Background

Migraines are often thought of as bad headaches which is a gross simplification of a condition which can be unpredictable and life changing [1]. Around 2–4% of people meet an epidemiological definition of chronic headache; that is, headaches on 15 or more days per month for at least three months [2]. The term chronic headache encompasses all chronic headache disorders, [3] including the primary headache disorders, which do not have secondary organic aetiology; chronic migraine and chronic tension type headache. It also encompasses medication overuse headache, a secondary headache disorder caused by using headache medication on either; ≥ 10 days per month if taking triptans (a migraine specific medication) or opioids, and ≥ 15 days a month if taking simple analgesia such as paracetamol or ibuprofen.

Between 2015 and 2019 we delivered the Chronic Headache Education and Self-management Study (CHESS) in the UK, funded by the UK National Institute of Health and Social Care Research (NIHR), Programme Grants for Applied Research. The main objective of CHESS was to test a supportive self-management approach to improve outcomes for people with chronic migraine or chronic tension type headache with/without episodic migraine. People with and without medication overuse headache were included. Recruitment was through NHS general medical practice [4]. Drawing on current best practice and the experiences of people with chronic headaches, a self-management support programme for people living with chronic headache was developed. This work is described in detail elsewhere [4,5,6,7,8,9].

As part of CHESS, we found a paucity of qualitative research looking specifically at the lived experiences of those who live with chronic headache. Our qualitative evidence synthesis (totalling n = 73 participants) identified only four papers meeting the inclusion criteria [10]. The synthesis of these papers showed three overarching themes: i) headache as a driver of behaviour (direct or indirect); ii) the spectre of headache (describing the emotional impact of this unpredictable condition); and iii) strained relationships (with friends, family and the medical profession) [10]. One of the recommendations from this review was that more research was needed to understand what it is like living with chronic headaches and how it affects people’s quality of life.

Whist the CHESS trial was large and adequately powered, no detectable effect was found for the intervention on the primary outcome (health-related quality of life) at 12-months [8, 11]. However, the CHESS programme of research included a considerable number of qualitative interviews. The interviews aimed to ensure that the voices of those living with chronic headaches informed both intervention development and contributed to the design, delivery, and interpretation of the study as a whole. This work is reported elsewhere [5]. During these interviews participants shared their stories of living with chronic headache. Here we present an analysis and synthesis of our participants’ experiences of living with chronic headaches.

Methods

The study is presented in line with the Consolidated criteria for reporting qualitative research (COREQ) [12]. Data presented in this paper are from interviews conducted during the CHESS study. Two study phases are described: i) Phase one – feasibility study (April 2016 – January 2017); and ii) Phase two—main trial (RCT; April 2017 – March 2019). Participants living with chronic headache were defined as experiencing headache on 15 or more days per month for at least three months. In both phases, interview participants were recruited via English GP practices. All participants in phase 2 of this interview study were also CHESS trial participants [4]. All participants gave separate informed consent to participate in the trial and for the interviews. No formal diagnosis was carried out with participants as part of the interview studies. However, each participant was given a classification of their headache type as part of the CHESS study overall.

Phase 1: Initial interviews

Overall, 131 people with headache disorders took part in the feasibility phase of the CHESS research programme. Participants who expressed an interest in taking part were purposively sampled to obtain a range of headache type, age, gender, and location. Participants in these initial interviews were not exposed to the CHESS interventions in any way. We used a semi structured schedule approach for face-to-face interviews exploring participants lived experience of chronic headache. Interview topic guides are provided as supplementary material. (Supplementary materials 1, pages 1–4).

Phase 2: Main study interviews

Of 736 participants in the main CHESS trial, 396 (54%) had chronic migraine, 331 (45%) had chronic tension type headache and episodic migraine, and nine (1%) had chronic tension type headache only.

Participants in the main trial were approached if they had expressed an interest in participating in an interview as part of the main trial consent process. Those who consented were first interviewed post baseline assessment and pre-randomisation (when they had not yet been informed as to which arm of the study they were allocated); they were also invited for follow up interviews at four and 12-months. We found however that because we recruited participants before they were randomised that the sample of those who were taking part was giving more participants from the control arm of the trial. Therefore, it was decided that additional participants in the intervention arm be invited for an interview at four months and twelve months to increase the diversity of our sample.

We purposively sampled both male and female participants with a range of ages and locations across different classifications of chronic headache. We sent this sample a participant information leaflet detailing the reasons for doing the interviews. Those who wished to participate were then booked for a face-to-face interview in their choice of convenient location (home, work, local community room). We used semi structured interview topic guides. (see supplementary material 1, pages 1–4) This was adapted/updated using our learning from the phase one interviews. Interviews asked about their lived experience of their headaches and different aspects of the impact of this on their quality of life.

