics, antimuscarinics and opioids) are commonly used in older people. However, because of concerns about frailty, some drugs, which with careful monitoring would be safe to use, are sometimes inappropriately omitted (e.g. ACE inhibitors in heart failure).6

About polypharmacy

The greater the number of medicines prescribed, the greater potential for harm through drug interactions, adverse drug reactions and non-adherence;1 such risks will be accentuated in frail people.7 Any use of over-the-counter drugs, herbal products, vitamins and mineral supplements may compound the problem. In addition, there may be considerable potential for medication errors due to confusion over regimens, poor dexterity or problems with vision. It is increasingly recognised that the burden related to the time and effort required to organise complex medication regimens can have a significant impact on a patient's quality of life.9–11

In some cases, a threshold number of drugs has been used to define polypharmacy and to trigger a medication review.1 Guidance from the National Institute for Health and Care Excellence (NICE) uses the definitions of appropriate and problematic polypharmacy proposed in the King's Fund report on the subject of polypharmacy and medicines optimisation (see Box 2).1,12

Box 2


Appropriate polypharmacy

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.

Problematic polypharmacy

The prescribing of multiple medicines inappropriately, or where the intended benefits of the medicines are not realised.

Research into appropriate polypharmacy in older patients was the subject of a Cochrane review that highlighted the paucity of good-quality intervention studies.13 Twelve studies were included, but these were limited by small sample sizes and were of poor quality owing to risks of bias. Assessments were based on surrogate markers of appropriate polypharmacy and the clinical significance of these outcomes is unclear.13

The scale of multimorbidity and associated polypharmacy

The proportion of older people in the UK is increasing. Estimates from the Office for National Statistics predict that by 2040, the number of people aged 75 years or older will have increased by 89% to 10 million and the number of people aged 85 years or older will double to 3.6 million.14

In the UK, there is relatively little published information on the scale of multimorbidity and associated polypharmacy. Research from Scotland has shown that, in older people, multimorbidity is common and more than 70% of people had at least one condition by the time they reached the age of 60 years.2 Other studies from Scotland have demonstrated the rapid growth in prescribing over the last decade (a threefold increase in the number of patients receiving 10 or more items)15 and examined the relationship between age and multimorbidity.16 In one study, 17% of the adults assessed received four to nine medicines, and 5% per cent received 10 or more.16 The number of prescribed items increased with age and was directly related to the number of morbidities. In patients with six or more comorbidities, 42% received 10 or more medicines.16

‘Monomorbidity’ in clinical trials and guidelines

Most clinical trials focus on single interventions in people who have been chosen because they are not frail or old and have relatively few comorbidities. However, this is not representative of routine clinical practice and the impact of combining multiple interventions is rarely examined. Clinical guidelines are largely written as though patients have a single condition and the impact of treatment recommendations from multiple clinical guidelines is not generally considered.17 In people with several conditions, application of recommendations from multiple single-disease clinical guidelines can result in complex and problematic polypharmacy.17,18 For patients in whom polypharmacy is causing problems, it is important that the most important interventions are identified and prioritised.

About deprescribing

The authors of a recent systematic review concluded that there was a lack of consensus on defining deprescribing but suggested a working definition: “Deprescribing is the process of withdrawal of an inappropriate medication, supervised by a healthcare professional with the goal of managing polypharmacy and improving outcomes.”19

Deprescribing does not mean stopping all medicines but refers to reducing the dose or discontinuing drugs that may be causing harm, may no longer be providing benefit or may be considered inappropriate for other reasons (e.g. interacting with more essential treatment). The aim of deprescribing should be to reduce medication burden and harm, while maintaining or improving quality of life. This process may be more challenging than that involved with initiating drug treatment.

Medication review has been described as, “a structured, critical examination of a person's medicines with the objective of reaching an agreement with the person about their treatment, optimising the impact of medicines, minimising the number of medication-related problems and reducing waste”.12 During these reviews, patients (or their advocates) should be given appropriate information about the harms, benefits and goals of such treatment so that they can be actively involved in the decision-making process. The reviews should include:

  • identification of the patient's priorities,

  • discussion of the acceptability of treatment and how it relates to the patient's beliefs and expectations, and

  • the option of stopping treatments.12

A discussion about stopping preventive chronic disease medication should highlight the potential impact on the anticipated long-term outcomes for the individual.6

Evidence to guide deprescribing

Clinicians are taught to prescribe but there has been limited training to help make decisions about stopping medicines and there is relatively little evidence to support such decisions.20,21 Prescribers may feel uncomfortable with stopping medicines, concerned by perceptions about denying people medicines. Some medicines are started by specialists and the prescriber may not wish to contradict 'expert' advice.21

Reasons for withdrawing a medicine include adverse drug reactions and a lack of clinical response.22 Drugs may need to be reviewed in the light of restrictions imposed by the licensing authority because of emerging safety concerns. A change in the circumstances of a patient or their disease state may change the risk-benefit profile of certain medicines unfavourably, and in frail people (whose condition can change rapidly) this should always be considered.20 A change in a patient's condition may mean that a drug that they have taken for many years can become problematic for them, and may even result in the patient falling outside the terms of the drug's product licence. New evidence and changing guidelines may also affect the desirability of using a particular medicine.22

