Development of an Ehealth and GP Assisted Programme for Persistent Physical Symptoms: Qualitative Exploration of Users’ Needs

play a in both assessment and management of PPS. However, a indicates GP and ways to in their to an and GP assisted for in


Background
Patients with persistent physical symptoms (PPS) such as pain, dizziness and fatigue visit their general practitioner (GP), medical specialists and emergency departments twice as often as other patients. The frequent attendance ranks their medical costs among the highest of all patient groups (1,2). The prevalence of PPS in adults is estimated to be 10% in the general population according to standardised interviews (3) and 17% in general practice according to questionnaire screening (4). In general, PPS are associated with high morbidity, distress to patients and their families, high healthcare costs (5) and high risk of work capacity loss (6).
PPS refers to the experience of bodily sensations perceived as possible signs of disease and consequently lead the patient to contact the healthcare system. PPS represent a spectrum of severity ranging from mild symptoms over moderate conditions to severe and chronic disorders (7).
Often the term medically unexplained symptoms (MUS) is used, because the majority of PPS are based solely on subjective complaints and not on veri able clinical or para-clinical ndings (8). However, this term is arguably misleading according to newer research, and it is unhelpful to patients with PPS (9,10). Instead, a recent survey among the general population argues that the term persistent physical symptoms may be the most appropriate and acceptable term (11). Consequently, we have chosen to use the term PPS as an umbrella term including MUS and similar terms and excluding recent-onset symptoms.
General practice is the rst point of entry to healthcare, and hence GPs act as frontline in the healthcare system and play a central role in both diagnostic assessment and management of PPS (12). A qualitative study of GP consultations on PPS showed that patients reported a mismatch between their GP's and their own agenda. Patients indicated that the GP evoked an uncomfortable feeling in them during the consultation and did not provide a speci c management plan for their symptoms. In general, patients found the GP inadequately prepared for the consultation, and they perceived prejudices in the GP during the consultation (13). In line with this, GPs often experience this type of patients as di cult to treat (14), and negative attitudes towards patients with PPS are prevalent among GPs. (15). A meta-synthesis describing challenges faced by GPs when managing patients with PPS found that GPs struggle with the incongruence between patients' symptom presentations and the explanatory models for biomedical disease (16). Likewise, Rasmussen and Bø conclude that the bio-psycho-social frame helps GPs understand and handle PPS better than the biomedical frame does (17). The lack of treatment strategies for patients with PPS is a problem as a growing body of literature highlights that improved early management of PPS in general practice could reduce the burden on patients, clinicians and society (18). eHealth refers to health services or information provided by the internet and related technologies (19). eHealth interventions have shown to be effective in promoting healthy behaviour, and their effectivity may be further improved by integrating self-regulation techniques (e.g. personal goal setting and coping strategies). In doing so, eHealth may assume some of the tasks of GPs prompting them to address selfmanagement in the consultation (20). Online self-help interventions designed to be conducted independently of specialised healthcare might be a new way of overcoming the challenges and form a valuable addition to current treatment options for patients with PPS in general practice (21). However, there is a lack of knowledge about how such eHealth tools should be designed to meet the needs of both GPs and patients. Studies have shown that eHealth programmes, which have not involved GPs in the development process, are not readily accepted and are not properly integrated in routine care.
Additionally, one of the major challenges of eHealth is to make it bene cial and easy to use for healthcare providers and patients; otherwise neither professionals nor patients will use it (22).
Given the challenges within PPS and the potential of empowering patients with self-help approaches, we wish to develop an eHealth and GP assisted self-help programme for these patients in general practice. The overall aim with the research programme is to improve health and quality of life for adult patients with PPS. The present study is the rst phase of a larger four-phased mixed method study. Our speci c aim with this rst study was to identify the GPs' and patients' needs related to self-management of PPS. Identi ed key factors will subsequently be included in the development of an eHealth treatment programme.

