What matters most to adults with a tracheostomy in ICU and the implications for clinical practice: a qualitative systematic review and metasynthesis

6 Purpose : Tracheostomy is a common surgical procedure in ICU. Whilst often life-saving, it can have 7 important impacts on patients. Much of the literature on tracheostomy focuses on timing and 8 technique of insertion, risk factors and complications. More knowledge of patient experience of 9 tracheostomy in ICU is needed to support person-centred care. 10 Materials and Methods : Qualitative systematic review and metasynthesis of the literature on adult 11 experience of tracheostomy in ICU. Comprehensive search of four bibliographic databases and grey 12 literature. Title and abstract screening and full text eligibility was completed independently by two 13 reviewers. Metasynthesis was achieved using thematic synthesis, supported by a conceptual 14 framework of humanised care. 15 Results : 2971 search returns were screened on title and abstract and 127 full texts assessed for 16 eligibility. Thirteen articles were included for analysis. Five descriptive and three analytical themes 17 were revealed. The over-arching theme was ‘To be seen and heard as a whole person’. Patients 18 wanted to be treated as a human, and having a voice made this easier. 19 Conclusions: Voice restoration should be given high priority in the management of adults with a 20 tracheostomy in ICU. Staff training should focus on both technical skills and compassionate care to 21 improve person-centred outcomes. 22


Introduction
Tracheostomy is a common procedure in intensive care units (ICU) and head and neck cancer, with an estimated 12 000 to 15 000 insertions per year in the UK prior to the COVID-19 pandemic (1,2), The pandemic has seen figures much higher than this due to the dramatic increase in patients requiring mechanical ventilation and higher rates of tracheostomy in patients with COVID-19 (3,4).
Whilst tracheostomy is often a life-saving procedure, negative impacts on patients occur and include fear and anxiety; temporary or persistent inability to talk; difficulty swallowing; pain due to the tube or stitches; increased work of breathing; and trauma to the trachea which can lead to tracheal stenosis (3)(4)(5)(6)(7)(8)(9).Management decisions and clinical practices such as choice of type and size of tracheostomy tube (TT), cuff deflation, tracheal suction technique, facilitation of communication and assessment of pain, may influence the experience of these (5)(6)(7)(8)10).COVID-19 has presented additional challenges including higher incidence of upper airway swelling, infection exposure risk to staff, and the impact of restricted family visiting and staff use of personal protective equipment on patients (3,13).
Various initiatives have sought to improve the care of people with a tracheostomy over the past ten years (1,14,15).It is now widely accepted that high quality healthcare must be patient-centred and that this requires understanding of patient experience and perspectives (16)(17)(18)(19).However, much of the evidence guiding tracheostomy management focuses on incidence, timing and technique of insertion, risk factors and associated complications.A number of quantitative, measurementfocused studies have addressed quality of life (QOL) and mental health outcomes in patients with a tracheostomy (20)(21)(22).While these studies capture prevalence and patterns of symptoms, they are not designed to present accounts of patient experience, which can help to shape future care.
Qualitative methods have been used to provide insights into ICU patient experience of delirium (23) and mechanical ventilation (24)(25)(26).Nakarada et al's (27) mixed-methods systematic review described challenges facing patients with a tracheostomy and their caregivers such as poor basic care, speech and communication difficulties, altered body image and reduced social interaction in the community.However, their review was not ICU specific.Tolotti et al's (2022) (28) qualitative scoping review addressed nurse-patient communication experiences in patients with a tracheostomy and/or on mechanical ventilation, and Whitmore et al's (2020) (29) mixed-methods scoping review of post-insertion ICU tracheostomy management concluded that more research was needed into patient experience of events relating to or impacted by tracheostomy.Further robust evidence and conceptual understanding of the global experience and priorities of ICU patients with a tracheostomy could help improve clinical care and patient-centred outcomes for this group of people..The aim of this qualitative systematic review and metasynthesis was to describe what matters most to patients with a tracheostomy in ICU, and to consider the implications for clinical practice.

