Sage Journals: Discover world-class research

Abstract

Oxygen is one of the commonest health-care interventions worldwide. This might suggest that health-care professionals (HCPs) would be knowledgeable and familiar with its uses and limitations. Yet it is apparent, through clinical audit, that oxygen is probably misunderstood by many HCPs. The aim of this critical interpretive synthesis (CIS) is to explore qualitative and quantitative literature in relation to HCPs beliefs and perceptions of oxygen therapy. A systematic search in Medline, Cinahl, Embase, British Nursing Index and PsychInfo using search terms, such as, oxygen therapy, chronic respiratory disease, HCPs and perceptions yielded 1514 studies of which 12 contained data relevant to the review question. Two reviewers independently screened the articles for eligibility against inclusion and exclusion criteria, and data were selected and synthesized with an integrative and interpretive approach using CIS. This allowed diverse empirical evidence to be synthesized to develop existing and new interpretations of data.Three synthetic constructs were interpreted from the available literature, namely, oxygen for symptom relief, levels of knowledge and understanding and oxygen as a therapy for HCPs. The literature alludes to deep-seated beliefs that exist. In order to enhance practice, these beliefs and cultures need to be challenged. Further research is needed to explore HCPs’ perceptions of oxygen therapy in order to inform the seemingly resistant adoption of evidence based practice in relation to oxygen.

Keywords

Oxygen health-care professionals perception critical interpretive synthesis respiratory care

Introduction

Oxygen is one of the most prolific health-care therapies used in the developed world. Its abundance might suggest that health-care professionals (HCPs) would be knowledgeable and familiar with its uses and limitations. Yet it is apparent, through poor prescribing and administration practices, that oxygen is probably misunderstood by many HCPs. This has been demonstrated in acute care, where despite reported prevalence of 24% use in all inpatients,¹ and in emergency settings where 34% usage² has been observed, suboptimal quality of care has been recorded.^1

–4

Similarly, a report produced by the Royal College of Physicians⁵ in response to spiralling costs of domiciliary oxygen therapy uncovered poor prescribing practices and follow-up of patients, which led to major changes in prescription and provision of home oxygen services in England and Wales.⁶ Yet despite these changes, poor practice and variations in practice persist.^7,8 The reasons for failure to alter practice in accordance with emerging evidence remain an enigma.

The British Thoracic Society (BTS) suggest that a high proportion of medical oxygen is possibly administered because most clinicians believe that oxygen can alleviate breathlessness,⁶ but there is no evidence supporting this claim. Indeed the evidence base for oxygen therapy per se is lacking, with most criteria and guidelines for oxygen therapy is supported by evidence that has evolved based on individual cases and consensus opinion rather than sound experimental research. This has often led to the adoption of a priori knowledge with the need for robust controlled clinical trials overlooked.

It is further suggested, possibly as a consequence of a lack of empirical evidence, that a major problem contributing to poor prescribing practices is that HCPs often receive conflicting information and advice about oxygen therapy during their training and clinical careers. There appears to be confusion about the entire area of oxygen prescription and use,⁶ but again there is no supporting evidence. The problem appears to become self-perpetuating as anecdotal evidence suggests that the same erroneous beliefs are exposed to individuals receiving oxygen therapy, their carers and the general public. Subsequently, this may lead to unrealistic expectations and poor adherence to prescribed oxygen therapy.

Despite growing acknowledgement of this issue, it is not clear from the literature where the roots of these fallacies lie, and indeed, whether this is a result of tradition, lack of knowledge regarding the indications and administration of oxygen therapy, or misunderstanding of basic physiological principles of oxygen per se. In order to address this clinical paradox, there is a need to identify reasons for enduring poor practices. The aim of this critical interpretive synthesis (CIS) therefore is to explore the literature in relation to HCPs’ beliefs and perceptions of oxygen therapy in order to provide a platform for further investigation.

Method

The dearth of literature addressing the review question directly, together with the diversity and complexity of the literature available, directed the review towards an integrative and interpretive approach.

CIS, a variant of meta-ethnography was utilized,⁹ allowing explicit integration of qualitative and quantitative evidence through an interpretative process. CIS draws upon conventional systematic review methodology, whilst allowing discretion in study selection to include articles that may contribute to findings whilst not necessarily answering the review question directly. The use of the ‘authorial voice’, in relation to the first author’s experience and knowledge of oxygen in the clinical context, in both selection of literature and interpretation is a further defining feature of CIS.⁹

Search strategy

A systematic approach to searching, locating and retrieving relevant literature was adopted.¹⁰ The initial search identified articles relating to both patients’ and HCPs’ perceptions of oxygen therapy. Findings from HCPs’ perceptions are reported here. Findings from patients’ perceptions have been reported separately.

