Integrating physical and mental healthcare in primary care for people with multimorbidity

Dr Peter CoventryDr Peter Coventry, Senior Research Fellow at the University of Manchester’s Centre for Primary Care, blogs about the rise of mental-physical multimorbidity.

53 million people live in England and more than 15 million of them live with a long-term condition. Put another way, about one in three people live their lives with a long-term health problem such as diabetes. If that statistic isn’t staggering in its own right, consider that by 2018 nearly three million people will be living with three or more long-term conditions. So called multimorbidity, the state of having two or more long-term conditions (with neither one being more important than the other) is a growing problem for many countries with ageing populations, such as the UK. But it would be wrong to think it is a problem of old age alone.

Barnett and colleagues showed that multimorbidity affects people 10 to 15 years earlier in areas of social and economic deprivation. What’s more, they showed that mental disorders were common in people with more physical long-term conditions, especially in the most deprived areas. These findings add to the broader understanding that when combined with a long-term condition, depression leads to the greatest reductions in quality of life. Depression and anxiety are known to drive unscheduled care in the health service too, and the health economic impact of mental and physical multimorbidity is significant. Work by the King’s Fund has shown that depression increases the cost of care for patients with long-term conditions by at least 45%, or from £3910 to £5670 a year.

A single-condition health service

Managing mental and physical multimorbidity is difficult and poses challenges for all health services. In the UK, the NHS has been historically built up as a service that deals in single conditions. This position has become entrenched in the era of evidence-based medicine driven by quality standards and clinical guidelines. Primary care, the home of ‘family medicine’, has become a highly performance-managed theatre of pay for performance activity as signalled by the management of long-term conditions in accordance with checklists and treatment algorithms. As Professor Chris Salisbury wrote in The Lancet, clinical guidelines and incentives may well push up quality or at least even out quality across the health service, but they do little to meet the needs of patients with multimorbidity whose lives are not well served by fragmented and disjointed healthcare.

Managing mental and physical multimorbidity

People with mental and physical multimorbidity get a particularly raw deal from the health service. Despite the availability of effective pharmacological and psychological therapies for depression, people with a long-term condition get poorer mental healthcare. Antidepressant medications are not the answer for everyone and many people prefer psychological or talking therapies but, as Professor Linda Gask has eloquently written in her blog, ‘Patching the Soul’, antidepressant medication can be an important component of managing mood. But as Kendrick and colleagues showed, in general practice patients with a long-term condition, including diabetes and heart disease, are less likely to be prescribed antidepressant medication despite incentives to screen for depression in long-term conditions.

A similar tale was told in a qualitative study by our own team in the CLAHRC GM. I led an investigation which showed that far from being treated, depression was very rarely even talked about in general practice consultations for long-term conditions: patients and doctors alike talked around the edges of depression, preferring instead to focus on what they knew best or what they felt comfortable discussing i.e., physical health. Checklists reinforced this strategy, with so much time spent on completing tasks for physical health we found that even in longer clinics run by nurses, mental health took a back seat. And even if patients and health professionals did find the space to talk about depression they struggled to come up with an agreed strategy to deal with it. Of course, these are generalisations based on a snapshot of activity in primary care and there is great work done in primary care to support people with mental health problems. But our findings resonate with international efforts to ramp up support for better and integrated mental healthcare for people with long-term conditions – the World Federation for Mental Health Day in October 2010 is a case in point. And in 2011 our paper published in BMC Family Practice was awarded research paper of the year in mental health by the Royal College of General Practitioners.

Collaborative care: A promising approach

So if we agree that people with mental and physical multimorbidity deserve better care, how do we achieve this? Well, one promising approach is collaborative care. It is a deceptively simple label but a complex intervention. At its heart, collaborative care shares many of the tenets of the Chronic Care Model, the most important being a belief that the delivery of healthcare for people with multiple health problems has to shift away from a reactive approach of only doing things when people are ill to a proactive one whereby patients are kept well. To do this, system-level change is required, including enhanced ways for professionals to share information and make decisions. And for their part, patients need to be activated to become better self-managers. When applied to mental healthcare, collaborative care includes dedicated non-medical case managers who provide proactive and structured healthcare, along with regular supervision from mental health specialists and a multi-professional approach.

Collaborative care originated in the United States under the auspices of Professor Wayne Katon’s team based in Seattle. Over a decade Katon and colleagues have tried and successfully tested collaborative care in a variety of guises, first for older adults with depression (IMPACT), then for people with diabetes and depression (Pathways), and then for people with diabetes or heart disease and depression (TEAMCare). The trial evidence from the States is impressive, not least from the TEAMCare trial which showed that people managed with collaborative care reported remarkable improvements in both their mental and physical health.

Since then members of our CLAHRC GM team have written the definitive word on whether collaborative care is more effective than usual care for managing depression and anxiety. The answer, shared in a Cochrane review, is seemingly unequivocal: collaborative care achieves better outcomes than usual care for adults with depression and anxiety over the short, medium, and long-term. And that review included the first test of collaborative care in UK primary care – the CADET trial. Led by Professor David Richards, CADET demonstrated that the benefits of collaborative care first shown in the US do indeed translate to the NHS. Patients had better mental health outcomes 12 months after treatment and they preferred collaborative care.