Materials and methods
Our qualitative approach to enquiry allowed exploration of behaviours and perspectives that would be unachievable through quantitative research.Our inductive methods encouraged the emergence of descriptive themes from the data rather than reviewers prior knowledge and beliefs (30).We included a metasynthesis to go beyond an aggregative synthesis of primary studies and develop new concepts and insights (31)(32)(33).Thematic synthesis as described by Thomas and Harden (2008) (33) allowed us to synthesise multiple datasets whilst retaining individual participant voices.The study protocol was registered prior to data collection with PROSPERO (reg.227554).

Search strategy
Our systematic searches were aided by an expert librarian and included four major bibliographic health databases (Medline, Embase CINAHL and Web of Science).A 'tracheostomy' search construct was developed using alternative spellings, key words and medical subject headings.This was combined with a 'patient perspective' construct, consisting of phrases such as 'patient report', 'patient opinion' and 'lived experience'.The search was first conducted and refined in OvidSP Medline (see supplementary material A for full search strategy), then translated in other databases.
We also completed grey literature, citation and journal searches and asked expert ICU clinicians for references of articles on patient experience of tracheostomy in ICU.Searches took place between 23 December 2020 and 18 January 2021.Further re-runs of bibliographic database searches were completed on 29 June 2021 and May 2022.

Study selection and data extraction
Included articles were written in English and reported qualitative data from the perspective of adults in ICU with a tracheostomy or their non-professional carers (see supplementary material B for inclusion criteria).No date limits were applied.All articles were uploaded to EPPI-Reviewer software for systematic reviews (34), and each was independently screened on title and abstract by the first author (HN) plus one other reviewer (GC, SW, or CS).Disagreements were resolved by referring to a third reviewer (NP).The same process was followed in full-text screening.We piloted a data extraction tool to capture study characteristics (see supplementary material B).No modifications were necessary.All text from 'Results'/'Findings' onward was treated as data for analysis.

Data synthesis strategy
We followed Thomas and Harden's (2008) (33) thematic synthesis as follows: 1. Line-by-line inductive coding of text; development of new codes and translation between texts as each set of findings from studies were coded 2. Descriptive themes were developed by identifying similarities and differences between the initial codes, then grouped into hierarchical structures 3. The synthesis of selected findings was reviewed and newly developed descriptive themes applied to the review question to develop the analytical themes and metasynthesis from across the dataset, moving the synthesis beyond a collection of reported themes and drawing out collective inferences During the third stage, a conceptual framework relevant to the preliminary review findings was identified (35) and used as a lens through which to view and shape emerging analytical themes, translate findings into clinical implications and provide a deeper level of analysis.

Reflexivity and rigor
We followed the ENTREQ checklist (36) to ensure transparency of reporting (see supplementary material C).Search results and study selection have been presented using the PRISMA flowchart.To enhance dependability and credibility, two reviewers undertook these stages independently, coming together to review codes and themes and discuss potential new analytical themes, repeating this in an iterative cycle (37).
Our research team included researchers with extensive experience in qualitative methods.Feedback on descriptive and analytical themes from the wider study team and patient and public involvement (PPI) group was incorporated into the findings.The PPI group was convened to support the first author's PhD project.PPI members had all experienced tracheostomy on ICU and had either responded to a request for support via ICUSteps, been a patient of the first author or were recruited via word of mouth.Together, the measures described above enhanced Lincoln and Guba's (1985) (38) concept of trustworthiness, credibility, confirmability, and transferability.The lead author/researcher was motivated by clinical experience as a Speech and Language Therapist in ICU.

