CHARMER is testing a deprescribing ‘behaviour change intervention’ – what does this mean?

By Yan Ching Cheung (Tiffany), student Physiotherapist on placement with CHARMER

I’m a healthcare student on a research placement with CHARMER, which is a large national trial of a hospital deprescribing behaviour change intervention. In my six week placement, I am supporting CHARMER’s dissemination efforts. Dissemination is about promoting the project and sharing the findings so far. To enable me to effectively support dissemination, I first needed to understand exactly what a ‘behaviour change intervention’ was and why one was being used for deprescribing in the hospital setting.

What is the problem that the CHARMER behaviour change intervention is trying to solve?

CHARMER aims to ensure that older people (aged 65 years and over) who take several medicines have them all reviewed whenever they are admitted to hospital. This is so that any potentially harmful or inappropriate medicines can be stopped before they cause harm. This is called ‘proactive deprescribing’. Proactive deprescribing does not happen regularly in hospitals at the moment. Because of this, if we want it to start happening in hospitals, healthcare professionals need to change what they do – this is called ‘behaviour change’.

What is a ‘behaviour change intervention’?

Behaviour change interventions are packages of support that help people change their behaviour. In the case of CHARMER, the intervention is designed to support hospital doctors and pharmacists to change from not doing proactive deprescribing to doing it. Behaviour change interventions usually have 3-5 different parts to them. Each part is called a ‘behaviour change technique’ or BCT for short. BCTs are essentially strategies that are designed to support the person to overcome the things that hinder them from doing the behaviour, or make the most of what helps them to do the behaviour.

How are behaviour change interventions developed?

Behavioural scientists are experts in designing behaviour change interventions. They use behaviour change theory to help find out the things that help and hinder people doing the behaviour.

There are lots of possible behaviour change theories to pick from to help design a behaviour change intervention. The CHARMER behavioural scientists used the ‘Theoretical Domains Framework’ (or TDF for short) as their theory. The TDF is actually not just one theory but a combination of theories covering lots of things that might help or hinder people doing a behaviour. This is why the CHARMER behavioural scientists like the TDF! Examples of things the TDF covers are “skills” (whether or not a person has the skills to do the behaviour) and “beliefs about consequences” (a person’s belief about the benefits/disadvantages of doing the behaviour).

The CHARMER deprescribing behaviour change intervention

Using the TDF to structure our findings, we found one thing that helps and four main things that hindered pharmacists and hospital doctors from proactively stopping medicines. These are:

  1. Drawing attention to staff who are succeeding in stopping unnecessary or risky medicines (helps)
  2. Misconception by hospital doctors that patients and carers don’t want to stop their medicines (hinders)
  3. Pharmacists being unavailable when stopping decisions could be made (hinders)
  4. Pharmacists thinking that doing nothing is safer than stopping a medicine (hinders)
  5. Perception that stopping medicines is not a hospital priority (hinders)

We held workshops with hospital doctors, pharmacists, other hospital staff (e.g., Improvement Managers) and patient representatives to design the Behaviour Change Techniques (BCTs) to address these things. The BCTs were designed as:

  1. A hospital action plan that sets deprescribing as a priority for the hospital
  2. A workshop for pharmacists to help them overcome the belief that it is safer to do nothing than proactively deprescribe medicines
  3. A workshop for doctors to help them overcome the idea that patients and family don’t want deprescribing to happen in hospital, which we know is not true
  4. A regular meeting each week for pharmacists and doctors to discuss and plan deprescribing for patients under their care
  5. A weekly report giving pharmacists and doctors information on how much deprescribing they have done so far and how they compare to other hospitals.


You made it to the end of the blog! Thanks for reading. I hope this helps you understand more about behaviour change interventions and, most importantly, CHARMER.

We are now undertaking a trial to establish whether the CHARMER behaviour change intervention is safe, effective and good value for money. Find out more about this here.

You’re doing what? A placement student’s guide to CHARMER’s stepped-wedge cluster randomised trial design

By Abhya Kapoor, student Physiotherapist on placement with CHARMER

As a healthcare student on a research placement, it can be overwhelming to come into a large research project like CHARMER. During my first week, I was introduced to so many new terms all whilst trying to get my head around the scale of the project itself. The team has been excellent and incredibly supportive in explaining how the past four years’ worth of work has got us to where we are now –the CHARMER definitive trial. But I’ve also had to do a lot of reading to be able to understand the ins and outs of the trial to enable me to contribute productively to getting tasks done!

