To determine the cost and clinical effectiveness of offering 6-10 sessions of remotely delivered cognitive behaviour therapy (CBT) via video calling or over the telephone for health anxiety in repeated utilisers of unscheduled/urgent care versus treatment as usual.

To optimise the delivery of CBT for health anxiety delivered remotely by systematically identifying and then acting on barriers and enablers to the intervention through a network of practice.

Background: Health anxiety costs £3 billion per year in unnecessary expenditure, much of it on unscheduled care and in-patient admission. CCGs are incentivised to reduce emergency care use and the Department of Health is spending up to an additional £400 million per year to provide psychological treatment for this problem. Yet patients with health anxiety are reluctant to accept face to face psychological treatment. There is strong evidence that delivered in secondary acute care as a liaison psychiatry service psychological therapy it can be clinically and cost effective for two years. Government policy is to deliver this intervention in primary or community care where there is little evidence of clinical or cost effectiveness. Face to face delivery of this intervention through secondary care mental health and IAPT services has not been acceptable to these service users. Remotely delivered psychological treatment designed to assist coping with symptoms can be delivered by mental health services and may be both more acceptable to service users than face to face treatment in IAPT services and just as effective as in secondary acute care.

Aims: To determine the clinical and cost effectiveness of remotely delivered cognitive behaviour therapy for health anxiety in repeated users of unscheduled primary or secondary care for physical symptoms without a physical health cause. To determine barriers and drivers to delivering such remote treatment and how such treatment might fit into a wider care pathway to enhance patient experience of care.

Methods of research: Randomised controlled trial of 6-10 sessions of cognitive behaviour therapy for health anxiety delivered by telephone or through the internet versus treatment as usual. Primary outcome is change in health anxiety from baseline to 6 months; secondary outcomes are persisting change in health anxiety to 12 months, emergency care use, generalised anxiety, depression, somatic distress, work and social adjustment and quality of life. We will assess economic outcome and qualitative analysis of barriers and drivers to delivery of intervention and view the intervention as part of an overall care pathway to provide alternatives to emergency care use. This will help network leads, practitioners and service users to shape the research so that it can provide the most information to enable putting the intervention into practice.

Methods of implementation (putting into practice). A network lead, a networking practitioner and an associated network of practice will bridge work between the research team and practitioners, commissioners, strategic clinical networks, Health England (education) and the Academic Health Science Networks (AHSN) across the East Midlands to put the intervention into practice. The process will be iterative, identifying core features of the intervention that should not be varied and those parts of the intervention that may be adapted allowing a degree of adaptation to local requirements.

Research plan:

Design: Pragmatic individually randomised controlled trial (RCT) of CBT versus treatment as usual, stratified by site. Economic evaluation and qualitative analysis of barriers and drivers to the research, intervention and its implementation into practice will also be performed. There may be differences in the uptake and retention to both the intervention and the trial if the service user is recruited to the study through their own practice or through an urgent care service. In order to refine the efficiency of recruitment and retention into the RCT and the intervention, we will conduct a 12 month feasibility phase in two parts of the East Midlands, then roll out the study within the East Midlands before starting the study in other AHSN areas.

Setting. Recruitment of service users from primary care, unscheduled primary or secondary care in the East Midlands, including rapid access services for problems such as chest pain. Delivery of intervention remotely by mental health services.

Interventions. CBT delivered remotely. We will use an experienced CBT therapist, who was trained in the engagement and delivery of CBT to health anxious patients in primary care to deliver CBT and to supervise up to four therapists also delivering the same intervention one day per week, one from each mental health trust in the East Midlands (lead from Nottinghamshire Healthcare NHS Trust). The lead therapist will receive supervision from the lead therapist in the CHAMP study (Tyrer, 2013) to ensure consistency with the approach in that study and to understand any adaptations that are necessary to deliver this intervention remotely and to high utilisers of care. The number of sessions is dependent on the pace of engagement with the patient and the complexity of their problems, and is tailored to the individual formulation of the patient's problems by the therapist (Tyrer, 2013). The treatment intervention will address health anxiety with reassurance seeking and/or phobic avoidance with or without management of panic attacks, generalised anxiety or depression, communication with health professionals, current psychosocial problems and differentiation from other physical health problems.

Treatment as usual will be that decided by the patient with their general practitioner and health professionals they consult for unscheduled care.

Outcomes. Outcomes will be assessed single blind at baseline, 6 and 12 months. The primary clinical outcome is change in the 14 item self-rated short week Health Anxiety Inventory (Salkovskis et al, 2002) from baseline to 6 months. Secondary clinical outcomes are contacts with unscheduled or emergency care, change from baseline to 12 months on the short week the 14 item HAI, 7 item GAD-7 (generalised anxiety; Spitzer et al, 2006), 15 item PHQ-15 (somatic distress; Kroenke et al, 2002), 9 item PHQ-9 (depression; Kroenke et al, 2001), 8 item social function (WASA; Mundt et al, 2002), and 5 item quality of life (EQ5D; EuroQol Group, 1990).

Economic analysis. A detailed resource profile will be established for the intervention versus usual care. A cost profile will then be attached to each arm. Cost utility and cost effectiveness assessment will be made from health and social care perspectives as recommended by NICE.

Inclusion Criteria:

• Two or more consultations, referrals or hospital admissions with any provider of unscheduled or emergency care (including urgent same day appointment with own general practice) in the last 12 months for common presentations not attributed to identified pathology such as: cardiac e.g. chest pain, respiratory e.g. breathlessness, neurological e.g. dizziness, gastrointestinal e.g. abdominal pain or genitourinary e.g. pain on passing urine.
• Scores above the threshold for severe health anxiety of 18 or more on the 14 item short version of the Health Anxiety Inventory (HAI; Salkovskis et al, 2002).
• The participant is aged 18 or over.
• Sufficient understanding of English (spoken and written).
• They give oral and written informed consent to participate in the study.

Exclusion Criteria:

• Pathological medical condition requiring further assessment or acute management, or pregnancy.
• Other severe mental illness (schizophrenia, bipolar disorder, severe major depressive episode, eating disorder) ascertained by the Structured Clinical Interview for DSM-IV Disorders (SCID, Spitzer et al., 2002) or anyone at immediate risk of harm to themselves or other people through their mental state
• Organic mental disorder (dementia, delirium, substance use disorder, organic mood disorder).
• They are already receiving specialist mental health intervention, including psychological treatment as part of specialist medical care e.g. pain clinic.

All of the above require a different clinical approach to the treatments being tested in the study. Unipolar mild to moderate depressive episodes, other anxiety disorders or stable physical illness are not exclusions to the study as they are readily addressed by the intervention and are necessary to include if the study is pragmatic and going to generalise to clinical practice.