Design
We conducted a qualitative study by performing eld studies in general practice and semi-structured individual interviews with GPs and patients with PPS. The COM-B model, a framework that enables a systematic evaluation of behavioural aspects, guided our interviews and analyses. The COM-B model proposes three interacting conditions for behaviour change to occur: Capability, Opportunity and Motivation (23). Capability is de ned as the individuals' psychological and physical capacity to engage in the activity concerned. It includes having the necessary knowledge and skills. Opportunity is de ned as all factors outside the individual that enable the behaviour or prompt it. Motivation is de ned as all those brain processes that direct behaviour, e.g. goals, conscious and analytical decision-making, habitual processes and emotional responding. A given intervention might change one or more of these components.

Recruitment
To include GPs who varied in age, gender and level of interest in PPS and eHealth, we adopted a purposeful sampling strategy (24) including GPs with high/low interest in PPS, high/low interest in eHealth, GPs of different sex and age and GPs with different practice types. Firstly, GPs were recruited and informed about the study. Secondly, the patients were recruited by the GPs if they ful lled the inclusion criteria listed in table 1. The patients could be undergoing diagnostic workup concurrently. If the patients ful lled the inclusion criteria, they were given an information letter about the study and what participation entailed. Furthermore, all patients were asked to complete a validated symptom checklist: the 25-item Bodily Distress Symptoms (BDS) checklist to identify relevant cases and to obtain further information on the patients' symptoms (25). The 25-item BDS checklist is developed and validated as a self-report instrument that can be used for case nding in both clinical practice and in research. It was previously developed on the basis of large sample of patients from primary and from neurological and internal medicine departments. It contains questions within three symptom groups; a cardiopulmonary group, a gastrointestinal group, a musculoskeletal group and a general symptom group.
The checklist opens with the question: 'Have you in the last four weeks been bothered by' followed by a list of symptoms. The BDS checklist has a ve-point response scale for each listed symptom ranging from 0 ('not at all') to 4 ('a lot') (25). A patient was included in the study if 4 or more symptoms were scored at or above 2.
The patients were contacted by phone by the rst author and were informed verbally about the study prior to accepting participation. If they agreed to participate, a time for interview was scheduled.

Data collection procedure
Field studies Prior to the individual semi-structured interviews, a total of 7 days of eld studies were conducted in the four included general practices. The aim was to make the GPs familiar with the study and to inform them about the patients we wished to include in the study and the recruitment process. Furthermore, the eld studies were conducted to achieve a clear understanding of the eld to be able to develop an interview guide. Field notes were made during the eldstudies.

Interview guides
The interview guides were developed in an iterative process by the research team and tested by performing pilot interviews with one GP and one patient who were not enrolled in the study. The interview guides included general questions related to aspects of capability, opportunity and motivation among both GPs and patients as well as the participants' general experiences of persistent physical symptoms and their attitude towards using an eHealth programme for this purpose.

Semi-structured interviews
The rst author (PRJA), a female researcher with a PhD in health science, conducted the eld studies and semi-structured interviews. All interviews were initiated with a presentation of the study and the researcher's background and interest in the research topic. The GPs were interviewed in their practice, and the patients were interviewed in their homes (n=13) or at their workplace (n=1). In all interview situations, only the participant and the interviewer were present in a quiet and uninterrupted environment. No repeated interviews were carried out. In the GP interviews, the focus was on how they experienced and managed patients with PPS based on capability, opportunity and motivation to identify barriers and enablers in providing self-help through an eHealth intervention. In the patient interviews, the focus was on describing their symptoms and their experiences of having PPS, again based on capability, opportunity and motivation to identify barriers and enablers in promoting self-management of PPS. The interviews lasted 37-87 minutes with an average length of 54 min.
Analytical procedure All interviews were recorded and transcribed verbatim and additional eld notes were written down immediately after each interview. Data were subject to a thematic analysis by the research team, guided by Braun and Clarke's ve step framework (26) (table 2). Analysis was done inductively and separately for patients and GPs in an iterative process. Finally, the identi ed themes were mapped to the COM-B framework to allow translation from GPs and patients needs to development of the eHealth intervention (23). Saturation was reached since no new themes and codes were found during the last coding process. Step Description

