Background

Paediatric chronic fatigue syndrome/myalgic encephalopathy (CFS/ME) is relatively common (0.6–2.4% of children) [1,2,3,4,5,6], diagnosed by extreme disabling fatigue and one or more physical and/or cognitive symptoms such as sleep problems, pain, problems concentrating, headaches, sore throat and dizziness persisting for 3 months in children [7, 8]. Children can become bedbound [9], miss school [10, 11] and develop mood disorders [12, 13]. In a large cross sectional study, adolescents with CFS/ME were found to have significantly lower health-related quality of life (HRQoL) (particularly in physical and school functioning) compared to healthy controls [10]. UK clinical guidelines [8] recommend that children are offered cognitive behavioural therapy (CBT), graded exercise therapy (GET), activity management or components of each. There are no objective tests for diagnosis or recovery in CFS/ME. Therefore, patients’ subjective perceptions and experiences of their symptoms and functioning are important outcomes. Evidence of the effectiveness of treatment is hindered by the lack of well-developed patient-reported outcome measures (PROMs) for children. A review of PROMs completed by children with CFS/ME identified 13 PROMs, six were child specific and seven were not. No CFS/ME child-specific measures were identified. Evidence of essential measurement properties such as test–retest reliability, structural validity and data quality was missing for all 13 measures. The authors failed to recommend a PROM due inadequate evidence of quality and acceptability to children with CFS/ME [14].

Patient-reported outcome measures which seek to assess HRQoL are increasingly used to measure outcomes in clinical trials [15, 16]. HRQoL focuses on the impact of a condition and its treatment on a patient’s physical, social and psychological functioning [17, 18]. The first stage in developing a PROM is to produce a conceptual framework with input from the patient population, to ensure the outcomes measured have face and content validity so that the PROM includes concepts that are important to children [19, 20]. Child-specific PROMs have traditionally been developed with input from health professionals or parents alone [21,22,23] rather than seeking to understand what is important to children themselves. Children have been recognised as “effective content experts” in PROM development [18]. Previous research with children with CFS/ME has produced a conceptual model of what it is like living with CFS/ME and broader contextual factors [24]. However, only outcomes that are most meaningful to patients should be included in a new HRQoL PROM [25]. Therefore, the aim of this study was to understand what is most important to include in the final conceptual framework specific to HRQoL to underpin a new short paediatric CFS/ME PROM.

Methods

Study design

This was part of a larger qualitative study exploring how ‘recovery’ should be defined in paediatric CFS/ME. We wanted to understand the most important outcomes to adolescents to include in a short PROM as well as gather enough detail for each outcome to develop questionnaire items. Interviewing adolescents with CFS/ME who are very tired and have cognitive difficulties poses a challenge for traditional qualitative methods. Card ranking is used in qualitative research as an ‘enabling technique’ to aid discussion and reasons for priorities [26]. Ranking has also been used in various studies with children to prioritise the order of items in a new amblyopia PROM [27] and determine which outcomes are ‘more important’, ‘somewhat important’ or ‘not important’ to children with neurodisability using a ‘Talking Mat’ [28].

The semi structured topic guide was developed with a Young Person’s Advisory Group (YPAG) [29] who felt card ranking (already used in school) would allow children to prioritise the most important HRQoL outcomes [26]. Important areas of life to children affected by CFS/ME were identified from previous qualitative work [24, 30] and used to produce ‘outcome cards’ that could be ranked in order of importance within the interview (Table 1, Fig. 1). We ensured the cards included the range of HRQoL domains (physical, social and psychological). ‘Tiredness’ as the key diagnostic feature of CFS/ME was separated from other ‘symptoms’. ‘Daily activities’ was separated from ‘general hobbies’ to identify potential functional differences. We also used cards such as ‘family’, ‘friends’, ‘boyfriend and girlfriend’ and ‘independence’ that have been shown to be important to adolescents, to allow possible age differences to emerge [31]. Authors who employed a Q-sort task to prioritise 33 health outcomes for paediatric neurodisability reflected that this was challenging due to “the large number of concepts” [32]. Therefore, we were mindful to keep the list of outcomes short and broad as not to burden adolescents with CFS/ME who experience fatigue and problems concentrating.

Table 1 Extract from topic guide
Full size table
Fig. 1
figure 1

Outcome cards ranked within the interview

Full size image

Adolescents and their parents were given the cards and then asked to rank the 15 different outcomes in order of importance. The interviewer then probed further on why participants had ranked in that order and discussed each card in further detail in order to identify the dimensions of each domain (Table 1, Appendix 1).

