Developing a clinical pathway to identify and manage cognitive problems in Multiple Sclerosis: Qualitative findings from patients, family members, charity volunteers, clinicians and healthcare commissioners


 Background: Cognitive problems are a common and debilitating symptom of multiple sclerosis (MS). Screening and treatment for cognitive problems are recommended, however these are not routinely delivered in UK clinics.
 We collected and synthesised stakeholder perspectives to develop a care pathway for cognitive problems in MS and produce a logic model, illustrating how this pathway might operate.
 Methods: Forty-nine stakeholders, including people with MS and care providers, participated in semi-structured interviews and focus groups. Participants viewed information that illustrated how the pathway might work and provided feedback. Data, transcribed verbatim and analysed using Framework Analysis, were mapped onto a preliminary logic model and accompanying thematic framework.
 Results: The proposed pathway was perceived as helpful in providing standardised support for a neglected MS symptom. Training packages, online cognitive screening, and triaging decisions were viewed as crucial activities. Shared responsibility, a person-centred approach, and addressing the complexity of cognitive problems were important engagement mechanisms. Allocating time during clinic appointments and within staff workloads were essential resources for implementation.
 Conclusion: Our co-constructed MS cognitive screening and management pathway will be evaluated for clinical and cost-effectiveness in a trial. However, in the interim, clinicians can adapt and implement this pathway in their own services and evaluate it locally.


• No established care pathway exists for screening and managing cognitive problems. • Based on stakeholder interviews, we developed a logic model for the pathway. • The logic model illustrates how a new clinical care pathway could work.
• To work, the pathway relies on shared responsibility and a person-centred approach.

Introduction
Cognitive problems affect up to 70% of people with MS (pwMS) (1) and can negatively impact quality of life and vocational activities (2)(3)(4). Consequently, routine screening and management for cognitive problems in MS has been internationally recommended (5,6), with addressing cognitive problems a 'top 10' research priority for pwMS (7).
Despite these calls to action, UK MS services do not have an established care pathway which integrates these recommendations. A UK-wide survey of clinicians found variation in cognitive assessments used, often with inappropriate screening tools used rather than the recommended tests (8). We also found a lack of consistency in reporting cognitive rehabilitation interventions, particularly regarding the content of interventions and their underlying framework (9). These issues make it difficult for healthcare services to consistently and systematically implement cognitive screening and rehabilitation (10). To address this gap, we aimed to develop a multi-agency, co-constructed, clinical pathway to forge a consensus on screening and managing cognitive problems in MS and how to go about this.
Medical Research Council (MRC) guidelines stress the importance of theory in developing and evaluating complex interventions (11). Here we propose an initial logic model ( Figure 1) which depicts the theory underpinning a screening and management pathway for cognitive problems in MS. Our model was informed by literature reviews (12)(13)(14), theory (e.g., Behaviour Change Wheel (15)), Patient and Public Involvement (PPI), clinical experience, and service realities. It adopts a Situation-Inputs-Outputs-Mechanism-Outcome configuration (16).
Situation describes the contextual features that pre-date the introduction of the pathway; including the high prevalence of cognitive problems, lack of standardised screening and support for cognitive problems (particularly mildmoderate problems (17,18) , and healthcare recommendations which drive the focus of clinical care (e.g., NICE -recommends screening and managing cognitive problems, but is not based on robust evidence and does not refer to a particular assessment or treatment (19)). Inputs are the resources required to support the pathway. Outputs describe the products created by the activities of the pathway.
Mechanisms are mediating factors and obstacles between the introduction of the pathway and the resulting outcomes (e.g., stakeholders need to be engaged and supported; and the pathway accessible and flexible for implementation). Outcomes describe what results from the pathway. The specifics of these factors can be found in Figure 1.
Logic models are dynamic, they represent working hypotheses and are modified iteratively as new insights emerge based on primary (e.g., stakeholder consultation, evaluation studies) or secondary data (e.g., existing literature/policy) (20). Conducting primary research with key stakeholders can enhance theoretical understanding of the processes of change (21). Exploring the views of key stakeholders at an early stage also helps to produce a co-constructed output (22) that empowers and engages stakeholders (23).
Here, stakeholder perspectives were used to develop a multi-agency cognitive screening and management pathway. This study is part of the NEuRoMS project (24) which will evaluate the efficacy of this pathway across six UK MS clinics.

