Experiences and perceptions of people with headache: a qualitative study

Participants were sampled from respondents to a population-based study [18, 19] conducted in the Grampian region of Scotland, UK which investigated the prevalence and natural history of chronic pain (pain or discomfort, present either all the time, or on and off, which had persisted for three months or longer [20]). A total of 660 respondents in 2000 reported having chronic pain in their head. Of these, 73 individuals identified the head as either the only site, or the main site, of their pain and were defined as having chronic head pain. A total of 59 of these individuals indicated a willingness to participate in further research. From this group, a purposive sample of 30 individuals was selected to include males and females aged 65 and under, with different levels of severity of chronic pain (as measured by the chronic pain grade [21]) and self reported use of different healthcare services. The 30 individuals were sent letters detailing the purpose of the study and inviting them to participate in an interview about their headaches. Individuals who did not, or no longer, suffered from headaches were given the opportunity to opt-out if their participation was not felt to be appropriate. Reminders were sent four weeks later to non-respondents. Ethical approval for the study was given in advance by the Grampian Research Ethics Committee.

A semi-structured interview schedule (Table 1) was developed, covering issues identified as potentially important from a review of the literature. The interview started with an open question inviting participants to tell us about their headaches. This allowed the participants to raise any issues that they felt to be important. If issues arose that weren't on the topic guide the interviewer ensured that these issues were discussed further during the interview. The closed questions detailed in the schedule were not directed towards the participants, but simply used as a guide by the interviewer to the topics to be covered and to identify further relevant areas of interest. Open-ended questions were used throughout the interviews to encourage participants to answer openly and unrestrictedly. The schedule was piloted on four researchers with experience in conducting qualitative interviews so that areas for improvement could be identified. This led to some amendments to the schedule. Formal headache diagnoses were not sought from GP records. Participants were asked to self-report what types of headache they suffer from.

The interviews were conducted by DAL between October and December 2002. All of the interviews took place in participants' homes except for one which took place on University premises at the participant's request. Informed consent was sought and granted in all patients interviewed and participants' anonymity and confidentiality were ensured. All interviews were audio-taped with the participants' permission and lasted an average of 45 minutes. Immediately after each interview, brief notes were made by DAL, detailing interesting points made, or behaviours displayed, by the participants. Within two days after the completion of the interview the tapes were transcribed verbatim by the interviewer. A sample of tapes and transcripts were reviewed by AME to check that the interviews had been accurately transcribed. Tapes and transcripts were marked with ID numbers to ensure anonymity and were stored securely to maintain confidentiality.

The interviews were analysed using the Framework approach: a thematic analysis consisting of five stages [22]. The first stage, 'familiarisation', was achieved by reading each transcript and noting down recurring issues. All transcripts were read twice by DAL and once by at least one other member of the research team (AME and PCH). Having done this, the second stage, 'identifying a thematic framework' was completed by arranging the recurrent issues that had emerged into themes and sub themes (Table 2). Interesting and relevant quotes were also identified at this stage to illustrate the themes. Themes and sub-themes were given unique codes, which were marked on the margins of the transcripts to identify interesting issues arising – this process constitutes the third stage, 'indexing'. For the fourth stage, 'charting', charts were constructed for each of the main themes that emerged, with sub-theme categories placed along the top of the charts and participants placed down the sides. By conducting each of these four steps accurately and rigorously, the fifth step, 'mapping and interpretation', was simply achieved by placing relevant sections of the transcript into the appropriate charts. The Framework approach makes between and within subject analysis possible enabling researchers to compare and contrast participants' opinions and perceptions in each sub-theme, therefore illustrating the similarities and differences between them. Reliability of the data and interpretation of the findings were checked at each stage of the process (AME and PCH).

Nineteen out of the thirty individuals contacted agreed to participate in the study, 5 respondents indicated that they did not wish to participate, and the remaining 6 did not respond. Reasons for non-participation were not sought. A total of 17 interviews were conducted. Mutually convenient interview times could not be arranged for the last two individuals within the timescale of the study.