Participants

Qualitative studies used to develop a PROM should recruit participants that reflect as closely as possible the patient population that will use the new measure [33]. CFS/ME is more prevalent in adolescents [34], and those who are mild to moderately affected can more reliably self-report their health compared to those that are severe (housebound) where a proxy report may be required [8, 35,36,37]. Therefore, adolescents aged 12–17 years, diagnosed with mild to moderate CFS/ME (not housebound) [8], were recruited from a specialist paediatric chronic fatigue service in South West England. We aimed to recruit a range of participants (age, gender, and disease severity) through maximum variation purposive sampling [26].

Data collection and analysis

Families were recruited in outpatient clinics and written consent was obtained from both parents and adolescents in person, prior to the interview. Participants were made aware that the researcher was not part of the clinical team as to avoid influencing their responses [26]. Participants were offered interviews in their own home or in the hospital and parents and adolescents were interviewed separately. Interviews continued until data saturation was reached, where new interviews produced little or no change in themes in the data [38].

Interviews were recorded on an encrypted digital audio recorder and transcribed verbatim. Thematic framework analysis [39,40,41] was used to develop an a priori ‘framework’ of codes based on the overall HRQoL outcome domain cards identified from previous work and used within the interview. This created an overall structure to help organise and index the data. A top-down approach was used to begin with, coding the data deductively according to the thematic framework (our knowledge of the overall outcome domains). The framework was then further developed through reading and re-reading transcripts to identify additional domains and subdomains. Inductive coding was then undertaken, with new codes derived from participants own words, or existing codes modified to provide more detail and the dimensions of the main domains. This expanded the coding framework, which served as the basis for the conceptual framework. Transcripts for the subgroups (adolescents vs parents, ages, and mild–moderate severity) were coded separately in NVIVO [42] and the groups were then compared to search for similarities and differences in the data [43, 44]. A reflective journal was kept in NVIVO to note down differences.

Analysis of the card ranking exercise

We identified the top five domains ranked by adolescents and parents using the card ranking. This identified the most important aspects of health impacted by CFS/ME that they wanted to improve and should be included in the conceptual framework. The qualitative data for each domain were then analysed to explore the dimensions of each domain and how to form questionnaire items.

Quality assurance

The first three interview transcripts were reviewed in a meeting with the three interviewers (RP, NA and DB) and a consultant CFS/ME paediatrician (EC) for interview practice (e.g. avoiding leading questions). Ten transcripts were double coded (NA) and discussed in a meeting between RP and NA to check codes were not missed as well as compare coding and interpretation in order to improve the trustworthiness of the analysis [45].

Results

Participants

We interviewed 43 participants: 21 adolescents and 22 parents. The age of the adolescents ranged between 12 and 17 years old, (mean 14.4 years) and the majority were female (16/21). Twenty mothers and 2 fathers participated, and one interview included both parents in a pair. Adolescents and parents were interviewed separately; however, in 4 adolescent interviews, a parent was present and mainly observed and provided support. Most interviews took place in participants own homes, one in hospital and lasted between 14 and 42 min (median 25 min).

Outcome card ranking (Table 2)

Four outcomes were consistently ranked as important by adolescents and their parents and included in the final conceptual framework: ‘symptoms’, ‘tiredness’, ‘payback and crashing’ and ‘activities and hobbies’. ‘School’ and ‘future’ were ranked highly by adolescents. ‘Mood’ and ‘how your child feels about him/herself’ were ranked highly by parents and are also included in the final framework. Parents generally felt that their child’s health was most important to improve and that participation in school would follow (Fig. 2). Parents were more likely than their child to refer to the psychological impact of the condition. They highlighted how their child had changed from outgoing to quiet, had mood swings and experienced anxiety and depression.

Table 2 Top ranked outcome cards (adolescents and parents)
Full size table
Fig. 2
figure 2

Outcome card ranking most important at top (girl 15 left, mother right)

Full size image

Differences between subgroups: age and gender

Younger adolescents (12–13 years of age) ranked ‘family’ higher than older adolescents and referred to activities such as playing with toys or outside (Table 3). Older adolescents talked about leisure activities such as the cinema and 16–17-year-olds referred to wanting to be able to drive, get a job and babysit. Fourteen to 15-year-olds ranked ‘school ‘and their ‘future’ highly and referred to not being able to complete their General Certificate of Secondary Education (GCSEs) and the resulting impact of not being able to go to university or meet their full potential.