Data collection and processing of all interviews

All interviews were audio-recorded on an encrypted digital device (OLYMPUS DS-7000), kept in a digitally safe environment, given pseudo identifiers before being transcribed by a member of the team who removed any identifiable data from the text. The researcher who did the interviews (VN) then checked the transcripts for accuracy. Researcher notes were written immediately after each interview to capture aspects which may not have been apparent from the recording, such as body language and the emotional tone of the interview as well as the researcher’s views, to promote researcher reflexivity and cross case comparison. We used NVivo software to manage and organise the data.

Analysis

For Phase 1 interviews we used thematic analysis [13], structuring the themes by the experience of interviewees from becoming aware of a headache through decisions about what to do in response to the headache. Two researchers (VN, FG) read and familiarised themselves with the transcribed text, developed codes, discussing, and updating the codes as analysis continued (VN, FG & DE).

In Phase 2 we had learnt from the phase one interviewees that how they experienced their headaches changed over time. Our focus was on exploring this change over time using serial interviews, e.g., those who gave interviews at baseline, four months, and 12-months. To do this, we brought into our analysis only data where the participant talked about their current headaches. For each interview we wrote a pen portrait. Pen portraits are a way in which longitudinal health research data can be concentrated into a focused account [14]. The methodology is a four stage process: i) understand and define what to focus on; ii) design a basic structure relevant to the dataset in question; iii) populate the content; and iv) interpretation. Stages one and two were informed by the data emerging from the phase one interviews providing a structure with the following five headings: context, different headaches, medication management, non-medication management, and quality of life. Initial pen portraits were reviewed and refined by the team until we established a consistent approach. Pen portraits were compiled by VN.

The team read the pen portraits and developed categories, a method we have previously used to identify change in categories of experience with pain over time [15]. These categories can be considered as ‘categories of headache impact’ whereby participants may match mostly with one category but have characteristics of other categories to varying degrees [16]. Each team member independently categorised a third of the pen portraits with VN categorising them all. Subsequently meetings were held to compare categorisations, discuss any discrepancies, to agree a final category for each pen portrait and give the categories titles.

The final set of interviews included participants recruited within the intervention arm of the trial at four months and again re-interviewed at 12-months (as noted in the methods above). Four additional participants (one intervention arm; three control arm) were interviewed just twice—at baseline and four months – and are included in this group. Again, pen portraits were crafted (as detailed above). However, as a validation check, the categories defined in the previous group analysis were applied to check if they resonated across this subsequent sample.

Results

Phase 1: Initial interviews

We interviewed 15 participants (10 female) age range 29 to 69 years (median 47 years): two had chronic tension type headache, six chronic migraine, and seven probable chronic migraine. Additionally, six of the 15 had medication overuse headache. Participants lived across a range of localities in the English Midlands: three rural, eight in towns and four in a city (Table 1).

Table 1 Phase 1: Interviewee characteristics
Full size table

Medication used by participants was, on the whole trial and error from over the counter (OTC) medicines at the outset. If these didn’t work then they would seek help from their GP for suggestions of other medications to try. The OTC medications used at the time of interview ranged from paracetamol preparations/ ibuprofen/, and opioid analgesics most of which helped but didn’t abolish the headaches. A few were on preventatives such as pizotifen, beta blockers or amitriptyline prescribed by the GP or a neurologist. Triptans were also prescribed which were useful for some. Others used medication prescribed for other conditions for their headaches such a naproxen, propranolol, or codeine. Around half described having tried preventatives in the past which were mostly unsuccessful.

Our participants recounted often long complicated medical/ personal/ headache histories. Most talked about experiencing different headaches, dichotomising them as ‘normal’/ ‘not severe’ or ‘migraine’/ ‘severe’ headaches. At onset, sometimes they were aware immediately which headache type it was but sometimes they had to wait to see how it progressed before identifying the headache type, gauging its intensity or its nature. The presentation and the context of the headache were important considerations as to what they would then do in terms of taking medication, altering plans, considering how it would impact others. Sometimes they had to leave their current environment because of the headache. Decisions would depend on whether they were at work, home or out somewhere and whether they felt they could manage the headache where they were. Participants described an often-continual search for triggers or an acceptance that there were no identifiable triggers. They described how the experience of headaches impacts on their lives including the frustration at people not understanding how it affects them and how fed up they were with having headaches. Table 2, provides illustrative quotes of their experiences of chronic headache.

