Factors influencing decisions people with motor neuron disease make about gastrostomy placement and ventilation: A qualitative evidence synthesis

Abstract Background People with motor neuron disease (pwMND) are routinely offered gastrostomy feeding tube placement and (non‐invasive and invasive) ventilation to manage the functional decline associated with the disease. This study aimed to synthesise the findings from the qualitative literature to understand how individual, clinical team and organisational factors influence pwMND decisions about these interventions. Methods The study design was guided by the enhancing transparency in reporting the synthesis of qualitative research (ENTREC) statement. The search of five bibliography databases and an extensive supplementary search strategy identified 27 papers that included qualitative accounts of pwMND, caregivers and healthcare professionals' (HCPs) experiences of making decisions about gastrostomy and ventilation. The findings from each study were included in a thematic synthesis. Findings Making decisions about interventions is an emotional rather than simply a functional issue for pwMND. The interventions can signal an end to normality, and increasing dependence, where pwMND consider the balance between quality of life and extending survival. Interactions with multiple HCPs and caregivers can influence the process of decision‐making and the decisions made. These interactions contribute to the autonomy pwMND are able to exert during decision‐making. HCPs can both promote and threaten pwMND perceived agency over decisions through how they approach discussions about these interventions. Though there is uncertainty over the timing of interventions, pwMND who agree to interventions report reaching a tipping point where they accept the need for change. Conclusion Discussion of gastrostomy and ventilation options generate an emotional response in pwMND. Decisions are the consequence of interactions with multiple external agents, including HCPs treading a complex ethical path when trying to improve health outcomes while respecting pwMND right to autonomy. Future decision support interventions that address the emotional response and seek to support autonomy have the potential to enable pwMND to make informed and timely decisions about gastrostomy placement and ventilation. Patient or Public Contribution The lead author collaborated with several patient and participant involvement (PPI) groups with regards to the conceptualisation and design of this project. Decisions that have been influenced by discussions with multiple PPI panels include widening the scope of decisions about ventilation in addition to gastrostomy placement and the perceptions of all stakeholders involved (i.e., pwMND, caregivers and HCPs).

decision support interventions that address the emotional response and seek to support autonomy have the potential to enable pwMND to make informed and timely decisions about gastrostomy placement and ventilation.
Patient or Public Contribution: The lead author collaborated with several patient and participant involvement (PPI) groups with regards to the conceptualisation and design of this project. Decisions that have been influenced by discussions with multiple PPI panels include widening the scope of decisions about ventilation in addition to gastrostomy placement and the perceptions of all stakeholders involved (i.e., pwMND, caregivers and HCPs). per 100,000 population. 1 MND is associated with a 2-5 year survival after symptom onset with a lack of therapeutic options that can delay disease progression. 2,3 The clinical management of MND focuses on compensating for the progressive loss of vital physiological functions including respiratory failure and swallowing difficulties. 4 People with MND (pwMND) are routinely presented with options to start interventions aimed at improving quality of life and/or extending survival. 3 Interventions offered include gastrostomy placement, providing an alternative route for administering nutrition and hydration, and ventilation (non-invasive and invasive) for respiratory support. 5,6 Clinical guidance recommends discussing such interventions early in the disease course and the timely initiation of interventions to reduce procedural complications and promote better outcomes. [7][8][9] Though these interventions have the potential to extend survival, [10][11][12][13] they can also address the negative impact that functional losses have on pwMND quality of life. 3 In addition to the proposed benefits, the risks and burdens should also be considered during decision-making. 14,15 pwMND make decisions in the context of an often rapidly progressive disease with no hope of a cure. The risks and benefits associated with interventions are continually changing as the health threats presented by symptom progression escalate. King et al. 16 developed a cyclical model of decision-making describing how pwMND repeatedly respond and adapt to the relentless stepchanges in their condition. There are a range of contextual and relational factors that influence how, what and when, pwMND make decisions about their care. 17,18 Decisions are the result of interactions with a range of external agents including caregivers and the multidisciplinary team (MDT) supporting them. [18][19][20] In addition, cognitive and behavioural changes prevalent in MND impact on the abilities of the individual to make decisions and the support they need to do so, adding to the complex context in which pwMND make decisions about their care. 21,22 There are a growing number of qualitative studies that have captured the perspectives of pwMND, caregivers and healthcare professionals' (HCPs) during decisionmaking about gastrostomy and ventilation. This synthesis and further conceptualisation of decision-making about these interventions in MND care from multiple perspectives, will inform the future development of contextually sensitive decision support strategies.
The aim of this qualitative evidence synthesis is to understand how individual, clinical team and organisational factors influence the decisions that pwMND make about gastrostomy and ventilation (for the purpose of this paper 'gastrostomy placement and ventilation [non-invasive and invasive]' will be referred to as 'interventions' from this point onwards).