Table 3 details the participant characteristics, headache characteristics and associated health care utilisation in the last year of the participants. Most respondents reported suffering from migraines or tension-type headaches. The frequency of headaches in the sample ranged from daily to monthly. Many participants reported that the frequency of their headaches changed depending on the time of year and their current circumstances (i.e. levels of stress at work, family commitments, other health problems etc). Most of the individuals interviewed had been suffering with headaches for several years. Most participants reported that their headaches were of mild to moderate severity, but some reported severe headaches.

A number of interesting issues were raised throughout the interviews, many of which were inter-related and overlapping. However, three broad themes emerged: 'impact on life', 'knowledge and understanding', and 'management'. A summary of the themes and sub-themes emerging from the interviews is shown in Box 2.

We found that headaches frequently adversely affect the lives of sufferers, with some wide-ranging effects on work, family life, social activities and mood. Many individuals had developed their own explanations for their headaches and many reported worry about the cause of the headaches. As a result most participants were searching for an increased understanding of their condition. Although the GP was seen as an important source of advice and treatment, many individuals were concerned about taking medication and many were trying complementary therapies or using their own management techniques.

This study interviewed men and women from the general population who suffered from different types of headache with various causes, frequency and severity. It is one of the first qualitative studies to explore the experiences and perceptions of people with varying headache types and provides important in-depth understanding of the range of problems headache sufferers face including the impact of headaches, understanding and knowledge about headaches and experiences and expectations associated with healthcare and treatment. By studying the broad spectrum of headache types, rather than limit ourselves to one specific type, we were able to identify a range of issues relevant to headache and its impact. Although this approach means that some of our findings will not be relevant to all headache sufferers the approach has provided a useful insight into the differences in perceptions and experiences of headache sufferers. It has also highlighted some issues that will be worth exploring in sub-groups of individuals, particularly with a view to developing new treatments. Information on the cause of the head pain was not determined in the original cohort and so we do not know if the individuals interviewed in this study had a primary headache disorder or whether the headache was a result of some other cause such as trauma. Our results can only apply to the age range of individuals included in the sample. Specific issues important to groups outwith this age range will not have been identified. In addition, since participants were sampled from a group of individuals with self-reported chronic pain (pain persisting beyond three months), our findings may not reflect the experiences of people who's headaches have not persisted this long. Local availability of different services may have driven use of certain services, and this may vary by geographical location across the UK. The researcher's roles and opinions may have affected both the data collection and interpretation. This was minimised where possible by having checks on the data by a second member of the research team.

Several important issues emerged from our study, some of which have been reported in previous studies and some of which have not been documented before. The impact of headaches on life emerged as a key theme in this study and has been well documented in previous quantitative studies [4‐11] and in the qualitative studies that have been undertaken [15‐17]. Our findings on the impact of headache on daily functioning (including impact on work, social and family life and mood) are broadly in line with previous qualitative studies of headache, despite differences in the type of headache investigated and the populations studied. In a focus group study in the US with 24 people who had been experiencing migraines for at least 6 months Cottrell et al [15] reported that all aspects of family, recreational and social activities had been hampered by their headaches. The effect on work was substantial, with participants reporting having to go home early and feelings of guilt for not carrying a fair share of the workload. They also found that many participants expressed apprehension about losing their jobs because of the time they missed. This did not emerge as a clear issue in our study and may be more relevant to migraine sufferers who are likely to take more time off work. Ruiz de Velasco et al [16] in a Spanish study on quality of life in migraine patients reported similar impacts on work and studies, life within the family, social relationships, and leisure time. They found that respondents had difficulties in performing their duties at work and felt they couldn't fulfil their real potential. Many participants reported difficulties in going out with other people and some reported that they had declined to participate for fear of having a migraine attack. They found a greater impact on family relationships than we reported, particularly in terms of the effect on children. This may have been because the age range of our participants meant that fewer respondents had young children at home. Like us Ruiz de Velasco et al found that the impact on mood was an important issue with their respondents reporting mood swings, unhappiness and hopelessness and a greater lack of emotional control. Contrary to this Cottrell et al [15] reported that their participants denied emotional distress despite the emotional charge obvious in many of their statements. Our study adds some understanding to the wide-ranging and extreme effects headache can have on peoples lives, for example forcing people to change career or leading to severe depression and thoughts of suicide. Further work is necessary to identify ways that impact can be minimised and quality of life improved.