Medline (1950–2014), Cinahl (1981–2014), Embase (1980–2014), British Nursing Index (1985–2014) and PsychInfo (1806–2014) were searched via Evidence Search (www.evidence.nhs.uk) using keywords oxygen therapy, chronic respiratory disease, healthcare professionals and perceptions. Term variants and synonyms were combined and searched using the Boolean ‘OR’ (e.g. oxygen therapy OR O₂ therapy). The Boolean ‘AND’ combined different facets. Truncation (*) was utilized to capture plurals and spelling variants (e.g. Oxygen therap*). To improve focus advanced search operators for phrase searching (“”) were used. Studies were restricted to English language only with no date or publication type restrictions applied.

A search of grey literature and the wider Internet was conducted to minimize publication bias. Hand-searching key journals together with key respiratory conference proceedings and ‘citation snowballing’¹¹ supplemented database searches, ensuring the inclusion of literature not yet indexed in databases. The search was undertaken on 21 December 2011 and rerun on 12 March 2014 to capture further relevant studies published since the initial search.

Inclusion criteria and study selection

The initial search included articles relating to both HCPs’ and patients’ perceptions of oxygen therapy. This resulted in 1503 articles identified, and titles and abstracts were considered for these articles. From these 1503 articles, 179 articles were selected for further consideration. Two reviewers independently screened the articles for eligibility against inclusion and exclusion criteria (Table 1) and relevance to the research question. The decision deferred the quality appraisal phase in the event of disagreement. Fifty-nine full-text articles were finally selected for quality assessment. The flow chart (Figure 1), adapted from PRISMA (Preferred Reporting Items for Systematic review and Meta-Analysis),¹² provides a summary of the search outcome.

Figure 1.

Flowchart of study selection.

Table 1.

Inclusion and exclusion criteria.

Inclusion criteria	Exclusion criteria
Any recorded perceptions of administering or prescribing oxygen therapy by HCPs (any profession, any setting, acute or chronic and any country) or respiratory patients (any disease catagory – acute or chronic; adults >18 years, in any setting).	Carers, lay health-care workers and non-respiratory patients and studies concerned with the efficacy of oxygen rather than perceptions.
Intervention – oxygen therapy either prescribed or delivered as part of medical management including acute oxygen therapy, domiciliary oxygen therapy and oxygen for palliation of symptoms.	Any article detailing oxygen as a complementary therapy.
Empirical studies with a clear, methodological stance, although actual method is unimportant.	Any article without an explicit stated methodology.
English language.	Articles unavailable in English language.

HCP: health-care professional.

Quality assessment

Exclusion of studies on the grounds of poor quality is a leading contention in relation to quality appraisal in integrative reviews and in particular CIS.^9,13,14 Qualitative research is regarded as an interpretative act¹⁵ and appraising the quality of articles requires more than a simple scoring system. Dixon-Woods et al. (2006) argue that studies that can potentially add to a review’s findings should therefore be quality appraised on their own merit and within the context of the review question, and that content and relevance of findings is an additional key consideration. The pro forma therefore included the key aspect of implications and usefulness to the review question. As the principal focus of most studies was not directly related to the review question, this was an important aspect.

A hybrid quality appraisal/data extraction sheet was developed based on the criteria for disparate data as suggested by Hawker et al.¹⁶ This allowed identification of various methodological features without excluding studies of poorer quality.¹⁷ Using Hawker et al.’s protocol for assessment a score of 1–4 is assigned to each of 10 criterion resulting in an accumulative score that indicated the overall assessed methodological rigour of each empirical study (ranging from 10 (very poor) to 40 (good)). Second checking by an independent reviewer (DL) allowed differences to be resolved by discussion and consensus. The process of quality appraisal, including relevance, reduced the number of studies from 59 to 51.

At this stage, articles were separated into two categories, namely, patients (42 studies) and HCPs (12 studies), with a total of 51 articles (three articles being eligible for both). Findings from HCPs’ perceptions are reported here. Findings from patients’ perceptions are available in a companion article.¹⁸

Data extraction strategy

Data extraction was facilitated using a pro forma which detailed characteristics and quality appraisal of included studies, including relevance. Data extraction was undertaken by the lead researcher and independently checked by a second reviewer (DL).

Synthesis of the extracted data

The current review adopted criteria for data synthesis set out by Flemming.¹⁴ This involved the article being read and understood in relation to both itself and the research question. Relevant data were extracted and translated, through identification of concepts, themes and metaphors and developed into synthetic constructs. Exploring relationships between constructs allowed explanatory accounts to be suggested in the form of synthesizing arguments. Integration of evidence in this way, from across studies, allowed new interpretations of data, which is demonstrably grounded in existing evidence.¹⁷

Findings

None of the studies addressed the research question directly. Studies were therefore selected on the basis that some aspect of, or reference to, the study’s findings included HCPs’ perceptions of oxygen therapy.

The final selection of 12 articles consisted of varying methodological approaches, quality, countries of origin and professional groups. The number of participants in each study ranged from 12 to 1051. The methodological quality of selected studies varied, but no articles were excluded on the basis of quality as discussed previously. The final selection of articles related to HCPs’ perception is summarized in Table 2. In addition, representation from the literature is detailed in Table 3. This allows assessment of the grounding of constructs in the original literature.

