Introduction
In the United Kingdom (UK) the proportion of people over the age of 65 is growing and it is estimated to reach nearly a quarter of the population by 2035.1 Prescribing for older people can be challenging because evidence is usually based on trials conducted in a younger population and age-related deterioration makes them more susceptible to adverse drug reactions.Reference Mallet, Spinewine and Huang2 Additionally, as time progresses, therapeutic momentum builds and patients are prescribed more medicines as they experience more ailments. Approximately 20% of people over 70 take five or more medicinesReference Rollason and Vogt3 and the average number of prescription items for people over 60 is now 40.8 for each person per year,4 with 60% of all medicines prescribed in England being for those people over the age of 60.5
The taking of multiple medicines (polypharmacy) has been linked to an increase in hospital admissions, morbidity and mortality in this group of patientsReference Hajjar, Cafiero and Hanlon6–8 and is a significant problem in the UK National Health Service (NHS). This may largely be due to an increase in the potential for adverse drug reactions, prescribing errors, poor adherence and insufficient monitoring,Reference Howard and Avery9, Reference Pirmohamed, James, Meakin, Green, Scott and Walley10 when multiple medicines are co-prescribed. One proposed method of addressing concerns with polypharmacy is by a process of medication review. Medication review is a structured, critical examination of a patient's medicines, with the objective of reaching an agreement with the patient about treatments, optimizing the impact of medicines, and minimizing the number of medicine-related problems. Medication reviews were formally recommended in the 2001 National Service Framework (NSF) for older people11 and the 2005 NSF for long-term conditions12 and since then they have been firmly established in NHS contracts for community pharmacists and general practitioners in the UK.13, 14 Internationally, medication reviews are included as a component of the collaborative home medication review (HMR) service in Australia and Medication Therapy Management (MTM) services in the USA.
With widespread interpretation and implementation of what constitutes a medication review the National Prescribing Centre defined three levels of medication review:Reference Clyne, Blenkinsopp and Seal15
• Prescription review – technical review of a patient's medicines without patient present;
• Compliance and concordance review – exploration of patient's medicine taking and belief about medicines;
• Clinical medication review – treatment considered in light of patient's conditions with full notes and patient present.
In practice it can be difficult for General Practitioners (GPs) to review medication regularly and alternative approaches have been tested, with many involving the use of the pharmacist,Reference Royal, Smeaton, Avery, Hurwitz and Sheikh16 as advocated by the NSF for long-term conditions.12 Pharmacist training in clinical skills (e.g. knowledge of therapeutics and counselling) and technical skills (e.g. supply and product knowledge) suggest that they are uniquely and appropriately trained to provide medication reviews.Reference Petty, Zermansky and Raynor17 While it is possible for pharmacists to detect and resolve drug-related problems through medication review,Reference Krska, Cromarty, Arris, Jamieson, Hansford and Duffus18, Reference Lau and Dolovich19 systematic reviews prior to 2008 have failed to identify their effect on hospital admission and mortality in older people.Reference Royal, Smeaton, Avery, Hurwitz and Sheikh16, Reference Holland, Desborough, Goodyer, Hall, Wright and Loke20
In 2010, the UK Government introduced wide-ranging reforms of the NHS that focused on implementing changes in the way services were commissioned.21 These changes became a reality in 2013 and now mean that ‘any qualified provider’ can supply services to the NHS.
There is no doubt that medication review is a key component of the care of patients prescribed multiple medications. It can be used to identify sub-optimal adherence, reduce medicines wastage, optimize drug therapy and ensure appropriate monitoring. With the changing NHS landscape being implemented primarily to achieve efficiency savings it may be appropriate to re-examine the evidence of the role of the pharmacist in providing a service centred on medication review. Therefore, this paper seeks to update and discuss the available evidence for pharmacist-led medication reviews in primary care and how these can be operationalized to maximize patient benefit in the post-2013 NHS structure. Whilst this review will focus on research conducted in a UK setting, it will also consider the international perspective.
