Abstract
Oxygen therapy is frequently prescribed for the palliation of breathlessness, despite lack of evidence for its effectiveness in people who are not hypoxaemic. This study aimed to compare and contrast patients’, caregivers’ and clinicians’ experiences of palliative oxygen use for the relief of chronic breathlessness in people with advanced life-limiting illnesses, and how this shapes prescribing.
A systematic review and meta-synthesis of qualitative data was conducted. MEDLINE, CINAHL and PsycINFO were searched for peer-reviewed studies in English (2000–April 2019) reporting perspectives on palliative oxygen use for reducing breathlessness in people with advanced illnesses in any healthcare setting. After data extraction, thematic synthesis used line-by-line coding of raw data (quotes) to generate descriptive and analytical themes.
Of 457 articles identified, 22 met the inclusion criteria by reporting perspectives of patients (n=337), caregivers (n=91) or clinicians (n=616). Themes common to these perspectives were: 1) benefits and burdens of palliative oxygen use, 2) knowledge and perceptions of palliative oxygen use beyond the guidelines, and 3) longitudinal trajectories of palliative oxygen use.
There are differing perceptions regarding the benefits and burdens of using palliative oxygen. Clinicians should be aware that oxygen use may generate differing goals of therapy for patients and caregivers. These perceptions should be taken into consideration when prescribing oxygen for the symptomatic relief of chronic breathlessness in patients who do not quality for long-term oxygen therapy.
Abstract
Patients’, caregivers’ and clinicians’ differing perceptions of the benefits and burdens of palliative oxygen should be considered when defining the goals of therapy and determining clinically relevant interventions offered for chronic breathlessness http://bit.ly/2ZE285P
Introduction
Chronic breathlessness is a cardinal symptom across many life-limiting illnesses, including cancer, respiratory, cardiac and neuromuscular diseases [1], and one of the most feared aspects of dying [2]. At a population level, approximately 9% of people live with some degree of chronic breathlessness and 2.6% of people are housebound due to extremely debilitating breathlessness [3]. In later disease stages, breathlessness prevalence increases greatly [1] and tends to intensify as death approaches [4]. Chronic breathlessness caries a significant symptom burden affecting people's physical and psychosocial wellbeing, resulting in decreasing functionality and increasing social isolation and dependence on others [5]. Caregivers’ quality of life is also detrimentally affected chronically [6].
Oxygen therapy is frequently prescribed for the palliation of breathlessness in people who do not fulfil current criteria for long-term oxygen therapy (LTOT), despite lack of evidence for its effectiveness in providing symptomatic relief [7, 8]. Current guidelines limit the funded prescribing of LTOT to hypoxaemic patients (arterial oxygen tension <55 mmHg) [9]. Palliative oxygen continues to be initiated for the management of breathlessness in nonhypoxaemic patients with advanced life-limiting illnesses despite growing evidence for its inappropriateness [10, 11], often with little or no reference to the underlying pathophysiology [11]. A recent multinational survey to which 440 respiratory and palliative medicine specialists responded found that palliative oxygen prescribing is decreasing, but that the majority of palliative and just over half of respiratory physicians continue to prescribe it for the symptomatic relief of chronic breathlessness in the palliative care setting, outside current guidelines [12].
Oxygen therapy carries appreciable burdens. People using LTOT may face physical, psychological and emotional challenges, including having to adapt to new life circumstances, living in a restricting world, and submission to and dependency on a chronic therapy [13]. Self-immolation (for people who smoke) can also be a problem [14], as can the psychological burden of having to rely on a machine [15]. Data on quality of life and patient preferences using oxygen are conflicting, often finding that people prefer to discontinue their oxygen due to poor tolerability [16].
The multidimensional nature of chronic breathlessness and individual responsiveness to treatment require an understanding of the broader context in which oxygen is prescribed and used. Ideally, this should include the person receiving the oxygen therapy together with their caregiver and the clinicians providing care for them. Individual patient and clinician perceptions of oxygen therapy have been synthesised [13, 17, 18], but are lacking for caregivers. Importantly, there is a gap in the literature that compares and contrasts perspectives that capture the complexity around oxygen prescribing, assessment and impact on patients’ and caregivers’ everyday lives. Combining patients’, caregivers’ and clinicians’ perspectives, and comparing them with the evidence base, is important to help determine the most relevant patient and caregiver clinical outcomes, and the drivers for prescribing and adherence to oxygen therapy in clinical practice [19]. This, in turn, will help inform future research interventions for this population, ensuring they are aligned with people's needs, values and preferences.
