An official website of the United States government
NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
NICE Medicines and Prescribing Centre (UK). Medicines Optimisation: The Safe and Effective Use of Medicines to Enable the Best Possible Outcomes. Manchester: National Institute for Health and Care Excellence (NICE); 2015 Mar. (NICE Guideline, No. 5.)
Medication review, as an overarching term, has been considered to be an important intervention for many years. Medication review can have several different meanings. It could be a review of medicines carried out every day when a prescriber sees a patient and there is a decision to prescribe or stop a medicine, or a multidisciplinary medication review, with the patient (and their family members or carers where appropriate) present, using a comprehensive and structured approach supported by the patient's full medical records. Medication reviews are carried out in people of all ages.
In 2001, the National Service Framework for older people included a milestone stating that ‘all people over 75 years should normally have their medicines reviewed at least annually and those taking 4 or more medicines should have a review 6 monthly’. The National Service Framework (NSF) did not provide information as to what this medication review should entail or how it should be carried out. To support the implementation of this milestone, it was incorporated into the General Medical Services Quality and Outcomes Framework (QOF) from 2006 until 2013.
Room for review – a guide to medication review: the agenda for patients, practitioners and managers was published by the Taskforce on Medicines Partnership, The National Collaborative Medicines Management Services programme in 2002. This document suggested key principles for the process of medication review:
This document, along with a subsequent National Prescribing Centre document A guide to medication review (2008), aimed to clarify the different types of medication review. Box 1 summarises the different types of medication review, adapted from Room for Review (2002) and A guide to medication review (2008).
Different types of medication review.
In 2005, the National Service Framework for long-term conditions recommended in its quality requirement 1, that ‘people have timely, regular medication review’. Again, there were no specific details about what this medication review entailed.
The National Prescribing Centre issued A guide to medication review (2008), which gave further guidance for medication reviews to commissioners and providers of services. This document further emphasised the need to involve patients in discussions about their medicines and stated that patients ‘had a varied experience of the review and varied perceptions of its benefits’.
In 2013, the GMC issued updated guidance for Good practice in prescribing and managing medicines and devices. The updated guidance provides more detailed advice on how to comply with the principles outlined in the GMC document Good medical practice (2013). The guidance considers the process of reviewing medicines, and suggests that reviewing medicines is particularly important for patients who may be at risk, who are frail or who have multiple illnesses (which may increase polypharmacy). It is also important for those patients who are taking medicines with potentially serious or common side effects (including high-risk medicines) or those taking controlled drugs or other medicines that may be abused or misused. Reviewing medicines that require regular monitoring or blood tests is also important. The guidance also considers the role that pharmacists can play in medication review to ‘help improve safety, efficacy and adherence in medicines use, for example by advising patients about their medicines and carrying out medicines reviews’.
Over the years the process of medication review has evolved, but it is not a new approach. It can be carried out in a wide range of settings and the type and depth of medication review carried out varies. The type of health professionals who carry out medication reviews has also changed, with the changing medical model of prescribing, supplying and administering medicines (see section on medicines-related models of care). Regardless of which health professional is carrying out the medication review, there appears to be variation in how and when reviews are carried out.
Furthermore it is important to consider the cost of medication reviews to the NHS. Cost effectiveness analyses of medication review have been carried out over a number of years, often alongside RCTs or other clinical studies. Costing studies for medication reviews have been carried out in several different countries. Such analyses calculate the costs of performing medication reviews and some aim to determine the subsequent costs and health benefits of the intervention.
Medicines use reviews are different from medication reviews because the pharmacist carrying out the review does not have access to the patient's medical records. A medicines use review can complement a medication review. Information about medicines use reviews is available on the Royal Pharmaceutical Society website. In addition to medicines use reviews, community pharmacists also provide a new medicines service to support people with long-term conditions who are newly prescribed a medicine to improve adherence. Several other NICE guidelines recommend carrying out a medication review. Again in many of the guidelines details about how the medication review should be carried out, who should be involved and the overall process are not specified.
For the purpose of this guideline, when the term medication review is used, this is ‘a structured, critical examination of a person's medicines with the objective of reaching an agreement with the person about treatment, optimising the impact of medicines, minimising the number of medication-related problems and reducing waste’ (National Prescribing Centre 2008).
Medication reviews can be offered at different levels, by different health professionals working in different settings. Therefore the aim of this review question was to review whether the evidence for a full, structured medication review led to a reduction in suboptimal use of medicines and medicines-related safety incidents.