Search methods
A literature search was performed, covering date of inception to April 2013, in Embase, MEDLINE, PsychINFO, AMED and Web of Science. Interventional studies of any methodological design in English language were included that focused on medication reviews conducted by pharmacists in primary care on patients with an average age over 60 years. This included community pharmacy, medical practices, patient's home or long-term care settings. All studies conducted in secondary care were excluded. The authors also considered reference lists from identified reviews to complete this narrative review.
The evidence for pharmacist-led medication reviews in primary care
There are a variety of settings in which medication review can occur in primary care: general practice, care homes and the community pharmacy. Each of these settings will be considered in turn.
General practice
The first randomized controlled trial (RCT) of pharmacist medication reviews in the UK was published by Mackie et al. in 1999.Reference Mackie, Lawson, Campbell, Maclaren and Waigh22 The objective of this study was to compare pharmacist medication reviews in general practice to usual care. Patients were randomized to intervention or control groups; all patients were seen and drug-related issues identified, but recommendations were only made in the intervention group. The GP agreed with recommendations in 84% of cases in the intervention group and after 6–12 months 87% of issues identified were resolved, but only 34% of issues in the control group (with no recommendations made) were resolved.
A similar RCT that measured clinical outcomes with a 1-year follow-up was conducted by Zermansky et al.,Reference Zermansky, Petty, Raynor, Freemantle, Vail and Lowe23 who tested whether one highly trained pharmacist could review repeat medication in clinics at four general practices. Invited patients were 65 years or over on at least one repeat medication living in the community. After one year, medication reviews were conducted for 97% of the intervention patients compared with 44% of controls, and there were significantly more changes per patient reviewed in the intervention group, 2.2 versus 1.9 (P = 0.02), respectively. Although medication costs rose in both groups, the rise in the intervention group was lower. There was no evidence that the pharmacist's review affected consultation rates, out-patient appointments, hospital admissions or mortality. This study provided evidence that a pharmacist can take over the review of medication for the elderly in the UK general practice setting and in its first year has the potential to save more money than the cost of the pharmacist. However, it is limited by only involving one pharmacist in the review process.
A similar study was published in 2001 by Krska et al.Reference Krska, Cromarty, Arris, Jamieson, Hansford and Duffus18 and recruited 322 patients aged over 65 years with a least two chronic conditions and four regularly prescribed medicines. All patients were interviewed in their own home and their medical records examined to identify pharmaceutical care issues. This process demonstrated that only around 50% of pharmaceutical care issues could be identified from prescriptions, 25% from medical records and 30% from patient interview. GPs agreed with 96% of the issues identified and accepted 90% of the recommendations to resolve them. After three months, 70% of identified issues had been resolved in the intervention group and only 14% in the control group. This study additionally reported no differences in medication costs, quality of life or hospital admissions, although intervention patients did have more contact with primary care professionals usually to resolve some of the identified issues. These studies demonstrated a pharmacists’ ability to identify pharmaceutical issues and work in close contact with GPs to resolve them; however, the short follow-up was not able to demonstrate any clinical or cost benefits. The high acceptance and implementation rate demonstrates that GPs feel these recommendations were worthwhile for the patient.
Holland et al. attempted to overcome some of the limitations of previous medication review studies in the UK by using a large number of pharmacists to determine whether home-based medication review by pharmacists post-discharge affected hospital admissions (the HOMER trial).Reference Holland, Lenaghan, Harvey, Smith, Shepstone and Lipp24 The study recruited 872 patients aged over 80 years prescribed two or more drugs on discharge who were returning to their own home or warden-controlled accommodation. At six months follow-up there was a 30% increase in readmissions in the intervention group. The study found no significant differences in mortality, quality of life or nursing home admissions. The authors proposed explanations for the counterintuitive findings that focused on the review increasing knowledge and adherence, which resulted in iatrogenic disease and health-seeking behaviour. The design of the trial meant that pharmacists unknown to the patient visited them rapidly after discharge, and that most likely resulted in confusion and health-seeking behaviour. It is important to note that while the primary outcome of hospitalization was counterintuitive, there was a non-significant decrease in mortality in the intervention arm.
