Future Hospital Programme case study comes from King’s College Hospital NHS Foundation Trust. It explores the need to integrate both psychological and social interventions into diabetes care for patients with complex psychological needs.Key recommendations
• Create a model of care that deals with the medical, psychological and social implications of having a chronic condition.
The challenge
From a patient’s perspective diabetes has biological, psychological and social consequences – it may be associated with significant distress and is associated with a 2-fold rate of depression. Indeed, diabetes is most common in inner-city areas with high levels of deprivation. For some individuals, addressing the biological aspect cannot be done until the latter two impacts have been addressed and are under firm control by the patient.
It is recognised in the literature that there are diabetes-specific psychological problems which can have a significant impact on diabetes management and glycaemic control. The work specifically incorporates diabetes-related themes, for example, if the reason that a patient has poor glycaemic control is because they are running high blood glucose levels to avoid hypos, then there will be some focus on managing ‘fear of hypos’.
Our solution
To aid in this process we have devised the ‘3 Dimensions of Care for Diabetes’ (3DFD) care model; its underlying principle being parity of esteem between physical and mental health. As the name suggests there are three core aims set out in this model of care:
• improve physical health
• improve mental health
• improve social functioning.
The team were not funded per additional session, instead it was block funded as an additional team which integrates with the diabetes team in the hospital and in the community. The cost was £190 per annum which was initially research-funded and then later supported by the CCGs. Patients are seen both in hospital and community clinics (at the ThamesReach base in the community and in people’s homes).
Staffing
Our 3DFD team comprises of clinical psychologists working alongside third-sector support workers, under the leadership of a consultant liaison psychiatrist; all integrated into diabetes teams found across primary, community and secondary care organisations in South London. We accept referrals of diabetes patients with notable issues with glycaemic control and psychosocial difficulties. Referrals are triaged and assessed by a multi-disciplinary team (MDT) in weekly MDT meetings. The patient is a key member of the team and we promote the individuals autonomy in managing his/her diabetes. We hold patient-led case conferences for complex patients to help us to shape their care according to their needs.
Key learning
We overcame some barriers in setting up this service – mainly attitudes within the organisation that the patients we were targeting would be too difficult to engage, and that it was a ‘luxury’ service rather than basic clinical care, and we found that by addressing patients’ mental health and social care needs alongside their diabetes care they were more motivated to engage in self-management. This is demonstrated by improved attendance at diabetes appointments (attendance rates increased from 72% to 79% in the year following the intervention) and improved glycaemic control (HbA1c). In addressing these we benefitted from having a steering committee with wide representation, including CCGs, to ensure we developed in line with the organisational vision, of both King’s College Hospital and the King’s Health Partners (South London’s Academic Health Science Centre), a vision with strong emphasis on integrated care and joined up working.
The patients referred were at high risk of developing diabetes complications: in the first pilot 120 patients were referred, and since then we have received a mean of 300 referrals per year. Referrals were accepted if patients were resident in Lambeth and Southwark, with poorly controlled diabetes (HbA1c> 9%) and psychological or social co-morbidity. Some were patients who had failed maximisation of therapy, however many had adherence problems secondary to their mental health/social problems. The proportion of type 1 to type 2 diabetes was 1:2, with the average HbA1c being 11.5%. We found that the majority of patients were from ethnic minorities: 42% black; 19% Asian/other; 39% Caucasian, and although the majority had depression, we identified a wide range of psychiatric morbidity including psychosis, dementia and eating disorder.
Outcomes
1. Medical
We observed clinically significant reductions in HbA1c with a mean reduction of 1.6%, which has persisted at a bi-yearly follow-up. This is a three-fold reduction compared to a new diabetes drug. Although not statistically significant, because of smaller numbers, there was a trend in improvements in blood pressure, lipid control and weight control.