What is the effectiveness and cost effectiveness of medication reviews to reduce suboptimal use of medicines and medicines-related patient safety incidents, compared to usual care?
A systematic literature search was conducted (see appendix C.1), which identified 1565 references. The review protocols identified the same parameters for the review question on medication review and medicines reconciliation. Therefore a single search was carried out. After removing duplicates, the references were screened on their titles and abstracts and each included study was identified as being relevant for inclusion for the review question on medicines reconciliation or medication review. Sixty-five references were obtained and reviewed against the inclusion and exclusion criteria as described in the review protocol for medication review (appendix C2.4).
Overall, 52 studies were excluded because they did not meet the eligibility criteria. A list of excluded studies and reasons for their exclusion is provided in appendix C5.4.
Thirteen studies met the eligibility criteria and were included. In addition, 7 systematic reviews of studies (RCTs and observational) were identified. The references included in these systematic reviews were also screened on their titles and abstracts, to identify any further studies that met the eligibility criteria. Fifteen additional studies were included.
All the included 28 studies were RCTs investigating the effect of medication reviews compared with usual care (see appendix D1.4 evidence tables for details). The studies were carried out in an adult population; there were no studies identified that looked at medication reviews in children. Twelve of the studies targeted medication reviews for patients with long-term conditions such as hypertension, angina, asthma, hyperlipidaemia, diabetes, coronary heart disease, arthritis and osteoporosis. Twenty-one RCTs looked at pharmacist-led medication review, 6 RCTs looked at multidisciplinary team-led medication review and 1 RCT looked at physician-led medication review.
The studies were quality assessed using the NICE methodology checklists for RCTs (see NICE guidelines manual 2012). Appraisal of the quality of the study outcomes was carried out using GRADE.
See appendix D1.4 for evidence tables summarising included studies.
See appendix D2.4 for GRADE profiles.
There was some pooling of studies, although this was limited because the outcomes measures used differed and the follow-up periods reported varied between the studies.
Mean differences were calculated for continuous outcomes and odds ratios for binary outcomes, as well as the risk ratios for dichotomous data. When a meta-analysis was possible, a forest plot was presented (see appendix D2.4).
Summary of included studies.
A systematic literature search was undertaken (appendix C.1) to identify cost effectiveness studies comparing medication reviews to reduce the suboptimal use of medicines. This search identified 1507 results, of which 1480 records were excluded following screening of their titles and abstracts. The full papers of 27 studies were assessed for relevance against the inclusion and exclusion criteria. Sixteen studies did not meet the inclusion criteria, the reasons for which are listed in appendix C.6.4. A further 3 studies were identified as relevant from the clinical evidence review, resulting in a total of 14 studies meeting the inclusion criteria.
The studies meeting the inclusion criteria were quality assessed using the NICE quality assessment checklists for cost effectiveness studies. Following this, 4 studies were judged not applicable to the guidance because better quality studies (considering health-related quality of life), more relevant to a UK NHS setting, had been identified. An additional study (Burns et al. 2000) was judged by the GDG to be not relevant to the current UK NHS because of the age of the study. This study was a cost comparison study with a short time horizon and therefore evidence more relevant to the decision problem had already been included. These 5 studies judged not applicable to the guidance were excluded from further analysis. Three further studies were considered to have very serious limitations and were excluded on that basis.
Table 22 shows the economic evidence profile based on the 6 included studies. These studies were judged to be partially relevant to the guidance. The evidence was of variable quality: 4 of the 6 included studies were of low quality and 2 were of high quality. The economic evidence on medication review was limited to review by pharmacists in different settings with no analysis beyond a 12-month time horizon (appendix E1.4).
Economic evidence profile – medication reviews.
A health economic model was not pursued in this area because of the lack of data on medication reviews carried out by health professionals other than pharmacists. There is already an evidence base of published cost–utility studies relating to pharmacist review, so the GDG judged that a de novo model in this area would not help their recommendations. The limited data availability relating to other types of medication review meant that a robust economic model could not be constructed. To aid the GDG discussions a simple costing analysis was undertaken to compare the costs of medication reviews undertaken by a variety of health professionals. A summary of this is provided after table 23.
Estimate of cost per medication review delivered.
A summary of the simple cost analysis carried out for this area of the guidance is provided. See appendix F for a full report of the cost analysis carried out for this clinical guideline topic.
