Development of an education and self-management intervention for chronic headache – CHESS trial (Chronic Headache Education and Self-management Study)

We completed three systematic reviews to help inform the development of the intervention. Full details of the methodology and results of these reviews are reported elsewhere [22, 35, 36]. Here we have provided a summary of the key finding and the influence these had on our intervention design (Table 2).

We conducted qualitative interviews to inform the intervention design. This interview data and the lived experience systematic review both aimed to ensure that the intervention design included a strong patient focus. Migraine Action,¹ one of the charity partners in CHESS, sent letters on the trial’s behalf to their members within a predefined geographical region for ease of travel (100 members resided in this area). Interviews were face to face in people’s homes and were audio recorded after taking informed consent. Topic guides explored their experiences of what might be helpful, or unhelpful, treatment strategies and where they sought information regarding their headaches. Interviews were transcribed, anonymised and analysed thematically. All data were collected and held in accordance with data protection guidelines.

From the 100 invitation letters sent out, we received 21 responses. Of these responses, five had headaches for < 15 days per month, three had no headaches, three were under the age of 18 years and two were not interested in the study. Of the eight that met our inclusion criteria of headaches on > 15 days per month for at least three months, one had new daily persistent headache and was therefore excluded. We interviewed seven people; the results of these interviews were presented at the intervention development day.

When participants were asked about things that they had tried which were helpful for their headaches they spoke about a vast array of treatments and strategies. These included; belonging to organisations for information and support, seeing approachable and knowledgeable doctors, seeing different therapists (physiotherapists, acupuncturist, counsellor, craniosacral therapist), meditation/relaxation, distraction techniques, being outdoors, having social support from people who understand, having a positive mind-set, Yoga, Pilates and breathing techniques. However, what was helpful to some was often unhelpful to others and many of the things tried were out of the scope of our intervention.

When describing their medication use, interviewees spoke of ‘mixing and matching’, and ‘trial and error’, some medication working at one time and not at another. Side effects were an important feature and two interviewees voiced worries about their medication being rationed. Exact medication use was difficult to ascertain at interview. This was an area for us to explore further in the feasibility study interviews.

We asked participants for their views about what we should consider for inclusion in an education and self-management group intervention. Suggestions included getting peer support (group meetings); developing skills in stress management, relaxation/meditation; learning about triggers, lifestyle factors and medication; having information in one place; having education for others such as family and employers; gaining expert support and finding out about the latest research and advances.

They also gave us some practical points about running a group for people with frequent headache, this included running of the course in a neutral environment so that it was not too medical (such as a hospital setting). They felt small groups were important and there should be regular breaks throughout the day. An option to have groups in the evening was also suggested for those that work or cannot attend during the day.

With the associated links between chronic headaches and those with chronic pain, we have drawn upon the knowledge and experience of intervention design and delivery from those involved in the design and running of the COPERS study [37]. Key members of that study are part of the CHESS study team. This was a randomised control trial of a group education and self-management intervention for those living with chronic pain. This was a complex intervention designed using underpinning psychological theory and the guidance of the MRC framework for designing complex interventions [32, 33]. This team conducted two reviews to help inform their intervention design and the results of these reviews have been examined as part of the CHESS intervention development [38, 39]. The COPERS intervention was proven to be acceptable and effective in the medium-term for depression, anxiety, social integration and support, pain acceptance, and self-efficacy in pain management. There are also long-term positive effects for depression, and social integration and support.

The basic structure and content of the COPERS intervention was used to inform the CHESS intervention as was the use of groups so that people could learn from each other and the use of lay facilitators to jointly facilitate with the nurse. Having two facilitators allows for easier management of challenges or difficulties without the need to disrupt the rest of the group. Other experience used to inform the CHESS intervention was holding courses in familiar accessible locations and that clinicians from different disciplines were capable of delivering the intervention.

We have drawn upon the core theoretical principles from several psychological theories including, many of which have been used in other self-management interventions; cognitive behavioural theory [40, 41], social cognitive theory [42, 43], acceptance and commitment therapy [44], theory of planned behaviour and reasoned action [45, 46] and the health belief model [47, 48]. to guide the development of the intervention aimed at our specific population. The behaviour change wheel and taxonomy [29‐31] have been used to guide our thinking about the sources of behaviour that could be targeted, the rational for each topic area and the most effective strategies for implementing and encourage behaviour change. Table 3 summarises the theoretical underpinnings and the behaviour change techniques that have driven the design and components of the CHESS intervention.