This review brings a triangulated perspective (patients, caregivers and clinicians) on the use of oxygen for the symptomatic relief of chronic breathlessness in people with an advanced life-limiting illness who do not quality for funded LTOT, hereafter called “palliative oxygen”. The review defines chronic breathlessness as breathlessness that persists despite optimal treatment of the underlying condition(s) and life-limiting illness as any incurable malignant or nonmalignant illness from which a person is likely to die [20, 21].
Our aim was to compare and contrast patients’, caregivers’ and clinicians’ experiences of palliative oxygen use for the relief of chronic breathlessness in people with advanced life-limiting illnesses, and how this shapes prescribing.
Methods
This study was a systematic review and meta-synthesis [22, 23] of patients’ with advanced life-limiting illnesses, their caregivers’ or clinicians’ narratives reporting their perceptions and experiences of using palliative oxygen. The conduct and reporting followed the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analysis) guidelines [24]. The systematic review protocol was registered at PROSPERO with identifier number CRD42019140813.
Eligibility criteria
Studies were included if they were peer reviewed, published in English from 2000 onwards, and reported the perspectives of patients, caregivers or clinicians on the use of palliative oxygen for the symptomatic relief of chronic and acute-on-chronic breathlessness in people with an advanced life-limiting illness. Studies had to report empirical qualitative data, defined as raw data, i.e. direct quotes (e.g. verbatim patient statements). Articles were required to contain at least one sample of raw data in line with minimum reporting standards [25]. Data collection methods were unrestricted (e.g. interviews, focus groups, open-ended survey questions, etc.) [26–30], as were approaches to data analysis (e.g. phenomenology, grounded theory, etc.). Quantitative and mixed-methods studies were included if they reported raw qualitative data separately in their findings. Studies reporting quantitative results only were excluded, as were (systematic) reviews, commentaries, editorials, case reports, dissertations and abstracts. Primary studies in (systematic) reviews were screened for eligibility.
The study population comprised adults (≥18 years old) with advanced life-limiting illnesses who had been prescribed oxygen for the symptomatic reduction of their breathlessness. The use of oxygen could be for chronic and acute-on-chronic breathlessness, and in the inpatient, outpatient or community (home and aged care) settings. Studies reporting the use of oxygen in populations other than people with advanced diseases were excluded.
Information sources
Databases searched included MEDLINE, CINAHL and PsycINFO, augmented by desktop searching of CareSearch for PubMed and CareSearch Grey Literature (if articles resulted in a peer-reviewed publication). Lateral searching of Google Scholar and the reference lists of included studies and relevant reviews was conducted manually. Database and desktop searches were conducted on 5 and 7 April 2019.
Search
The search strategy was informed by two systematic reviews on the use of oxygen [7, 16], and included Medical Subject Headings (MeSH) terms and text words for these domains of interest: 1) dyspnoea/dyspnea/breathlessness/shortness of breath, 2) oxygen/oxygen inhalation therapy/oxygen therapy/long-term oxygen therapy/supplemental oxygen/home oxygen/domiciliary oxygen and 3) palliative care (using the Palliative Care Search Filter for MEDLINE and text word for all other databases) [31]. CareSearch for PubMed used the search for “dyspnoea” combined with “oxygen” as key words. CareSearch Grey Literature was searched using “oxygen” and “breathlessness” as key words. The MEDLINE search strategy is provided in supplementary table S1.
Study selection
Results were imported into EndNote X9 (Clarivate, Philadelphia, PA, USA) and duplicates removed. Title/abstract screening with full-text review was independently performed by two researchers (S.K. and D.F.) using Covidence systematic review software (Veritas Health Innovation, Melbourne, Australia).
Data collection and data items
Data extraction was two-fold. Raw quotes relating to people's experiences and perspectives on the use of palliative oxygen for the symptomatic relief of breathlessness were imported into NVivo 12 (QSR International, Doncaster, Australia) for coding and analysis (M.G.). Additional data items were extracted (S.K. and D.F.) using electronic proformas (Word/Excel 2016; Microsoft, Redmond, WA, USA). These included author, year, country, aims, sample characteristics (patients/caregivers/clinicians, sample size setting and sociodemographics), study characteristics (design, data collection method and theoretical framework), themes/findings and conclusions. For studies with patient participants, data were extracted on the person's health condition (e.g. type/stage/duration of disease, was oxygen prescribed to people who do not qualify for funded provision under current international guidelines [32] and length of oxygen therapy). For studies with caregiver participants, data were extracted on the nature of the relationship to the person receiving oxygen, the patient's health condition and the length of caregiving. For studies with clinician participants, data were extracted on any professional characteristics reported (e.g. discipline and years of experience).