Simple costing calculations were carried out to provide the GDG with information around the cost per medication review undertaken dependent upon the health professional delivering the review. These are displayed in table 23. The length of time utilised for each medicine review was estimated by the GDG and various scenarios are displayed. Health professional costs were sourced from the Personal Social Services Research Unit (PSSRU) (PSSRU, 2013).
A variety of cost options are displayed, which include salary costs only, PSSRU unit cost per health professional and PSSRU unit cost per hour of health professional contact with patients, for consideration by the GDG. It is important to note that an NHS and PSS perspective should be taken for all NICE guidance (NICE, 2012). The costs provided in table 23 are limited in that they provide no information on the quality and impact of the review, nor the long term cost savings resulting from the review.
Low-quality evidence from 10 RCTs showed no significant difference in mortality between patients receiving medication review or usual care. The patient populations in the 10 RCTs were people with a mean age of 65 years or over, and who had medication reviews carried out by pharmacists (8 RCTs), a physician (1 RCT) or a clinical pharmacologist (1 RCT).
Low-quality evidence from 2 RCTs pooled together to show the effect of pharmacist-led medication review on targeted hypertensive patients showed that there was a significant change in mean blood systolic pressure in the intervention group compared with control group to meet the target blood pressure.
Low-quality evidence from 1 RCT showed that pharmacist-led medication reviews in a population with a mean age of 65 years significantly increased the percentage of intervention patients achieve target blood pressure compared with usual care. The same RCT showed a significant improvement in low-density lipoprotein (LDL) levels, reaching target INR with anticoagulation and improving diabetes control by meeting HbA1c targets with medication reviews.
Moderate- and low-quality evidence from 2 RCTs showed that medication reviews compared with usual care in a population with a mean age of 60 years reported no significant differences in the following clinical outcomes (as reported in the studies): proportion of patients prescribed cardiovascular medicines for secondary prevention, 5-year risk of cardiovascular death score, target lipid levels or reduction in LDL levels and changes in cardiovascular risk factors.
Low-quality evidence from 1 RCT showed that medication reviews in patients with asthma significantly improved the optimisation of asthma medicines, asthma severity and inhaler technique, but there was no improvement in spirometry results compared with usual care.
Moderate-quality evidence from 1 RCT showed that medication review significantly reduced the need for rescue medicine and night-time awakenings in patients with asthma. However, it did not improve the peak expiratory flow, asthma control test (ACT) scores or occurrence of severe exacerbations of asthma.
Moderate-quality evidence from a meta-analysis of 2 RCTs showed that medication reviews in a population with a mean age of 84 years significantly reduced the number of falls. One low-quality evidence RCT reported that medication reviews did not significantly reduce the number of fractures in the elderly. Another moderate-quality evidence RCT carried out in a population with a mean age of 68 years showed that medication reviews when compared with usual care significantly improved pain scores, pain severity score, arthritis self-efficacy pain scale scores and clinical response within the first 3 months of the intervention.
Low-quality evidence from 1 RCT showed that medication reviews did not significantly improve the reporting of adverse drug events compared with usual care.
Moderate-quality evidence from 1 RCT showed no significant differences in the number of adverse drug events per 1000 days between the groups that received medication review or usual care.
Moderate-quality evidence from 2 RCTs showed that medication reviews identified more medicines-related problems compared with usual care, but there was no significant difference between the 2 groups. In 1 RCT, medication reviews significantly reduced the mean number of potential medicines-related problems from baseline to endpoint compared with usual care. Another RCT identified medicines-related problems in 71% of patients who received usual care before having a medication review.
Moderate-quality evidence from 1 RCT in an elderly population showed that medication reviews significantly reduced the prescribing of inappropriate medicines compared with usual care.
Moderate-quality evidence from 1 RCT showed that medication reviews in patients with a mean age of 65 years reduced the Medication Appropriateness Index (MAI) scores in all 10 domains compared with usual care. However, the statistical significance of the results was not reported.
Low-quality evidence from 2 RCTs pooled together showed that medication reviews significantly reduced the MAI scores compared with usual care. One moderate-quality evidence RCT showed that medication reviews reduced the number of potentially inappropriate medicines per patient compared with usual care, but this was not significant. Moderate-quality evidence from 1 RCT showed that medication reviews identified more inappropriate medicines (as per Beers criteria) prescribed (although not significant) and also identified significantly more underused medicines (as per ACOVE criteria) compared with usual care.