Study selection
The search returned 2971 records.Following de-duplication, 2395 articles were screened on title and abstract.Full text screening was conducted on 127 articles.Thirteen articles were included in the review and metasynthesis (see Figure 1.).The main reasons for exclusions were insufficient focus on tracheostomy (for example, mentioned only as an outcome) or no qualitative data.One article was excluded due to difficulty distinguishing between data relating to intubated or tracheostomised patients (39).

Risk of bias assessment
Quality assessment of selected articles was completed using the Critical Appraisal Skills Programme (CASP) checklist for qualitative research (40) (see supplementary material E).Following the example of Lachal et al (2017) (41), a relevance weighting of one to three points was added depending on how closely studies met the criterion of reporting on patient experience of tracheostomy in ICU (all studies included some data on this).No articles were excluded, however, a sensitivity analysis reviewed the impact of including studies with lower CASP or relevance scores (see supplementary material F).Weighted scores ranged from 13-18 (out of 20).One study had only three participants and was completed by a single author, limiting credibility of findings (42).The balance of participant quotes to author interpretation and provision of contextual information varied across studies, impacting credibility and transferability of findings.There were nuanced differences between studies based in long-term settings versus acute settings in terms of patients' experience of anger due to communication difficulties and issues of autonomy and self-identity.

Descriptive Themes
The five major Descriptive Themes (DTs), developed from participant quotes and study author interpretations from the 13 selected studies, are presented with illustrative quotes in Table 1 below.
Figure 2 shows the analytical process post initial coding, including thematic maps of high-level descriptive and analytical themes.
Opposing priorities of patients and care teams may lead to focus on physical safety over independence (44,48) Self-determination as 'having control' and/or a mental attitude (43,41,54) Perseverance in communicating a message vs giving up/withdrawal affected by the perceived importance of message or sense of futility (42,45,48,49,53) Families as key patient advocates and motivators (43,53) The self and connecting with others Feeling incomplete without a voice (42,(44)(45)(46) Participants understood the concern for their safety, yet as those concerns prevailed, they spoke of a changing sense of freedom, independence, and a self that was, once again in jeopardy: "I was feeling caged, being very smothered and stifled, and that I was losing my own sense of who I am and that, you know.And it was making me hurt inside" p61 (44) I had expected to feel better.But it was just the opposite.I was laid down, now and then washed and combed.At six in the morning someone came and just started…did all sorts of things that I didn't want.p1106 (51) Some care assistants are more concerned about the ventilator than me p334 (47) Participants wanted to connect with others, and to do so, they found various voiceless ways to fill the void left from the impaired communication process.Participants attempted to express themselves in physical ways, often using a variety of communication means, even though this was arduous for them.p1170 (45) Risk of dehumanisation and focus on physical needs in absence of voice (42,44,47,51) Importance of relationships with others to mitigate isolation (44,45,48,53) Families act as intermediaries between patients and staff (43,53) Praise for staff when patients felt well-looked after both physically and emotionally Several of the participants also highlighted the importance of their relatives being familiar with their usual body language and gestures.This familiarity put them in a good position to understand and relay the participants' concerns to others p2300 (48) Nobody came…nobody was near me (surely they did come)…nobody...no, there wasn't anybody and then my desperation grew more and more…but probably they also didn't have the time… p28 (53) Speech gives patients a sense of freedom, control, ability to join in, request information and ask for things to be done (45,49,51,53) In Families provide huge support to patients and act as translators and advocates for patients (43,47,48,50,53,54) The constant presence of family members and health care professionals provided solace, comfort, and reassurance during a time wrought with uncertainty, stress and fear.In addition to providing emotional support, vigilant family members in this study may have served as a communication conduit for information delivery and reinforcement, thus reducing anxiousness and ambiguity.p332 (43) Several participants mentioned the importance of having eye contact and physical contact: When the anxiety comes, they were there straight away.They tried to calm you down, held your hand, and spoke to you.Communicated even though I didn't have a voice.And eye contact, that was very important for me then.p2300 (48) ...I really trusted them, the nurses don't know how good they are, ...they don't realize how important they are.... they treated me like a baby, they reassured me... p29 (53) Families are a source of comfort and motivation to recover (43,53) Staff Analytical Themes (ATs) The analytical themes and inferred answers to the research question were derived through methods described above (33).Similar to "best fit" framework synthesis (55), we combined inductive and deductive methods in our review.In contrast to this method, we selected and chose to incorporate a conceptual model after developing and in response to the DTs from stage 2 of our thematic synthesis.We had noted human aspects of experience and care in the descriptive themes and sub- The three analytical themes identified are: 'Being seen as a whole, unique, autonomous person'; 'Making sense of it, coping, and connections'; and 'Patients' voices as a key currency in humanising care' (see Figure 2.).These are described further below and presented in Table 3 with corresponding implications for practice.The overarching analytical theme is defined as 'To be seen and heard as a whole person'.Patients wanted to be treated as a human, and having a voice made this easier.
AT 1: Being seen as a whole, unique, autonomous person The evidence suggests that it is greatly important to patients to be seen for who they are as a person, not just for their medical needs.In the studies reviewed, not being able to speak threatened this through changing interactions between patients and staff.It made patients feel invisible (53), not valued as a human being (45-47, 51, 53), or treated as 'just a "body" on which people act' (53).
'they talked as if I weren't there at all.As if I were deaf, or not quite right in the head.. that sort of thing, a bad situation.It was degrading' (direct participant quote) ( 'Common to all participants were the communication challenges that occurred while they were in ICU on the ventilator.Their compromised ability to communicate resulted in feelings of frustration, vulnerability, isolation, and a diminished sense of self' (author quote) ( Perceptions of staff not making an effort to communicate with voiceless patients, not attempting to correct misunderstandings, or jumping to conclusions about a message ( were deeply upsetting to patients, and can be seen as dehumanising .In contrast, Flinterud and Andershed's (2015) (48) study highlighted the value of caring, attentive staff who acknowledged communication difficulties, took time to try to understand, were present at the bedside and communicated through verbal and non-verbal means.Patients wanted to have some control or influence over care and management decisions.This extended from immediate concerns, such as getting basic needs met (43-45, 48, 53), to longer term decisions around rehabilitation or discharge destination (44,49).Family was seen to be hugely important to patients (43,48,50,53).In the terms of humanisation, families afforded patients agency through interpreting communication more easily and acting as advocate for patient needs.Functional activities such as eating and drinking were seen to be important to patients for physical and psychological reasons and symbolised not only being human but also milestones in recovery (46,50,52,54).