Material and Methods
Ethical approval was granted by the University of Nottingham Faculty of Medicine and Health Sciences Ethics Committee (reference: 263-1903). All participants provided informed consent prior to data collection.

Recruitment
We sought to interview a broad range of stakeholders with complementary perspectives on a screening and management pathway for cognitive problems, we selected purposively from those willing to be interviewed to generate this. Participants were contacted by email or phone and invited to a focus group or an interview (in-person, or via telephone/video conferencing), based on their preference.

Data collection
Semi-structured interviews were conducted by authors JMM, RdN, NE and HE (see Supplementary Materials for interview schedules). The focus group with pwMS was co-facilitated by our PPI partner CB, to enhance data richness (25).
Interviews were audio-recorded and transcribed verbatim.
Participants were shown several resources to illustrate how the pathway might work and were asked to provide feedback. Making Test (32)); and, a self-report questionnaire (Multiple Sclerosis Neuropsychological Questionnaire (33)). The initial logic model ( Figure 1) was also shared.

Analysis
Anonymised transcripts were analysed on NVivo 12 using the Framework approach (34). The logic model informed a working analytical framework (see Figure   1 and Supplementary Materials for coding scheme) and data were mapped onto this.
Review of the mapped and organised data informed a revision of the logic model based on the findings below.
Yardley's evaluative characteristics for good qualitative research were applied (35). Regular team discussions were also held to modify the coding scheme to better represent the data (36). Summary data were presented to a PPI group to sensecheck our interpretations.

Study participants
Forty-four participants were interviewed (25-

Overview
There was consensus on the current situation-cognitive problems were reported as prevalent and disruptive, and stakeholders recognised the need to address these problems using a standardised pathway. Discussions therefore focused on how the pathway would work.
Data were organised into overarching themes (pre-defined logic model configurations), themes (core patterns) and sub-themes (further depth). Key results are presented for each overarching theme (inputs, outputs, mechanisms, outcomes) and supporting data are displayed in Tables 2-5, with superscript numbers linking relevant quotes to the text. When multiple participant groups endorsed a sub-theme the term 'stakeholders' is used, otherwise the relevant group of stakeholders is specified (e.g., pwMS).

Inputs (Resources)
Clinical Staff were recognised as an important resource. Stakeholders pointed to 'competition' between symptoms during clinic appointments, with limited time to address multiple MS symptoms 1 . Clinicians commented on the need to monitor drug treatments during the appointment, which took precedence over dealing with cognitive problems 2 . Physical symptoms including spasticity and bladder and bowel issues were often prioritised over cognitive problems 3 .
Pressured workloads were also raised, particularly those of MS nurses 4 .
Stakeholders reflected that any new pathway would impact on staff time 5 and that it is unclear who might have the capacity to deliver this 6 .
PwMS thought they would need to expend mental and physical resources to engage in the pathway. Travelling to hospital requires time, effort and planning 7 .
Digital technologies are also needed to access cognitive screening 8 . Further time commitments may be necessary from those attending multiple rehabilitation sessions 9 . Clinic facilities, including technology to host the cognitive screening (e.g., computer tablet, WiFi access 10 ), and a clinic room where screening and support sessions could be completed 11 , was also identified as a key resource.
Clinicians and commissioners thought understanding existing costing and commissioning frameworks and having a strategy in place to show how the proposed pathway addresses these drivers 12  PwMS understood the relevance of the cognitive measures presented to them and thought these were appropriately challenging 8 . However, some found the mental arithmetic task (PASAT) unpleasant 9 . Everyone agreed that cognitive screening should be brief. Clinicians acknowledged the need to balance the sensitivity and brevity of screening tasks 10 .
Clinicians thought that screening results should be digitised within patients' medical records 11 . They felt that the feedback report should include cut-offs to help identify individuals who may require support 12 and enable discussions around the type and severity of the cognitive problem 13 . Most stakeholders thought the results should be communicated face-to-face, at a routine clinical appointment, rather than over the telephone or via letter/email 14 . Communications should be initiated by a neurologist where problems are severe 15 .
Stakeholders recognised the complexity of triage decisions and referrals.
Where present, concurrent symptoms such as low mood and fatigue needed to be interpreted in relation to cognitive problems 16 . Stakeholders also highlighted the importance of the perspective of pwMS. Triaging should consider how the person was feeling during cognitive screening and any extenuating circumstances (e.g., relapse, technology problems 17 ). Stakeholders thought that pwMS should engage in these discussions to reflect upon the functional impact of cognitive problems 18 .
Data relating to the cognitive rehabilitation/management programme reiterated that concurrent symptoms need to be addressed by the pathway, through the provision of relevant information 19 . Stakeholders felt the content should include compensatory strategies that can be implemented at home 20 and gave examples of strategies they thought might work well (e.g., digital technologies 21 ) or be less effective (e.g., abstract visualisation 22 ). <<Table 3 here >>