Moderate-quality evidence from 1 RCT carried out in patients with dyslipidaemia showed that medication reviews reduced the number of patients being prescribed high-intensity statins compared with usual care. Moderate-quality evidence from 2 RCTs and low-quality evidence from 4 RCTs which reported on the mean number of medicines per patient for the study showed that there was no significant difference between the usual care and medication reviews after the follow-up period.
Moderate- and low-quality evidence from 2 RCTs showed that medication reviews significantly reduced the mean number of medicines prescribed compared with usual care. Moderate-quality evidence from 1 RCT reported a significantly smaller increase in the number of medicines prescribed to elderly patients who had medication reviews compared with usual care.
Low-quality evidence from 2 RCTs and moderate-quality evidence from 1 RCT showed that medication reviews significantly increased compliance compared with usual care. Moderate-quality evidence from 3 RCTs showed no significant difference in patient compliance between the groups that received medication reviews or those that received usual care.
Moderate-quality evidence from 2 RCTs and low-quality evidence from 1 RCT showed no significant difference in patient satisfaction between medication review and usual care. Moderate-quality evidence from 2 RCTs showed that medication reviews received significantly high patient satisfaction rates compared with usual care.
Moderate-quality evidence from 1 RCT showed no significant difference in the change in depression and anxiety scores for patients who received medication reviews or usual care. The same study showed that medication reviews significantly improved patients' clinical response to knee pain management (using the OMERACT-OARSI responder criteria) at 3 months compared with usual care. There was no significant difference at the end of the 12-month study.
Partially applicable evidence from 2 studies with minor limitations, built on RCT data, suggests that pharmacist review is not cost effective, with ICERs above £50,000/QALY, compared with no intervention.
Partially applicable evidence from 4 studies with potentially serious limitations provided conflicting evidence, with 3 studies finding pharmacist review to be cost incurring and 1 finding it to be cost saving compared with no intervention.
No evidence was identified informing cost effectiveness of medication reviews by health professionals other than pharmacists.
Linking evidence to recommendations (LETR).
Medication review can have several different interpretations and there are also different types which vary in their quality and effectiveness. Medication reviews are carried out in people of all ages. In this guideline medication review is defined as ‘a structured, critical examination of a person's medicines with the objective of reaching an agreement with the person about treatment, optimising the impact of medicines, minimising the number of medication-related problems and reducing waste’. See also recommendation 33.
Consider carrying out a structured medication review for some groups of people when a clear purpose for the review has been identified. These groups may include:
Organisations should determine locally the most appropriate health professional to carry out a structured medication review, based on their knowledge and skills, including all of the following:
The medication review may be led, for example, by a pharmacist or by an appropriate health professional who is part of a multidisciplinary team.
During a structured medication review, take into account:
To be read in conjunction with the NICE Research recommendations process and methods guide)
This review question looked at the clinical and cost effectiveness of medication reviews to reduce the suboptimal use of medicines and medicines-related patient safety incidents, compared to usual care or other interventions. The systematic review provided no evidence for medication reviews carried out in children.
Uncertainties may be related to:
The searches did not identify any randomised controlled trials (RCTs) looking at medication reviews carried out in children. Although RCTs have been carried out in an adult population, the GDG found the outcomes used to measure the effectiveness were mixed. The GDG agreed that, although the principles of medication reviews carried out in adults can be applied to children, further research is needed in children because of the different levels of engagement and because it often needs parent or carer involvement where appropriate when carrying out the medication review. RCTs may not have been carried out in children because of ethical aspects.
The GDG also discussed and agreed that there was uncertainty about the following factors, which may affect the clinical and cost effectiveness of medication reviews:
The GDG found that the above factors varied between the studies that were carried out in adults.
The key uncertainty is whether medication reviews can reduce the suboptimal use of medicines and medicines-related patient safety incidents compared with usual care or other interventions in children.
This uncertainty can be answered by conducting a study that will deliver good quality evidence, such as an RCT.
The research should be carried out in children that use services where medication reviews can be carried out.
Study methodology can be based on other well-conducted RCTs that have been carried out in adults, the difference being the age of the population. Approval from ethics or other committees would be needed given the young age of the population. ‘Usual care’ or other interventions would be used as a comparator. ‘Usual care’ would need to be defined in the study. A follow-up period of 1–2 years or more would capture longer-term outcomes. The outcomes for this research question should be patient-centred and include suboptimal use of medicines, medicines-related patient safety incidents, patient-reported outcomes, clinical outcomes, medicines-related problems, health and social care resource use and cost effectiveness.