AT 3: Patients' voices as a key currency in humanising care
Much of what mattered to patients depended on having a voice, both literal and metaphorical.For staff to see the patient as a whole person; understand their worldview; meet their physical, psychosocial, and emotional needs; and for patients to have a say in care, establishing communication was vital (42, 44-49, 51, 53).Patients communicated in a range of ways.Voice was valued over Alternative and Augmentative Communication (AAC), however, partly due to the efficiency of communication it afforded but also due to its contribution to patients' identities (42,45,46,(48)(49)(50)(51)(52)(53).Literal voice re-enforced patients' uniqueness as well as facilitating agency.
'It was a relief; just so, so good being able to speak...It was so good to be able to communicate normally again instead of trying to mouth words.It was just so much easier' (direct participant quote) (49) 'When this woman no longer had her tracheostomy tube and therefore was vocal at all times, she said "Look, I'm free!I'm so happy"' (author/direct participant quote) (45) *author suggestions and recommendations in primary study ** author data and/or interpretation from primary study support this interpretation

Discussion
This review suggests a discrepancy between the evidence base on tracheostomy management and the primary concerns of patients; while researchers have largely focused on technical issues of tracheostomy insertion and epidemiology, we found that patients' priority was to be seen and treated as a human.This discrepancy has significant and potentially detrimental impact on the clinical care offered to patients.