Two further studies from the authors of the HOMER trial have evaluated the effects of pharmacist medication reviews on hospital admissions.Reference Holland, Brooksby, Lenaghan, Ashton, Hay and Smith25, Reference Lenaghan, Holland and Brooks26 The first study used a similar design to the HOMER trial but focused on patients over 18 years discharged with heart failure as an ongoing clinical condition.Reference Holland, Brooksby, Lenaghan, Ashton, Hay and Smith25 The HeartMed trial included some intermediate outcomes measures such as self-reported adherence and satisfaction not included in the HOMER trail. There were no significant differences in the outcomes, although 74% of patients reported to find the visits extremely or very useful. This study had similar limitations to the HOMER trial: only conducting a prescription medication review, the pharmacists may not have had sufficient experience or training to deliver the intervention appropriately and only 50% of recommendations were enacted by prescribers.
The POLYMED trial was a small RCT at one medical practice in Norfolk using one experienced medication review pharmacist.Reference Lenaghan, Holland and Brooks26 Recruited patients (n = 136) were over 80 years old, living in their own homes and were prescribed at least four medications with at least one other risk factor for medication misadventure. For intervention patients the pharmacist had copies of their medical history and current prescribed medication to aid the medication review process at the patient's home. Following visits, regular meetings were held with the medical practice lead GP and recommendations were discussed; control patients received standard care. After six months follow-up there was no significant difference in hospital admissions, mortality or quality of life, but the mean number of prescribed medicines had fallen in the intervention group (P = 0.03). Although this study did not report the proportion of recommendations enacted, it does concur with previous research which has demonstrated a reduction in prescribed medication when a specialist pharmacist in close contact with prescribers conducts medication reviews.Reference Zermansky, Petty, Raynor, Freemantle, Vail and Lowe23
In a final study from the UK, an enhanced pharmacy review of pain medication was performed by an experienced community pharmacist in general practice with access to the patients’ medical records and ability to make changes to therapy according to an agreed protocol.Reference Hay, Foster, Thomas, Peat, Phelan and Yates27 This study allowed the patient and pharmacist to have a consultation up to six times over a 10-week period and focused on their risk of adverse events from non-steroidal anti-inflammatory drugs (NSAIDs). At the end of the 12 month follow-up there was no difference between the control and intervention groups in terms of pain, but there was a significant reduction in NSAID use and increase in simple analgesic use in the intervention group. This study supports the role of the pharmacists as a supplementary prescriber and demonstrates their potential to reduce potentially inappropriate prescriptions.
Internationally, there were six papers identified as part of the review. Most of these involved the pharmacist identifying drug-related problems from a combination of the medical records and patient interviews.Reference Laucka, Webster and Kuch28–Reference Sellors, Kaczorowski, Sellors, Dolovich, Woodward and Willan31 Two of these studies involved face-to-face discussion with the primary care team regarding implementation of recommendations,Reference Leikola, Virolainen, Tuomainen, Tuominen and Airaksinen29, Reference Sellors, Kaczorowski, Sellors, Dolovich, Woodward and Willan31 and one study relied on the nurse referring patients to the pharmacist after symptom evaluation.Reference Milos, Rekman, Bondesson, Eriksson, Jakobsson and Westerlund30 A final study addressed two major areas of care with the patient in a 30–40 minute consultation: health education and drug therapy.Reference de Lyra, Kheir, Abriata, da Rocha, dos Santos and Pela32 This was conducted by a single researcher pharmacist and they met with each patient on a monthly basis.