| Design
A qualitative evidence synthesis was selected to systematically identify and synthesise the findings from the published qualitative literature, to gain a broad and rich understanding of the factors that influence decision-making about interventions in MND care. 23 The study design was guided by the enhancing transparency in reporting the synthesis of qualitative research (ENTREC) statement (Supporting Information: Appendix SA). 24 The protocol was registered on PROSPERO on October 14, 2021 (CRD42021283314).

| Search strategy
The search strategy was informed by a scoping search and in consultation with subject experts, academic information specialists, patient and public involvement groups and the research team. All

| Inclusion criteria
Studies were included if they contained qualitative accounts of pwMND, caregivers or HCPs making decisions about interventions.
Caregivers were defined as being unpaid people who support pwMND. Only studies in the English language were included and no date restrictions were set.

| Study screening and selection
A total of 3781 references identified by the database and supplementary searches were imported into EndNote 20 reference manager. Following the removal of 937 duplicates, the title and abstracts of 2844 references were screened for inclusion by S.W.
A total of 95 full-text papers were screened for inclusion in the review by S.W. L.C. screened 10% of the references at both the title/ abstract and full-text screening stages. Any disagreements were discussed and resolved between S.W. and L.C., with further discussions with C.M., A.C. and V.H. when required. Finally, 26 papers met the inclusion criteria to be included in the review. The supplementary searches identified one further paper for inclusion. 25 See Figure 1 for the full results of the study screening and selection process.

| Quality assessment
The Critical Appraisal Skills Programme (CASP) qualitative checklist was used to assess the quality and rigour of the included papers (Table 1). 26 The quality assessment did not guide study selection due to the documented concerns relating to the consistency and subjectivity of such appraisals. [52][53][54] Two studies scored poorly across most domains of the CASP checklist. Leslie 38 did not include any methods, analysis or discussion, but the transcripts from the interviews provided interesting insights that were relevant to the aims of this review. Versalovic and Klein 49 paper lacked information about the study design and analysis, but the richness of the accounts addressed the review aim. Only six studies discussed the researchers' relationship (i.e., reflexivity) with the participants. 27,36,37,42,45,51

| Study characteristics
This review includes a synthesis of the findings from 25 studies reported in 27 papers, (three papers included the qualitative analysis of the same participant cohort from one study [29][30][31]  female) and 85 were HCPs (genders not stated). One study did not include any information about participants. 49 All the studies used interviews, with two studies also including observations 43,45 and another group interviews. 17 Table 2 for the study characteristics.

| Data synthesis
A scoping search of the current evidence and the resources available informed the decision to perform a thematic synthesis. 55 Thematic synthesis allows for the generation of analytical findings that conceptualise how decisions are made in the context of MND care.
The methods described by Thomas and Harden, 56  Papers were imported into QSR NVivo (Version 13) for coding and analysis by S.W. Only the text present in the findings section of the papers were extracted for the analysis. 56,57 All of the papers were read and re-read to increase familiarity with the data before coding.
Codes were organised into descriptive themes that remained close to the original data. The final phase involved moving beyond the categorisation of descriptive themes and onto the development of analytical themes.  Table 3 describes the contribution each paper made to the analytical themes. Discussions about interventions acted as an acute reminder for some pwMND of disease progression. 27,37,39,[42][43][44] Acceptance of the need for interventions was associated with giving in to the disease 39 and signified a stage of the disease associated with a poor quality of life or nearing the end of life. 27,37,39 (H) '…A very obvious sign of being ill…a very concrete representation that he's seriously ill.' 39 An emotional response was generated when learning about the procedures involved, 33,38,39,42,43,46 a requirement to come into hospital 27,36,39 and threats to pwMND physical integrity. 27,33,38,40,43 F I G U R E 1 Results of the study screening and selection process.

| An emotional response to interventions
For some pwMND their response to the prospect of interventions was influenced by previous negative healthcare experiences. 27,33,39 Additionally, pwMND were concerned about how interventions may impact on their current sense of normality. pwMND reported contrasting views about how an intervention may enable or hinder their ability to engage in social activities 27,38,39,51 or alter their reliance on care from others. 34,36,37,39,50,51 While some pwMND were able to visualise the gains interventions could facilitate in terms of freeing time or reducing burden on caregivers, 36,39,43,46 most studies reported how they represented changes to 'normal life' that they were unwilling to accept easily.  36 or, in contrast, facilitated them to be with or support family for longer. 33,34,39,43 For some pwMND extending life was an overriding goal, particularly when faced with a direct threat to life such as in crisis situations. 30,33,37,39,42,48,51 T A B L E 1 Quality assessment of the included papers using the CASP checklist. 26 Study year 1 2 3 4 5 6 7 8 9 10 Versalovic and Klein (2020) 49 Results of the quality assessment using the CASP qualitative checklist. 26