Descriptive Themes
The descriptive themes in stage 2 of our analysis provided the building blocks for our analytical themes and the practice implications we inferred.The first theme concurred with previous studies that have found an inability to communicate is one of the hardest things ICU patients have to face and leads to anxiety, frustration, anger, and untreated pain (38)(39)(40)(41)(42). Alternative forms of communication and staff training have been shown to improve patient communication (43)(44)(45)(46)(47).
However, supporting previous research findings we highlighted that AAC often fails (51), and our data corroborate previous assertions that patients most highly value having their own voice (28,48).
In line with theories of stress and coping (57,58), our second descriptive theme showed that sense of autonomy and self-determination aided coping whereas lack of control could lead to loss of coping and withdrawal, with implications for rehabilitation and recovery.This theme also described the impact of being on ICU with a tracheostomy on patients' sense of identity and inter-personal connections.We found little other evidence of this in the literature.The third descriptive theme found many patients were fearful and anxious, which is known to be common in ICU patients (59)(60)(61)(62).We found a lack of information and situational understanding contributed to fear and anxiety, and is intensified by voicelessness.Our fourth theme relating to physical experience identified sources of pain and discomfort that have been described elsewhere.However, pain and difficulty breathing were less dominant than experiences of thirst, swallowing difficulties and sense of physical restriction that resulted from being voiceless.We found that the fear of not being able to flag physical needs or call for help could cause more distress than the physical experience itself.Our fifth theme described positive influences on patient experience.

Analytical Themes
The analytical themes moved beyond the initial findings to develop a cohesive, conceptual picture of patient experience and provide interpretations of what matters most to ICU patients with a tracheostomy.This aspect of our metasynthesis of study findings was supported by a model of humanisation (Todres, 2009) (35).
Our finding of the fundamental need to be seen and treated as a whole person fits with philosophical theories of humanism (35,63), phenomenological embodiment (64) and personcentred care (65,66), and contradicts Cartesian views of mind-body separation or Maslow's hierarchy of needs model (67).We found that psychological, social and emotional needs of ICU patients with a tracheostomy were of fundamental importance, and that ignoring this risked patient dehumanisation.We suggested that the provision of adequate information that is tailored to patients' needs helps satisfy the fundamental human need to make sense of what is happening.This is consistent with the theory of 'facilitated sense-making', which, though developed to guide interventions to support families of ICU patients, states that when faced with crisis humans need to make sense of the situation and of their new role (68).Participants in our review placed great importance on relationships with others.Family presence brought solace and could facilitate communication, consistent with previous literature (69)(70)(71).In contrast, Halvorsen et al (2020) (72) found that family presence could in some circumstances be a source of distress, for example when patients were aware of the impact of their own illness on their family, and advocated a tailored approach to family visiting.Additionally, Broyles et al (2012) (73)identified that families often lacked skills to support their non-vocal relative to communicate, which could be upsetting to both family and patient.Interestingly, a recent development of facilitated sense-making has added 'patientfamily communication' to the model (74).Staff relationships were also important, and patients distinguished between two types of care: competent completion of tasks versus caring attitudes and behaviours, including efforts to communicate with them and being present at the bedside, echoing previous research (75)(76)(77)(78).Presence in turn supports communication and information exchange, allowing staff to get to know the 'person' in the patient.The Humanisation Value Framework (35) concept of 'togetherness' may help explain why nurse and family presence and caring attitudes and behaviours were significant to patients.We found that non-verbal communication from staff such as touch was an important means of connecting which fits with studies of the role of touch in human social bonding, stress and pain relief (79)(80)(81).
Our review supports Happ's (2000) concept of 'voicelessness' in intensive care (82), which describes the complex impact of communication impairment on the feelings and actions of patients, clinicians and families.Additionally, we highlight the importance of voice to identity and autonomy, lending support to the theory of voice as an embodied entity as described in one of the selected studies (45).
Interesting parallels are drawn between our study and Pound and Jenson's (2018) (83) investigation of humanising and dehumanising aspects of care reported by aphasic patients.In common with them, we found good communication between staff and patients was key to achieving humanised care.Interestingly, while we found that AAC was often associated with failed attempts at communication, non-verbal communication such as touch and eye contact was powerful in conveying caring and safety.It may be that this fits with findings of patients valuing naturalness of communication.Future research would be useful to investigate the apparent contradiction and explore ways to harness the broader spectrum of human modes of communication (84) to mitigate the impact of voicelessness on adults with a TT in ICU.