None of these studies used hospital admissions or mortality as an outcome measure, but they did chart the number of recommendations made and implemented as a result of the pharmacist intervention. Pharmacists recommended a significant number of changes to patient's therapy and implementation by the physician ranged from 41 to 86%. Pharmacist interventions also reduced the number of prescribed itemsReference Laucka, Webster and Kuch28, Reference Milos, Rekman, Bondesson, Eriksson, Jakobsson and Westerlund30 and no significant difference in health-related quality of life was detected when compared with a control group.Reference Sellors, Kaczorowski, Sellors, Dolovich, Woodward and Willan31
A significant proportion of the research into pharmacist-led medication reviews in primary care has occurred with pharmacists working collaboratively with general practice. These studies demonstrate that pharmacists can review patients’ medication, identify and resolve pharmaceutical care issues. These reviews do not have an adverse effect on patients and are viewed positively by those who receive them. They are also capable of reducing prescription costs. Studies which used less experienced pharmacists with more remote relationships to the patient and GP have had less positive outcomes.
Care homes
Pharmacist-led medication reviews in care homes have demonstrated the ability to reduce the number of prescribed medicines, and have not detected significant changes in morbidity or mortality.Reference Furniss, Burns, Craig, Scobie, Cooke and Faragher33, Reference Roberts, Stokes, King, Lynne, Purdie and Glasziou34 Zermansky et al. performed an RCT on 661 care home residents.Reference Zermansky, Alldred, Petty, Raynor, Freemantle and Eastaugh35 One clinical pharmacist performed the medication reviews and made recommendations to the GP. After six months follow-up the number of medication changes in the intervention group had significantly increased and the number of falls was reduced, and no other effects on healthcare resource use were detected.
In the US when the MTM service has been provided in assisted living facilities it resulted in interventions that covered the cost of providing the service.Reference Maack, Miller, Johnson and Dewey36 In Norway, Kersten et al. conducted medication reviews in 22 nursing homes focusing on reducing anti-cholinergic drug burden using an anti-cholinergic drug scale (ADS).Reference Kersten, Molden, Tolo, Skovlund, Engedal and Wyller37 The pharmacist conducted a paper-based review with a view to reduce ADS scores and then discussed this with the doctor before changes were implemented. While ADS scores were reduced in the intervention arm this did not result in improved cognitive function or anti-cholinergic side-effects for the residents. This study further supports pharmacists improving the appropriateness of medication for residents in care homes through medication review.
Nishtala et al. retrospectively analysed medication reviews provided to care home residents as part of the nationally remunerated Residential Medication Management Review (RMMR) service in Australia.Reference Nishtala, McLachlan, Bell and Chen38 The aim was to investigate the number and nature of pharmacist-identified drug-related problems (DRP) and their acceptance and implementation by GPs. While their acceptance (72%) and implementation rate (58%) was similar to other studies, further analysis identified that GP acceptance of recommendations was dependent on disease category. Recommendations linked to mental health were less likely to be implemented and the authors suggest that GPs may need more specialist support to be willing to implement recommendations in this patient population.
In care home residents the patient population is often more frail and has more complex pharmaceutical needs than a general primary care population. Consequently, outcomes from research demonstrate that experienced pharmacists can perform medication reviews, which have the potential to improve the quality of prescribing and may improve clinical outcomes (falls). When the pharmacist is more distanced from the prescriber, in the case of the Australian model, recommendations in mental health patients may be less likely to be accepted.
Community pharmacy
Most of the research thus far has centred on the role of the primary care pharmacist in medication review. However, in the UK the community pharmacist also plays a role in the review of a patient's prescription. Medicines Use Reviews (MURs) are designed to give pharmacists the opportunity to address physical barriers to adherence, provide memory cues if necessary, simplify regimens and address patient beliefs about medicines.14 A nationally determined fee is paid for these reviews, which are funded by an overall reduction in reimbursement for community pharmacists. These reviews are not full clinical medication reviews with access to patient notes, but would more likely to be considered a compliance review. MURs, however, are not performed in isolation, and many GPs will be providing paper-based reviews to these patients to reach their NHS targets.Reference Celino, Levenson and Dhalla39 If MURs contribute to identifying patient issues, concerns and expectations they may provide the additional face-to-face component needed for a clinical medication review to help achieve the desired goal of improved patient health.
