Article Text

Download PDFPDF

Medication review in hospitalised older people: what have we learnt?
  1. Nina Lee Barnett1,2,
  2. Lelly Oboh3,4
  1. 1 Pharmacy, London North West Healthcare NHS Trust, Middlesex, UK
  2. 2 Medicines Use and Safety, NHS Specialist Pharmacy Service, London, UK
  3. 3 Pharmacy, Guy's and St Thomas' Hospitals NHS Trust, London, UK
  4. 4 Medicines Use and Safety, NHS Specialist Pharmacy Services, London, UK
  1. Correspondence to Professor Nina Lee Barnett, Pharmacy, London North West University Healthcare NHS Trust, Middx, UK; nina.barnett{at}

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.

Person-centred care is considered essential to clinical practice by regulatory and professional bodies around the world. For example, in the UK, the General Medical Council, General Pharmaceutical Council, Care Quality Commission and National Institute for Health and Care Excellence strongly advocate for it. The paper by Thevelin and colleagues1, in this edition of BMJ Quality & Safety, offers us the opportunity to reflect on medication review from the perspectives of both patients and healthcare professionals. In addition, it challenges us to consider how we can do better for our patients using a patient-centred approach through making shared decisions about changes to their medicines.

The OPERAM trial2 evaluated the impact of a complex intervention of medication review with shared decision making on drug-related readmissions for older people with multi-morbidity and polypharmacy (>5 medicines) who were inpatients in four European countries (Belgium, Ireland, Switzerland and The Netherlands). The study demonstrated that these reviews, undertaken jointly by pharmacists and doctors, reduced inappropriate prescribing but had no significant effect on drug-related hospital admissions.

To improve our understanding of the contextual factors and mechanisms that influence the effectiveness of medication reviews, Thevelin and colleagues used a multi-centre mixed-methods study, embedded in the OPERAM trial, exploring the experiences of patients in relation to hospital-initiated medication changes in depth. The investigators performed semi-structured interviews with a purposive sample of 48 patients across the four countries, starting 12 months into the OPERAM trial. The interviews were undertaken within a month of discharge and included a subset of patients, from the OPERAM trial, who had medication changes during their hospital stay and were willing to share their experiences. The interviews covered relevant aspects of the UK NHS Patient Experience Framework3 and the OPERAM intervention components. As patients’ beliefs about their medicines can affect their acceptance of changes, patients’ beliefs about medicines were also assessed quantitatively at the end of the interview, using the Beliefs about Medicines Questionnaire4 to complement the qualitative interviews. For those in the intervention arm, the results from the shared decision making intervention were extracted to understand the patients’ experiences of shared decision making. A subsample of the doctors and pharmacists from the Belgian and Swiss sites, who had recommended the drug changes for the OPERAM trial, were asked to complete the physician version of the SDM Questionnaire (SDM-Q-DOC)5 to obtain their perspectives.

Thevelin and colleagues’ analysis reveals helpful insights about the importance of appropriate information provision and patient–clinician collaboration as part of medication review. What makes this study unusual is the extraction of the results of the shared decision making intervention with patients and the concomitant use of a shared decisionmaking tool5 (the 9-item Shared Decision-Making Questionnaire 3) by 17 clinicians in the Belgium and Switzerland sites. Findings suggest a clear difference in perspective between patients and clinicians on the use and effectiveness of shared decision making. The following five paragraphs address each of these subjects in turn: the role of trust between patient and clinician in effectiveness of medication review, information exchange, patient satisfaction, perspectives on shared decision making and patient attitudes to medication review.

Although the study was conducted in a hospital setting, some of the patients interviewed highlighted the importance of the long-term, trusted relationship with their general practitioner. Patients reported that trust in their GP was a facilitator to them agreeing and implementing the medication changes and raised the issue of balancing advice from different healthcare professionals to make a decision about which medicines they choose to take post-discharge. Lack of coordination of information for patients, leading to mixed messages about medicines, is very familiar to both of the authors of this editorial. Based on Thevelin’s findings, clinicians would therefore be wise to take notice of the importance of trust in relationships between themselves and their patients in relation to effective medication review. We are reminded of studies from almost 20 years ago, when early exploration of medication reviews6 in 2001 demonstrated the value of pharmacists undertaking medication review if embedded within the general practice team. The HOMER study7 that followed taught us that patients were less comfortable discussing their medication with a pharmacist they did not know, and who was not part of their general practice team.8 This alerts us to the possibility that patients may feel uncomfortable with medication changes in hospital as they are less likely to have a long-term, trusting relationship with their ward-based doctor. The findings of a recent review suggests that while one-off medication reviews may reduce the number of medicines, they have little effect on quality of life.9 These issues together may lead us to question the value of one-off hospital-based medication review on patient experience, adherence and outcomes.