The study would need to take into account:
The GDG recognised that the key focus of the medicines optimisation agenda is to make care person-centred. In line with this and to ensure the best use of NHS resources, the GDG agreed that research needs to be carried out in children to identify the benefit from them having medication reviews. There may be some longer-term gains with this approach, as from a young age the child would become more aware of the intervention, develop a relationship with the health professional and be encouraged to understand their medicines.
Research into this area will provide guidance to organisations who may want to, or already provide, medication reviews as part of their care and enable better use of resources (for example, health professional cost and time and health and social care resources). This information would be useful to commissioners who may consider whether or not to commission providers to carry out medication reviews.
Proposed format of research recommendations.
To be read in conjunction with the NICE Research recommendations process and methods guide.
This review question looked at the clinical and cost effectiveness of medication reviews to reduce the suboptimal use of medicines and medicines-related patient safety incidents, compared to usual care or other interventions. The GDG found that the systematic review provided no economic evidence for medication reviews carried out by health professionals other than hospital or community pharmacists.
Uncertainties may be related to:
Most of the clinical and economic evidence identified related to medication reviews carried out by community or hospital pharmacists. There were randomised controlled trials (RCTs) that looked at the clinical effectiveness of medication reviews carried out by a doctor and by multidisciplinary teams (comprising a doctor, pharmacist and/or nurse), but there was no economic evidence for these. Therefore, the GDG was uncertain about the cost effectiveness of professional-led (other than hospital or community pharmacists) and multidisciplinary team medication reviews as no relevant data were available to feed into an economic model.
In addition, the included studies varied in quality and were carried out in Europe, Australia, Canada or the USA. The GDG was aware of the limitations relating to the applicability of some of these studies to the UK setting given the differences in healthcare systems and processes and populations.
The GDG discussed and agreed that the clinical and cost effectiveness of medication reviews would depend on several factors such as:
Although the GDG was presented with a high number of RCTs for this review question, there was uncertainty about the above factors, which varied in the included studies.
The economic evidence presented to the GDG on medication review was limited to review by hospital and community pharmacists. The GDG found that the evidence was conflicting and of varying quality. The GDG was uncertain about the cost effectiveness of medication reviews carried out by multidisciplinary teams or professional-led by health professionals other than hospital or community pharmacists, compared to usual care or other interventions. No economic evidence was found for primary care pharmacists carrying out medication reviews. The GDG agreed that there was uncertainty about this and that the research recommendation should include primary care pharmacists.
These uncertainties can be answered by conducting a study that will deliver good-quality evidence, such as an RCT.
Is a medication review more clinically and cost effective at reducing the suboptimal use of medicines and improving patient-reported outcomes, compared with usual care or other intervention in the UK setting?
The study should consider the cost effectiveness of the health professional(s) carrying out the medication review.
The medication review should be carried out by a multidisciplinary team or be professional-led by any health professional other than a community or hospital pharmacist to provide data to develop an economic model for cost effectiveness. There is already economic evidence available for community and hospital pharmacists (see section 8.4).
Research can be carried out using an RCT. Study methodology can be based on other well-conducted RCTs that have been carried out looking at medication reviews. ‘Usual care’ or other interventions would be used as a comparator. ‘Usual care’ would need to be defined in the study. A follow-up period of 1–2 years or more would capture longer-term outcomes. Outcomes for this research question should be patient-centred and include the suboptimal use of medicines, patient-reported outcomes, clinical outcomes, medicines-related problems, health and social care resource use and cost effectiveness.
The study would need to take into account:
The GDG recognised that the key focus of the medicines optimisation agenda is to make care person-centred and to have services that support people in the optimal use of their medicines. Medication reviews can be offered to people by different health professionals at different levels, working in different settings. Resources (for example, staff and time) needed to enable routine medication review may vary locally depending on the setting and health professional availability.
Research into this area will provide guidance to organisations who may want to, or already provide, medication reviews as part of their care and enable better use of resources (for example, health professional cost and time and health and social care resources) and facilitate service delivery. This information would be useful to commissioners who may consider whether or not to commission medication reviews by providers.
Proposed format of research recommendations.
All rights reserved. This material may be freely reproduced for educational and not-for-profit purposes. No reproduction by or for commercial organisations, or for commercial purposes, is allowed without the express written permission of NICE.
Your browsing activity is empty.
Activity recording is turned off.
See more...