Mechanisms
The complexity of cognitive problems was raised as an important mediator.
Stakeholders highlighted the interdependence of symptoms and recognised that stress and fatigue can exacerbate cognitive problems 1 . Similarly, cognitively demanding activities left pwMS feeling fatigued and drained 2 . Stakeholders thought the pathway should recognise that some cognitive problems will stem from brain damage driven by MS, whilst others are a secondary reaction to living with MS 3 .
Clinicians thought this was an important distinction to recognise 4 . Stakeholders suggested the pathway should also acknowledge that individual differences could influence cognitive performance and the effectiveness of support programs, and the perspective of related informants need to be addressed 5,6 .

Engaging pwMS addresses their reactions towards the pathway.
Stakeholders felt pwMS should be informed about the rationale for the pathway.
Explanations should reassure them that the pathway is meaningful 7 and clarify how it will inform their clinical care 8 . Stakeholders thought information should be clearly communicated avoiding medical jargon 9 .
The timing of the pathway is an important mediator for pwMS. Stakeholders thought invitations to complete the screening should be aligned with a routine appointment where the results can be communicated without delays 10 . Most pwMS supported the idea of being told about the screening tool in advance of their appointment 11 but acknowledged some might worry about this 12 . Some thought receiving information about the pathway might overwhelm newly diagnosed patients 13 , while others felt cognitive problems should form part of these early conversations 14 .
Stakeholders indicated that home-based approaches (e.g., online screening, telephone follow-ups for the cognitive management programme) would be convenient and less stressful for most pwMS 15 , particularly those in employment who cannot attend multiple appointments 16 . However, home-based approaches were not perceived as feasible for all 17 ; stakeholders thought access to additional support (telephone or face-to-face in-clinic) would promote engagement 18 .
Engaging clinical staff encompassed clinicians' reflections on which team members they thought responsible for addressing cognitive problems. Some thought MS nurses and neurologists would not consider cognitive problems as part of their The pathway should empower pwMS to be proactive and to take actions which target milder cognitive problems 29,30 . However, stakeholders also acknowledged some pwMS will not be motivated to engage with this because of fatigue and cognitive problems 31 .
PwMS thought the pathway should promote a positive outlook and inspire them 32,33 . However, clinicians recognised the need to manage expectations, e.g. manage problems rather than restore abilities 34,35 . These discussions can be frustrating for pwMS who want their cognitive abilities restored 36 . Stakeholders thought appropriate information might temper unrealistic expecations 37 . <<Table 4 here >>