Table 2 Phase 1 interviews, showing themes and illustrative quotes
Full size table

Phase 2: Main Study interviews

Comparison across all time timepoints (developing and testing the categorisations from the pen portraits)

A total of 26 participants were interviewed, from a variety of locations in the Midlands or London. (Table 3) These 26 participants make up the two groups below.

Table 3 Phase 2: interviewee characteristics
Full size table

Fourteen (median age group 50s, range 20s – 70s; 9 females) were interviewed three times (baseline, 4 and 12-months; total 42 interviews). Participants were randomly assigned to control (n = 7) and group intervention (n = 7) arms of the trial. However, one participant did not engage with the intervention (did not attend DNA) and was classified as ‘control group’. These interviews informed the development of pen portraits and proposed impact categories.

Twelve participants were interviewed twice (total 24 interviews; median age group 40s-50s 11 females). Four were interviewed at baseline and four months, three of these were in the control arm, with one in the intervention arm (but they too did not attend any intervention elements DNA). Eight were interviewed at 4 and 12-months (age etc.; 7 female); all were in the intervention arm. These interviews were used to test the impact categorisations.

The developed categorisations from the pen portraits (From the dataset of the 14 triple interviews).

Figure 1 above shows the four categories developed from the longitudinal interview data. It was very clear that as time goes by people often change their categories and sometimes exhibit characteristics of some of the other categories at the same time hence the diagrammatic overlap and close relationship between all of the categories.

Fig. 1
figure 1

Concept model of the relationship between categories of headache experience

Full size image

To present our findings we have written short summaries of the pen portraits for each participant and presented these as a table of how they were categorised at each time point.

Tables 4 and 5 below show summary pen portraits from each of the interviewees who gave interviews over the three timepoints (Baseline, four and twelve months). Each participant at each timepoint is coded with one of the four categories. Examples of actual pen portraits, illustrating each of the four categories, are provided in supplementary materials. (see supplementary material, pages 5–21).

Table 4 Headaches over time – Summary pen portraits at baseline, 4 ms and 12 ms (CHESS Control participants)
Full size table
Table 5 Headaches over time – Summary pen portraits at baseline, 4 ms and 12 ms (CHESS intervention participants)
Full size table

Table 4 shows interviewees who were not exposed to the CHESS intervention. At baseline four interviewees were categorised as their headaches ruled their lives, two whose headaches controlled – not ruling their lives, two were not letting headaches rule their lives and none were categorised as Headaches out of control. Six of the eight interviewees changed category at least once over the twelve-month period.

Table 5 shows the interviewees who were in the CHESS intervention arm. At baseline three interviewees were categorised as not letting headaches rule their lives, two where their headaches ruled their lives, one where they were catergorised as headache out of contol and none were categorised as headaches controlled – not ruling their lives. Four of the six interviewees changed category at least once over the twelve-month period.

Relationship between categories

Whilst a predominate category is defined for each interviewee at each timepoint it was very clear that many exhibited elements from others demonstrating the complex nature of living with chronic headache. For example, interviewee #08 (Table 5) who is not letting headache rule their life but also notes that at times they put things off because of it. Whichever the predominant category we decide on there are elements of another category. Some people are at the boundary of a category, so #09 at Timepoint 2 is categorised as not letting headaches rule their life but bordering on their headaches being controlled which by Timepoint 3 they are.

Those sitting in the upper left quadrant of the conceptual model are more likely to be managing with the strategies they have, with headaches impacting little on their day-to-day functioning. Those sitting in the lower right quadrant are experiencing headache effects which either limit or dominate what they can and can’t do.

Validating the categories (interviews at two points in time)

Table 6 Provides summaries of the pen portraits from the twelve pairs of interviews. We were able to categorise participants at each time point using the proposed categories, and as before, some participants expressed elements of another but less dominant category. Of the four who gave interviews at baseline and four months two, at baseline, were categorised as their headaches were ruling their lives and two as their headaches were out of control and something needed to change. At four months all were no better and indeed all were categorised as their headaches were out of control and something needed to change. Of the eight who gave interviews at four and twelve months: three, at four months, were categorised as not letting their headache rule their lives, three were categorised as their headaches were ruling their lives and two were categorised as having headaches that were controlled. At twelve months all but two of the eight participants changed categories, with two re-categorised as ‘headaches out of control’ and four as their ‘headaches are now controlled’. The final two are categorised as, not letting headache rule their life and headaches not ruling life.