Participant characteristics
During January and February 2019, a total of six GPs from four general practices were recruited. They all agreed to participate. Four GPs were female and two were male. The age of the GPs ranged from 46 to 62 years (mean 52.8). All participating general practices were either partnership or group of solo practices. One practice was rural and three were in urban areas.
From January to March 2019 16 patients were asked to participate in the study. Two patients were excluded due to severe mental disorders. In total 14 patients were recruited (table 3). Ten patients were female and four were male. The age ranged from 20 to 58 years (mean 35). The BDS score ranged from 5 to 25 (mean 13.4) and the duration of symptoms from 2.5 to 108 months (9 years) (mean 29 months). Ten patients were employed, one was unemployed and three were on sick leave. The analysis identi ed 11 associated subthemes in the patient interviews and 7 subthemes in the GP interviews that were mapped into the three different COM-B components (table 4). In the following, we will describe the identi ed themes rstly in the patient interviews and secondly in the GP interviews. The patient interviews provided information about how the patients experienced PPS. The three domains in the COM-B model were used to identify patients' needs in relation to self-management of their symptoms.

Theme 1: Capability
Capability was explored with patients to establish whether they felt they were physically and psychologically capable in self-management of their PPS.

Physical capability
A jungle of management strategies The patients in general described a lack of skills and knowledge about how to manage their symptoms. They managed their symptoms in many different ways by making use of various offers in the community. Ten of the patients were seeing a psychologist due to psychological problems. One had tried metacognitive therapy. Some used a physiotherapist, a masseur, an osteopath, an acupuncturist or a chiropractor when they experienced muscular skeletal pain or symptoms. Others participated in mindfulness or yoga sessions or exercised in hot water.
They all described a jungle of information and asked for speci c recommendations in relation to their own symptoms. They experienced that the guidance from GP wasn't su cient and did not meet their needs.

Psychological capability
Understanding causes of PPS Psychological stress was often presented as a starting point for the symptoms. Stress could be caused by different circumstances in the participants' life and be present for a long period of time with subsiding and recurring symptoms before PPS occurred, e.g. triggered by stress at work. However, when the symptoms occurred, the patients were unsure whether they were caused by stress or by physical factors. This was not further elaborated on in the consultation, and the patients were often referred to physical investigations at the hospital as described by this participant: "It was a tough period for me. Small kids and busy at work. Suddenly, I felt like my legs could not carry me anymore. My GP referred me to the hospital, but the neurologist could not nd any explanation. Later I realised myself the causality between stress and physical symptoms" (P6, female, aged 37) Other patients described that their present symptoms might be a result of severe life events such as divorce (P5), childbirth (P12), car accident (P9), mid-life crisis (P8), establishing a family (P10, P14), unbalanced body (P2), concerns about the future (P4, P13) or stress in general (P6, P7). However, this was not broached during the consultation since the focus was on the physical aspect of the symptoms, which is described as an admission ticket to visit the GP. Participant 14 suggested that if the GP had been more focused on how he was doing in general instead of only on the stomach symptoms, it would have been more helpful in identifying the root cause of the symptoms: Dealing with uncertainty Nine of the participants described uncertainty about whether a serious health condition might explain the symptoms and had thoughts about not being examined su ciently. As this participant, a woman aged 32 with persistent low back pain, described: "I really need some more physical investigations to be sure that nothing severe is wrong with me, and if they show that everything is ne, we can take it from there" (P9, female, aged 32) If the patients were referred to specialists, the uncertainty was more about whether the doctor was thorough enough as described by this participant, a woman who was referred to a gynaecologist because of persistent symptoms from her stomach: "I keep on thinking about whether the gynaecologist was thorough enough. He was very busy that day, and there were a lot of patients in the waiting room" (P11, female, aged 38).
Even though some of the patients believed that an underlying disease was not the cause of their symptoms, they required the physical investigations to be on the safe side prior to bringing other explanations such as stress into the consultations. To some patients, the more physical investigations, the better. Some of the patients described the ambivalence between being on the safe side and being worried when they were referred to physical investigations.