Table 2.

Grid of main points from literature reviewed – study characteristics and HCPs’ perceptions.

	Author	Year	Methodology	Context	Participants	Focus of study	QA score	Relevance and Supporting Evidence
1	Abernethy¹⁹	2005	Survey	Australia and New Zealand	214 Physicians	Pilot to define clinical efficacy of palliative oxygen	21	Outlines most frequently reason for prescribing oxygen as retractable dyspnoea. Majority believed oxygen was beneficial.
2	Atiş et al.²⁰	2001	Q survey	Turkey	379 Physicians	Characteristics and compliance of LTOT patients	26	Relates education (from a physician) to increased compliance; 33.2% patients told to use oxygen only when breathless.
3	Austin et al.³	2010	RCT	Australia	HCPs, number not stated/405 patients	High-flow versus titrated oxygen	30	Finding post-trial deviation from protocol. Alludes to attitudes and culture but no empirical data to support.
4	Barr et al.²¹	2005	Survey	United States	1051 Primary care physicians	Patient and physician perception of COPD	32	Symptoms as criteria for prescribing.
5	Considine et al.²²	2005	Quasi- experimental controlled	Australia	88 Emergency department nurses	Effect of education on knowledge	32	Education increased knowledge; increased knowledge was a predictor of independent decision-making.
6	Considine and Botti²³	2006	Quasi-experimental	Australia	20 Registered nurses	Effect of education on clinical decisions	33	Effect education on clinical decision-making in emergency oxygen Some changes in management post-intervention/other aspects of assessment no change.
7	Glaab et al.²⁴	2006	Mail survey	Germany	845 Physicians	Compliance to guidelines	25	Limited discussion of knowledge of effect on survival towards respiratory specialists.
8	Neri et al.²⁵	2006	Survey	Italy	1504 Patients and physicians. Patients LTOT >6 months	Behaviour and knowledge regarding oxygen therapy	31	Mostly relates to compliance. Two questions were relevant: perception of improvement; why prescribed, 38.5% believed it was for dyspnoea. Exact rationale for prescription unclear.
9	Pepin et al.²⁶	1996	Q Survey	France	310 Physicians	Daily use of oxygen in COPD LTOT patients	35	Evidence of poor compliance to prescription criteria and inconsistent education – related to patient compliance.
10	Reinke et al.²⁷	2008	Grounded theory	United States	84 Physicians /55 patients oxygen dependant	Comparison of perceptions re-transition of illness	31	Very limited, regarding transition of illness mainly. Oxygen seen as a restriction, Oxygen as a sign of deterioration.
11	Roberts et al.²⁸	1993	Mixed methods	Canada	10 Late-stage cancer patients/12 nurses	Dyspnoea last year of life – patients and nurses perspective	33	Limited to two short but relevant points. Patients’ frustration. From nurses the psychological benefit is acknowledged for patients, carers and HCPs. 66% used oxygen as a measure for alleviating dyspnoea.
12	Stringer et al.²⁹	2004	Telephone survey	Canada	21 Physicians	Prescribing practices of palliative oxygen	28	Variable p`ractice attributed to lack of evidence and guidelines in palliative care. Ambivalence/uncertainty.

RCT: randomized controlled trial; HCP: health-care professional; LTOT: long-term oxygen therapy; COPD: chronic obstructive pulmonary disease.

Table 3.

Grid of themes – HCPs.

Major theme	Representation from the literature
Oxygen for symptom relief	Abernethy’s¹⁹ email survey of 214 physicians showed that palliative care physicians were more likely to prescribe palliative oxygen (29% vs. 9%, p = 0.009) and that the most frequently (65%) cited reason for prescribing was intractable dyspnoea.
	Roberts et al.’s²⁸ mixed-methods study of 10 palliative care nurses showed 66.7% reported using oxygen as an intervention to manage their breathlessness.
	Barr et al.’s²¹ large survey of 1051 physicians found 51% of GPs and 10% of respiratory specialists thought that presentation of symptoms was additional criteria for prescribing oxygen therapy.
Knowledge and understanding	Atiş et al.’s²⁰ survey reported that only 58.3% of patients were educated by a physician about oxygen therapy at the beginning of treatment.
Knowledge and understanding	Considine et al.’s²² demonstrated that oxygen administration was improved as a result of education and that increased knowledge was identified as a predictor of independent decision-making.
Oxygen as a therapy for HCPs	The initiation of oxygen therapy was just one more therapy that might help (p. 206).²⁷
	I often feel that nurses do it for themselves … because we’re doing something … It’s a major thing to do (p. 317).²⁸
	Austin et al.’s³ RCT demonstrated 37% non-compliance to study protocols by paramedics, feedback suggested concern regarding insufficient delivery of oxygen in distressed patients as the reason for non-adherence to protocol and referred to the common conception that ‘more is better’.

HCP: healthcare professional; GPs: general practitioner; RCT: randomized controlled trial.