The purpose of the CHARMER trial is to test a new intervention designed for pharmacists and geriatricians (hospital doctors who work with older people). The intervention aims to ensure that people over the age of 65 who take several medicines have them reviewed whenever they are admitted to hospital and that any medicines that are unnecessary or that could be harmful are stopped. The trial started on the 1 February 2024 and will run until Summer 2025 across 24 NHS hospitals in England. The largest hospital trial of its kind internationally, it will involve around 100 hospital doctors and pharmacists across these hospitals, aiming to improve the care of 24,000 older inpatients.

So, the CHARMER trial seems quite simple at first. But then someone casually mentions that CHARMER is using a ‘stepped-wedge cluster randomised trial design’…! It can be difficult to know where to start unpacking this, so let me share my learning on what this actually means.

What is a stepped-wedge cluster randomised trial?

Stepped-wedge trials are a type of research design used in clinical trials or public health interventions (in this case CHARMER). This involves organising participants into clusters (aka groups). In the case of CHARMER, each step is a cluster of hospital sites. The clusters transition from the control phase (where participants do not receive the intervention) to the implementation phase (where participants receive the intervention) at different times. These timings form a step-like pattern and allow for a gradual roll-out of an intervention to evaluate its effectiveness over time.

A breakdown of how the process works:

1. Clusters (e.g., groups of participants) are identified for the study.

2. Clusters are randomly assigned the order in which they will transition from the control phase to the implementation phase.

3. The intervention is introduced sequentially to each cluster in the randomly assigned order.

4. Data are collected at each step, including baseline measurements during the control phase and subsequent measurements during the intervention phase.

5. Specific methods are used to analyse the data. These methods look for differences in the outcome measures between the intervention and control periods. Additionally, they account for patterns over time and any cluster-specific effects. This helps us make sense of the information in a way that takes these factors into account.

Why is CHARMER using a stepped-wedge design?

CHARMER has used this method to allow all trial hospitals to receive the intervention. Using a stepped-wedge design mirrors the real-life implementation of new interventions through graded exposure, aiding the acceptability of CHARMER in the future. Indeed, if the trial confirms the CHARMER intervention is safe, acceptable, and cost-effective, we will work with relevant stakeholders to roll it out nationally.

Because of the large scale of the trial, it was also more efficient to use a stepped-wedge design. This increases statistical precision because the clusters partly act as their own control (this is different from traditional trials where there is a completely different control group that never receives the intervention) and results in the design requiring fewer clusters.

Issues inevitably arise during a trial as complex and large as CHARMER. A stepped-wedge structure allows most issues to be addressed, as learning from earlier steps means that we can adapt the trial’s processes for future steps.


CHARMER uses a stepped-wedge structure to mimic the real-life implementation of new interventions and increase statistical precision. Participants partly act as their own control groups, transitioning from control to implementation phases at different times. These timings form a step-like pattern allowing for evaluation of effectiveness over time and process adaptation. I hope this guide helps in understanding stepped-wedge trials and why they’re used in CHARMER.

Why bother with research?

By Dr Victoria L Keevil, Consultant Physician (Dept Medicine for the Elderly, Addenbrooke’s Hospital) and CHARMER co-applicant

Equity in research and clinical practice

At various points in our professional lives, it is reasonable to consider why we do what we do. One chance for such reflection materialised when I was asked to consider joining the CHARMER study team as a co-applicant. I had research training, a PhD in Epidemiology, and I had used this training to conduct large scale, data driven service evaluations as a full time NHS consultant. Despite this, I felt I was not fulfilling my obligations to fully capitalise on the skills learnt through my doctoral experience. This professional guilt was balanced against the ever-increasing clinical demands on Consultant Geriatricians as the complexity of inpatient caseloads increased year on year. Why would anyone want to make professional life more challenging, by adding in the demands of clinical research and academia?

As I pondered this, I considered an editorial in Age & Ageing that picked out a review of rehabilitation interventions after hip fracture. The authors found that over a quarter of all potential participants in trials of rehabilitation interventions after hip fracture were excluded due to equity factors; things like place of residence (i.e., living in a care home), age, occupation, disability, or socioeconomic status. A major criterion excluding potential participants was cognitive impairment, despite one third of patients who are admitted to hospital with a hip fracture having cognitive impairment.

This imbalance between the research and patient populations raises the possibility, as discussed in the editorial, that inequitable research might inadvertently contribute to inequitable healthcare. If certain population groups are not included in research data, and the NHS continues to prioritise evidence-based care as one of its constitutional commitments, it is going to be hard to develop treatments and services to meet the needs of all patients. Thus, older adults along with other vulnerable population groups may not have equal opportunity to reach their full heath potential compared to the general population.