Timing in bringing PPS into play
When the patients were asked if they would like to participate in the study, the GP had explained about PPS, and for some patients it was the rst time that they were introduced to this focus. This was an eyeopener for some of the patients, and they argued that it would have been helpful if PPS had been provided as an alternative explanation during concurrent physical investigations as suggested by this participant: "My GP could have told me that of course we have to investigate whether it is a critical condition but maybe you should just learn to manage your symptoms and understand what might cause them" (P8, female, 58 years).
By bringing PPS into play earlier, the patients thought that they would be more willing to accept it as a reasonable explanation if physical investigations showed no ndings and explanations. However, the timing was important since some of the patients argued that if the GP had introduced PPS as a possible cause the rst time they showed up in the consultation and prior to physical investigations, they would not have accepted it, the reason being that the patients need to be sure that nothing serious is wrong with their body. As the same participant explained: "If it had been introduced to me the rst time I had symptoms from my heart in connection with my father's death of a heart attack, I do not think that I would have been prepared to accept it. At that time, I really needed to be sure that there was nothing wrong with my heart" (P8, female, aged 58)

Theme 2: Opportunity
Physical opportunity was explored with PPS patients to understand how the environment, resources and the locations in general practice affected their self-management of PPS. Social opportunity was explored to better understand how the cultural norms and the interpersonal relations affected the experience and self-management of PPS.

Physical opportunity
One consultation -one symptom Visiting the GP with diffuse persistent symptoms was a challenge because of the time available in the consultation. A GP has in average 10 minutes for each patient, and the consultation is structured in a way so that the patient is supposed to address only one symptom per consultation. This was a challenge to patients with PPS since most of the patients explored diffuse, multiple and vague symptoms, which they found time-demanding to explain to their GP.

Social opportunity
Shift in social roles due to PPS Patients described how their role in relation to their social network and families had changed after the symptoms started and described a feeling of guilt since they were not able to be the friend, mother/father or wife/husband they used to be.
"And I have to be there for my little son, but when I get dizzy, I have to go into the bedroom and relax, and I really think that it is hard to accept" (P14, male aged 33) Some experienced social isolation since they only to a limited extent were able to participate in normal activities with their friends and families due to their symptoms.
Patients described how they had delegated different daily activities to other family members. As an example: P1 had asked her husband to prepare the meal when they invited guests, P2 asked her son to do the laundry, and P3 asked her boyfriend to do the cleaning.
The GP as a discussion partner Overall, the patients voiced a good relationship with their GP, and ve patients described that they used their GP as a discussion partner, like P2: "It helps to visit my GP even though we only talk together" (P2, female, aged 42) Some of the patients had a high degree of trust in their GP and trusted their ability to deal with a possible serious disease.

Theme 3: Motivation
Various factors were explored in relation to motivation for self-management among patients with PPS through an eHealth intervention. Central subthemes were as follows: (1) Guidance from the GP, (2) need for hope, (3) tracking of symptoms and (4) easy access to self-help strategies.
Guidance from the GP Guidance in how to manage the symptoms, especially by the GP, was a key motivator for most of the patients. Patients felt that being offered an eHealth programme by their GP would increase their motivation for using it, but even though it was meant to be a self-help programme, patients wished the GP to guide their activity in the programme and follow up on their activity. Some patients found the follow-up important (P5, P8 and P10).
"I think it could be nice if you could get a sort of homework in an online programme and then your GP would follow up on whether you have done it or not" (P10, female, aged 20) Overall, the patients underlined the importance of the self-help programme being a part of a continuity of care in collaboration with their GP rather than an alternative to GP visits.
Need for hope Some patients described a feeling of hopelessness concerning PPS. They worried about their future and whether they would be able to work, take care of their family and whether the symptoms would persist for their entire life. They felt that the GP and other healthcare professionals focused overly on acceptance and learning to live with symptoms. P2 and P9 were directly told by their GP that the symptoms would be chronic. They described a need for seeing other patients with PPS and how they successfully managed their daily life as described by participant 3: "It could be really helpful if you could get a feeling of not being alone by talking to or seeing others in the same situation and see how they manage their symptoms" (P3, female, aged 23) Six patients suggested a forum where they could meet peers or some videos or case stories where the patients could see how people like themselves have managed PPS. It was important to the patients that the focus was on hope and positive experiences and strategies.