Literature pertaining to perceptions from the HCPs’ perspective is very limited. The final selection of studies identified 13 articles of which 4 were excluded at quality appraisal. The reasons for exclusion included the article’s focus was chronic obstructive pulmonary disease (COPD) generally with no specific mention of oxygen, two articles did not relate to perceptions and one was a duplicate publication of the same study. Further details and references of excluded studies are detailed in Table 4. In addition, three articles ^25,27,28 selected in the patients’ perception review were also included in the review of HCPs, bringing the total of included studies to 12. Heterogeneity of design and methods was apparent in the studies selected (Table 2).

Table 4.

Reasons and references for exclusion of studies.

Reasons for exclusion of studies	Number of studies
Initially excluded on the grounds that no mention of perception of oxygen therapy identified.	59
Exclusions following quality appraisal (including relevance):
Four patient studies further excluded: 1 involved lay carers; 2 focused on HCP perceptions of COPD – no oxygen; 1 practice focused – no perceptions.	55
Four studies of HCPs excluded: 1 did not specifically mention oxygen (related to COPD generically); 1 duplicate publication of the same study;²³ 2 did not relate to perception specifically	51
References list of excluded HCP studies
Brown HB and Manring EA. A comparison of oxygen administration practices of EMTs and emergency physicians. Am J Emerg Med 1997; 15: 648–651.
Considine J, et al. The effect of education on hypothetical and actual oxygen administration decisions. Nurse Educ Today 2007; 27(6): 651–660.
Kesten S and Chapman KR. Physician perceptions and management of COPD. Chest 1993; 104: 254–258.
Yawn BP and Wollan PC. Knowledge and attitudes of family physicians coming to COPD continuing medical education. Int J COPD 2008; 3(2): 311–317.

HCP: health-care professional; COPD: chronic obstructive pulmonary disease.

Findings were very limited for HCPs due to the paucity of evidence. Three synthetic arguments were constructed from the available literature, namely, oxygen for symptom relief, levels of knowledge and understanding and oxygen as a therapy for HCPs. These constructs, together with the number of articles that contributed to each, are outlined in Tables 3 and 5.

Table 5.

Grid of themes – HCPs.

Author	Year	Oxygen for symptom relief	Knowledge and understanding	Oxygen as a therapy for HCPs
Abernethy¹⁹	2005	X
Atiş et al.²⁰	2001	X	X
Austin et al.³	2010			X
Barr et al.²¹	2005	X	X
Considine et al.²²	2005		X
Considine and Botti²³	2006		X	X
Glaab et al.²⁴	2006		X
Neri et al.²⁵	2006	X
Pepin et al.²⁶	1996		X
Reinke et al.²⁷	2008			X
Roberts et al.²⁸	1993	X		X
Stringer et al.²⁹	2004	X

HCP: health-care professional; COPD: chronic obstructive pulmonary disease.

Oxygen for symptom relief

The notion of HCPs perceiving oxygen to relieve breathlessness was common in the literature and appeared in half of the considered articles. Abernethy¹⁹ published the results of an email survey of 214 physicians (93 palliative care physicians and 121 respiratory physicians). Primarily intended to define the necessary duration of a clinical efficacy study regarding palliative oxygen therapy, the survey captured habits of prescribing oxygen therapy for palliative care including the frequency and indications. The findings showed that palliative care physicians were more likely than respiratory physicians to prescribe palliative oxygen (29% vs. 9%, p = 0.009) and that the most frequently (65%) cited reason for prescribing was ‘intractable dyspnoea’. Despite a low response rate (33%) and crude methodology, the survey demonstrates that the majority of responding clinicians believe that oxygen relieves dyspnoea.

Stringer et al.²⁹ similarly surveyed physicians’ prescribing practices of palliative oxygen therapy. This telephone survey reported variability in prescribing practices attributed to both a lack of evidence and clear guidelines. Based on hypothetical scenarios, physicians were assessed in response to specified cues. Cluster analysis revealed three patterns of response, namely, those who prescribed in the presence of hypoxia regardless of symptoms, those who prescribed only when both hypoxia and symptoms were present and those prescribing for breathlessness alone. Adding a ‘dummy factor’ on the effect of spousal concern revealed an increasing tendency to prescribe, probably related to expectation from carers and perhaps patients. Ultimately the study, although a small sample of 17, demonstrated that uncertainty and inconsistency exists when physicians are faced with the decision whether to prescribe oxygen to palliative care patients. The article’s authors suggest that this may be a reflection of inadequate understanding of pathophysiology and treatment and that the decision to use oxygen is often based upon individual past experiences and biases.