Quality appraisal
The Standard Quality Assessment Criteria for Evaluating Primary Research Papers from a Variety of Fields [33] was used to appraise studies of varying design. Study quality was assessed independently by two reviewers (M.G. and D.F.); disagreements were resolved by a third reviewer (S.K.). For qualitative studies, the sum of all scores (“yes”=2; “partial”=1; “no”=0) was divided by the highest possible score of 20. For survey studies with free-text comments, the sum of all scores (“yes”=2; “partial”=1; “no”=0) was divided by the highest possible score of 28, minus any nonapplicable fields where the survey was purely descriptive.
The quality appraisal did not eliminate studies but provided an overall view of the quality of data collection and reporting. The threshold for high quality was set at ≥0.75 [33].
Synthesis of results
A three-stage thematic synthesis [22] was facilitated by: 1) free line-by-line coding of extracted raw data, 2) organisation of the “free codes” into “descriptive” themes, commencing the synthesis process, and 3) development of “analytical themes” informed by the “descriptive themes” to answer the research question.
Initial line-by-line coding was performed for the raw data of each included article (M.G.), and then discussed until consensus was reached on its accuracy and consistency (S.K. and D.F.). The development of descriptive and analytical themes was carried out by S.K. and D.F. in the first instance, and then discussed with the remaining authors (M.G., J.P. and D.C.) until consensus was reached. Descriptive and analytical themes were derived iteratively. All qualitative data (raw quotes) contributed to theme development.
Combining the three perspectives used the framework analysis [34] consisting of: 1) familiarisation (reading the quotes and results from included articles), 2) identifying a thematic framework (developing a coding system for descriptive themes consistent across different participants’ groups), 3) indexing (open coding of the findings section of each article and comparison of findings between articles), 4) charting (developing analytical codes and regrouping data into overarching themes), and 5) mapping and interpretation (identifying similarities and differences between codes from different groups, i.e. patients, caregivers and clinicians, using a framework matrix).
Results
Study selection
Of 457 articles identified, 22 met the inclusion criteria and were included in the meta-synthesis (figure 1) [35–56]. The characteristics of the study populations and included studies are summarised in tables 1 and 2, respectively.
Study population
Most studies were undertaken in high-income countries [35–49, 51–56], other than one study from China [50]. Of the 1044 participants, most were clinicians (n=616), with under a third being patients (n=337) and less than 10% being caregivers (n=91). The majority of studies only included patients (n=12) [35, 36, 39, 41, 42, 44, 46, 47, 49–51, 54, 55]. A smaller number included patients only with caregiver quotes (n=1) [35], caregivers only (n=2) [38, 43] or clinicians only (n=4) [37, 48, 52, 56]. Two studies included mixed patients/caregivers/clinicians [45, 53] and one study included mixed patents/clinicians [49].
The primary underlying conditions generating breathlessness were chronic obstructive pulmonary disease (COPD) (n=200 (59%)), followed by cancer (n=63 (19%)). Of the caregivers, 42% were spouse/partner (n=38) or relative (n=38); relatives included adult children, siblings or in-laws. Across studies, 77% of clinicians (n=477) were doctors, followed by nurses (n=122 (20%)).
Study design
Data in the contributing studies were collected from published semistructured interviews (n=16) [35, 36, 38, 39, 42–44, 46–51, 53–55], combined interviews and focus groups (n=1) [45], combined quantitative methods and interview (n=1) [40], interviews, phone calls and field notes (n=1) [41], focus groups (n=1) [37], unstructured interviews (n=1) [54], survey combining quantitative methods with open-ended questions (n=1) [52], and survey combining case vignette with free-text comments (n=1) [56].
Approaches to data analysis included phenomenology (n=6) [41, 42, 44, 48, 49, 54], grounded theory (n=5) [35, 38, 40, 51, 53], symbolic interactionism (n=1) [37], constructivism and naturalistic theory (n=1) [36] or was not reported (n=9) [39, 43, 45–47, 50, 52, 55, 56].