The current evidence behind MURs is limited, with some studies demonstrating that it can have a positive impact on medicines knowledge, patient confidence and an improvement in disease outcomes,Reference Colquhoun40–Reference Youssef, Hussain and Upton45 as well as providing a mechanism to refer patients back to the doctor if necessary.Reference Colquhoun40, Reference Cree46 However, the majority of published studies demonstrate a weak study design, rarely conducting a follow-up appointment or appropriate clinical measurements to monitor particular conditions. These studies rely on the self-reporting of outcomes by the pharmacist or satisfaction surveys given to patients after they have experienced the MUR. There has been a call from the medical profession to improve the evidence base behind these services to support their continued use within the NHSReference Richardson and Pollock47 and general practitioners have highlighted them as frustrating and potentially leading to a duplication of work.Reference Bryant, Coster, Gamble and McCormick48, Reference Wilcock and Harding49
Despite this lack of robust evidence for community pharmacy medication use review services, the UK Government has repeatedly stated that it wants greater pharmacist involvement in the management of patients with long-term conditions.21, 50–52 As a result, a number of research papers have been published that examine a more extensive medication review by community pharmacists.
The MEDMAN study, was conducted in patients with coronary heart disease from a large number of community pharmacies across England.53 The intervention involved an initial consultation with the community pharmacist (with patient medical data) to assess therapy, adherence, lifestyle factors and social support. Further consultations were provided according to pharmacist-determined patient need. The concurrent implementation of coronary heart disease targets in the new General Medical Services contract may have reduced pharmacist's capacity to improve therapy. Consequently after 12 months no significant differences in appropriateness of treatment (according to the NSF for coronary heart disease), use of healthcare resources (GP visits, hospital admissions or medicines) or self-reported adherence were identified. This study highlights the importance of professional collaboration to avoid unnecessary duplication of work when conducting medication reviews. The RESPECT study used the patient's own community pharmacist to provide medication reviews to 760 patients aged over 75 years on five or more medications.Reference Richmond, Morton, Cross, Chi Kei Wong, Russell and Philips54 The study used a randomized interrupted time series design in which five Primary Care Trusts (PCT), in a random order, introduced pharmacist-led medication reviews at quarterly intervals. Patients were followed for 36 months acting as their own controls with a 12-month intervention period. For the intervention period pharmacists met with patients monthly following the preparation of a pharmaceutical care plan. Over the year there was a small but non-significant increase in medication appropriateness with no significant effects on hospital admissions, primary care usage, mortality or quality of life. It would appear that pharmacists can positively influence prescribing, but that this does not translate into detectable outcome differences. This research group also published a cost-effectiveness analysis of their study and despite a lack of statistical significance in the original findings they determined that this type of service was probably cost-effective from an NHS perspective.Reference Bojke, Philips, Sculpher, Campion, Chrystyn, Coulton, Cross, Morton, Richmond, Farrin, Hill, Hilton, Miles, Russell and Wong55
Two further international studies have been identified. In New Zealand the community pharmacist had a consultation with the patient to address their concerns and expectations about their medicines and provide lifestyle and diet advice.Reference Bryant, Coster, Gamble and McCormick56 They also undertook a clinical assessment of the medications and made recommendations to the physician. This assessment led to the identification of 462 drug-related problems, of which 46% were fully and 16% partially implemented. This study recruited 498 patients and demonstrated a significant improvement in the medication appropriateness index (MAI) over 12 months in the intervention group.
In the Netherlands, community pharmacists were asked to review recommendations based on a computer screening programme of patient's medical records.Reference Denneboom, Dautzenberg, Grol and De Smet57 These recommendations were put to the GP either via a case-conference or a written report, and a comparison was made between the two approaches. The computer screening tool identified 62% of the recommendations put to GPs, with pharmacists identifying the remaining 38% on their own. Pharmacists in the case-conferencing group identified more recommendations that those randomized to the written reports and recommendation made via case-conferencing were more likely to be implemented than those from written reports. The additional cost of case-conferencing was offset by the increased recommendation implementation compared with the written reports group.