Roberts et al.²⁸ studied the experience of dyspnoea in the last year of life. The mixed-methods study of 10 patients with late-stage lung cancer and the nurses providing their care provided data to investigate the phenomenon of dyspnoea, as experienced during the last weeks of life. A triangulated approach included patient self-report survey, chart audits (to record incidence and management of dyspnoea) and patient and HCP interviews. Of the 10 patients interviewed, 7 recalled no suggestions made by the nurse regarding how to manage their breathlessness, although 66.7% of nurses reported using oxygen as an intervention. The study revealed an inconsistent understanding; with reference to oxygen therapy, some nurses reported that it was helpful and that it should be ordered as soon as dyspnoea was apparent, whilst others claimed that although it wasn’t clinically therapeutic they believed it had a symbolic benefit (p. 317). Several nurses were of the opinion that oxygen provided psychological comfort to both patients and their families. Whether this notion can be extended to HCPs themselves will be explored later.

Barr et al.²¹ recorded patient and physician perceptions of COPD in a large survey (1050 physicians and 1023 patients) in the United States. Although the main focus of the study was COPD, in relation to oxygen therapy, physicians reported confusion regarding treatment options; 51% of general practitioners and 10% of respiratory specialists thought that presentation of symptoms was additional criteria for prescribing oxygen therapy.

Adding to the synthetic construct of oxygen for symptom relief, a Turkish questionnaire survey aimed at exploring issues of compliance of long-term oxygen therapy (LTOT) by patients, provides some insight into the messages that HCPs provide.²⁰ Thirty-three percent were told only to use oxygen when they were short of breath and only 28.2% reported that they used oxygen for more than 15 hours per day (the recommended evidence-based duration necessary to obtain long-term clinical benefit). Although limited by the response rate (34.5%) and its geographical specificity (in Turkey patients purchase their own oxygen), this study adds to the elusive, but anecdotally common, belief that oxygen therapy relieves breathlessness.

The limited insight found in the literature relating to how HCPs prescribe and use oxygen provides an inconsistent approach. It does however appear prevalent that it is frequently used for, and HCPs appear to believe it relieves dyspnoea. It has been demonstrated that pressure from carers influences prescription²⁹; a phenomenon that may be linked to expectations of both patients and carers.

Levels of knowledge and understanding of HCPs

Half of the articles considered referred to HCPs’ knowledge and understanding of oxygen therapy. Glaab et al.²⁴ undertook a national mail survey of 845 physicians (486 generalists and 359 specialists) in Germany to investigate compliance to guidelines when prescribing. Although primarily concerned with generic COPD guidelines, there is some limited reference to oxygen therapy. LTOT was generally regarded as an effective measure for improving the quality of life and symptoms rather than prolonging life expectancy. The knowledge of effect on survival rate was higher in respiratory specialists than generalists (p < 0.0001).

Atiş et al.’s²⁰ survey reported that 58.3% of patients were educated by a physician about oxygen therapy at the beginning of treatment, whether the rest received any form of education is not clear. Logistical regression identified the likelihood of achieving compliance increased 4.5-fold (confidence interval 2.27–9.13, p < 0.001) when education was provided. This relationship between patient compliance and the level of education supports the findings of an earlier study by Pepin et al.²⁶ This French survey questionnaire of 219 physicians and 564 patients assessed patient compliance and prescribing practices for LTOT. Although 87% of patients were prescribed over 15 hours per day of oxygen therapy (as reported by the physicians), only 45% reached this target. Again logistical regression showed that patients receiving a follow-up education session increased the likelihood of receiving effective treatment 4.5-fold (CI 2.3–9.1). The study concluded that supplementary education regarding LTOT, given by a nurse or physiotherapist, was an important factor for increasing patient’s compliance to therapy. Clearly then education of the patient is central to understanding the therapy and consequently compliance, but fundamental to that must be the knowledge and understanding of the HCPs themselves.

Considine et al.²² demonstrated that the use of supplemental oxygen was improved as a result of educating the HCPs in the acute setting and that increased knowledge was identified as a predictor of independent decision-making. The quasi-experimental design sets out to test the assumption that an increase in nurse’s knowledge, though a self-learning package, improved clinical decisions. Pretest scores were comparable (p = 0.091), whilst posttest scores between the experimental and control group showed a statistically significant improvement (19.31 ± 3.56 vs. 13.05 ± 3.76; p = <0.001). Eighty-seven percent of nurses in the study reported making clinical decisions about oxygen therapy on a daily basis (91% of those decisions were autonomous). The calibre of these decisions clearly has the potential to influence patient outcomes and therefore it is important to consider factors that influence the acquisition of knowledge.

A further study by the same Australian group²³ explored specifically the effect of education on clinical decisions regarding emergency oxygen therapy. Utilizing a similar design, this smaller study of 20 emergency department nurses tested nurses’ decision-making skills, rather than knowledge, following completion of the educational intervention. The results from this study were variable with some changes in hypothetical management of patients seen, in particular device selection, but in other aspects (e.g. the parameters used for assessment) no change was demonstrated.

It is apparent from the literature that education of HCPs is important and that education given to patients has the potential to improve compliance and, therefore, hypothetically, clinical effectiveness.