Study focus
More than half of the studies focused on the use of ambulatory oxygen or home oxygen or LTOT (n=12) [35–40, 45, 47–49, 52, 56], while a smaller proportion focused on advanced or severe COPD at the palliative stage or at the end of life (n=6) [42–44, 46, 51, 53]. The remaining studies focused on advanced COPD and LTOT (n=1) [41], severe hypoxic COPD (n=1) [54], breathlessness in advanced lung cancer (n=1) [50], and episodic breathlessness across different aetiologies (n=1) [55].
Length of oxygen therapy was reported in one study where patients (n=4) received oxygen 16–24 h·day−1 for 1 year (n=1) or 2 years (n=3) [41]. In one study oxygen use lasted 8–48 months (mean 22 months) [54]; in another study the duration of therapy was not specified but oxygen was received for 16 h·day−1 [42].
Methodological quality of studies
The qualitative studies scored generally well, ranging from 0.6 to 1.00, with a median (interquartile range) of 0.8 (0.2), with the majority of studies scoring low in the use of verification procedures and in reporting reflexivity. The surveys [52, 56] scored generally high at 0.9 and 0.8, with a median of 0.85.
Themes
The synthesis generated 60 line-by-line codes, 10 descriptive themes and three analytical themes (table 3). These include: 1) benefits and drawbacks of palliative oxygen use, 2) knowledge and perceptions of palliative oxygen use beyond the guidelines, and 3) longitudinal trajectories of palliative oxygen use.
Theme 1: Benefits and drawbacks of palliative oxygen use
Participants reported various benefits for palliative oxygen use, but not without drawbacks and impacts on daily living. Four subthemes informed this broader theme.
Benefits
Most patients, caregivers and clinicians (health professionals (HPs)) identified psychological [38, 47] and therapeutic benefits of palliative oxygen use [45, 55], including a reduction in breathlessness [55].
“You know, because honestly when you see someone's lips going blue and they are trying to breathe and can't, you know, it is horrible. You know, especially if it is your dad or your mum or somebody you love. It is horrendous. It is horrendous. So at least with the oxygen you quickly stick it up his nose and it helps a bit. It takes a while to kind of get in them or something but yeah.” [Caregiver] ([38], p. 35)
“Some patients find oxygen helpful but I think this is not because of increasing the SpO2, I think it is psychological and one can argue the pros and cons of this, although I consider it inappropriate to give oxygen therapy in these circumstances.” [HP] ([56], p. 254)
Patients reported palliative oxygen reduced their usage of health services [45], and along with their caregivers, noted improvements in sleep, independence and quality of life [45].
“Prior to that I would about 1 or 2 hours and now I sleep all night so that's made a big difference.” [Patient] ([45], p. 5)
“It is not going to change his life but it has picked him up. It has given him a better quality of life.” [Caregiver] ([45], p. 6)
Clinicians reflected that palliative oxygen provided emotional comfort [56] or sense of security [37], while caregivers reported it eased anxiety and distress [38]. However, some patients reported a lack of benefits [54].
“I've been on oxygen three years and I still get progressively worse. I don't know if that is my fault or the illness.” [Patient] ([54], p. 41)
Burdens/harms
Patients noted that handling and managing oxygen equipment was burdensome [41]. The judgement or pity associated with the need for palliative oxygen impacted on patient's self-perception and generated social stigma [35, 36, 40, 45, 50]. For others, the benefits of the oxygen outweighed these burdens [47].
“Wearing an oxygen cannula just looks like someone addicted to drugs. How can I face others? They think that I am an addict.” [Patient] ([50], p. 5)
“… it's just that I feel that people would stare at you and it would make me feel very uncomfortable” [Patient] ([35], p. 4)
“I am willing to put up with the nuisance for the benefit of oxygen; the burden does not mean much.” [Patient] ([47], pp. 767–768)
Caregivers reported they take on the responsibility of handling and carrying oxygen equipment to reduce the burden for patients though this increased their responsibility and added to their caregiver load [35, 38].
“She doesn't think it's a problem – the weight – because she doesn't carry it, I put it in the car between us and she uses it as she needs to when we're going along. I sort out all the oxygen, we have one in the car and this one in the house is a back-up. I have to check them all the time ‘cause she leaves them on.” [Caregiver] ([35], p. 4)
Impact on daily life
Patients (and sometimes caregivers) who were extremely limited by breathlessness tended to report that palliative oxygen was beneficial, increasing the physical and social space available to them and enhancing their independence [38, 45, 54].
“I feel as though I've got some independence back whereas before I was 100% reliant on other people doing things.” [Patient] ([45], p. 6)
Conversely, for patients with higher levels of function, palliative oxygen was considered to hinder daily activities and restrict freedom of movement [45, 53].