| HCPs: A source of knowledge and guidance
The style, amount, and pace of information shared by HCPs influenced pwMND experience of the decision-making process.
HCPs focusing on delivering comprehensive information about interventions, may prevent pwMND from having the opportunity to understand and explore the issues most important to them. 25,33 Some HCPs reported following a structured approach to supporting pwMND decision-making through seeking permission to initiate discussions, presenting options available, exploring preferences and distributing discussions over multiple interactions. 37    Caregivers were reported to play a supportive role, through seeking information, expressing opinions and deliberating about options. 31,33,37,38 Through these interactions caregivers 'influenced not only the decision but also the process of decisionmaking'. 39 Caregivers required information early in the disease process to facilitate their role supporting pwMND decisionmaking. 28,37 While caregivers recognised the pwMND right to self-determination, their contributions to the decision-making process were not always described as being neutral and were informed by their own preferences and acceptance of the diagnosis. 32 The involvement of multiple external agents raises the question of who is responsible for making decisions and how interactions with others may influence the loci of control.

| Control
pwMND wanted to be in control of the decisions they made about interventions; a preference that could be facilitated or threatened by interactions with others.
3.3.1 | Retaining autonomy over decisions pwMND consistently reported a preference to remain in control over decisions about interventions. 27,29,33,[37][38][39][40]51 In line with pwMND preferences, HCPs and caregivers sought to protect pwMND right to self-determination and empower pwMND to make their own decisions. 35 34,39,50 or framed alternatives in a way to make them not even appear to be a choice. 44,48 (H) 'I haven't discussed noninvasive ventilation because… how would he ever cope? So, I've made, on best interests, not to start discussing those issues…. His anxiety, it's just going to raise his anxieties.' 39 pwMND were able to enact agency over decisions while they were coping with symptoms and choosing to delay or decline WHITE ET AL.
interventions. However, when symptoms presented a significant threat to pwMND health or life, interventions were often perceived to be the only choice available to prevent serious consequences. [32][33][34]37,39,42,46,48 3.3.3 | Planning for the 'future self' Decision-making often involved pwMND attempting to visualise how their future self may value an intervention. pwMND reported struggling to imagine how they would feel about an intervention when living with increased disease burden. 37,43,46,49 Acceptance of disease progression was associated with pwMND planning ahead through either commencing interventions before them being needed or making advanced decisions to decline interventions altogether. 28,29,33,36,37,[39][40][41]46 Planning ahead could allow pwMND to remain in control of their destiny, including those at risk of cognitive decline. 25,35,39,51 However, other pwMND feared that interventions would be commenced or continued despite them experiencing a poor quality of life. 30,39,50,51 pwMND wanted to remain in control of decision-making.
Interactions with HCPs and caregivers influenced pwMND perception choice and agency over their decisions. The timing of interventions is a challenging issue for all involved.

| Tipping the balance
The timing of initiating discussions and commencing interventions was a source of conflict for all stakeholders. Commencing interventions often required pwMND to reach a subjective tipping point informed by disease progression, acceptance of need and recommendations of HCPs.

| Readiness to start the conversation
While HCPs advocated for an individualised patient-centred approach, HCPs supported introducing intervention options early during the disease course. 17,39 In addition to referring to signs of disease progression, HCPs reported making subjective assessments of pwMND psychological readiness to discuss interventions. 39,45 Some pwMND reached crisis point without having the opportunity to understand their options. 34 While early discussions about interventions allowed more time to consider the options, 33,38,39,43,46 for some pwMND having these discussions conflicted with a personal preference for focusing on the present day challenges of the disease. 29

| Autonomy and control
The review captured how external agents can facilitate individual autonomy, 69,70 an ethical principle valued by pwMND, through protecting the right to self-determination and provision of information. Caregivers play a multilayered role during decision-making.
Consistent with a critique of autonomy in end-of-life care, pwMND made decisions in collaboration with and through concern for their caregivers. 71 When pwMND decisions are based on concern for caregiver burden or prioritisation of caregiver preferences, an individualistic conceptualisation of autonomy could be perceived to be threatened. The social and interactional context in which pwMND make decisions in collaboration with caregivers may be better aligned with a relational conceptualisation of autonomy and is consistent with the principles of shared decision-making. 71

| Strengths and limitations
A limitation of this review was the absence of a second reviewer to shortlist the references for inclusion, extract data and independently code the data which may limit the credibility of the study. However, this is mitigated to a degree by some of the methodological strengths of the review such as the systematic approach to the study design,

| Implications for practice
The emotional response to interventions highlights the importance of exploring the value pwMND place on the implications and outcomes of an intervention beyond the concern for clinical outcomes.
Decision-making should address the issues that matter most to