Oxygen as a therapy for HCPs

The synthetic argument of oxygen as a therapy for HCPs is probably the most ambiguous construct derived from the literature. Nonetheless, it is evident in four of the reviewed studies that such a notion exists and seems a familiar explanation for the common prescription of oxygen therapy for non-hypoxic patients. Relating to patients’ fears and restrictions with palliative oxygen, a study by Reinke et al.,²⁷ though very limited in its reference to HCPs, refers to physicians’ recognition of oxygen therapy as a milestone in a patient’s condition; one physician stating, ‘The initiation of oxygen therapy was just one more therapy that might help’ (p. 606). This quote seems to epitomize the often felt frustration from HCPs regarding the management of the intractable symptom of dyspnoea.

Roberts et al.’s²⁸ mixed-methods study of lung cancer patients’ experience of the last year of life draws on this idea further. With reference to the notion of oxygen providing palliative comfort to keep patients at home and ‘happy’, one nurse wondered whether the psychological benefit of oxygen was as relevant to the patients and families as it was to the nurses. ‘I often feel that nurses do it for themselves … because we’re doing something …’.²⁸

This rare recording of an anecdotally common perspective gives a potential insight into HCPs’ rationale for prescribing or recommending oxygen therapy. This insight seems to be potentially the most elusive with regard to the variability and inconsistencies of oxygen prescription and would support Considine’s and Botti’s²³ notion that factors other than knowledge and education may influence clinical decision-making.

A further tentative reference to the prospect of oxygen being given to relieve HCPs’ need to help patients manage dyspnoea, and the possibility of an entrenched culture, is alluded to by Austin et al.³ This well-designed and conducted randomized controlled trial set in Tasmania tested outcomes in relation to high-flow versus titrated oxygen in the pre-hospital care of COPD patients. The main findings from this landmark study demonstrated that titrated oxygen significantly reduces hypercapnia, respiratory acidosis and mortality. The researchers discovered a lower than expected adherence to study protocols. Of the 214 patient records, 37% showed that received treatment did not comply to study protocols (56% in the titrated arm and 21% in the high-flow arm). In the titrated arm, all violations involved administration of high-flow oxygen at some point in the pre-hospital care. The authors muse that this is probably a result of entrenched culture and training in emergency medicine. Chart reviews found no evidence of equipment malfunction, requests from patients, or lack of protocol understanding, but the authors report that feedback from some paramedics indicated concern regarding insufficient delivery of oxygen in distressed patients and referred to the common conception that more is better. This study is potentially important in that it does expose, intangibly, the notion of culture and the need to explore and substantiate the reasons why this exists is manifest. Although fundamentally relating to the efficacy and detrimental effects of oxygen in the acute setting, the existence of persistent beliefs and entrenched practices is apparent.

The literature pertaining to perceptions from HCPs is very limited but from what evidence is available there appears to be a persistent belief that oxygen is useful for management of dyspnoea. There is also some degree of evidence that using oxygen for patients in such a way helps HCPs and has the potential to offset guilt and frustration at not being able to help patients.

Discussion

Inconsistency of reported beliefs, understanding and variations in clinical practice, even with respiratory specialists, dominate this limited sample of empirical evidence. The reasons for inconsistency in the indications and use of oxygen have been cited as a result of a lack of clear guidelines, endorsed through the obscure nature of lucid information and directives.

The relationship between knowledge and clinical practice is clear. Transfer of this knowledge to patients is part of the therapeutic relationship, but it can be contested as beneficial if that knowledge is not grounded in evidence but in misplaced beliefs and misconceptions. It is suggestive that the faith in HCPs¹⁸ that patients have, on occasion, may be poorly placed.

It has been suggested that factors other than knowledge may influence clinical decisions.²³ What these factors are remains obscure and further research is warranted. Attempts have been made previously to uncover perceptions of oxygen therapy and the implications that this may have on clinical decision-making, but contributing factors identified tend be those that affect efficacy of oxygen³⁰ and patient compliance^20,25,26 rather than culture.

Respiratory specialists appear to have a greater knowledge and understanding regarding oxygen, as may be expected, it can be argued however that with the prevalence of prescriptions and use from non-specialist HCPs that the prescriber should always have the relevant knowledge and understanding to enable safe clinical decision-making.

The overuse and misunderstanding of oxygen therapy by HCPs has been alluded to in other literature,^31

–35 nevertheless there is very little evidence to verify this and a dearth of empirical evidence to substantiate why these misunderstandings appear to be so prevalent among HCPs. It is often cited that a culture exists whereby oxygen is given automatically to patients who present critically ill or dyspnoeic.³⁶ It appears that this culture may be deep-seated and may in fact be so ingrained that it actually provides relief for HCPs themselves as they feel they are doing ‘something’.

This review overall offers some, albeit limited, insight into the inconsistencies of oxygen prescription and administration, the importance of education and its potential relationship to clinical efficacy and a glimpse at a culture and an innate response that may be obstructive to HCPs practicing evidence-based care.