Resources on polypharmacy produced for clinicians in Wales and Scotland provide guidance to support deprescribing and have also focused on the needs of frail people.23,24 To aid prioritisation of treatments, these provide tables describing estimates of absolute benefit expressed in terms of number-needed-to-treat (NNT). The availability of information on absolute risk using number-needed-to-harm (NNH) is more limited. However, it should be noted that estimates of NNTs and NNHs are largely derived from individual clinical trials that are often too heterogeneous to aggregate. Also, in the context of frail, older people who are not well represented in trials, they are likely to overestimate benefit and underestimate risk. Despite these limitations, NNTs and NNHs can give crude estimates of benefit and harm and can be used to illustrate for an individual patient the time over which the benefit or harm is expected to occur. For example, the benefits associated with many preventive interventions may not be seen for several years. Some researchers have suggested using an estimate of the likelihood of ‘time until benefit’ when discussing options with older people.25

Other tools that can be used to aid decision making are Screening Tool of Older Person's Prescriptions (STOPP) and Screening Tool to Alert doctors to Right Treatment (START), which have been designed for use in people over the age of 65 years.26 Last updated in 2014, these tools were developed by consensus. They represent indicators of potentially inappropriate prescribing, rather than focused purely on safety. Furthermore, START highlights risks of under-prescribing (e.g. not prescribing anticoagulation to older patients with atrial fibrillation who are at risk of stroke).26

Patient decision aids (PDAs) can also be used to help support decision-making in prescribing. However, optimal use of PDAs has not been established and requires further research and development.27

End of life considerations

In some instances, the deprescribing process may be clinically and ethically challenging. As patients move toward the end of their lives, the need for long-term preventive treatments (e.g. statins for cardiovascular disease, drugs to manage osteoporosis) may become less relevant. Deprescribing is an important intervention when a patient's life expectancy is limited, and drugs that are intended to prevent events over several years may be less relevant. This requires a sensitive approach and careful discussion with the patient, their relatives and carers. It is necessary to establish whether they wish to continue with all their medicines or would prefer to limit the number of drugs that they take.4 The transition from prevention towards a palliative approach is not easy. The UK Gold Standards Framework Centre and Royal College of General Practitioners' prognostic indicator guidance for people nearing the end of life proposes a system for judging the potential for continued benefit from treatment.28 One of the triggers that may suggest that a patient is nearing the end of the life involves addressing the question,4,28 ‘Would you be surprised if this patient were to die in the next few months, weeks, days?’ If the answer is ‘no’, this could lead to a discussion with the patient and carers to review the goals of care, revise treatment options and limit further investigations, unless needed for palliative care.28 For patients (and possibly their carers) who have frequent blood tests, have to attend the surgery or hospital regularly and take a complicated drug regimen, this may make life easier.4

Initiating deprescribing

The deprescribing process should be considered when treatment is first initiated. All new prescriptions should be regarded as a trial of therapy and treatment goals, review dates and criteria for stopping a drug agreed with the patient; the benefits and adverse effects should be reviewed before further prescriptions are issued.29 In general, drugs should not be designated as ‘repeat prescriptions’ that can be obtained without seeing the prescriber until these steps have occurred and stable use is established. Repeat prescriptions may perpetuate long-term prescribing, and a recent study in an urban UK setting suggested that as many as 77% of all prescription items were issued in this way.30 Concern has been raised over a lack of relevant agreed processes and standard operating procedures for repeat prescribing.31 All GP practices should have robust systems for managing repeat prescribing in place and should audit them regularly.

Discontinuation reactions

It is important to take into account the possibility of discontinuation reactions. Care is needed with some drugs as abrupt withdrawal may cause serious problems in frail people. Withdrawal reactions are fairly well recognised with benzodiazepines, opioids and antidepressants but may be less well recognised with other drugs (e.g. beta-blockers).22

In some instances, the dose should be gradually reduced before the drug is stopped.22,29 Where more than one drug needs to be discontinued, it is sensible to stop one at a time. Patients and carers should be warned about potential discontinuation reactions and given advice on how long these may last and how they can be managed. Deprescribing needs to be planned carefully with follow-up appointments to identify adverse consequences. Some practical considerations to guide deprescribing are provided online (

Organisational issues

A major challenge for UK general practice is that the processes of prescribing and deprescribing for multimorbidity and polypharmacy (particularly in the context of frailty) requires dedicated time spent with patients and is augmented by maintaining continuity of care through a named clinician.32 It may also require a different set of skills and identification of learning needs.2 Multidisciplinary assessment and care planning is needed to support the process and the enhanced role of appropriately trained clinical pharmacists could be valuable for this work.33


Polypharmacy is common and may be appropriate, but for many people can be problematic. Frailty means that a person's response to an adverse event is poor and recovery is impaired, so that any harms of medication use are accentuated. The need for medication changes over time and requires regular reassessment to identify when continued prescribing may be unnecessary or has become unsafe. Care is required to identify the treatments that are likely to provide the outcomes most valued by the patient, recognising that these may differ from the prescriber's viewpoint. When stopping medication, consideration needs to be given to the likelihood of discontinuation reactions, and these may have more serious consequences in frail people.

Deprescribing requires careful counselling, and a shared decision needs to be reached with the patient and carers. This is a complex process, requiring a similar level of skill to that needed to prescribe a drug in the first place. Deprescribing should be considered as part of routine clinical care.

We believe that there needs to be a greater focus on collecting evidence, developing guidance and training healthcare professionals on how to manage multimorbidity, polypharmacy and deprescribing.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.