MND exceptional review 2023 for consultation.docx

Cognitive assessments

Recommendation 1.3.1 was made based on consensus of the guideline committee and co-opted expert as no clinical or economic evidence was identified. The committee acknowledged that identification of cognitive change would be of clinical benefit to people with MND and unlikely to be harmful. Identification of cognitive impairments and behavioural change in people with MND at the point of diagnosis could have impact on ensuring that useable and individually tailored equipment is appropriately provided for their needs. Ensuring equipment is tailored can reduce costs to the NHS and improve health outcomes by avoiding provision of inappropriate equipment that eventually has to be replaced. No specific evidence was found for frequency of assessment, however, the committee included cognition and behaviour as an area to be covered at the MDT review in their discussion of the evidence. At the time of development of the guideline, the committee acknowledged that there was no single validated tool specifically to assess people with MND and that a variety of tools were used by psychologists when conducting formal assessments.

The enquirers highlighted that the current wording of recommendation 1.3.1 implies that frontotemporal dementia is the only cause for lack of mental capacity, and that it does not reflect people with MND without dementia could lack mental capacity due to other causes. Additionally, the enquirers highlighted that validated screening assessment tools, such as the Edinburgh Cognitive and Behavioural Screen (ECAS) or the Beaumont Behavioural Index (BBI) have been developed since publication of NICE guideline NG42, that assess behavioural changes in people with MND. The enquirers have suggested that evidence for these assessment tools have differing strengths in identifying cognitive and behavioural changes and could be more effective at differing points of MND stages.

We assessed the studies suggested by the enquirers which included a relevant systematic review on the validity and diagnostic accuracy of Amyotrophic Lateral Sclerosis (ALS)-specific screening tools (Simon et al. 2019). Other studies suggested by the enquirers did not meet the required study design criteria (Gosselt et al. 2020; Gray et al. 2022; Hodgins et al. 2020; Radakovic et al. 2020). Nguyen et al. 2021 was highlighted by the enquirers and was also identified from the focused search, and is described below.

The focused search of the evidence for cognitive or behavioural assessment identified 14 studies. Simon (2019) and 13 other publications identified were validation studies or diagnostic studies (Aiello et al. 2022; Aiello et al. 2023; Beeldman et al. 2021; Elamin et al. 2017; Greco et al. 2022; Iazzonling et al. 2022; Lule et al. 2014; Niven et al. 2015; Pinto-Grau et al. 2017; Saxon et al. 2020; Tiokrowijoto et al. 2023; Turon-Sans et al. 2016; Wooley et al. 2010). Mostly the population in the studies was people with ALS. The types of assessment tools that were identified in the studies included the Beaumont Behavioural Inventory (BBI), ALS Cognitive Behavioural screen (ALS-CBS), Frontal Assessment Battery (FAB), Edinburgh Cognitive and Behavioural ALS Screen (ECAS), Motor Neuron Disease Behavioural Instrument (MiND-B), Consortium to Establish a Registry for Alzheimer's Disease (CERAD) and Montreal Cognitive Assessment (MoCA) test. These were compared to neuropsychological assessments or another assessment tool as the reference standard. The ALS-CBI, BBI and ECAS were tested among the studies.

Cognitive deficiency with ALS-CBI was detected with a range of 71-86.2% sensitivity and 62-85% specificity compared to neuropsychological and/or neurological assessment as the reference standard in two studies (Turon-Sans et al. 2016; Woolley et al. 2010). Behavioural changes on ALS-BCI had 80% sensitivity and 92% specificity compared to neuropsychological testing (Woolley 2010).

The BBI had a sensitivity of 87.9% and specificity of 78.9% compared to the Frontal Systems Behaviour Scale (FRSBe) at a cut off 7 in detecting mild behavioural changes (n=85) (Elamin et al. 2017). Against the ALS-FTD questionnaire and FAB, BBI had 100% sensitivity and 92% specificity in detecting severe behavioural changes, however, for mild behavioural changes sensitivity and specificity were 50% and 76% respectively (Pinto-Grau et al. 2017).

ECAS had 85% sensitivity and 85% specificity in detecting cognitive impairment compared to neuropsychological assessment (Niven et al. 2015). Saxon et al. (2020) reported the sensitivity and specificity of ECAS in detecting cognitive impairment in people with ALS-FTD (90% and 35% respectively), and in people with ALS behavioural variant frontotemporal dementia (ALS-bvFTD) (78%% and 35% respectively). ECAS had high specificity compared to CERAD (75% to100%) but sensitivity was lower, 33% to 50% (Turan-Sans et al. 2016). Two studies evaluated behavioural impairment or cognitive decline in people with ALS using MiND-B and ACE scores respectively on survival time. In Kaplan-Meier survival analyses the studies showed that deficits in behaviour or cognition were associated with shorter survival time (Nguyen et al. 2021; Xu et al. 2017).