Evidence was also emerging that research active hospitals had lower risk-adjusted mortality for acute admissions than centres that were not research active, which persisted after adjustment for staffing and other structural factors. Embedding research into everyday practice appeared to benefit all patients, whether they were directly involved or not. Consistent with this, being research active was also associated with a happier workforce and a more efficient healthcare system, leading to the Royal College of Physicians and National Institute for Health and Care Research to publish a joint position statement entitled ‘making research everybody’s business’.

So there seemed a lot of positives about engaging with research opportunities as a Consultant Geriatrician. It was important to support development of evidence-based interventions tailored to the needs of older adults and collaboration could also enhance my own professional well-being.

The role of CHARMER – why did I take up this opportunity and has it lived up to expectation?

The CHARMER programme is working with older adult inpatients, a patient group undoubtedly underserved by clinical research, to develop and test a behaviour change intervention for hospital doctors and pharmacists to encourage proactive deprescribing. This means stopping medicines before they cause harm.

Currently half of older people are prescribed a medicine with a potential safety risk. The CHARMER team’s previous research found that these medicines are rarely stopped; less than 1% of hospital admission medication is proactively deprescribed. Yet 99.8% of patients and carers want doctors to initiate deprescribing discussions in hospital.

We can only speculate as to why older adult inpatients are often excluded from research. In order to enable participation, the CHARMER research team adopted a collaborative approach from the study’s inception. Consultants like myself, specialising in Medicine for Older People, as well as public and patient representatives were included as co-applicants and have been embedded in all CHARMER study activities. The work packages offered further opportunities to engage Geriatricians, both consultants and higher specialist trainees, other practitioners working in healthcare for older people and importantly older patients themselves, in research across multiple sites. For example, Work Package 2 involved researchers, patients and healthcare professionals, from both district general hospitals and tertiary centres, co-designing the final behaviour change intervention. So, as well as investigating important research questions around prescribing, I saw CHARMER as an important resource; helping our workforce and patients to engage with and participate in clinical research.

As we now approach the definitive trial stage of CHARMER, it has the potential to be the largest and most significant trial within the Ageing Specialty of the Clinical Research Network’s portfolio across England. The trial aims to recruit approximately 20,000 older inpatients across 20 hospitals. To achieve this, there will be challenges and more boundaries to push. In comparison to professionals from other specialties, it is not routine for consultants and pharmacists working in Medicine for Older People to take on the role of site Principal Investigator. Specialties such as Cardiology and Oncology have had decades of lived experience of participating in multi-centre clinical trials, building strong relationships with hospital Research and Development departments as well as national research networks.

But, if we are to advance healthcare for older adults through research, it is vital that clinical research is part of our ‘business as usual’. Through clinical academic leadership and research activity we can better serve older patients in the future, helping to shape research agendas and identify where there is uncertainty of practice and need for more evidence. This has been an important consideration in my own motivation to re-engage with research, both via collaboration on CHARMER and my own independent research activity, facilitated through award of a Clinical Academic Research Partnership grant from the MRC.


The CHARMER study has delivered on its promise to better involve older adult inpatients and professionals across the specialty in clinical research. I hope we can continue this success as we enter into the final programme of activity, the CHARMER trial. It is important that we can develop evidence-based interventions for this underserved patient group and to do this we need high quality research that focuses on them.

Developing the CHARMER trial: The experience of a Geriatrician and Pharmacist as Co-Principal Investigators

By Sai L. Duraisingham (Consultant Geriatrician) and Paresh Parmar (Lead Clinical Pharmacist for Care of Older Persons and Stroke), London North West University Healthcare NHS Trust

The CHARMER research programme aims to develop and test a behaviour change intervention for hospital doctors and pharmacists to encourage proactive deprescribing – this means stopping medicines before they cause harm.

In summer 2022, London North West University Healthcare NHS Trust, Northwick Park took part in the CHARMER feasibility study (Work Package 3). We really wanted to be involved because we recognise the importance of deprescribing in older people and know there needs to be more focused research into this area. We thought our Trust would have a lot to offer because we are a busy hospital in North West London with a large in-patient cohort.

We originally signed up for the feasibility study in April 2022 but could not participate due to lack of workforce at the time. The CHARMER research team were very understanding of our disappointment having worked hard to participate up until this point. They promised to involve us in the future if at all possible. We were then offered a chance to be part of the study as a site purely implementing the CHARMER intervention in August 2022 and jumped at the chance to be involved in a small way.