Tracking of symptoms
The patients found it very helpful to ll in the BDS checklist prior to the interviews, since it gave them a visible and precise overview of their symptoms. They suggested that the BDS checklist could be used to track their symptoms and provide them with a continuous status of their symptoms as described by P9: "It is really di cult to tell how you are doing from day to day. Therefore, it is really helpful to have this checklist to monitor if you are going in the right direction" (P9, woman aged 32) One patient (P6) suggested that the BDS checklist could be lled in prior to seeing their GP. This would help patients to better prepare for the consultation and participate in treatment decision-making. P5 stated that having the opportunity to ll in the BDS checklist at home could be a way of 'parking the symptoms in storage' and then take them out when visiting the GP. According to P5, this might help to not 'having to' sense the symptoms all the time out of fear of not being able to remember all the symptoms when presenting them to the GP.

Easy access to self-help strategies
Generally, the patients seemed motivated to self-manage their symptoms, but since PPS were not discussed as a reasonable explanation for the symptoms in the consultation, it was not easy to talk about self-management strategies with their GP. They proclaimed that being prescribed a self-help programme by their GP would be a safe and easy way to increase their self-management of PPS since they had a high degree of trust in their GP and thereby trust in the content in the program.
Identi ed themes in the GP interviews: GP interviews provided information of PPS from the healthcare professionals' perspective in general practice. The three domains in the COM-B model were used to identify GP needs in relation to support patients with PPS to self-manage their symptoms.

Theme 1: Capability
Capability was explored with GPs to establish whether they felt they were physically and psychologically capable to support patients with PPS in self-management of their symptoms.
Physical capability

Missing biomedical reassurance
The six GPs participating in the study agreed that their roles as a GP are to diagnose, treat and reassure the patients in the consultation. During eld studies, we observed that the GPs often conducted a simple screening of their patients in the consultation by listening to the heart and lungs with the stethoscope and making referrals to blood tests. However, when it came to patients with PPS, simple screening with the traditional biomedical methods did not reveal anything, and therefore the GPs were challenged when applying solely a biomedical approach to the patients. Hence, the patients were often referred to further tests at the hospital, even though the GP was quite sure that everything was normal: "And I have to admit that sometimes you refer the patient to further tests at the hospital even though you have a clear expectation that everything is normal. But when blood test and scanning con rm this, the patient is reassured to a higher extent" (GP 2).
However, when symptoms persist, negative test results may prompt further healthcare seeking and an Odyssey of referrals and investigations. Most of the GPs were aware of this dilemma and re ected on the challenges of the missing biomedical reassurance to patients with PPS and claimed that they really needed a valid investigation and diagnostic tool to help the patients understand and accept the presence of PPS.

Psychological capability
Dealing with uncertainty Diagnostic uncertainty is well known in general practice, and all the GPs described this problem. As an example, GP1 described that she always conducted some biomedical tests to be sure that the symptoms were not explained by any physical condition prior to introducing PPS as a possible diagnosis: "I do not feel a huge degree of uncertainty when it comes to patients with PPS. However, we have to hedge our bets, because we are so afraid of missing severe illness in the patients. Therefore I always start with biomedical investigations" (GP1) The GPs described different opinions about expressing the uncertainty to the patients. GP2 argued that she sometimes expressed her thoughts to the patients when she referred the patient to physical investigations at the hospital. She could sometimes tell the patient that she did not expect any critical ndings and that the investigations were just conducted to be sure that no underlying disease was causing the symptoms. GP5 argued that he was dealing with the diagnostic uncertainty by sometimes expressing his uncertainty this way: "When you have been a GP for almost 20 years, you have a certain experience and then you are coping with diagnostic uncertainty and you are also able to express your uncertainty to the patient. I often say to these patients that I am not sure what is causing your symptoms, but I am sure based on the ndings that it is nothing severe" (GP5) Timing in bringing PPS into play All GPs agreed upon the importance of bringing PPS into play in the consultation as early as possible to avoid unnecessary investigations, and they described that they often considered alternative diagnoses early in the consultation when the patients presented symptoms which were not easily diagnosed.
However, the GPs in general experienced di culties in early introduction of their considerations and the concepts of PPS. Some described a barrier related to worries about how the patient would receive the message and whether the patient would accept the term PPS (GP1).