Prognostic indicators

Only one study on an externally validated tool (ALS Prognostic Index) for predicting survival in people with MND was identified during development of the guideline. The study was of low quality due to its observational study design. While the ALS Prognostic Index was considered a useful tool for predicting survival, the guideline committee thought the tool needed to be validated in a UK cohort of patients who were receiving NHS care before it could be recommended. The guideline committee considered the importance of site of MND onset, age at diagnosis, weight loss, forced vital capacity at diagnosis, and lower ALSFRS or ALSFRS-R scores as predictors of reduced survival.

The enquirers highlighted that the NICE research recommendation on prognostic tools was no longer relevant as research on validating ALS-PI had stalled and new prognostic tools were available that should be considered in the guideline. The studies highlighted by the enquirers were checked for inclusion against the current evidence review protocol. Two studies were identified that were relevant. Caga et al (2016) investigated and established apathy as a critical prognostic factor for shorter survival in people with moderate to severe ALS (hazard ratio 3.8, 95% confidence intervals 1.9 to 7.5, P=0.0001; Cambridge Behavioural Inventory-revised (CBI-R tool); adjusted for cognitive status, disease status and symptom duration at start of study). The second study, Nguyen et al. (2021) showed that the risk of death was 2.53 times higher in people with ALS with behavioural impairment on MiND-B compared to those without behavioural impairment (95% CI 1.3 to 14.6, P=0.003). Behavioural dysfunction using the MiND-B assessment predicted survival and demonstrated that for every unit decrease in the MiND-B, the risk of death increased by 3%. Similar results for cognitive impairment using ACE-R assessment were observed, with shorter survival times in people with cognitive impairment compared to those without (HR 2.0, 95%CI 1.04 to 3.3, P= 0.042), and demonstrated that reduction of 1 unit on ACE was associated with increased risk of death by 4% (Nguyen et al. 2021).

Organisation of care

During development of the guideline, the committee acknowledged that access to health professionals outside of the core MDT was important and could be involved in more targeted care. The evidence identified in the review was based on 8 studies (one RCT, one before and after study, and 6 cohort studies). Most studies compared multidisciplinary care to general neurological care. One study compared multidisciplinary care plus coordinator care to multidisciplinary care alone. Two studies included a psychologist in the core MDT. The evidence was limited because of outcomes being rated as very low quality and the composition of the MDT being varied among the studies. However, there was a clinical benefit of the involvement of a MDT in a community or hospital-based setting on survival time and hospital stay. There was no clinical benefit of having an independent, additional coordinator to an MDT. The committee did not wish to specify who should coordinate care for the person with MND as this may differ for different geographical locations.

The economic model evaluated the trade-off between benefit observed in the clinical evidence and additional costs associated with MDT care such as additional staff time, increased NIV use and Riluzole plus additional costs of general MND care associated with increased survival. The committee discussed the assumptions made in the economic model: 1. cost of MDT care was simply added on to the average costs associated with treating an individual with MND (i.e., usual care costs); 2. the model assumed no cost savings generated from MDT care; 3. the model assumed no quality of life improvement for people with the MDT arm. Quality of life may be underestimated by the EuroQol-5 dimension (EQ-5D) because the model assumes no quality of life improvement for people in the MDT arm. Thus, any intervention that extends life in an MDT cohort is unlikely to be cost-effective even if it only costs a small amount. The committee agreed that cognition and behaviour was important to assess, manage and review regularly.

The MDT composition in the health economic model was the same as that identified in the clinical studies. The cost of MDT was calculated by the committee using expert consensus. The health economic model for MDT configuration was not shown to be cost-effective at £20,000 per QALY threshold in the base case, and it was acknowledged that the model undervalued quality of life of individuals with MND. The model was sensitive to small increments in quality of life, any psychological improvement brought by clinical psychologist in an MDT could potentially bring the incremental cost-effectiveness ratio (ICER) within NICE's acceptable quality- adjusted life year (QALY) threshold.

In the committee discussion within the guideline, it was acknowledged that a multidisciplinary team (MDT) with the core members currently recommended would be beneficial for patients, although there were uncertainties in the health economic model. Any changes or addition of more core members would require good quality new evidence to allow re-analysis of the health economic model. Current recommendations on the core members are not exhaustive or exclusive and recommendation 1.5.5 does allow prompt access to clinical psychology and neuropsychology. There is no new evidence that specifically evaluates the effectiveness of a MDT with psychologist as a core member.

The enquirers challenged the absence of a clinical psychologist in the core multidisciplinary team (MDT) (recommendation 1.5.4) because they believe new evidence suggests psychological wellbeing is a key factor for improving quality of life in people with MND.