The Thevelin study discusses a number of themes, one of which is the importance of information exchange. There are many tools to support effective information exchange between clinicians and patients (eg, Aqua’s ‘Ask Share Ask’,10 WHO’s ‘Five moments of medicines safety’ tool,11 the ‘Five Es’12) and processes to structure conversations,13 14 but how many of us use them in routine practice? In spite of the availability of such tools, how many patients actually receive prompts about questions they could ask us before they meet us for medication review?15 The Thevellin paper suggests that patients and clinicians need to prepare for collaborative medication review conversations in hospital, and hospital clinicians need to consider what information the patient wants, how and when they wish to receive it, and how to communicate with their primary care colleagues in a timely way. This is because a trusted patient–doctor relationship facilitates shared decision making to whatever degree the patient is willing to be involved.

The theme of cross-sector communication is also discussed by Thevelin and colleagues, underlining the need to provide appropriate information and consistent follow-up for medication review conversations after discharge from hospital. We suggest that post-discharge, healthcare professionals in primary care should receive a summary of content discussed in hospital and suggestions of what might be helpful to consider in their next conversation. This is preferable to solely providing recommendations for action, which may reinforce the historical challenges of hierarchy among sectors of care.

The patients in the OPERAM study appeared to have mixed views about their level of satisfaction with information. While some patients clearly received the information they needed in a way they felt comfortable with, others cited barriers including lack of understanding, lack of recall and lack of time. Being given information when feeling unwell was also cited as unhelpful. Patients felt that involvement of carers or loved ones was a facilitator to both being well informed and patient participation. Avoidance of all jargon when speaking to patients was highlighted as important. Added to this, the study describes how many patients ‘passively’ accept their role as recipients of care rather than being more active participants. This adds to the disempowerment that being in hospital can cause—the impact of wearing a hospital gown, not being in your own home, and not in control of your food and environment cannot be underestimated in reducing the feeling of overall control, including in managing medicines. Separately, some patients, keen to be discharged, may not wish to spend time discussing medicines while they are in hospital.

Two of the most interesting findings of Thevelin and colleagues’ study are the patients’ mixed views on paternalism in decision making and the clinicians’ views on shared decision making. The paper suggests that while most patients preferred to be ‘informed rather than involved’ others were dissatisfied with this and wanted to be more involved. The study’s results also suggest that clinicians thought they had facilitated shared decision making for 70% of interventions, but 77% of patients reported that the care they received was paternalistic. We need to understand this divergence of viewpoints. Thevelin and colleagues call this result ‘paradoxical’. How can it be that clinicians appear to believe that they are routinely practicing shared decision making and yet the majority of patients describe the care as paternalistic? The Care Quality Commission in the UK produced a report in 201616 identifying that there was no significant change between 2005 and 2014 in how involved patients felt in their care, despite almost universal consensus across health and social care, that patient involvement, as part of shared decision making, should be routine practice. Here, we have an interesting question: can paternalism and shared decision making coexist as part of delivering person-centred care? If so, how do we ensure that we offer patients the amount of involvement they want?

The core value of shared decision making is to conduct consultations in a person-centred way. We argue that person-centred care means that our role as healthcare providers is to ensure that patients have both choice and control around the extent of shared decision making. Paternalistic care is person-centred when the patient has told us that this is what they want. So perhaps the result of the OPERAM study is not paradoxical, rather it offers us two different viewpoints of the same thing.

Finally, the Thevellin study offers some insight and encouragement around patient attitudes to medication review and changes. Many patients wanted to take fewer medicines, and this is encouraging as it aligns with the multiple initiatives around reducing inappropriate polypharmacy and over prescribing (NHS England 2021,17 NHS Scotland and the Scottish Government 201813) and promotion of safe deprescribing.18 This finding concurs with a recent review19 that found older hospital inpatients were willing to stop medications if possible. The results of the OPERAM study suggest that we need to be mindful of the context of medication changes in relation to the patient’s situation: one example is given of a patient who felt that starting a proton pump inhibitor was minor in comparison with the cancer they had, reminding us how clinicians need to take care to avoid assumptions about what is perceived by patients as ‘major’ and ‘minor’. A good example of this is shown in a study of older people’s responses to ‘minor’ side effects20 where only 3% of older people would be prepared to take a medicine for primary prevention of cardiovascular disease if it negatively affected their functioning. Other Thevelin study results, from the structured interviews and the Beliefs about Medicines Questionnaire, showed that the vast majority of patients were motivated to take medicines, their necessity beliefs outweighing their concerns, with over two-thirds ‘accepting’ (high necessity low concerns). This concurs with views of older people after discharge.21

The Thevelin study reminds us to adapt information for patients according to what they tell us they need and to work with our patients so that both clinicians and patients are better prepared to undertake medication review. We are also tasked with working to resolve the challenge of effective communication about medication review across care settings: we suggest that one-off hospital medication review events are likely to be much more effective if secondary and primary care are more seamlessly linked, to create a process of ongoing medication review.

Ethics statements

Patient consent for publication

Ethics approval

Not applicable.



  • Twitter @NinaLBarnett

  • Contributors NLB wrote the first draft. LO contributed to further drafts, editing and finalising.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Linked Articles