Theme 2: Opportunity
Physical opportunity was explored with the GPs to better understand how the environment, resources and the locations in general practice affected their support of patients with PPS in self-management of their symptoms. Social opportunity was explored to better understand how the cultural norms and the social cues affected the way GPs think about and manage patients with PPS.

Physical opportunity
Lack of resources The GPs described that consultations with patients with PPS in general are time-consuming, and often patients present a variety of multiple, vague and diffuse symptoms not tting into the construction of a consultation concentrating on one or few symptoms. Hence, the GPs described that they need more time and more concrete and easy accessible tools in the management of patients with PPS.
"I think that I omit patients with PPS because we have only few options to help these patients. And because of that I feel much more hesitant" (GP6) Being a GP without effective tools to diagnose, treat and reassure patients with PPS was described as frustrating and time-consuming, since it often leads to less effective consultations as described by GP2: "As a GP you really want to diagnose and reassure the patient satisfactorily, since they come in my practice with the expectation that I can tell them what is the matter with them and give them the needed treatment. However, when it comes to patients with PPS, I neither have the time nor the tools to help them su ciently" (GP2) GP6 described that he often manages patients with PPS by talking to them regularly and offers them continuous follow-up appointments in the absence of better treatment strategies: "I think that the only way to manage those patients is to offer them some follow-up appointments and just talk with them" GP6

Social opportunity
Cultural norms in the healthcare system GPs described a general negative attitude and feeling of fatigue towards patients with PPS in their colleagues. GP 4 described an attitude in the specialised healthcare system that patients with PPS are not their business and that the responsibility lies with the GPs. However, they do not collaborate with the GPs but start an Odyssey of biomedical investigations and treatments rather than bringing PPS into play.
Furthermore, GP4 requested a better collaboration with the specialised departments at the hospitals and expressed a strong wish that the departments would bring PPS into play rather than keeping referring the patients to more examinations: "I think it is a huge problem that we keep examining the patients just because none of us are able to say that there might not be a physical explanation of their symptoms. And instead the patients become more and more nervous because we all keep searching for an answer that is not there" (GP4)

Theme 3: Motivation
Various factors were explored in relation to motivation for supporting patients with PPS to self-manage their symptoms through eHealth. Central subthemes were as follows: (1) Fewer and more bene cial consultations and (2) better prepared consultations.
Fewer and more bene cial consultations Combined with the frustration about the lack of time and missing diagnostic and management tools, the GPs wished that the patients would show up less frequently in the practice. The frequent consultations with no strategy generated a feeling of insu ciency in the GPs. Therefore, the GPs were highly motivated by the thought of having easy access to self-management tools for the patients, potentially resulting in fewer consultations.
The GPs agreed that an online self-help programme could be a great supplement to the face-to-face encounter with the patient.
Better prepared consultations Some GPs mentioned that an online self-help programme with a symptom screening could result in the patient being better prepared for the consultation as the GPs could use the data provided by the programme as a starting point for the dialogue with the patient. In contrast, the GPs explained how they often use half of the consultation to create an overview of the patients' symptoms. The GPs were generally worried about the increasing work load in general practice. Thus, they saw great potential in online symptom screening prior to the consultation allowing for more time for dialogue in the consultation, which was a high priority for the GPs. This was seen as improved quality of the consultation and thereby improved patient treatment.