Medication reviews conducted from community pharmacies appear to have demonstrated less effectiveness than those in other settings. This is probably a result of less clinical information being available about the patient, and care not being provided in a joined up approach with other healthcare professionals. However, patients appear to like the interaction with community pharmacists and value the role they play in their care.
Discussion
This review has considered the growing body of evidence for pharmacist-led medication reviews in primary care. Studies have demonstrated their ability to identify and resolve drug-related problems and reduce inappropriate prescribing. While there have been many attempts to measure the impact of pharmacist-led medication reviews on outcomes such as hospital admissions and mortality, these outcomes may not be the most appropriate measure of effectiveness. Medication review should be a key component of the care of patients prescribed multiple medicines due to the need to ensure current therapy remains suitable.11, 12 Therefore, with the recent NHS re-organization, demonstrating efficiency savings and risk reduction through reducing inappropriate prescribing should be considered suitable measures of effectiveness. Importantly, pharmacist-led medication reviews do not have an adverse effect on patient-orientated outcomes and particularly in care homes, have the potential to improve them.
Conducting a full clinical medication review is a very complex intervention requiring a number of advanced skills. It is therefore unsurprising that the most successful studies have used a small number of highly trained or specialized pharmacists in close collaboration with the prescriber.Reference Zermansky, Petty, Raynor, Freemantle, Vail and Lowe23, Reference Lenaghan, Holland and Brooks26, Reference Hay, Foster, Thomas, Peat, Phelan and Yates27, Reference Zermansky, Alldred, Petty, Raynor, Freemantle and Eastaugh35 Many medication review studies have focused on process outcomes regarding the implementation rate of recommendations. Higher implementation rates are reported when pharmacists have full access to information and are able to communicate face-to-face with prescribers. Studies that have used a larger number of pharmacists and have had reduced collaboration with the prescribing team demonstrated less positive outcomes. To maximize the potential of these reviews on a larger scale, pharmacists may require further training in communicating with patients and prescribers. Improving negotiation and influencing skills when making recommendations with prescribers may increase the acceptance rate. Furthermore, conducting a fully multi-professional medication review as described by the CAREMED trial protocol may offer additional advantages,Reference Desborough, Houghton, Wood, Wright, Holland and Sach58 particularly in the most complex patients where full discussion of the case would be beneficial.
When communicating with patients, research which analysed the consultations in the HOMER trialReference Holland, Lenaghan, Harvey, Smith, Shepstone and Lipp59 identified weak communication skills by some pharmacists and some difficult interactions with patients.Reference Salter, Holland, Harvey and Henwood60 The study suggested that these interactions may have unsettled the status quo and raise doubts for the older patients in the trial, potentially leading to increased need for healthcare intervention.
An observational study of MURs also identified that pharmacists generally followed a rigid structure to an MUR, determined by the paperwork that needs to be completed.Reference Latif, Pollock and Boardman61 The observations revealed minimal open questions used by pharmacists and a focus on the pharmacist's agenda (medicines) rather than the patient's illness. This may be one of the contributing reasons for a lack of demonstrated effectiveness.
Therefore, with the appropriate communication skills and increased collaboration with the wider primary care team, there should be little reason why pharmacists cannot play an increased role in the care of this group of patients.
Conclusions
This review has highlighted some key considerations for designing pharmacist-led medication review services in primary care. With an increasing number of older patients on complex medication regimens the need for medication review has never been greater. Pharmacist-led medication reviews must be integrated into existing healthcare systems and the training of the pharmacist is essential to maximize the reduction in inappropriate prescribing. If pharmacists can develop their communication skills, further benefits may be realised.
Conflict of interest
Dr Desborough is the lead investigator on the CAREMED study.