The CHARMER research team were very engaged and held regular meetings with us to check on our progress and support us with our requests to the Trust Research and Development department. They responded to email communications very promptly and did their best to answer questions we asked in a timely manner. They demonstrated a lot of patience with clinicians who aren’t regularly involved in Research and Development.

The CHARMER study was quite different to research we had previously been involved with. It was testing a system change; this required multiple members of the hospital team working together, including IT, managers, pharmacists and geriatricians to implement CHARMER. The research involved us:

  • planning how pharmacists and geriatricians could meet up at least once weekly to discuss proactive deprescribing opportunities
  • putting together a hospital action plan for proactive deprescribing
  • attending workshops to discuss deprescribing case studies
  • reporting the number of proactive deprescribing discussions undertaken.

Our biggest challenge was implementing CHARMER in addition to our normal workload being delivered at a higher intensity because of a workforce shortage. Despite this, our existing pharmacists and geriatricians invested the time and effort to engage with CHARMER.

Our involvement in CHARMER and this experience has now opened a wider conversation within our Trust about the role of Research and Development and how this should be part of everyone’s role. Interested clinicians should be given the time and resource to lead on these important studies.

CHARMER Away Day 2022

By Megan Pritchard, Trial Manager at Norwich Clinical Trials Unit, University of East Anglia

One of the great things about the CHARMER team is the variety of knowledge and experience each member brings to the group. We live and work across England in universities and hospitals as busy clinicians, researchers and senior representatives of patient groups. This, combined with the effect of delivering the early parts of the CHARMER programme during the pandemic, has meant that despite our regular meetings as a group over Microsoft Teams, we have rarely had the opportunity to meet up in person.

Winter 2022/23 marked an important stage in the CHARMER programme. We moved from the feasibility study in Work Package 3 (testing and fine-tuning the processes for delivering the CHARMER intervention and trial) into the intensive planning and set up of Work Package 4: the definitive clinical trial to test the effectiveness of CHARMER on a large scale. December 2022 was then the perfect time to arrange our first face-to-face meeting as a research group at the University of Leicester, the university sponsoring the CHARMER programme.

Having had an early start for the trip to Leicester we met for a brain boosting coffee and breakfast before knuckling down to study planning discussions.

Pictured here is Katherine Murphy (Public and Patient Involvement lead), David Taylor (Public and Patient Involvement member), Sion Scott (Programme Manager) and Megan Pritchard (Trial Manager).

The focus of the away day was to review successes, challenges and lessons learned during the feasibility study (a massive thanks again to Norfolk and Norwich University Hospital; Royal Salford Hospital; Wrightington, Wigan, and Leigh Hospital, and Northwick Park Hospital for their involvement in this study).

We reviewed the study data and discussed how we could improve the processes in preparation for the definitive trial, including things like patient recruitment materials, the hospital action plan template and presentation of benchmarking data.

It was wonderful to see everyone together and having a whole day’s focused event meant we made great progress towards finalising our approach to the definitive trial. There are still some details to iron out, but this experience has motivated all of us to meet more regularly as a team, which I hope will go a long way to making the next stages of the CHARMER programme a success.

Harnessing behavioural science to tackle overprescribing

By Debi Bhattacharya, Professor of Behavioural Medicine, University of Leicester / CHARMER Co-Chief Investigator

In October 2022, I delivered a talk at the Westminster Health Forum about how we can harness behavioural science to tackle overprescribing in the NHS. Here’s a summary of the key messages from the talk:

1. Overprescribing should be addressed during a hospital admission

The NHS National Overprescribing Review Report quite rightly focuses on primary care as this is where the majority of prescribing occurs. Yet 99.8% of patients and carers want doctors to initiate deprescribing discussions in hospital.

Currently half of older people are prescribed a medicine with a safety risk. The CHARMER team’s previous research found that these medicines are rarely stopped; less than 1% of hospital admission medication is proactively deprescribed. The CHARMER programme aims to develop and test a behaviour change intervention for hospital doctors and pharmacists to encourage proactive deprescribing.

2. Key stakeholders need to be involved in the development of deprescribing interventions

Existing deprescribing interventions have demonstrated marginal increases in deprescribing activity that are not sustained beyond the trial period. This can be partly attributed to how these interventions are developed and designed.