“I used to dance and travel; then I was diagnosed with COPD and oxygen therapy, everything stops – it's like you are on a leash, tied to a regimen. I feel sad I can't do the things I would like to do with my kids … go to the horse races. I haven't done anything I like to do.” [Patient] ([53], p. 606)
Safety concerns (clinicians, patients)
To clinicians, patients living alone were a prescribing risk [37]. This was amplified if the patient smoked, as this posed a fire hazard [49], particularly in larger households [37, 49]. Use of oxygen equipment alongside other tools or equipment where ignition was possible was also seen as a safety issue [37].
“For people who live on their own, that index of caution is just ratcheted up a little bit higher.” [HP] ([37], p. 271)
Oxygen equipment itself was a hazard due to its bulkiness [36], but concerns for safety also arose from the image it created. Patients were concerned that having an oxygen tank with them signalled vulnerability and the presence of their illness, making them a “target” [36].
“I'm supposed to go out with a bottle when I go shopping … but I don't take it out [with me] because I look at myself like I'm gonna be at target. If youngsters see me walking round with one of those [oxygen cylinder], they'll think, ‘well there's an old guy, he is stuffed, we'll roll him [steal from him] … I don't want anybody to know I'm crook.” [Patient] ([36], p. 86)
Theme 2: Knowledge and perceptions of palliative oxygen use beyond the guidelines
Knowledge of the evidence base and perceptions of palliative oxygen use influenced prescribing and decision making in various, often contradicting ways. Two subthemes informed this theme, drawing on data (largely) provided by clinicians.
Knowledge of the evidence base
Clinicians expressed scepticism about palliative oxygen outside of guideline recommendations, noting lack of evidence for its effectiveness in many populations [48]. Some considered that prescribing palliative oxygen to nonhypoxic patients was deceitful when it did not reflect the clinician's knowledge of the treatment [48].
“it just seems a little bit unethical” [HP] ([48], p. 955)
Others considered it to be just another treatment option [53], causing no harm [49]. Of note, many clinicians expressed a lack of competency for assessing, prescribing and managing their patients’ palliative oxygen [49].
“it's just oxygen” [HP] ([49], p. 624)
“[many HPs] don't really feel competent to prescribe and adjust” [HP] ([49], p. 623)
By contrast, patients had little prior knowledge of palliative oxygen prescribing and use, and deferred these clinical decisions to clinicians [49].
“Because I've been told to do that” [Patient] ([49], p. 627)
Reactive prescribing outside of current guidelines
Clinicians had difficulties in following guidelines [45], sometimes resorting to reactive prescribing of palliative oxygen in response to patient [49] and caregiver/family needs and expectations [52].
“… clinicians feel something's being done, because patients expect something to be done, and when it's not done can feel very abandoned …” [HP] ([49], p. 626)
“I will typically provide information on the benefits/burdens of supplemental oxygen and allow families control of this, especially if it seems they need this.” [HP] ([52], p. 11)
Clinicians also acknowledged that palliative oxygen may be prescribed to frail patients to palliate their breathlessness outside prescribing guidelines [56], often for compassionate reasons. This prescribing practice was often justified by the widespread use of oxygen in acute care settings [37] and especially at the end of life [48].
“You wouldn't want anybody to die not being able to breathe … it must be awful to die with a sense of breathlessness, must be so frightening …” [HP] ([48], p. 954)
Resistance to change and practice culture [49] also influenced clinicians’ prescribing and palliative oxygen use.
“Reflects on entrenched practice: ‘anybody and everybody had an oxygen mask on, no matter what’.” [HP] ([49], p. 624)
Theme 3: Longitudinal trajectories of palliative oxygen use
Once prescribed, adherence to palliative oxygen therapy is influenced by people's experience, beliefs, dependency and safety concerns, resulting in dynamic re-assessment and individualised patterns of use. Four subthemes informed this broader theme, incorporating patients’, caregivers’ and clinicians’ views.
Beliefs
Patients’, caregivers’ and clinicians’ beliefs influenced adherence [38, 39, 48] and nonadherence to prescribed palliative oxygen therapy [45].
“If you know it's there you seem to relax but if you say to yourself, ‘I have no oxygen’, then you'd start to panic I think and that's when the attacks come on.” [Patient] ([39], p. 792)
Patterns of oxygen use (patients)
Many patients reflected on their use of oxygen over time, having to become accustomed to its use at the beginning but finding it easier to use over time [54]. Patients reported titrating the oxygen flow rate to accommodate their activities of daily living [44]. Some were resigned to the fact that they would have to use palliative oxygen forever [53].