The second theme to emerge from our study was knowledge and understanding about headaches which showed that many participants had formed their own ideas about their headaches, which had often led to considerable worry, and many were searching for an increased understanding of their headaches from a variety of sources. In a qualitative study of patients decision making for migraine and chronic daily headache management Peters et al [17] reported similar issues to us. They found that an individual's knowledge about headaches was acquired through the participant's own and other people's experiences and through information gathering. Patients sometimes had to recognize the limitations of his/her own ideas for headache management and resort to GP consultation. They found that individuals had sought information from a variety of sources including health professionals, family and friends, the media and specialized organisations. Cottrell et al [15] also found that people with migraines often need more general information about migraines and migraine management. It is clear then that people are interested in understanding more about their headaches and securing relevant information about it. This desire for further information needs greater investigation. The issue of worry raised in this study has not been highlighted clearly before. Most participants in our study had, at some point, worried, often substantially, that their headaches were caused by serious pathology, even patients who had been given a formal diagnosis. The fact that participants who had received brain scans stated that their worry had reduced and in some cases their headaches had diminished as a result merits further enquiry.

The third theme to emerge focussed on issues about management of headaches. We found that most participants had used or were still using a variety of management strategies including conventional, complementary and personally developed management techniques and that satisfaction with different management strategies varied widely. Our findings on issues related to management were largely consistent with previous studies [15, 17]. Cottrell et al [15] found that while some of their group members were happy with the medical care they received others reported feeling 'dismissed' by physicians who did not appear to take headaches seriously. In addition they found that headache sufferers were taking responsibility for researching new treatment options including medications and alternative treatments. Peters et al [17] found similar reasons to us for GP consultation with headache severity, patient preferences, previous experience of healthcare services and other people's experiences of health care all playing a role in management decisions. Some of the issues that emerged from our study around headache management have not been investigated in qualitative studies of headache, but are consistent with qualitative studies of individuals with other chronic conditions, particularly with respect to general practitioner involvement and medication [23‐27]. For example, previous qualitative studies have clearly highlighted limitations with the GP consultation, [23‐25] while other studies have found that chronic users of medication often worry about the side effects of their medication and possible drug interactions [26, 27]. The full range of management techniques used by headache sufferers including unconventional techniques developed by the sufferers has not been highlighted before and this needs further investigation to optimise appropriate management of headache in the community.

An in-depth understanding of the experiences of people with headache is important if we are to develop new treatment or interventions that might reduce the burden of the disorder. This study has highlighted some issues that require further investigation and provided useful information that might guide the development of such interventions. For example, the frequent wish to increase understanding of headache suggests that many people with headache could benefit from detailed information about the causes and management of headaches. Research is needed to determine whether such educational strategies result in beneficial health outcomes. Educational strategies are being developed for patients and health care professionals, particularly with reference to migraine [28, 29] and will require future evaluation. Open accessing scanning may be a useful intervention to rule out serious underlying pathology and reduce anxiety in headache sufferers. Previous work has shown that a single visit to a neurological clinic for headaches not due to structural disease results in the number of attendances at general practice for headache dramatically falling [30, 31]. However, numbers have been small and it may be that patients present to general practice just as often, but with different symptoms. There is little evidence that scanning is of any use in non-anxious patients [32] and it is likely therefore that the most cost effective route would be to scan only those who present with anxiety.