Primrose Study: Supporting the whole person: Why physical health matters for patients with severe mental illness


by Dr Philippa Shaw, Dr Zuneera Khurshid & Gemma Copsey

The Primrose project is all about reducing cardiovascular disease risk in people with severe mental illness. We are the new research fellows (Philippa and Zuneera) and mental health integrated care research lead (Gemma), working on the Primrose project. As the new starters joining the Primrose team, we needed to start by understanding what had been done in the Primrose project before – this blog brings you on the journey to understand the history with us. 

Where it started: cardiovascular disease risk in people with severe mental illness

People with severe mental illness, such as people with a diagnosis of schizophrenia and bipolar, die about 15-20 years earlier than people who don’t have severe mental illness. This long-standing finding continues to be a troubling statistic. There have been quite a few theories of why this might be, such as severe mental illness related deaths including suicide and an increased rate of physical health problems. 

For a long time researchers have been particularly interested in how severe mental illness is connected to cardiovascular disease – a broad grouping for physical health problems of the heart or blood vessels (e.g. heart attacks and strokes). A research study that looked at patient records from UK general practices found the risk of death from cardiovascular disease was 3x higher for people under 50 with severe mental illness, when compared to people without a severe mental illness. This is concerning because much of this can be prevented through timely diagnosis and appropriate care.


Therefore, researchers developed computerised tools for doctors to use to predict cardiovascular disease risk specifically for people with severe mental illness. This looked at both the usual factors (such as age, sex, ethnicity, smoking status, and body mass index) and included additional risk factors like if patients were taking medication for symptoms of severe mental illness. Such risk assessments could be used in physical health checks, which the NHS recommends are offered to people with severe mental illness once a year. As these tools looked at physical health and mental health together, they were better at predicting cardiovascular disease in people with severe mental illness and could save the NHS money if used consistently.

This told us how to identify risk, but not how to reduce and manage risk. In other words, what was also needed was some kind of prevention. Many of the risk factors that can lead to poor heart health are preventable, and there had been some success in supporting patients with severe mental illness to reduce these risk factors, such as weight management and stopping smoking, but there had been limited research into other risk factors like type 2 diabetes.

Logically, researchers asked patients, healthcare staff, and carers about what needed to be done. What should support look like for people with a severe mental illness who are also at risk of cardiovascular disease? Some of the suggestions for what a service should include were continuity of care (a patient having ongoing support from an allocated member/team of healthcare staff), goal setting with positive feedback, and elements that increased awareness and changed knowledge around severe mental illness, including healthcare professionals’ knowledge and beliefs, to reduce stigma.

So now the team, led by Professor David Osborn of University College London (UCL), had a way to predict risk and the foundation of ideas for a cardiovascular disease management service. Therefore, you might see how an acronym was reached, Primrose – PRedIction and Management of Cardiovascular Risk in peOple with SEvere mental illness (we agree – Primrose is much catchier).

What is Primrose, the integrated primary care service?

Building on this foundation, the team of researchers based at UCL worked with patients, behavioural change experts, and healthcare professionals (like GPs and nurses) to develop a new service. The work on this project was guided by a lived experience advisory panel – a group of 27 patients and carers who advised and were active partners in developing Primrose, supported by the charities Rethink Mental Illness and The McPin Foundation. The Mental Health Research Network (MHRN) even awarded the project a national prize [SP4] in 2013 for how well the project had involved patients and the public.

The result was an integrated (bringing mental health and physical health together), primary care (based in GP practices, delivered by nurses) service (an intervention to reduce cardiovascular disease delivered to patients with severe mental illness) called Primrose.

Manuals for healthcare staff were developed to guide them on how to carry out Primrose. These manuals covered key components like the timeline (8-12 sessions over 6 months) and how to reduce cardiovascular disease risk (setting collaborative goals such as improving diet and doing more exercise or stopping smoking, with an action plan for how the patient could achieve this and points for positive feedback).

Whilst reducing cardiovascular disease risk was the focus through intensive behaviour change sessions, importantly holistic support was also provided (seeing the patient as a whole person), so the service also included things like signposting to additional services if needed, such as to help with financial worries for example.

This all sounds great, but the key question was: does Primrose work?

To try to answer this question, the researchers compared Primrose to the care patients would normally get when they had high cardiovascular risk and severe mental illness across 76 GP practices in England. Patients would usually be provided with leaflets made by the British Heart Foundation. This was done in a way that was rigorous – a national randomised control trial – where half the GP practices provided patients with Primrose, and half provided patients with care as usual.  

The headline findings: people saw the need for Primrose and therefore healthcare staff were engaged, sessions well attended, and patients liked it. Importantly, patients who were supported through Primrose were less likely to be admitted to inpatient hospital care, which may indicate that these patients were staying well or making better use of other support. Researchers found that building positive relationships between patients and staff, involving supportive others, and using what was already available to patients to help them reach their goals supported engagement and success with Primrose. Plus, when looking at how much care costed, Primrose was estimated to save the NHS £895 per patient per year through mental health costs compared to if they received usual care. 