Table 3 Top ranked outcome cards (age groups)
Full size table

Girls ranked ‘symptoms’ higher than boys; boys ranked ‘activities and hobbies’ as more important (Table 4). Boys talked more about computer games and sports. Girls additionally valued sports but also described activities such as cooking, baking and embroidery. Boys more often referred to being ‘frustrated’ at not being able to do what they wanted to do. This range of activities is included in the dimensions as part of the final conceptual framework (Fig. 3). ‘Frustration’ is also included as a dimension of mood. Few children selected the ‘independence’ card or talked about not being able to do things without their parents. As a result, these aspects are not included in the final conceptual framework.

Table 4 Top ranked outcome cards by gender
Full size table
Fig. 3
figure 3

Conceptual framework of paediatric CFS/ME HRQoL

Full size image

Conceptual framework of paediatric CFS/ME HRQoL

The final paediatric CFS/ME HRQoL conceptual framework includes 4 domains and 11 subdomains (Fig. 3). All domains and dimensions are described below with supporting quotes (Table 5) to illustrate how the final dimensions are derived from qualitative themes.

Table 5 Qualitative quotes from children and parents for each domain of the conceptual framework
Full size table

Adolescents and parents described a range of symptoms: sleep problems, tiredness, cognitive symptoms and individual physical symptoms.

  1. (1)

    Sleep reversed sleep (sleeping in the day and not at night), problems getting to sleep, waking up in the night and trouble waking up in the morning- feeling “dead”.

  2. (2)

    Tiredness feeling “worn out”, “heavy”, “weak” and “drained”. Tiredness was often constant and could get worse at times. Only doing activities for a short time, suddenly needing to stop/sit down, being very sedentary.

  3. (3)

    Problems concentrating “Brain fog”- being unable to think, forgetting things, unable to concentrate enough in school, to read or do homework.

  4. (4)

    Individual physical symptoms pain all over the body or specific locations (head, throat, arms, legs, back). Pain, dizziness, nausea and problems eating were more frequent on worse days and very debilitating.

The presentation, impact and priorities for improvement of symptoms in adolescents were complex. Adolescents wanted to improve how severe and frequent symptoms were and the interference with their daily lives. Some adolescents specified being able to cope with mild constant symptoms but not with more severe symptoms. For some, tiredness was most important to improve whereas others wanted to get rid of headaches or nausea. All participants wanted to reduce the fluctuation of the symptoms (more good days) and the how symptoms often got worse after activity (payback).

Physical functioning

Adolescents and their parents described how symptoms interfered with their ability ‘to do things’ and participate in daily life. They referred to a range of activities they found problematic: self-care, movement, going up and down stairs, walking, sport and managing several activities in one day. Some adolescents could only do activities for short durations and often did not leave the house. Parents accommodated adolescents to help conserve their energy, always having to plan family outings, alter or stop plans altogether. Differences between those with mild versus moderate severity were demonstrated in the qualitative data. Moderately affected adolescents reflected on problems with self-care (washing, dressing) and frequently needed help from their parents. They also described being too tired to walk and many were using a wheelchair when on outings out of the house. Some of those who were mildly affected talked about problems managing several activities in a day.

Social functioning

Education and social life were disrupted for all adolescents in this study. Time at school had been reduced and they reported not being able to go out, stay out for long and thus missing out on hobbies and social events.

  1. (1)

    Participation in school/college: Reduced attendance and needing to take extra breaks. The concentration and attention required in school often led to adolescents becoming very tired, falling behind with their work.

  2. (2)

    Participation in leisure activities and social life: Stopped/reduced sport, hobbies or leisure activities. Younger adolescents (12–13 years of age) reported restrictions to ‘play’ with siblings, toys at home or outside in their free time. Older adolescents reported not being able to go into town, shopping or to the cinema. Adolescents often became isolated as they no longer saw friends at school, through sleepovers or parties and there was a lack of understanding from friends. Parents commented that family life revolved around the ill child’s restrictions and mood. Adolescents were unable to participate in family activities (parties, long walks, going to the cinema).

Psychological wellbeing

All adolescents in this study described an impact on their psychological wellbeing due to symptoms limiting their usual activities, with friends and family, ranging from: low mood, frustration, feeling anxious and lacking confidence.

  1. (1)

    Mood feeling “down”, “miserable” and “upset”. Parents described “mood swings”, “tantrum” and “agitated”. Being unable to do things contributed to frustration (more often expressed by males). Adolescents often conveyed a lack of enjoyment of activities and fluctuation and payback led to feelings of hopelessness.

  2. (2)

    Worry and anxiety general anxiety and specific worries relating to: fluctuating symptoms- not knowing when symptoms will be worse, activity causing payback and being unable to cope with symptoms in social situations. Stress and anxiety in relation to GCSE’s and being unable to fulfil their potential. Adolescents often worried about the future (more evidence in 14–17 year olds) being left behind, going to university, getting a job or being able to live independently.