Outcomes
We elicited 'short-term' outcomes of the pathway from the data, including improved access to standardised clinical care for cognitive problems. Clinicians spoke about the lack of formalised guidelines for MS services 1 and thought the pathway would address this 2 . Clinicians and commissioners felt the pathway would facilitate conversations about cognitive problems by providing information about patients' symptoms 3,4 . These conversations were also perceived positively by pwMS 5 , who valued the opportunity to discuss cognitive problems 6 and have their concerns validated 7,8 .
Some stakeholders acknowledged that anticipating the screening results could worry pwMS 9 . Moreover, being reminded of symptoms and noticing cognitive decline might be upsetting 10,11 . Consequently, some pwMS may disengage with the pathway 12 .
'Longer-term', stakeholders thought being able to understand and manage cognitive problems would improve quality of life amongst pwMS 13,14 . Earlier detection and management of problems might even prevent deterioration 15 and prolong independence 16 . Stakeholders indicated that early detection and management could promote efficient use of NHS resources 17,18 . Home-based cognitive screening was also considered efficient, reducing the time required with a clinician 19 . Stakeholders reflected on a potential increase in referrals to psychological services 20 , which could overwhelm already pressured services 21 .

Discussion
Stakeholder feedback confirmed that staff time, training packages, brief online screening tasks, and person-centred support for cognitive problems are important elements in the pathway. Improved access to care and clearer referral pathways for cognitive problems were still thought of as possible outcomes (as in Figure 1).
However, our data also offers new insights that challenge the initial logic model and enabled us to consider implementation more directly. Figure 2 reflects a revised logic model that could inform subsequent service development.
<<Insert The 'Activities' component (previously labelled 'Outputs') supports recommendations that computerised testing is a means to implement cognitive screening in routine care (13). Electronic data gathered in this way may be integrated within patient records to promote routine monitoring (37). Our study builds on this by demonstrating how and when cognitive screening could be undertaken Nonetheless, our results are based on data gained from a research sample.
Other stakeholders might offer further nuance to our understanding -for instance, we did not include IT specialists who would be responsible for data linkage from the screening results to the patient notes and ultimately integrating the pathway within existing IT systems. This, and insight from subsequent pathway piloting, will contribute to the iterative development of the logic model.
Our model is thus a blueprint that healthcare professionals could adapt to suit local circumstances, with local pathways being designed in the nexus of our logic model and local needs/resources. We view the training packages, screening tool, triaging and cognitive management programme (resulting in an integrated care pathway) as core elements to be retained across healthcare systems/pathways.
However, availability and commissioning of resources will differ across healthcare models. For those low-income and middle-income countries where patients do not routinely receive disease modifying therapies, this might be prioritised in terms of funding (42). Access to technology (e.g., tablet and Wi-Fi access for screening inclinic) may also be limited here. Transferring the screening results to the clinical team will also require contextual strategies to integrate the results within existing record keeping systems (e.g., digital software, paper-based files) and ensure accessibility. Engagement mechanisms of timing screening to coincide with an upcoming appointment and providing the option of in-clinic support will also vary, with healthcare systems showing considerable variation in the frequency of consultations (e.g., quarterly versus annually (43)), which will influence the timing of appointments. Our model offers MS clinics the flexibility to work within the constraints of their systems and integrate existing successful practices.

Conclusions
Existing MS clinical pathways have not been co-constructed with stakeholders and are not based on robust evidence (19,44).

Data Statement
Anonymised data that support the findings of this study are available from the corresponding author upon reasonable request.
To facilitate reproducibility and date reuse, we also share our logic models (Figure 1 and 2), coding schema (Supplementary Materials) and interview schedules (Supplementary Materials).

Funding Information
This report is independent research funded by the National Institute for Health

Disclosure
The Authors declare that there is no conflict of interest.   Table 2. Key Input (Resource) themes and sub-themes with supporting data.