Table 6 Summary pen portraits testing the categories defined earlier in pairs of interviews (two timepoints)
Full size table

All four categories could be influenced by many external contextual factors e.g. if they were retired, if they were able to work around their headaches, where they were when a headache started. Where one person had to battle through another had the flexibility to take time out for it to ‘work through’ with or without medication. Others kept going even with severe headaches that others wouldn’t have been willing or able to do.

Discussion

All participants involved in these interviews attested to meeting the CHESS study criteria for chronic headache defined as headache on 15 or more days per month for at least three months with most reporting these as migraine. In a pre-randomisation telephone interview, we classified just over half as having chronic migraine. The phase one interviews revealed that their headaches had often developed over long periods of time and were embedded into their lives giving a complex fluctuating presentation. Headaches appear to determine how people can function in their everyday lives and a great deal of time and energy is spent anticipating the severity of the next headache and gauging effective strategies such as medication use and behavioural approaches to lessen their severity and impact. Different headaches can feature as a difference in severity or nature which can alter the strategies used.

Throughout we see that people are often struggling to function and ‘keep going’ most often in flux and dependant on many factors. This complex picture keeps people gauging what to do, when and how to do it whenever a headache presents, often learning from trial and error.

What we see in these results is the often overwhelming and comprehensive effect that living with this condition has on people. Our pen portraits illustrate this complex picture demonstrating that these are not marginally troublesome issues rather they are something at the heart of health, well-being, and quality of life. Our findings are similar to those of a study in Spain looking at the views and experiences of women living with chronic migraine [17]. They identify five key themes: i) the shame of suffering from an invisible condition; ii) treatment: between need, scepticism and fear; iii) looking for physicians’ support and sincerity and fighting misconceptions; iv) limiting the impact on daily life through self-control, and v) family and work: between understanding and disbelief. However, whilst our participants often hid their condition and there was some stigma associated with it, they didn’t speak of feeling ashamed. Other studies have highlighted the negative impact that living with chronic headache/chronic migraine can have, emotionally and psychologically, on relationships, careers and finances [18, 19].

Improvement, or perceived improvement, is possible but deterioration also occurs. These fluctuations may in part be the result of life events (e.g. employment issues, family problems etc.) not directly related to their condition. Such events seem to impact on the persons abilities to live with and cope with their headaches. We see that people move in and out of different phases of coping and not coping with their headaches. Although contextual factors make a huge difference in their headache management, our data suggest that the severity and/or frequency of headaches is the main driver of the state of coping with their headaches. Participants reported migraines that were unpredictable and could not be sure how well they would be coping with their headaches at a future time. This is similar to the findings of a qualitative synthesis which highlights headache as a driver of behaviour either directly or indirectly affecting many aspects of life including their relationships often evoking emotional responses of guilt, worry or uncertainty [10]. Indeed, another study looking at developing a strategy to measure outcomes from the patients’ perspectives for use in evaluating preventive treatments for migraine found, that chronic migraine impacts physical functioning, social and leisure activities, and also has emotional impacts. These impacts are experienced during and between migraine attacks and vary considerably day-to-day [20].

A paper from 2022 suggests that management of migraine requires a biopsychological approach integrating non-pharmacological management alongside conventional biomedical treatments [21]. The experience of our participants supports this but also warns of the challenges of finding relief from migraine. The CHESS study was designed to address both the physical and psychosocial aspects of living with a chronic headache disorder but as noted in the introduction no detectable effect was found for the intervention on the primary outcome (health-related quality of life) at 12-months [7, 8]. A recent article taking a closer look at the outcomes of CHESS did find some positive effects on self-efficacy and noted that future work may want to look at other more relevant outcome measures highlighting again the complex nature of this condition and the need to consider all of these complexities [22].

Strengths and limitations

A particular strength of this paper is that it includes 41 individuals who live with chronic headache, and that from 26 of these we have longitudinal data. This has allowed us to explore the condition in more depth. All of those interviewed had self-reported chronic headache at the start of their involvement in the CHESS study (i.e. 15 or more days per month for at least three months) and it is clear from their stories that they do live with this disabling condition. However, it could also be seen that this is a limiting factor as many of those included may not have ever received a definitive, specialist delivered diagnosis. Nevertheless, we are reporting on the experience of people living with chronic headache disorders; few of whom will ever see a headache specialist.

Conclusion

“It’s just part of who I am…” is the title of this paper and we suggest that the sentiment expressed within this quotation reflects the impact that living with chronic headache can have. Indeed, this disabling condition is often more than just an aspect of their lives, it is all consuming in its unpredictability and complex in its presentation and management. Our findings imply that helping people who live with chronic headache conditions requires a holistic, whole person, approach encompassing lifestyles, medications, support and societal acceptance of a frequently misunderstood condition.