Clinical implications
There are notable similarities between the clinical implications we identified and recommendations from the International Research Project for the Humanisation of Intensive Care Units (HU-CI) (85-87), a Spanish-based group whose aim is to promote the humanisation of ICU through research,

Figure 1 .
Figure 1.PRISMA flow diagram of search results and article selection process

Figure 2 .
Figure 2. The analytical process and thematic maps of high-level descriptive and analytical themes -table 1 provides detail of each descriptive sub-theme

Table 1 .
(53)e 1 provides detail of each descriptive sub-theme Descriptive themes, sub-themes, and illustrative quotes Patients reported that they felt invisible in relation to the inability to communicate with others ad not being involved in the care plan, as if they did not exist.Some patients felt uneasy when the physicians did not speak directly to them as if the absence of their voice blocked any form of communication, including information."…thedoctors…never spoke to me.Until recently, they just didn't speak to me…now if I ask a question they give me an answer.All was well with the nurses, I didn't have any problems."p28(53) (53)rpretation/metaphorical: Communication or least the ability to speak is a fundamental need for most people; for Brian the stress of not being able to speak 'drove me nuts'.P1121(46)Several participants described how they were screaming inside because of pain, but were unable to communicate their pain to healthcare professionals.One participant described this sentiment well: Then it hurt so much I almost screamed my head off.Well, inwardly.But I didn't get out a damn thing.It really hurt.So, it sucks when you can't say anything.p2299(48)Theinabilitytospeakmadeitdifficultto interact with other people, and this generated in patients a sense of isolation linked to a feeling of 'counting for nothing' (i.e., worthlessness).''…IfeltasifIcountedfor nothing...I could not say anything...I couldn't interact..." p28(53)[on return of voice] I could say 'no, that I didn't want to be in this position', 'no, I didn't want that pillow behind my back', 'no, I didn't particularly want another nasogastric tube', things like that.p14(49)If it wasn't important, I wouldn't repeat it.If it was something important enough that it was bothering me or aches or pains, I would repeat it over and over and over until they got the message… And I'd keep persisting and persisting if it was important enough.If it wasn't important, I'd just forget it.p1171(45)…veryoftenthey didn't understand me…they thought I had a tic or a problem with my head instead I was trying to tell them that something was hurting me there… in the end I gave up p27(53) (45)rticipant reflected on the time when she was voiceless and stated that time passed "very badly.Slowly.And to me, it would never end.Then I was starting to think-I said, 'Oh, my God.This [pointing to tracheostomy, indicating not having voice] could go on and on and on forever.p1170(45)InICUthereisn'treallyanyroutine or pattern, it's just twelve hours of blur.Whereas, with eating, it breaks this up and makes it more like your normal day.The routine helped as I had a set pattern and knew what to expect.It also really helped me in getting ready to be transferred to the ward.p149(54)'It's a life-altering transition for me, but a positive life-altering transition in that I can be me again right now.You know, I take this whole disease day to day.p62(44)Several participants described how they were screaming inside because of pain, but were unable to communicate their pain to healthcare professionals.One participant described this sentiment well: Then it hurt so much I almost screamed my head off.Well, inwardly.But I didn't get out a damn thing.It really hurt.So, it sucks when you can't say anything p2299(48)Participants described impaired communication causing their needs for toileting, pain medication, and suctioning to go unmet..