Discussion
This study addresses both patient and GP needs when it comes to improved self-management of PPS.
The ndings highlight the need for certain aspects of capability, opportunity and motivation to be addressed in the design and development of an eHealth self-help programme for patients with PPS prior to the implementation in general practice. The main themes from patients and GPs corresponded and focused on understanding and explanation of symptoms, timing of this, uncertainty and consultation premises. Furthermore, important motivational elements for an eHealth programme were identi edespecially the need for hope was emphasised.
Regarding capability, it was apparent that there was a gap between patients' and GPs' explanatory models in the consultation. Even though the patients clearly had concerns regarding severe diseases, most of them were aware that other factors could explain their symptoms as documented in an earlier study (27). However, they presented physical symptoms as the starting point in the consultation, since this was considered an admission ticket to the GP. The GPs then tended to use a biomedical approach to rule out serious disease and sometimes referred patients to physical investigations in secondary care using a defensive biomedical approach. A consequence of this dynamic seemed to be that psychological and social factors were rarely discussed in the initial consultations.
Studies have shown that patients need support and compassion, but GPs provide somatic screening and interventions (28,29). Salmon et al concluded that patients with somatoform disorders feel satis ed and empowered by medical explanations that are tangible and involving. Empowering explanations has the potential to improve these patients' well-being and help reduce the high demands they make on health services (30). For the GPs, participation in this study meant that they employed the term PPS to include patients. The patient inclusion therefore became a tool for the GP to initiate a conversation about PPS using a biopsychosocial frame, which several of the included patients found helpful.
This highlights the potential of bringing PPS into play earlier in the consultation process in an easy and understandable manner. Such an approach might empower the patients to engage in self-management of their symptoms using a future eHealth programme.
The main reason for not bringing PPS into play earlier in the consultation is uncertainty in both patients and GPs. "Dealing with uncertainty" and "timing in bringing PPS into play" are central subthemes in both patients and GPs interviews. The patients described worry and expressed that they wanted to be sure that their symptoms were not caused by severe disease. Likewise, the GPs were afraid of overlooking severe disease and agreed upon the importance of conducting biomedical investigations rst and foremost prior to talking about differential diagnoses such as PPS. In addition, a study has showed that GPs face di culties recognising and labelling PPS. Even when there is no indication of a somatic problem, GPs still experience uncertainty and fear of overlooking a serious disease (31). Alam et al claim in a systematic critical review that diagnostic uncertainty is greatest in general practice due to the undifferentiated symptoms GPs are often presented with and is a routine inevitability (32).
Our ndings on capability show a need among both patients and GPs to acquire improved knowledge and skills with regard to their understanding of PPS, communicating about PPS and dealing with uncertainty. The GPs in our study argued that they did not communicate about PPS as they did not have any valid treatment if PPS was the diagnosis. This was a main barrier among the GPs as they perceive that their role as a GP is to diagnose, treat and reassure the patients. Henningsen states that good communication with the patients is essential at all stages of PPS and levels of care, including reassurance, anticipation of likely outcomes of diagnostic tests, positive explanations of the functional character of the disorder and motivation of the patient to actively engage in the coping with bodily distress (14). Since PPS is not communicated su ciently in the consultation, the patients in our study sought alternative management strategies than those provided by their GP. Therefore, many patients felt lost in a jungle of management strategies and various investigations when they really are in need of GP assisted self-help strategies.
Under the theme "Opportunity","Lack of resources" was an issue for GPs and especially when it came to patients with PPS as these consultations were found to be time-consuming. Furthermore, the GPs described cultural norms in the healthcare system causing a general feeling of frustration towards patients with PPS. This results in patients with PPS being referred to examinations continuously in search of the missing biomedical explanation leaving the patients in 'no mans land'. The patients described that they often used their GP as a partner of discussion. However, they were also aware of, and to some degree restrained by, the nite resources in general practice and requested more time to just talk to the GP to be able to discus the symptoms more broadly.  (13). However, our study pointed out that GPs are reluctant to share their diagnostic uncertainty with the patient.
When exploring opportunity issues with the patients, it became clear that the patients needed support as to how to involve their social network in the management of PPS since they experienced a shift in social roles due to PPS. They described that they missed seeing peers and other patients to get motivated on how to manage PPS in their daily life. The importance of social network for good health is also described by Holt-Lunstad, Smith and Layton (33). The patients also expressed a great need for hope. By having the opportunity to follow other patients' good examples, the patients thought that they could gain more hope for the future and become more motivated for self-management of their persistent symptoms.
Previous research has examined that self-help is associated with a reduction in symptom severity and improvement of quality of life among patients with MUS (21). Therefore, we discussed motivational factors relevant to an eHealth and GP assisted self-help program for persistent symptoms with both the patients and the GPs. In addition to need for hope, guidance from the GP, tracking of symptoms and easy access to self-help strategies were integral elements of the patient theme motivation. It became clear that the patients in general were motivated to self-manage their symptoms, but the GP would still have an important role in guiding the patients in the right direction. Despite face-to-face contact is essential to both patients and GPs, the GPs were motivated to have an eHealth program that could result in fewer and bene cial consultations and better prepared consultations. All motivational factors related to support the patients in self-management in their own environment to avoid the patients' dependence of continuous contact to the GP. This motivation is partly due to the well-known negative attitude of many GPs towards patients with PPS and the lack of effective management strategies in general practice (15