  3. (3)

    Self-esteem, withdrawal: staying at home, not going out with friends and wanting to be alone. Lack of confidence in carrying out activities they did previously.

Discussion

This study includes two novel aspects: the use of a child-centred interactive card ranking technique and the resulting conceptual framework comprising the most important aspects of HRQoL for adolescents with CFS/ME. This outlines four outcome domains and 11 subdomains: sleep, tiredness, problems concentrating, individual symptoms, fluctuation and payback, daily and general activities, participation in school, leisure and social life, mood, anxiety and self-esteem. This conceptual framework forms the foundation of a patient-centred PROM for paediatric CFS/ME.

Strengths and weaknesses

This was a large qualitative sample (n = 43) with adolescents recruited across the age range: 9 (12–13-year-olds), 6 (14–15-year-olds) and 6 (16–17-year-olds) to ensure the conceptual framework is representative of the range of experience. Adolescents were mostly interviewed alone (17/21) consistent with international guidance [18] and may have reduced socially desirable answers [38, 46, 47]. Parents provided an important complementary source of information incorporating their perspective in the conceptual framework [48,49,50,51]. However, patients were recruited from only one service and fewer males and moderately affected patients were recruited, which reduced opportunities to explore differences they might experience [26]. We did not sample for all patient characteristics such as socioeconomic status (SES) and ethnicity; therefore, further research is needed to identify if there are any additional HRQoL issues for different SES and ethnic minority groups.

This study employed an interactive child friendly technique following the increasing trend to use more innovative methods to understand patient experiences such as drawing, body mapping and ranking/rating [27, 28, 32, 52, 53]. The card ranking exercise, developed with feedback from adolescents, facilitated an understanding of adolescent’s priorities and preferences, and those of their parents, whilst providing a framework for deeper discussion within an interview [26]. Adolescents were in control as they ranked outcomes and this may have reduced interviewer effects [54]. The pre-defined outcome cards may have limited the domains expressed spontaneously by participants [55], however; those were carefully selected from previous extensive qualitative work and despite the option of blank cards, no new domains were added. The qualitative and quantitative data produced and the use of constant comparison helped identify differences between subgroups, such as specific problems with self-care in moderately affected adolescents [43].

Results in context with previous literature

The domains outlined in the conceptual framework as part of this study are similar to those found in generic HRQoL models [56,57,58] and consistent with outcomes measured in clinical trials: pain, fatigue, physical functioning, social role participation and emotional distress [59]. This study also highlighted key aspects described by adolescents with CFS/ME: sleep problems, problems waking, fluctuating symptoms, payback (feeling tired after activity) and problems with the sustaining activities (e.g. being out all day). Parents ranked ‘mood’ highly which has been suggested as a predisposing and perpetuating factor paediatric CFS/ME [60]. This study revealed the specific types of anxieties adolescents with CFS/ME have: worries about making symptoms worse after activity or coping with symptoms in social situations. Separation anxiety and social phobia have been reported as the most elevated anxiety types in paediatric CFS/ME [13].

‘Symptoms’ were ranked as most important to improve by adolescents and parents, however, the individual experience of symptoms and which were most important to improve, varied between participants, consistent with the heterogeneity found in CFS/ME [61, 62]. Adolescents in this study also ranked ‘school’ and ‘activities and hobbies’ highly. Adolescents with CFS/ME miss an average 1 year of schooling [9] and 90% quit their hobbies [63]. This study also demonstrated the different activities important to different age groups as well as older adolescents worrying more about the future. This is consistent with the unique social and emotional aspects for different age groups [31, 64, 65]. It is advocated that adolescent outcome measures should address the importance of separation from parents [31, 66]. However, few adolescents in this study selected and rated independence from parents highly. This may be because adolescents with CFS/ME do not necessarily lose their independence completely, as they can often see friends depending on a ‘good or bad day’ or restrict the amount of time. As a result, complete ‘independence’ may be less important in this condition. This may be different for adolescents who are severely affected.

Conclusions

The conceptual framework developed as part of this study defines the health outcomes domains that should be measured by a new paediatric HRQoL CFS/ME PROM. Adolescents describe specific impacts of symptoms (e.g. fluctuation and payback), activity (e.g. suddenly limited, problems with duration) and worries about causing payback or coping in social situations. These dimensions are not currently captured in PROMS used in paediatric CFS/ME and should be used to develop new questionnaire items [14]. When HRQoL PROMs include what matters to patients and this is reflected in healthcare, this can improve patient–clinician communication and patient adherence to treatment and promote shared decision-making for a better therapeutic alliance [67,68,69,70].