“Well, I get short of breath … so I set the oxygen higher … After I've finished bathing and drying myself and dressing, I can put it back down.” [Patient] ([44], p. 553)
Re-assessment of oxygen use (patients, caregivers, clinicians)
Caregivers were critical of the therapeutic oxygen monitoring process, noting that clinicians should provide more hands-on assessment of the dosage and need for palliative oxygen [43]. Clinicians encouraged self-monitoring as part of ongoing assessment for its effectiveness [37]. Patients reported that self-assessment sometimes resulted in their completely ceasing the therapy [36].
“If someone was severely dyspneic today and we put oxygen in, I would get them to phone me if things hadn't resolved.” [HP] ([37], p. 271)
Oxygen dependence (patients, clinicians)
For patients, increase in dosage was a concern, indicating they may be becoming addicted to palliative oxygen [40]. Clinicians noted that patients easily become over-reliant on palliative oxygen therapy [49], and that their dependency can also be linked to fear and anxiety that arises when oxygen is not on [48]. Patients made a conscious effort to reduce palliative oxygen use during periods of rest because they were worried that continuous use would result in ineffectiveness [40].
“It does bother me to feel this way, because I wonder, am I addicted to it? See, I started at 1 and I am up to 4 liters now.” [Patient] ([40], p. 752)
Discussion
This is the first study to compare, contrast and synthesise patients’, caregivers’ and clinicians’ perspectives on the use of palliative oxygen for the symptomatic relief of breathlessness in people with advanced life-limiting illnesses. The findings are presented within a proposed research agenda to advance the science of palliative oxygen use in this setting (table 4) [19].
What is the true role of oxygen therapy in clinical practice?
When considering the use of palliative oxygen for the relief of breathlessness, the goals of treatment need to be clearly defined based on patients’ and caregivers’ priorities [19].
Function and activities of daily living
Maintaining functionality and independence is crucial for people who are chronically breathless. This review found that palliative oxygen can facilitate or hinder their mobility, which strongly influences the net benefits of this therapy (weighing benefit and harms). Patients who are severely limited by chronic breathlessness and housebound are more likely to report an overall positive experience with palliative oxygen therapy. For this population, the burdens of palliative oxygen are small when compared with the gains in autonomy and freedom of movement. Conversely, patients who are less limited by their breathlessness report that the oxygen equipment reduces their mobility and ability to engage in social activities. These findings align with the body of evidence suggesting that people with severe breathlessness prioritise maintaining their autonomy and “life space” (i.e. the actual physical and social space available to them) [41] until later stages of their disease [57–60].
Psychological benefit
All groups in this review reported that palliative oxygen is sometimes helpful to reduce the psychological distress associated with chronic breathlessness, especially when experiencing acute exacerbations. This trust in the effectiveness of oxygen reduced the anxiety associated with breathlessness, breaking the breathlessness–anxiety–breathlessness cycle [61, 62]. Caregivers are often significantly distressed by patients’ breathlessness [6], so the availability of palliative oxygen may help to ease their anxiety by believing something is being done, particularly at the end of life [63].
Quality of life and health services utilisation
Some patients reported improved quality of life and fewer hospitalisations as benefits experienced with palliative oxygen therapy [45]. Patients and caregivers also reported improvement in sleep [45]. This may be a direct result of small gains in function (due to enhanced blood oxygenation) or reduced anxiety associated with breathlessness. Improved function and interruption of the breathlessness–anxiety–breathlessness cycle may also help reduce deconditioning, which is a key contributing factor for disability associated with breathlessness [45, 60, 64–66].
What are the clinically relevant burdens or harms associated with oxygen administration?
Environmental issues
For patients who tolerate higher degrees of exertion, palliative oxygen can hinder mobility and potentially lead to greater social isolation [19, 45, 53]. These patients report that the use of palliative oxygen may affect their entire personhood, posing physical, psychological and social challenges, and increasing reliance on others. Aspects related to equipment management and safety concerns feature prominently in the narratives of patients, caregivers and clinicians. Safety concerns involve tripping on equipment or fire hazards. The presence/absence of a caregiver can be a mitigating factor in how well people cope with these challenges daily and over the course of treatment.
