In the Woodland we celebrate well structured, delivered and analysed research; (hopefully) turning the outputs into easy to read blogs. The series of research studies in this blog come from a very impressive and detailed piece of work and I will highlight the abstracts and key outcomes only. I would however strongly recommend reading the original document especially if you are looking for an example of a structured, high quality, quantitative and qualitative study.

In May of this year, this programme of research into the relationship between coronary heart disease and depression in primary care patients was published as part of the Programme Grants for Applied Research (PGfAR) programme, which is part of the National Institute for Health Research (NIHR),

The PGfAR was set up in 2006 to produce independent research findings that will have practical application for the benefit of patients and the NHS in the relatively near future. The programme is a national funding scheme that aims to provide evidence to improve health outcomes in England through promotion of health, prevention of ill health, and optimal disease management (including safety and quality), with particular emphasis on conditions causing significant disease burden.

The overlap between long term conditions and mental health would seem to fit perfectly within this description; Coronary heart disease (CHD) and depression are predicted to be the two leading causes of health-related disability worldwide by 2020. In 2007, the annual societal costs of depression in England were estimated to be £7.5B, projected to be £12B by 2026. The King’s Fund estimated that £1 in every £8 spent on long-term conditions is for comorbid mental health.

Depression is more prevalent in patients with CHD, but the nature of this relationship is uncertain. Many people with CHD are depressed and research on people who have had a heart attack found that depression increases the chance of further heart attacks.

The UPBEAT-UK team conducted research with people on general practitioner CHD registers in 33 south London practices to:

• Examine any link between CHD, depression and worse future heart disease; and
• Develop case management by practice nurses for people with CHD and depression.

Tylee et al on behalf of the UPBEAT-UK team have pulled together the results from these 4 relate studies:

1. Meta-synthesis of previous research and a qualitative study of GP and Practice Nurse (PN) views on current management of patients with CHD and depression
2. Qualitative study of patients’ perspectives of their biopsychosocial needs
3. Pilot trial to assess the acceptability, feasibility and cost-effectiveness of an intervention based on nurse-led personalised care (PC) for people with CHD experiencing current chest pain and depression and determine best outcome measures
4. 3-year cohort study of patients on GP CHD registers to explore prevalence, incidence, course and costs of depression, course and pattern of chest pain and cardiac outcomes, and the relationship between depression and cardiac outcomes.

Methods

Participants

Adults aged ≥ 18 years, registered on GP CHD Quality and Outcomes Framework registers.

Patients from 16 surgeries were recruited to the cohort study and patients from an additional 17 practices were recruited for the pilot randomised controlled trial (RCT). Patients from the cohort study were recruited for the qualitative study, as were GPs and PNs for the qualitative study of professionals.

Interventions

The team developed an evidence-based intervention informed by patient and clinician preferences and established theory. This was a primary care-based nurse-led personalised care intervention. Following a face-to-face assessment, nurses trained in behaviour change techniques facilitated patients to address the problems that they perceived as most important to them and which related to their CHD or depression. Existing resources for CHD or depression-related problems were identified and used by nurses. Follow-up was by telephone.

Results

1. Meta-synthesis and qualitative study of general practitioners and practice nurses

Meta-synthesis identified seven qualitative and 10 quantitative studies, none of which concerned depression and comorbid physical illness.

GPs and PNs reported that:

• They were aware of a relationship between mood and social problems, but were unsure of their role in addressing this
• They considered that distress after a cardiac event resolves spontaneously; if it endured, or became severe, it was treated as depression
• Psychosocial problems were viewed as contributing to depression in CHD, but they expressed uncertainty about their role and responsibility in addressing these problems
• An individualised approach was favoured, but they were unsure how to achieve this.

2. Qualitative study of patients

30 patients with depressive symptoms on the CHD register were interviewed:

• A theme of loss, both before and after the onset of CHD, underpinned accounts (e.g. interpersonal loss, loss of health and of control)
• Participants felt ‘depressed’ by what they perceived as a ‘medicalisation’ of loneliness and by the experience of ageing and ill health
• Some believed that their GP would not be able to help with their complex health and social issues
• Talking therapies and interventions providing social interaction, support and exercise (e.g. cardiac rehabilitation) were thought helpful, whereas antidepressants were not.

3. Pilot randomised controlled trial

17 practices agreed to participate. Of 3,325 patients on CHD registers, 1,001 consented to contact, of whom 81 were eligible and randomised (41 intervention, 40 control).