Strengths and weaknesses
To the best of our knowledge, this is the rst study to address patients' and GPs' needs prior to the development of an eHealth and GP assisted self-help program for PPS. The use of qualitative interviews in this study allowed in-depth understandings of patients' and GPs experiences and opinions on dealing with PPS to better understand their needs.
The use of the COM-B inspired interview guides and the COM-B as framework for the analysis facilitated the identi cation of themes that are to be addressed in future eHealth intervention for PPS and furthermore highlighted areas that need to be taken into consideration when eHealth programs are to be implemented in general practice. However, the majority of interventions guided by the COM -B model have   been targeted at lifestyle including diet and exercise behaviours and there are only few examples of using the COM-B model within the bio-psycho-social area (34). Despite this, we found the use of the COM-B model helpful to guide the inductive research process, which is necessary especially in the eld of PPS where a lot of different perceptions and experiences are explicated.
The study was conducted by a multidisciplinary research group (general practitioner, psychologist, public health researcher and health service researcher). Multidisciplinary and continuous discussions throughout the whole research process, from development of the interview guide to analysis and preparation of the paper helped to minimise the risk of our ndings being dominated by preconceptions.
In general, we think that the above-mentioned factors strengthened the study. There are however some weaknesses that need to be considered. The transferability of the study results may be hampered. GPs in the study were recruited by purposeful sampling based on one of the author's personal knowledge of the participating GPs. This relationship may have caused the participating GPs to express a more positive attitude to patients with PPS than they would have otherwise. Another limitation of the study was that many of the included patients had suffered from PPS for longer than the intended target group. Thus, their attitudes and experiences may differ from that of patients with more recent symptom onset. We wished to have more newly diagnosed patients in the study. However, we think that the patients in our study share the same experiences and needs as more newly diagnosed PPS patients, and hence we still think our ndings give a valid and an in-depth understanding of both GPs' and PPS patients' needs.

Future research
This is the rst phase of development and test of the eHealth intervention eASY. The next phase will aim at designing and developing an eHealth intervention in an iterative process. Further extensive testing of the feasibility, acceptability, satisfaction and usability of the eASY program from the patients' and GPs' perspective will be assessed prior to quantitative evaluation of the programme.

Conclusion
In conclusion, several identi ed needs emerged from the in-depth qualitative work with PPS patients and GPs. It is clear that an eHealth and GP assisted self-help program for patients with PPS is desirable if it meets the users´ needs. Key learnings include 1) a need to bring PPS into play early on in the consultation in order to take a more bio-psycho-social approach in the management of PPS, 2) patients and GPs need better skills to communicate about PPS and manage uncertainty and 3) patients need guidance from their GP in how to self-manage their PPS. Our design of a new eHealth and GP assisted self-help program for patients with PPS will target these identi ed key factors. We will use these ndings to design an eHealth and GP assisted self-help program for patients with PPS that targets the identi ed key factors. All participants were informed about the study and provided written consent prior to being enrolled in the study. According to Danish law, it is not necessary to apply for scienti c permission to conduct qualitative studies, as they are of a non-bio-medical character. The study was approved by the Danish Data Protection Agency.

Consent for publication
Not applicable Availability of data and materials The datasets generated during and/or analysed during the current study are not publicly available due to transcripts in Danish but are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.

Funding
Innovation Fund Denmark and Trygfonden fund the study. The funders are funding research projects that aim to design, develop and test new digital health solutions. They were not actively involved in the study.
Authors' contributions PRJA designed the study, collected the data, participated in the analysis and wrote the rst and subsequent drafts of the manuscript. MTR, LF, MR and JC participated in the design and data analysis, and commented on all drafts, including reading and approving the nal manuscript. The corresponding author attests that all listed authors meet authorship criteria and that no others meeting the criteria have been omitted. PRJA is the guarantor.