Clinical staff
Clinic appointment Time allocated to cognitive symptoms during limited clinic appointment 1. And in our symptom management clinics where they have half an hour appointments, in that appointment we will be looking at bladder, bowels, fatigue, mobility, spasticity -you name it, it's addressed in that. And if you touch on cognition, but again we don't have time to sit there and go through a proforma or anything like that or any kind of referral. MS Nurse MS07 2. There's these new medicine, I give you this medicine, has it got any side effects, let me tell you about the side effects and so on. So, the competition for time is, do I -at the moment because we didn't have the evidence base for the effective rehabilitation, there seem to be some pressure and probably it will take second priority in people's minds. Neurologist N01

So I usually really just tell her conversation about medication and that's all there's, you know, we have a bit of a tete a tete about that! [Laughs] And that's all there's time for! Charity Volunteer CV05 Pressured Workloads
Capacity of multidisciplinary team members 4. Personally my experience of MS…yeah, I don't know how they would have the bandwidth to do this, it would be great if they do, but you know. Charity Volunteer CV05 5. I would just worry a little bit about timing, if you're doing it with everybody, because if they've got thethey're also very limited on time and they'll also be doing other OT roles as well. Neuropsychologist PS05 6. I think probably the Occupational Therapists are more used to delivering cognitive screening, giving advice on cognition and especially in the sense of how we're using the, you know, mechanismsso the strategy that you're suggestingincorporating with everyday life.
…I have the impression that they have possibly a bit more time than the MS nurses. Neurologist N01

People with MS Physical & Mental
Requires concentration 7. Because it's hard work when you have to get to a hospital appointment, even for those of us that drive and don't have too much physical problems at any one time. It is hard work. Focus Group Male M2 Planning. Focus Group Male M1 Yeah. To get to [hospital 1], plus the expense, it's a big thing. Focus Group Female F1 Physical Requires travel, time away from work, access to technology, mobility 8. They may not have the money to buy the machinery, machines to access the link, they may not be IT literate and they may just be plain right poorly, you know, too poorly to do it. PwMS P02 9. And also trying to get the timing right as well because obviously a lot of people with MS are working so doing -if it's one off group it's easier, but if it's a group over a number of sessions, doing it during working hours is difficult for people to commit to. Neuropsychologist PS03

Clinic facilities Technology
Availability of Wi-Fi, tablet, etc.

10.
We have a Wi-Fi. We have also NHS Wi-Fi which is free.

MS Nurse MS05
Clinic rooms Availability and management

11.
There is general mismanagement of rooms in the hospital. So once the MS nurses, they always find this space, because they find that the empty room and let's go there for the next half an hour. Neurologist N01  Table 3. Key Output (Activity) themes and sub-themes with supporting data.

Training packages
Screening & Triaging How to support pwMS to complete screening and interpret the results

1.
I'm not sure our OTs, even our specialist rehab OTs would have used the digit symbol and some of these, so it's kind of, you know, are introducing something new to people that then will need to interpret that, but I don't think with training, I think that's feasible isn't it. Neuropsychologist PS05

Cognitive management programme
How to deliver and set goals 2. Not all nurses have been through rehabilitation unit or rehabilitation training to kind of be aware of the notion of goalsetting and monitoring and motivationalyou know. Neurologist N03

Supervision
Ongoing monitoring and support 3. So if this is just an MS nurse working on their own then they might find it harder to make some of those decisions unless they've had some solid training and some supervision, ongoing supervision just to, you know, flush it out a bit. Neuropsychologist PS05

Screening tool
Administration How to enable completion of screening ahead of routine appointment

5.
That would be an ideal, digitalising the assessment pre-appointment. That would be wonderful…so when they come into clinic we have got everything there. MS Nurse MS05

Interviewer
So you wouldn't have any problems filling that in and accessing the link.

Carer C20
No, no, not at all, I'd be alright.

7.
Which is obviously brilliant because the whole thing about a paper link is that someone then has to type it in. And as much as possible, isn't it, you want someone to have the link on their device so they just have to click on it. Neurologist N03 Cognitive measures Include short, sharp assessments that avoid mathematics

Interviewer
Did those tests seem relevant? PwMS P20 Yeah, very much so, yeah. Especially the one with the colours and the -Interviewer The Stroop test.