Psychological issues
Psychological dependence on oxygen is a key harm reported by all groups in this review. Palliative oxygen has a psychological impact on patients and caregivers while using it, and on clinicians when deciding its appropriateness. Caregivers find palliative oxygen reassuring because it provides them with concrete tools to help patients with their breathlessness. For patients, however, palliative oxygen may carry a psychological and emotional burden as it restricts their social interactions and space [41]. Safety concerns can also be psychological when they are out and about in public, with patients reporting vulnerability and negative perceptions. This means burdens and harms need to be looked at in the broader context of the patient–caregiver partnership. For patients who do not have caregivers, these burdens will be amplified as patients would have to deal with them alone (e.g. when handling the equipment, being less mobile, in interactions with others, etc.).
Who are the drivers for oxygen prescribing by clinicians: patients or caregivers?
Evidence-based compared with reactive prescribing
The drivers for prescribing palliative oxygen are complex. The net benefits of palliative oxygen for the relief of chronic breathlessness in people with advanced life-limiting illnesses who are not hypoxaemic are still unknown [8, 59] and clinicians are often unsure of the outcomes they are trying to achieve. Often, their focus will be symptom relief, but compassionate prescribing means palliative oxygen will be given for patient/caregiver support in the hope that it will not do much harm. This is particularly true towards the end of life, when decisions to initiate/maintain palliative oxygen are often driven by caregivers and clinicians, potentially to the exclusion of patients.
This review suggests that prescribing is often for compassionate reasons: to ease caregiver distress, because the person is at the end of life or because palliative oxygen might provide emotional comfort and a sense of security to either the patient or the caregiver. This, in part, may be a reflection of more intense breathlessness experienced by the patient in the last weeks/days of life [67, 68]. Addressing patients’ and caregivers’ distress is an important aspect of providing care at the end of life [69, 70] and a common goal of care for all parties involved. Although palliative oxygen may play a role in such situations, its use needs to be considered within a larger framework with other evidence-based interventions [71–75]. There is also an element of palliative oxygen being an intuitive therapy (“if you are out of breath it means you need more oxygen”) and this needs to be addressed in clinical practice.
Established practices
This review found that established practice culture and (lack of) knowledge of current prescribing guidelines are contributing factors when initiating and assessing the use of palliative oxygen. Oxygen use in acute settings [7, 76] can be a major driver for clinicians to prescribe palliative oxygen outside of guidelines [77, 78]. Uncertainty about the evidence can also compound this practice. Importantly, improving the functional independence of the patient (e.g. their mobility or ability to perform daily activities) was not a stated consideration for clinicians when prescribing palliative oxygen. Although this aligns with current evidence that ambulatory oxygen compared with medical air in people with COPD does not improve function [59], it raises the question of how clinicians evaluate relevant outcomes for their patients and caregivers [19] when deciding treatment.
Patients defer to their doctors the decision of initiating palliative oxygen and there is a sense of “the doctor knows best”. Once prescribed, however, patients actively monitor the benefits/harms and reassess their need for palliative oxygen (when and in what circumstances), thus managing their oxygen use without seeking clinicians’ advice. For clinicians, knowledge of the evidence base is important when initiating and continuing palliative oxygen, and any lack of knowledge or uncertainty means palliative oxygen will be prescribed outside of the prescribing guidelines. Re-assessment is often not consistent.
Caregiver availability
A large cohort study of people on home LTOT for chronic breathlessness at the end of life found that not having a caregiver significantly reduces the likelihood of being prescribed oxygen [63]. This review provides potential explanations for these findings, while confirming that caregiver availability is one of the key factors influencing palliative oxygen prescribing and adherence. Palliative oxygen is often prescribed to ease caregivers’ anxiety; it helps them to have something with which to respond when the patient was seen to be breathless. Clinicians feel more confident prescribing palliative oxygen to patients with an available caregiver, mainly due to the risks associated with its use. A new finding from the current review is that caregivers play a significant role in ameliorating the drawbacks of palliative oxygen use (e.g. assisting with equipment), thus helping patients cope better with the prescribed therapy. In many cases, patients would not be able to cope with the demands of the therapy alone, a factor that needs to be taken into account by clinicians when considering palliative oxygen therapy.
Implications for clinical practice and education
The review's findings have implications for clinical practice and education (table 4).