The CHARMER team’s previous research with geriatricians and pharmacists working with older adults identified what things help and what things hinder them from proactively deprescribing medicines in hospital. The team also worked with them to identify and decide on what strategies would help to address these things and support proactive deprescribing. These strategies are called ‘Behaviour Change Techniques’ and are the active ingredients that are responsible for bringing about change. In other words, the CHARMER intervention has been designed in collaboration with the people who will deliver and receive the intervention.

3. Deprescribing interventions need to be supported by behaviour change theory

While stakeholder engagement is essential for achieving behaviour change, purely focussing on what people think they need to deliver the change often fails to address all barriers and enablers. Staff commonly cite the need for extra resource as the solution. However, given extra resource such as time and workforce, staff will often increase existing activities with which they are familiar and have a known pathway of recognition rather than undertake the desired new behaviour. Another commonly delivered ‘solution’ to effect a change in practice is education and training. Whilst having the required knowledge and skills is clearly essential, behaviour change is rarely achieved by addressing only these barriers/enablers.

Asking staff what help they think they need places the onus on them to correctly identify their barriers and enablers to undertaking the desired behaviour and then select the most appropriate solutions. The field of behavioural science has offered a scientific approach to garner more meaningful input from stakeholders to shape the development of strategies (or Behaviour Change Techniques) to support implementation of a new behaviour. In CHARMER, we have worked with geriatricians and pharmacists to select and design strategies that are likely to be acceptable to everyone, affordable for NHS organisations and that will not introduce inequity.

Combining research evidence, behaviour change theory and collaborative design methods increases the likelihood that deprescribing activity resulting from the CHARMER intervention will be sustained beyond the trial period.

Developing the CHARMER trial: The experience of an NHS Research Project Manager

By Ania Spurdens, Research Project Manager, Norfolk and Norwich University Hospital

The CHARMER research programme aims to develop and test a behaviour change intervention for hospital doctors and pharmacists to encourage proactive deprescribing – this means stopping medicines before they cause harm.

In summer 2022, Norfolk and Norwich University Hospitals NHS Foundation Trust (NNUH) took part in the CHARMER feasibility study (Work Package 3). The aim of the feasibility study was to test and then fine-tune processes for delivering the CHARMER intervention and trial. This was to ensure that hospitals across the country would be able to deliver the processes with ease in the definitive trial.

I was invited to work closely with the CHARMER team from an early stage. This was particularly useful as it meant I had access to information about the study early on, allowing plenty of time for us to plan ahead as an organisation and consider what resources would be needed. There were a few things that were different about the CHARMER feasibility study compared to studies I’d set up previously:

  1. It is rare to implement organisational changes as part of a trial – however, the clinical research team had previous experience in implementing organisational changes and we were able to follow our local processes. Senior staff in the hospital were also interested in and supportive of the study and its wider aims.

  2. Some data was collected through data linkage with routinely collected datasets from NHS Digital to minimise the burden on staff collecting data. This was something we were keen to make work. The CHARMER team, our Information Governance team, and our Senior R&D leaders worked to ensure patient data would be handled appropriately and kept secure.

  3. The Principal Investigator (PI) at the organisation was given training by the CHARMER team on how to facilitate workshops for staff taking part in the study.

  4. A benchmarking element of the intervention allowed NNUH to monitor and benchmark its progress with other hospitals taking part in the feasibility study. During set-up, we had to work out who would prepare these reports and we worked closely with specialist pharmacists to pull data from the hospital’s electronic prescribing programme.

I enjoyed setting up the CHARMER feasibility study and learned a lot along the way gaining experience and developing connections with teams across the hospital that I don’t usually work with. The CHARMER team were incredibly helpful and approachable throughout, and it was clear they wanted to support our staff and make things as easy as possible for us. There were a few things we hadn’t fully considered during set-up that came up during the active study window; however, working closely with the CHARMER team, we were able to feedback on our experience and improve the processes in preparation for the definitive trial.

Sign up to the definitive trial here.

Guest Blog: Patient and Public Involvement in Deprescribing Research

Members of the CHARMER team – Katherine Murphy (Patient and Public Involvement Lead), Doreen Pegg (Patient and Public Involvement member), David Wright (Professor of Health Services Research) and Caroline Smith (Research Associate) – have written about the importance of Patient and Public Involvement (PPI) in deprescribing research for

Patient and public members have been central to CHARMER programme of research and in the blog they describe their involvement in the project. Here’s a snippet:


“My motivation is to give suggestive ideas… to improve the service in which there is a focus on remedying the way repetition of medications can be reviewed at every opportunity which will be of value to those who are not reviewed in their circumstances.”

– CHARMER PPI member

Read the full guest blog here.