PwMS P20
Yeah, you know, it's quite profound how your brain works because, you know, somebody without MS you still kind of have to really think one thing but you're having to override it. I think that's a really good test 9. Because I've got to think of too many numbers so I can't focus on the number that I need to focus on because I've just done the total. So I will have forgotten that number. I just couldn't do that. Focus Group Female F1 10. Your balances a bit, isn't it, it's finding something which is quick but MS nurses will be able to do with everybody quickly as part of their clinical interview while still being meaningful enough. Neuropsychologist PS03

Screening results
Transfer Electronic transfer to clinical team 11. So probably for us it would be emailing to the MS coordinator, who would upload it onto the patient's EPR so that it was there as an electronic document and then it would remain so, as opposed to lost in someone's email and never available again. Neurologist N03 Feedback report What should this include 12. I think if we could categorise them really in a binary way or in these three categories, I think that would be excellent. That's very practical and I think we all know pretty well who we're thinking of when we're thinking of these categories. Neurologist N02 13. I mean, it might be useful to know in roughly what sort of domain we're talking about the deficit as being, you know, so whether it's a memory problem or whether it's -I don't knowbeen a processing problem or maybe something roughly categorising it a little bit further rather than just severity.  19. I am trained to give medicines. So, why to see patients and make sure that they have all medicine's correctly. So there will be some doctors that will not be keen doing this 2 minutes in a consultation. Neurologist N01 20. I think that it's not part of their role, they wouldn't see it as part of their role and it's slightly out of their competency I suppose...it's not traditionally seen as a nurse thing, I don't think, it's more a psychology, an occupational therapist, you know, an OT thing, yeah. Neuropsychologist PS01 21. I actually do think it needs to be you know, every symptom and management kind of needs to be sort of kind of responsibility for everybody. Occupational Therapist OT04 22. So it's really useful to have that team approach to cognitive difficulties. Neuropsychologist PS03 Willingness and motivation To adopt an alternative approach 23. Yes, we absolutely value the importance of cognition. And that would be no problem at all from our perspective, in terms of making that a key priority. Occupational Therapist OT02

Foster shared values of the pathway
Person-centred Individualised 24. Not the same thing works for everybody, so I think a kind of basic starting point and then personalise it after that would be a good idea. Charity Volunteer CV01 25. How much it's impacting somebody's life, so you can have, you know, quite mild difficulties but that really impact on somebody's life, or you can have moderate difficulties which aren't really impacting.

Neuropsychologist PS03
26. I suppose it's meaningful to that person, rather than it just being a lot of suggestions thrown at them, that it is meaningful to that person. Neuropsychologist PS05 Holistic Care for the person as a whole 27. We want to ensure there's parity, that these people get parity of esteem, that they're whole system.

Commissioner CM02
28. So there's always been very holistic, very focused on physical rehab as well as psychological and cognitive. Neuropsychologist PS04 Proactive Initiate and act 29. I feel like that could have quite a protective element to it for them if they understand those changes earlier and they can learn the strategies then potentially that might protect employment, it might protect…and it helps them, it empowers them to speak about their difficulties as well with language that they understand. Neuropsychologist PS05 30. It's to help and it'll help yourself and it'll help others and just go that way. Carer C20 31. So if somebody is really, really depressed, again through struggles with their memory and they will probably struggle with initiation as well because they'll be so low, so actually probably implementing some memory strategies might be difficult for them. Neuropsychologist PS03 Positivity Encourage a positive attitude 32. Somebody who's like inspiring, positive, you know, that's what I would love. Focus Group Female F1

33.
It's about retaining a positive mind and a positive mindset. PwMS P20

Manage Expectations
Coping with cognitive problems -not retraining cognitive skills

34.
It's about this is a way of managing and understanding, not a way of getting rid of difficulties. Neuropsychologist PS05 35. But I think it's very important in terms of the language we're giving MS nurses and other professionals who are then going to be feeding this back and doing cognitive rehabilitation is the idea that coming on these four sessions, it's not going to make this any better, but it may help you to live with it better. Neuropsychologist PS05 36. I want to recover, yes, I do want to compensate and find other strategies in the meantime, but I still want to feel that it's something I can regain and rebuild. Charity Volunteer CV05

37.
And there are really difficult conversations to have with lots of people we work with and you're going 'actually there's no evidence for that'. Neuropsychologist PS05