Grounded in the evidence base from our review, we built in ‘taster’ sessions around relaxation and mindfulness. The sessions provide participant with the opportunity to participate via instruction on how to engage in relaxation and mindfulness. Participants then have the opportunity to take away material to allow for home practice with the aim to encourage regular practice and habit formation in the long- term.

We have also produced a DVD for participants to share with family and friends. The DVD is aimed to be informative, reiterating the core messages from the course as well as portraying what it is like to live with chronic headaches. This was developed in response to suggestions from PPI members at the intervention design day as well as the qualitative review and interviews that suggested those living with chronic headaches feel others do not understand.

This meeting was attended by clinicians including a neurologist and two general practitioners, directors from two leading headache charities (Migraine Action; National Migraine Centre), three lay people living with chronic headaches, psychologists with expertise in self-management and behaviour change, academics and researchers (18 attendees in total). The multidisciplinary team bring together clinical expertise in the management of chronic conditions including headaches and pain as well as expertise in behaviour change and self-management. We drew on this experience to inform the development of the intervention.

The day comprised of short presentations on the results from the three systematic reviews, main findings from the qualitative interviews, an overall study summary from COPERS [37] and a summary of the main outcomes from a classification day which took place to inform the design of a logic model to allow chronic headaches to be classified. The classification day comprised of facilitated discussions on core questions to help inform the development of a logic model. The outcomes from these small group discussions were discussed in a large plenary session and this information was used to help develop and refine a logic model to be used in the study and intervention process. Full details have been published elsewhere [49].

During the intervention design meeting, presentations followed facilitated discussion about the results, what the tailored education self-management intervention should look like, what ongoing support participants would need and what the control intervention should be. We have had input from the three leading headache charities. We have also had the input from three lay members who attended the intervention design day and were involved in the discussions and decisions. After the meeting, a drafting document was circulated which provided all those that attended the opportunity to feedback. This combined input has helped shape the intervention.

We designed an education and self-management intervention delivered and facilitated by a non-headache specialist nurse and lay facilitator (someone who lives with chronic headaches). The intervention was delivered over two full days (10:00–15:00) followed by a one to one consultation with the nurse facilitator (agreed at a time convenient for the nurse and participant, approximately a week after the group sessions, lasting up to two hours). This was then followed by a half-day group follow-up session (10:00–12:30). The course was designed for between 8 and 10 participants.

Facilitators were recruited through adverts and local contacts. Lay people were recruited through our charity partners. We had three nurses and two lay people interested and available to attend a two day training course aimed to inform them about trial procedures, equip them with facilitation skills for running groups and familiarise them with the content of the intervention. An assessment of learning form was completed by the facilitators following the training to check their understanding, knowledge and confidence in delivering the intervention. Figure 1 shows the structure of the course.

The intervention was piloted in four groups in Warwickshire, England with a total of 18 participants. Three of these groups were delivered based on the course structure presented in Fig. 1, the fourth group was delivered using a two day structure (Fig. 2) based on the feedback that had been received. The groups were run in community settings.

We interviewed the five facilitators and 12 of the participants to capture their feedback and experience. We also collected written participant feedback at the end of the course and asked facilitators to provide written reflective logs after each day of facilitation.

Thematic analysis was used to identify common themes across the different components of the intervention. Overall, the results suggested the groups were liked and the material was deemed useful and interesting. Some people had gained new information which was personally useful and others had tried some of the self-management strategies discussed in the group. The opportunity to meet and participate in a group was much appreciated. The feedback from these interviews have enabled the intervention team to streamline the course. Table 4 describes the main changes made following the feedback.

Some other more practical issues identified by participants included the need to carefully consider venues for delivering CHESS especially in relation to seating, temperature and lighting. Each venue selected for the CHESS delivery is assessed for suitability by the trial team. The facilitator manual provides a reminder for to facilitators to assess the temperature and lighting and to accommodate needs where possible.

In light of the feedback the structure of the intervention was revised, Fig. 2 shows the structure for the RCT and Table 5 shows the final modules and content of the intervention package.