But (there always seems to be a ‘but’ in research) when looking at the medical measures of cardiovascular risk such as levels of cholesterol, these went down in both groups (groups of patients who attended Primrose and those who received their normal care). This is good – patients improved physically over time – but in answering the question of ‘did Primrose work?’, for this finding we can only say that Primrose worked as well as care as usual. 

Researchers suggested a few reasons for this, such as all GP practices involved being aware and supportive of reducing cardiovascular risk in people with severe mental illness (even when not delivering Primrose). Moreover, there were a few implementation challenges with delivering Primrose, including some healthcare staff and patients misunderstanding what Primrose was, healthcare staff being concerned with how well it fitted in with their normal work, and some recognised that stigma around serious mental illness was still present. 

One theory was that healthcare professionals and patients were not talking about statins (a group of medications that are commonly used to reduce cholesterol in people who are high risk of cardiovascular disease). It might be supportive to have these conversations as statins have been shown as effective for people with severe mental illness, and in some age groups statins are less likely to be prescribed to patients with severe mental illness. 

Adapting Primrose

Taking these findings on board, Primrose was further developed with the aim of improving outcomes (becoming Primrose-Adapted or as we will refer to it, Primrose-A). This included adding in checks to make sure patients were prescribed and took statins if they needed them, improving how Primrose was delivered such as training different members of staff to make it more available, and adding peer coach support (peer coaches are people with lived experience of mental health problems who could deliver sessions focused on different aspects of recovery). 

The new version was tested out in London’s Camden (the COVID-19 pandemic also led to sessions being delivered online instead of face-to-face) and is still being delivered to patients today. Patients and healthcare staff viewed Primrose-A positively, and the addition of peer coaches was found to support building additional therapeutic relationships. 

This was especially valuable during the pandemic, with patients commonly experiencing increased isolation. In the latest Primrose study, a peer coach when interviewed said: 

“I think especially in terms of people who might be isolated it’s just opportunity to see someone else and get that trust, it can be a real confidence booster and if you, if that client can relate to someone it definitely makes a difference”. 

There were some important learnings from Primrose-A too – such as the need to improve training and admin support and finding ways to actively reach out to patients to support them to attend their physical health checks.   

For this work, the Primrose-A team was highly commended in the mental health team of the year awards (BMJ awards 2021). 

So what are we doing now? Introducing UCLP-Primrose

Taking everything that has been learned about Primrose so far, and building on this to further improve Primrose, we have reached UCLP-Primrose (which is the combination of the ‘UCLPartners Proactive-Care Framework’ with Primrose-A. This is shortened to UCLP-Primrose. 

The obvious new bit here is the UCLPartners Proactive Care Framework. This is a tool that healthcare staff use to put patients into an order of who should be seen first, based on how many cardiovascular risk factors the person has. If a person has many cardiovascular risk factors, they will need to be seen for their annual health check much earlier than someone who doesn’t have any. If someone hasn’t had a physical health check in the past 18 months then they are automatically put higher up on the list to be seen, to make sure that people get the right support at the right time. More intensive Primrose behaviour change sessions are then offered based on the results of the physical health check. 

There have also been some other changes too, a greater diversity of staff will be trained to support the delivery of UCLP-Primrose (including pharmacists), there is a more defined pathway for patients which prompts further ways to support patients to engage (such as outreach), and UCLP have developed the manuals and training to address identified limitations. 

The UCL research team, funded by NIHR Applied Research Collaboration North Thames and East Midlands, are still working with the Camden and Islington NHS Trust (Philippa is based with the UCL team), but we are also exploring delivering UCLP-Primrose in Bradford (Zuneera is based in the Improvement Academy in the Bradford Institute for Health Research). Our national research team (supported by the brilliant patient and public engagement panel Diamonds Voice) are working collaboratively to explore how UCLP-Primrose is used by and integrated into the GP practices (the adaption and implementation of UCLP-Primrose to local settings across London and Bradford is being supported by Gemma and the UCLPartners team). 

We’ll keep you posted with what we find. 

Related Blogs

Implementation researchers’ perspectives on bridging the research-practice gap

In our blog earlier this year, we talked about simplifying implementation science, and making it more accessible to frontline staff. In this blog we will share critical insights that the implementation team, based at the Improvement Academy, has gathered while supporting, and facilitating putting evidence into practice.

We need a more realistic approach to implementation in healthcare (Part 1)

The basic origins of implementation science have always been a push approach. We as implementation researchers find and know the evidence-based practices that ‘need’ to be implemented. We tell health systems, hospitals, schools, communities, and clinics about these interventions. We ask them if they want to engage with us in a clinical trial or implementation trial. If they agree, funding is provided externally or internally, local managers and their teams are informed that the project is happening, researchers are funded to learn about the implementation, and the executives and policy makers who gave the green light expect findings that tell them how to scale up these interventions. Implementation begins and most pilots are usually pretty successful.

We need a more realistic approach to implementation in healthcare (Part 2)

There are a lot of people post Covid, and a lot of practitioners who are traumatised. In my field of work this was most evident on ICU wards. There is also the idea of traumatised systems. Data has shown how hard these systems were before Covid and the moral injury that practitioners experience in the service settings. Unfortunately, there’s just not enough in the literature and the big implementation science journals and conferences which talk about these burnouts and these traumas.