This was a pilot trial, and so was not powered to detect efficacy of the intervention, therefore there are no p-values within the report findings.

Depression:

• Both groups showed improvement in depression (HADS-D score) at all time points
• Mean scores moved from moderate depression at baseline to mild depression at 12 months. A mixed-effects model showed no significant differences between groups over time for any measure of depression and confidence intervals (CIs) were wide, so an effect in favour of either group cannot be ruled out.

No longer reporting chest pain:

• 6 months: 37% in personalised care (PC) vs 18% in usual care
• 12 months: 31% in PC vs 19% in usual care

Other results:

• PC participants made fewer accident and emergency visits (24% PC vs. 38% TAU) although missing data concerning the reason for these visits makes this difficult to interpret
• Self-efficacy was also improved more in PC
• Total costs in the intervention arm were lower than usual care but QALYs were also lower. These differences were not statistically significant
• Overall, personalised care seemed to be acceptable and feasible.

4. Cohort study

16 practices, 142,648 patients, 2% (2,938/142,648) were on GP CHD registers.

A total of 803, participated, representing 27% (803/2,938) of those on the CHD registers.

• Mean age was 71 years
• 70% were male and 87% were white
• Participants reported multiple social problems, multimorbidity and disabilities, including problems with general pain and discomfort (53%), poor mobility (49%) and difficulties with intimate relationships (38%).

573 patients (71.3%) provided complete data to 36 months:

• 7% had depressive disorder at baseline and 13% had depressive symptoms
• 12% had an anxiety disorder comprising: panic disorder (< 1%), generalised anxiety disorder (3%), and mixed anxiety and depressive disorder (8%). 25% had anxiety symptoms yet only 3% of people were recorded as having anxiety in the GP records
• The incidence of depression was 130 per 1,000 person-years at risk for men and 90 for women, so males were nearly 1.5 times more likely to develop depression
• Standardised mortality ratios compared with the general population were:
  • 1.13 (95% CI 0.62 to 1.90) for men
  • 1.87 (95% CI 0.81 to 3.70) for women.

Anxiety was an independent predictor of myocardial infarction and cardiovascular death (RRR 3.93, 95% CI 1.95 to 7.90). Depression did not predict any cardiac outcomes.

Risk profiles for continued reporting of non-exertional (Rose category 1) chest pain:

• Female sex [odds ratio (OR) 2.80, standard deviation (SD) 7.19],
• Asthma (OR 3.34, SD 1.98) and
• Anxiety (OR 1.33, SD 0.62).
• Good quality of life was protective (OR 0.98, SD 0.01).

Risk profiles for continued reporting of exertional (Rose category 2) chest pain:

• Exertional pain at baseline (OR 28.07, SD 7.14) and
• Anxiety (OR 0.65, SD 0.38).
• Good quality of life was similarly protective for exertional pain (0.98, SD 0.01).

The average cost over the 36 months for patients with depressive symptoms at baseline was double that for patients without depressive symptoms at baseline.

Statistically significant predictors of higher societal costs were: depressive disorder, white ethnicity, housing problems, relationship problems, self-reported current cancer and baseline health-care costs.

Conclusions

Chest pain, anxiety, depression and social problems are common in patients on CHD registers in primary care and predict adverse cardiac outcomes.

Together this combination of morbidities pose a complex management problem for GPs and PNs. The pilot trial of personalised care (PC) suggests a promising approach for treatment of these patients. Generalisation is limited because of the selection bias in recruitment of the practices and the subsequent participation rate of the CHD register patients, and the fact that the research took place in south London boroughs.

The paper’s conclusion is that further research is needed to understand the links between anxiety, chest pain and heart disease, and to further develop the promising findings that PC can be helpful in reducing chest pain in general practice. I would also add that there is a strong educational component required within primary care to improve clinician’s confidence in how best to manage CHD patients’ symptoms in the context of the many psychosocial problems.

The notion of PC promoting self-management fits with the principles of the Five Year Forward View and PC combining case management and self-management with an extra emphasis on anxiety needs to be further piloted and definitively tested with primary care staff. I’m upbeat, I feel good.

Links

Primary paper

Tylee A, Barley EA, Walters P, Achilla E, Borschmann R, Leese M, et al. (2016) UPBEAT-UK: a programme of research into the relationship between coronary heart disease and depression in primary care patients. Programme Grants Appl Res 2016;4(8)
http://dx.doi.org/10.3310/pgfar04080

Photo credits

• Alan Levine CC BY 2.0
• Beth Scupham CC BY 2.0