Studies suggested by the enquirers did not meet the inclusion criteria of both the review protocols for this topic. Ando et al (2022) and Zarotti et al (2019) were both qualitative studies, and Young et al (2019) was not a comparative study.

Psychological support

Recommendations 1.6.1 to 1.6.6 were based on qualitative evidence highlighting important areas included coping with diagnosis, understanding the disease, acceptance, coping with a changed life, change in relationship, carers, sources of support, and decision making. The committee were aware of the importance of psychological and emotional issues for all other areas of disease management, noting that acceptance of the disease helped people to cope with all aspects of symptom management. The committee recognised that healthcare professionals who deliver psychological support must adapt the content and delivery of this support to the needs of the person with MND. Everyone has different requirements for support, and therefore the interactions must be led by the person. The committee distinguished information emotional support and counselling from formal psychological support in the recommendations. Many people with MND and their families can be helped with informal support, including involvement with support groups and charities such as those associated with the MND Association. The most appropriate psychological intervention(s) will depend on the nature and severity of the individual's problems, any history of previous psychological problems and the quality of social support available. A range of psychological interventions can be offered by both the statutory and voluntary sectors. Health and social care professionals offering day-to-day care provide much general psychological support to patients and carers. They play a key role in psychological assessment, and in the prevention and amelioration of distress.

The committee highlighted that practitioners should be alert to the requirement for formal psychological assessment and support and the need to refer to psychological and neuropsychological services. They highlighted the need for appropriate treatment for people with MND who are also suffering from depression. They felt that the issue is covered by CG91 NICE guideline.

The enquirers highlighted the absence of recommendations on specific psychological therapies that may be clinically and cost-effective to improve quality of life and psychological wellbeing in people with MND

Two studies were suggested by the enquirers. One study had a cross-sectional study design and did not meet the inclusion criteria of the guideline review protocol (Lapin et al. 2022). The second study was a qualitative study that explored needs and preferences for psychological interventions of people with motor neuron disease (Weeks et al. 2019). No study on clinical effectiveness of psychological support was provided by the enquirers.

Several studies on clinical and cost-effectiveness of psychological interventions for people with MND and their carers were identified from the focused search of this surveillance review.

One RCT (N=15) compared 16 weeks of CBT (stress coping model) in psychologically distressed patients with ALS and their caregivers to usual care alone (Groenestijn et al 2015). CBT consisted of 5 to 10 sessions (individual or group) provided by three trained and instructed psychologists using behavioural and rational-emotive principles in addition to usual care. Outcome measures included the Mental Component Summary (MCS) of the SF-36 scale, ALSAQ-40-EF, Hospital Anxiety and Depression Scale (HADS) and Caregiver Strain Index (CSI). The MCS of the SF-36 score assessed in caregivers improved in the CBT group compared to usual care. Similarly, ASLAQ-40-EF in people with ALS improved in the CBT group compared to usual care. No significant differences between the CBT group and usual care group were found in psychological distress (Hospital Anxiety and Depression Scale) (p =0.26).

One RCT (N=74 people with ALS and progressive muscular atrophy (PMA) and their caregivers compared a support programme delivered by a psychologist based on Acceptance and Commitment Therapy (ACT) for 12 weeks to a control group (De Wit et al 2020). No significant differences were observed on psychological distress or quality of life, however, improvements in self-efficacy were observed in the intervention group including caregivers. Nearly 50% of participants did not complete the intervention which may have contributed to lack of efficacy observed in outcomes.

One RCT (N=100) compared an 8-week ALS-specific mindfulness-based stress reduction programme to usual care was identified (Paganini et al 2017). Significant differences between MBSR and usual care groups were observed in measures for quality of life, anxiety, depression, negative emotions and interaction with people and the environment.

Another RCT (N=48) compared an expressive disclosure intervention with a control group in people with ALS (Averill et al 2013). The patients in the intervention group showed better wellbeing status than the control group at 12 weeks after the intervention, but not at 26 weeks. Expressive disclosure therapy was only effective in those with low emotional expression which suggests that the intervention may not be affective in later stages of ALS.

One RCT (N=30) compared 12 sessions of active music therapy plus standard care to standard care alone in people with ALS (Raglio et al 2016). Quality of life improved on the McGill Quality of Life Questionnaire global scores in the active music therapy group compared to standard care alone (P <0.035).

Cough effectiveness

Recommendations 1.13.1 to 1.13.4 were developed based on the committee's experience. The committee highlighted that the evidence did not support one cough augmentation technique over another. In current practice, regular breath stacking was used for ineffective cough, however, there was no evidence to compare unassisted versus assisted breath stacking. There was no clear clinical difference between breath stacking and other cough augmentation devices. The committee agreed that doing cough augmentation was better than doing nothing, therefore their informal consensus was used to make recommendation for cough augmentation techniques first and are associated with reduced staff time and cost. Breath stacking (manual followed by assisted) and then mechanical cough assisting techniques should be considered when cough augmentation techniques are ineffective or inappropriate.