Summary and recommendations

The literature is not very forthcoming in enlightening beliefs and perceptions of HCPs regarding oxygen therapy. There appear to be clues, but these lack tangibility and verification. It is proposed that this missing perspective may serve to illuminate the problems associated with poor adherence to guidelines and recommended practice concerning oxygen therapy and clearly further research is this area is warranted.

It is interesting to note that overall the methodological qualities of studies pertaining to HCPs are poor, with a seemingly over-reliance on survey as means of data collection. Response rates, especially from non-specialist physicians are recorded as low, despite strategies to increase returns. The rudimentary survey approach to gathering information can overlook some of the more deep-seated roots of beliefs, culture and practice that may inform the seemingly resistant adoption of evidence-based practice in relation to oxygen.

HCPs can be considered to have immense power over patients’ lives in what they prescribe and advocate. With the possibility that perception is shaped by several influences including media, knowledge, cultural, historical, professional and social, these factors appear vague in the literature but should not be disregarded as potential sources of antecedents. The findings presented here highlight possible misconceptions and potential influences regarding oxygen therapy. Further exploration of the perceptions regarding oxygen therapy from both HCPs’ and patients’ perspectives are needed in order to explore these potential influencing factors in order that recommendations to address these can be made.

From the literature it can be deciphered that knowledge affects clinical practice with regard to oxygen therapy, yet practice is not always influenced by education. It appears that it is difficult, even in a controlled experimental situation, to rise above deep-seated beliefs, especially when a patient is struggling to breathe. Yet in order to improve practice, with regard to the safety and efficacy of oxygen therapy, these beliefs and cultures need to be challenged.

Strengths and limitations of the review

By casting a ‘wide net’, the high sensitivity of the initial search could be considered as the strength of the review. The incorporation of representation from the literature is a further strength that demonstrates grounding in the original data and is a characteristic of CIS.⁹ The use of the author’s experience in interpretation is a key to the explanation of data and construction of the synthesizing arguments. Whilst drawing upon personal experience and preconceived ideas that can be considered a threat to impartiality, it can alternatively be argued that the insight brought to the review through expertise was true to CIS and the fulcrum to developing the synthetic constructs.

The quality of the reporting of included articles was an inherent weakness with many studies not being well described and methods of analysis not always explained. This issue of poor quality in write-up, as opposed to poor quality of methods, has been identified by other authors.¹⁶ Ultimately, most literature focuses on diverse objectives, such as HCPs understanding of disease and treatments, illness transition and efficacy of interventions rather than actually what HCPs think or believe about oxygen, and this limits the review’s ability to address the research question directly.

The material selected displayed heterogeneity, being based on differing philosophical and scientific assumptions. In particular, most studies utilized a quantitative design with retrospective data and an over-reliance on survey. In the absence of directly relevant studies, the use of CIS enabled the construction of synthetic arguments, which informs and gives some foundation to this elusive phenomenon. The use of reflexivity, as advocated with this interpretative review methodology,¹⁷ facilitated consideration of the review methods and the process of synthesizing both original data and the researchers’ interpretations of these data.

Footnotes

Acknowledgement

I would like to thank Dr Dave Lynes, Edge Hill University, for his guidance with study selection and data extraction.

Funding

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

Conflict of interest

The authors declared no conflict of interest.

References

Eastwood

Peck

Young

. Oxygen administration and monitoring for ward adult patients in a teaching hospital. Int Med J 2011; 41 (11): 784–788.

Hale

Gavin

O’Driscoll

. Audit of oxygen use in emergency ambulances and in hospital emergency department. Emerg Med J 2008; 25: 773–776.

Austin

Wills

Blizzard

. Effect of high flow oxygen on mortality in chronic obstructive pulmonary disease patients in prehospital setting: randomised controlled trial. Br Med J 2010; 341: c5462.

Wijesinghe

Perrin

Healy

. Pre-hospital oxygen therapy in acute exacerbations of chronic obstructive pulmonary disease. Int Med J 2011; 41: 618–622.

Royal College of Physicians (RCP). Domiciliary oxygen therapy services: clinical guidelines and advice for prescribers. London: RCP, 1999.

BTS Emergency Oxygen Guideline Group. Guideline for emergency oxygen use in adult patients. Thorax 2008; 63(Suppl VI): vi1–vi73.

Davidson

Williams

Baxter

. P102: a survey of oxygen provision across London. Thorax 2011; 66 (4): A108.

Lee

McDonnell

Davidson

. Has the new contract delivered better ambulatory oxygen devices for patients? A London perspective. Thorax 2013; 68 (Suppl 3): A205.

Dixon-Woods

Bonas

Booth

. How can systematic reviews incorporate qualitative research? A critical perspective. Qual Res 2006; 6 (1): 27–44.

10.

Centre for Reviews and Dissemination (CRD). Chapter six: incorporating qualitative evidence in or alongside effectiveness reviews. In: Centre for Reviews and Dissemination (eds) Systematic reviews: guidance for undertaking reviews in healthcare. York: CRD, 2009, pp. 219–281.

11.