(….).One woman's silent screams were not heard when her oxygen tank ran out.This resulted in extreme anxiety for her because she felt the constant need to ensure that her oxygen tank was full.p1172(45) (49)ite the frustrations and discomfort that all the participants described, all of them appeared to understand the need for a tracheostomy.They emphasized that they had no choice.They viewed the tracheostomy as a means of survival p505(52)All I'm worried about is coughing this thing out, (indicates tracheostomy tube) which I nearly did!... p40(42)[on regaining voice] I could communicate.I could tell people what I needed and ask questions about what was happening and people don't think you want to know why, so they would say this is what happened but afterwards you could ask why, I , well, I could ask questions.p14 (49)When I came out of the coma it took me a while to fully understand what was going on.Also because you are in a confused state, you forget where the buzzer is sometimes and (I) was also getting medication, which was given to me that also doped me up a little bit more.Even though I was conscious I was very confused and so the lack of speech makes it worse (Roger) p13(49)This may help acceptance of restrictive ICU treatments such as TT but also engender fear, e.g. of TT blockageAhhh!That's when you get a bit of pain, because around this bit (indicates the tracheostomy stoma)… is very tender….You can feel it all the way around because when you cough, you're retching at the same time.. and then it is painful!p40 (42)all the participants commented on difficulties with swallowing and some problems with eating and drinking [Swallowing] was awkward too.Very difficult.It still is actually (Pt B) I was uncomfortable with it because every time you tried to swallow ...I must say that the nurses were really good and I will always say this, they were very professional... then we would also joke and laugh... look I say it from the bottom of my heart, because when they passed they said hello and asked how I was feeling and this was comforting for me... there were these people who really cared about me..." [on being told she would get her voice back] I said "Oh, my God answered by prayer."I was so pleased, so happy.And then the first day I got my voice, it's like a grand opening, you know…and I can relate to [others] what went on during the day.Just instead of sitting there like a bump on a log.p1169 (45) (46)ink the best way is once you've had it done once and you know you've gotta have is done a second time is just to grin and bear it, really attack it mentally otherwise it'd jut bloody drive you crazy I think p1119(46) (54)ons.He expressed the need to 'start shedding attachments', which enabled him to 'start believing that the end is in sight'.p144(54)Formany,commencingoral intake indicated that they were getting better.It was described as symbolizing the 'road to recovery'.Fred expressed it as "being very glad to have got to that stage, as it signified that things were moving in a positive way.Therefore, it was a step on the road to recovery, so I was delighted to have got there.It was terribly important."Itwouldappear that often it is less the need for food, than what eating signifies that is important."Eatingispart of normal life, it wasn't like I was dying for a plate of fish and chips, but it represented a return to normality and that was so important."p144(54)Charlesexplained that return of voice was positive but also a sign of recovery of general health status, ''It was pleasing to be able to speak again.Mainly you felt as though you were getting better, recovering from the worst of your illness".In agree-ment Rodger reported, ''In my opinion that slows your recovery down as you're frustrated and start to go into a shell and it adds to the confusion.In my opinion, talking helped me recover quicker.I'm no doctor but it made me feel a lot better, when you feel better you recover.
and used Todres et al's (2009) (35) Humanisation Value Framework for healthcare to help shape a cohesive conceptual model of the experience of having a tracheostomy in ICU.Todres et al's (2009) eight dimensions of humanisation are shown in Table2(see supplementary material G for a table showing DTs against eight dimensions of the Humanisation Value Framework).This step helped reveal the interplay between DTs and move from surface level findings to higher level themes and conceptual understanding of patient experience of tracheostomy in ICU.It also aided interpretation of implications for practice.Analytical themes were cross-checked with original texts and descriptive themes to ensure they captured and explained the core concepts identified.Implications for clinical practice were also identified.

Table 3 .
Analytical Themes and Clinical Implications