The findings suggest that the relevant outcomes of palliative oxygen therapy ought to be determined by and tailored to the individual(s), in the context of their clinical history, discussed at the start of therapy, and assessed over time to ensure that the intervention aligns with patients’ and caregivers’ needs and preferences. Acknowledging the burden (physical, emotional and psychological) experienced by patients and caregivers is important. Identifying the specific challenges and daily activities hindered by chronic breathlessness [60] may help to select patients that are more likely to derive benefit from palliative oxygen therapy, as would assessing desaturation associated with tasks that the patient finds important. Sleep assessments should also be routinely conducted. Safety concerns (environmental and psychological) should be discussed when therapy is initiated and as it progresses. These should include issues around the practical handling of equipment and home environment. Psychological assessment is critical for both patients and their caregivers, especially the impact of oxygen on their social interactions and independence. Establishing a shared understanding of the goals when considering prescribing outside the guidelines is important, especially at the end of life. Clinicians should be aware that their goals may not always align with those of their patients and caregivers, so eliciting patients and caregivers concerns should be a priority and incorporated in a shared decision-making process when prescribing palliative oxygen therapy. Once prescribed, clinicians should regularly assess oxygen use and monitor adherence. Burdens and harms should be assessed for the patient–caregiver unit and support offered to patients who do not have a caregiver to minimise those risks.
Strengths and limitations
The rigorous methodological approach to synthesise the qualitative data and identify themes (i.e. generate findings) enabled a triangulated perspective of the use of palliative oxygen in people with advanced life-limiting illnesses. Identifying what is unique and shared across the populations of interest (patients, caregivers and clinicians) has provided significant insight into the experiences, painting a more comprehensive picture of what life is really like for patients using palliative oxygen, their caregivers and clinicians. Generating themes based on reported quotes means the findings are directly representative of patients and caregivers voices.
Incorporating palliative care in the search strategy means that people with late-stage disease who have qualified for LTOT but are not receiving palliative care may have been missed. The majority of the patient population in the included studies had COPD as the primary condition generating breathlessness compared with other diseases, which limits insights into the use of palliative oxygen for other patient populations. Combining the various conditions and healthcare settings may also limit applicability, and future studies should look to delineate the use of palliative oxygen in these settings. The original authors were not approached to provide clarification about the numbers of patients receiving LTOT across the studies. The synthesis relied exclusively on the use of raw data (i.e. quotes) and did not consider the themes generated by the authors of the original articles, which means issues not reflected in the quotes (but summarised in those themes) may have been missed.
Conclusions
There are differing perceptions regarding the benefits and burdens of using palliative oxygen. Clinicians should be aware that oxygen use may generate differing goals of therapy for patients and caregivers. These perceptions should be taken into consideration when prescribing oxygen for the symptomatic relief of chronic breathlessness in patients who do not quality for LTOT. Future interventions should use outcome measures that are important to patients and caregivers, and make a difference to their wellbeing. Before initiating palliative oxygen, clinicians have an opportunity to explore the meaning of the therapy to patients and, separately, caregivers.
Supplementary material
Supplementary Material
Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.
Supplementary table S1 ERJ-04613-2020.SUPPLEMENT
Shareable PDF
Supplementary Material
This one-page PDF can be shared freely online.
Shareable PDF ERJ-04613-2020.Shareable
Acknowledgements
The authors would like to thank Debra Anderson (University of Technology Sydney, Sydney, Australia) for her critical reading of the manuscript, and Debbie Marriott (University of Technology Sydney) for her invaluable and patient assistance in preparing this manuscript for publication.
Footnotes
This study is registered at PROSPERO with identifier number CRD42019140813.
This article has an editorial commentary: https://doi.org/10.1183/13993003.00775-2021
This article has supplementary material available from erj.ersjournals.com
Author contributions: All authors contributed to the design and conduct of this study.
Conflict of interest: S. Kochovska has nothing to disclose.
Conflict of interest: D.H. Ferreira has nothing to disclose.
Conflict of interest: M.V. Garcia has nothing to disclose.
Conflict of interest: J.L. Phillips has nothing to disclose.
Conflict of interest: D.C. Currow is an unpaid advisory board member for Helsinn Pharmaceuticals, is a paid consultant and receives payment for intellectual property with Mayne Pharma, and is a consultant with Specialised Therapeutics Australia Pty Ltd.
Support statement: This research was funded by discretionary funds held by the academic team at IMPACCT, University of Technology Sydney (Sydney, Australia). Funding information for this article has been deposited with the Crossref Funder Registry.
- Received September 24, 2020.
- Accepted February 15, 2021.
- Copyright ©The authors 2021. For reproduction rights and permissions contact permissions{at}ersnet.org