2. Psychological health
There were significant improvements in psychological symptoms with statistically significant improvements achieved in measures of depression (PHQ-9 scale: reduction from 8.8 (SD 6.5) to 6.4 (SD 5.8) p= 0.021), anxiety (GAD7 scale: reduction from 8.9 (SD 5.3) to 4.9 (SD 5.3) p< 0.001) and diabetes specific distress (Diabetes Distress Scale: reduction from 47.3 (SD 11.5) to 61.6 (SD 14.9) p= 0.005). In addition 60% of patients received a new diagnosis of a psychiatric disorder: indicating significant previously unmet need.
If a large unmet need was identified, eg we found an unmet need that problems such as depression, anxiety, eating disorder are not being detected routinely. The team has therefore delivered education with the referring teams and plans to build on this in the future.
As this is a complex collaborative care type intervention, we did not dissect out the exact components. Previous studies of collaborative care have not done this so we did not see that it was necessary.
…its underlying principle being parity of esteem between physical and mental health
3. Social functioning
There were improvements in the measures of social functioning across multiple domains including personal responsibility, living skills, social networks, substance misuse, meaningful use of time and accommodation.
4. Qualitative evidence of patient satisfaction
A focus group of 3DFD patients conducted by ThinkPublic identified the “high levels of care and reliability” as being key to its success in engaging patients with their care. The accessibility of the 3DFD team, as well as the quality of communication between the team and other professionals involved in their care, were noted as being important in the patients’ experience.
This reduction in HbA1c is greater than the improvements seen with new diabetes drugs and by local community diabetes clinics in select individuals.
5. Economic outcomes
In our first pilot of 120 patients, we found that 3DFD patients had reductions in unscheduled care: A&E attendances by 45% (for the whole cohort, from 141 in the year preceding the intervention to 77 in the year following), hospital admissions by 43% (from 72 to 41)) and bed-days by 22% (from 381 to 300). Analysis undertaken by the Diabetes Modernisation Initiative (DMI) projected further savings of £102,000 for every 120 patients, year-on-year, in delaying or preventing diabetes complications.
Patient case study
Helen is a 61-year-old woman with a history of type 2 diabetes since 2006, and was on maximum oral therapy. In September 2012 she was seen by a diabetes specialist nurse (DSN) in her GP practice. Her HbA1c was high at 13.4% but she was unwilling to attend the intermediate clinic or try insulin. The DSN referred Helen to 3DFD, where she was seen by the support worker and the psychiatrist.
She reported a long history of depression, only partially treated on a sub-therapeutic dose of anti-depressants, and a history of self-harm throughout her life. In addition she had high levels of anxiety, had a life-long pattern of binge eating, an alcoholic husband and financial problems. She was diagnosed with depression and eating disorder not otherwise specified. The psychiatrist titrated her anti-depressants to a therapeutic dose and gave her 12 sessions of diabetes-focused cognitive behavioural therapy (CBT) which focused on diabetes-behaviours as well as her mood. The support worker also gave her advice around her husband's alcoholism. With this input Helen was able to make lifestyle changes and improved her diet, exercise and adherence to her diabetes medications. She eventually agreed to attend the community diabetes clinic; at her first appointment her HbA1c had dropped down to 8.9% (a 4.5% reduction). At the clinic she started a glucagon-like peptide-1 (GLP-1) agonist in group and 3 months later her HbA1c was at its target of 6.8% (an overall reduction of 6.7%; almost halving her original HbA1c). Her depression is in remission and her eating patterns are greatly improved.
Who’s involved?
Dr Anne Doherty, consultant liaison psychiatrist and project lead Dr Carol Gayle, consultant diabetologist
Professor Khalida Ismail, consultant psychiatrist and Professor of Psychiatry in Medicine Dr Anne Doherty MB MRCPsych MMedSc MD is a consultant liaison psychiatrist and project lead at King's College Hospital, London. Her research interests include mental disorder in the context of diabetes and suicidal behaviour in adjustment disorders. She leads the 3 Dimensions of Care for Diabetes service, awarded the BMJ Diabetes Team of the Year prize in 2014 and an NHS Innovation Challenge Prize 2015.