The enquirers highlighted more recent evidence on cough effectiveness interventions specifically peak cough augmentation. The evidence they suggested was based on efficacy of cough augmentation, through a combination of interventions rather than any one intervention in isolation. They also highlighted the importance of acknowledging the impact of cognitive and behavioural changes on people's acceptability of cough augmentation and their ability to carry out the intervention. The enquirers pointed out that people's cognitive and behavioural states need to be considered prior to carrying out cough augmentation. Cognitive and behavioural impact on cough augmentation was not discussed by the guideline committee during the development of recommendations on this topic area.

Studies suggested by the enquirers were checked for relevance. One RCT of 30 people with ALS by Nicolini et al (2022) did meet the inclusion criteria in the review protocol. Cough assist plus expiratory flow accelerator was compared to cough mechanical assisting. Outcomes included changes in respiratory function tests, gas exchanges, and peak cough expiratory flow. Improvement in respiratory function tests was reported in both groups but more so in the cough assist plus expiratory flow accelerator group. No differences in gas exchange between both groups was observed. Both groups reported similar increases in perceived cough efficacy and no adverse events were identified.

A further search of evidence identified limited data from one small, low quality RCT that reported improvements in respiratory function and peak cough expiratory flow when cough assist and expiratory flow accelerator devices were performed in combination compared to cough assist alone. No studies were identified regarding the impact of patient's cognitive and behavioural changes on how cough augmentation should be delivered.

Non-invasive Ventilation (NIV)

Enquirers suggested that a medical history check should be included as part of the discussion about NIV in recommendation 1.14.2. The evidence underpinning this recommendation was qualitative and did not include themes related to medical history checks. The committee did not discuss this issue when developing the recommendation. However, this suggestion is more relevant to recommendation 1.14.7 because recommendation 1.14.2 is relating to information and support. The evidence underpinning recommendation 1.14.7 was very limited and of low quality. The committee used the EFNS task force guideline (Anderson 2005) and the Motor Neurone Disease Association review (www. Mndassociation.org) on the management of respiratory insufficiency in patients with MND/ALS for discussion. Based on their knowledge, experience and expertise, the committee proposed a list of symptoms and signs for routine monitoring to detect potential respiratory impairment through committee informal consensus.

The enquirers also challenged that arterial or capillary blood gas assessments should be carried out as part of routine assessment of respiratory impairment rather than based on SpO₂ assessments as currently recommended in 1.14.11. This recommendation was based on the committee's expertise and knowledge due to absence of clear evidence about which respiratory function tests were best to detect respiratory impairment. The committee agreed to recommending SpO2 as a first-line routine initial screening because it was practical to use, less stressful for patients, and could be carried out by most health professionals. A cut-off of less than 90% was used as the reference standard in the limited evidence for nocturnal SpO2 (one cohort study), however, the committee felt that the cut off for daytime SpO2 for people with MND should be higher in order to detect respiratory impairments appropriately and early. Based on the committee's consensus, further arterial or capillary blood gas analysis was recommended if SpO2 was less than 92% (in people with known lung disease) or if SpO2 was less than 94% (in people with no lung disease).

The enquirers further challenged that the current wording of recommendations 1.14.10 and 1.14.17 'symptoms and signs' (instead of 'symptoms and/or signs') sets a high trigger threshold for further respiratory impairment assessments or respiratory function tests respectively and increases the likelihood that people with MND requiring further assessment are excluded from referral by practitioners. Both recommendations 1.14.10 and 1.14.17 were based on consensus and expertise of the guideline committee. During discussion of the evidence, the committee did not discuss the wording or threshold for assessment of respiratory impairment or respiratory function was identified.

The enquirers did not highlight any new evidence to support their queries about recommendations 1.14.7, 1.14.10, 1.14.11, and 1.14.17. For recommendation 1.14.2, neither the evidence included in the guideline, nor the committee discussion included any information about checking medical history as part of discussion about NIV. In addition, recommendation 1.14.2 refers to information and support therefore, the suggestion by the enquirers to check medical history may be more appropriate as part of identification and assessment of respiratory impairment (recommendation 1.14.7). A further search of the evidence did not identify any relevant new studies on this topic area.

No new evidence was identified that suggests arterial or capillary blood gas tests should be used instead of SpO2 (recommendation 1.14.11). Similarly, no new evidence was identified to support the change of wording of 'symptoms and signs' to 'symptoms or signs' in recommendations 1.14.10 and 1.14.17.