Bridging the collaborative care gap to multimorbidity

On the face of it, the evidence seems open and shut. Collaborative care is more effective than usual care for depression and anxiety, and based on the US evidence this should be true for people with long-term conditions. However, the trials in the States were relatively small and recruited selected populations who were relatively affluent and educated – not your typical patients in general practice. Furthermore, the US trials, and indeed CADET too, were reliant on the input from elite clinical academics to provide the clinical supervision of the case managers. And no collaborative care trial to date had included people with multimorbidity. So, many questions remained unanswered, not least could collaborative care deliver the goods for people with multimorbidity and in more routine settings?

COINCIDE logoCLAHRC GM took a bold decision to run a pragmatic clinical trial to find out. And to do that we worked with our partners in primary care and in Improving Access to Psychological Therapies (IAPT) to draw up a collaborative approach that became known as COINCIDE: Collaborative Interventions for CIrculation and Depression. Acronyms come and go but COINCIDE not only captured our efforts to target depression in people with long-term conditions (and our previous CLAHRC had a focus on cardiovascular disease) but it also signalled our desire to integrate mental and physical healthcare in ways that had not been previously tested.

In 2012, under the expert tutelage of Professors Karina Lovell, Linda Gask and Carolyn-Chew Graham, and in partnership with Clare Baguley (a cognitive and behavioural therapist and at the time educational lead for IAPT), CLAHRC GM trained psychological wellbeing practitioners (PWPs) and practice nurses in the COINCIDE care model (see figure below).

COINCIDE care model

In COINCIDE, PWPs were tasked with delivering brief talking therapies (known as low intensity psychological interventions) to people with depression and diabetes and/or heart disease. The care model resembled that provided in routine IAPT services except the treatments were modified to acknowledge the interactions between physical and mental health. Importantly, two treatment sessions were scheduled to take place between the patient, the PWP and the patient’s practice nurse, offering greater scope for integrated care.

The trial ran from January 2012 until November 2013, and was conducted across the North West in 36 GP surgeries and in partnership with 13 IAPT teams. We recruited 387 patients from diabetes and heart disease registers and they were allocated to GP surgeries that had been randomised to either collaborative care or usual care. By randomising practices rather than patients we ensured that health professionals trained in COINCIDE were not able to treat patients who may have been allocated to a control group. This so-called cluster design preserves separation between the intervention and the control group.

What sets COINCIDE apart from comparable trials of collaborative care is that we recruited a population with high levels of mental and physical multimorbidity from deprived areas. The mean age of the sample was 58.5 years (so adults of working age), but only a quarter of them were in work. Moreover, in addition to diabetes or heart disease the population had a mean of 6.2 other long-term conditions and much more severe forms of depression and anxiety than anticipated. In every sense the trial population in COINCIDE was representative of people with multimorbidity who ordinarily do not make it into clinical trials and whose mental health is often overlooked.

Exactly 90% of patients were followed up at four months when the research team evaluated their health, including depression and anxiety, along with physical quality of life and ratings about disability. The results were positive. Both depression and anxiety were significantly reduced in patients in the collaborative care arm compared with the usual care arm. The reductions were not quite as good as we had planned for but, given the profile of the people in the trial, the mental health gains achieved are significant. But perhaps of greater significance was the fact that patients in the collaborative care arm reported being better self-managers and deemed their care to be more patient-centred. They were more satisfied all round. While we could only show these effects at four months it is likely that for treatment effects to endure patients need to be activated to share ownership of their healthcare with professionals. While it is only one trial and there is still much to learn, COINCIDE showed that integrated mental healthcare for people with multimorbidity can be achieved in routine primary care.

Currently the National Institute for Health and Care Excellence (NICE) recommend that collaborative care be offered to patients with depression and long-term conditions only if they have not got better after getting either more intensive psychological treatment (such as cognitive and behavioural therapy), or antidepressants, or both. The COINCIDE trial data suggests that people with moderate to severe depression and multiple long-term conditions can in fact benefit from collaborative care much sooner along the care pathway than currently NICE suggest.

The next step is to further finesse the COINCIDE care model to meet the ever growing needs of primary care mental health providers to deliver patient-centred care to people with long-term conditions, including people with multimorbidity. At CLAHRC GM, we are now engaged in supporting a phased roll-out of the COINCIDE training and care model across IAPT and are looking to evaluate the impact of these training and clinical resources on the quality of mental healthcare for people with long-term conditions.

Read the full publication, “Integrated primary care for patients with mental and physical multimorbidity: results of a cluster randomised controlled trial of collaborative care for patients with depression comorbid with diabetes or cardiovascular disease” in the British Medical Journal.


Leave a Reply

Your email address will not be published. Required fields are marked *