Booth

. Cochrane or cock-eyed? How should we conduct systematic reviews of qualitative research? Sheffield: Education-line, http://www.leeds.ac.uk/educol/documents/00001724.htm (2001, accessed 1 May 2014).

12.

Moher

Liberati

Tetzlaff

. The PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009; 6(6): e1000097.

13.

Dixon-Woods

Agarwal

Young

. Integrative approaches to qualitative and quantitative evidence. London: NHS Health Development Agency, 2004.

14.

Flemming

. Synthesis of quantitative and qualitative research: an example using critical interpretative synthesis. J Adv Nurs 2010; 66: 201–217.

15.

Kuper

Lingard

Levison

. Critically appraising qualitative research. Br Med J 2008; 337: 687–689.

16.

Hawker

Payne

Kerr

. Appraising the evidence: reviewing disparate data systematically. Qual Health Res 2002; 12 (9): 1284–1299.

17.

Dixon-Woods

Cavers

Agarwal

. Conducting a critical interpretative synthesis of the literature on access to healthcare by vulnerable groups. BMC Med Res Methodol 2006; 6 (35): 1–13.

18.

Kelly

Maden

. How do respiratory patients perceive oxygen therapy? A critical interpretative synthesis of the literature. Chron Respir Dis 2014; 11: 209–228.

19.

Abernethy

. Prescribing palliative oxygen: a clinician survey of expected benefit and patterns of use. Palliat Med 2005; 19: 168–170.

20.

Atiş

Tutluoglu

Bugdayci

. Characteristics and compliance of patients receiving long-term oxygen therapy (LTOT) in Turkey. Monaldi Arch Chest Dis 2001; 56 (2): 105–109.

21.

Barr

Celli

Martinez

. Physician and patient perceptions in COPD: the COPD resource network needs assessment survey. Am J Med 2005; 118(12): 1415.e9–1415.e17.

22.

Considine

Botti

Thomas

. Effect of a self-directed learning package on emergency nurses’ knowledge of assessment of oxygenation and use of supplemental oxygen. Nurs Health Sci 2005; 7: 199–208.

23.

Considine

Botti

. The effects of specific educational preparation on emergency nurses’ clinical decisions regarding supplemental oxygen administration. Nurs Health Sci 2006; 8: 73–80.

24.

Glaab

Banik

Rutschmann

. National survey of guideline-compliant COPD management among pneumologists and primary care physicians. COPD: J Chron Obstruct Pulmon Dis 2006; 3 (3): 141–148.

25.

Neri

Melani

Miorelli

. Long-term oxygen therapy in chronic respiratory failure: a Multicenter Italian Study on Oxygen Therapy Adherence (MISOTA). Respir Med 2006; 100 (5): 795–806.

26.

Pepin

Barjhoux

Deschaux

. Long-term oxygen therapy at home: compliance with medical prescription and effective use of therapy. Chest 1996; 109 (5): 1144–1150.

27.

Reinke

Engelberg

Shannon

. Transitions regarding palliative and end-of-life care in severe chronic obstructive pulmonary disease or advanced cancer: themes identified by patients, families, and clinicians. J Palliat Med 2008; 11 (4): 601–609.

28.

Roberts

Thorne

Pearson

. The experience of dyspnoea in late stage cancer: patients’ and nurses’ perspectives. Canc Nurs 1993; 16 (4): 310–320.

29.

Stringer

McParland

Hernandez

. Physician practices for prescribing supplemental oxygen in the palliative care setting. J Palliat Care 2004; 20 (4): 303–307.

30.

Eastwood

O’Connell

Gardner

. Patients’ and nurses’ perspectives on oxygen therapy: a qualitative study. J Adv Nurs 2009; 65 (3): 634–641.

31.

Downs

. Has oxygen administration delayed appropriate respiratory care? Fallacies regarding oxygen therapy. Respir Care 2003; 48: 611–620.

32.

Troosters

. Oxygen: the good, the bad, and the necessary. Thorax 2004; 59: 1005–1006.

33.

Calverley

PMA

. Supplementary oxygen therapy in COPD: is it really useful? Thorax 2000; 55: 537–538.

34.

New

. Oxygen: kill or cure? Prehospital hyperoxia in the COPD patient. Emerg Med J 2006; 23: 144–146.

35.

Tin

. Oxygen therapy: 50 years of uncertainty. Paediatrics 2002; 110 (3): 615–616.

36.

Beasley

Patel

Perrin

. High-concentration oxygen therapy in COPD. The Lancet 2011; 378: 969–970.

How do health-care professionals perceive oxygen therapy? A critical interpretive synthesis of the literature

Abstract

Keywords

Introduction

Method

Search strategy

Inclusion criteria and study selection

Quality assessment

Data extraction strategy

Synthesis of the extracted data

Findings

Oxygen for symptom relief

Levels of knowledge and understanding of HCPs

Oxygen as a therapy for HCPs

Discussion

Summary and recommendations

Strengths and limitations of the review

Footnotes

Acknowledgement

Funding

Conflict of interest

References