Nurse-led randomised controlled trials in the perioperative setting: Nurse-led randomised controlled trials in the perioperative setting: A scoping

Purpose: Nurses provide care at each phase of the complex perioperative pathway and are well placed to identify areas of care requiring investigation in randomised controlled trials. Yet, currently, the scope of nurse-led randomised controlled trials conducted within the perioperative setting are unknown. This scoping review aims to identify areas of perioperative care in which nurse-led randomised controlled trials have been conducted, to identify issues impacting upon the quality of these trials and identify gaps for future investigation.


Introduction
Health care providers are facing pressure to provide effective services to an increasing population with often limited resources. 1 This pressure to provide more with less is evident within the provision of perioperative care. As morbidity increases, so does the complexity of surgery and the pressure upon resources in this highly technical, resourceintensive, fast-paced, acute clinical environment.
For most patients, the experience of undergoing a surgical procedure represents a significant life event. During this critical period, health care practitioners are entrusted to advocate for and maintain the safety of patients when they are removed from family and loved ones and unable to speak up for themselves due to anaesthesia. 2 A safe passage through surgery is the highest priority. However, it has been argued that -despite the amount of effort spent on developing interventions and policy in recent years -progress in optimising patient safety in perioperative care has been much slower than anticipated. 3 Internationally, perioperative care is described in four distinct phases: pre-admission, the immediate preoperative (pre-anaesthetic) phase, the intra-operative phase (during induction of anaesthesia and surgery itself) and the immediate post-operative phase of care (prior to patients returning to ward areas). 4 This multi-staged pathway necessarily involves care delivered by a range of health care professions: registered and enrolled nurses, surgeons, anaesthetists, technicians, orderlies and radiographers. However, nurses are a consistent presence at all phases of perioperative care and may work in multiple roles, including pre-operative care, anaesthetic assistance, intra-operative (scrub/ scout) and immediate post-operative care roles. In some countries, other professions such as registered operating department practitioners (ODPs) take on perioperative roles. 5 However, globally, nurses have a ubiquitous presence in health care teams that provide perioperative care and are uniquely placed to understand critical points of care and patient concerns across the whole perioperative pathway. It is imperative that nurses ensure they are both driving health care improvements and identifying research priorities in this specialised field.
Experimental research underpins the assessment of the effectiveness of interventions, yet it is widely acknowledged that randomised controlled trials (the gold standard of experimental research) are expensive, resource-intensive and time-consuming. 6 It is essential that time and finite resources are well spent on interventions that are effective, safe and acceptable to patients. Resources and funding to conduct research are difficult to obtain, and therefore it is imperative that resources are directed to areas where gaps in experimental research exist. Furthermore, there is a need to ensure that resources are directed toward research that will be conducted in a rigorous manner in order to ensure high quality and reliable findings.

Experimental research in the perioperative setting
The conduct of rigorous, randomised controlled trials (RCTs) is often inhibited by well-known factors such as cost, time and resources. There are also other challenges in conducting research within this complex, multidisciplinary field that are not widely acknowledged. For instance, e-51 many recent systematic reviews and meta-analyses of perioperative care lack sufficient detailed reports of individual elements of care which may impact on or confound outcomes. 7 Perioperative outcomes are influenced by a wide range of factors throughout the preoperative journey and need to account for the truly multidisciplinary nature of perioperative care, by including nursing as well as medical interventions during each phase of care in study designs. 6,8 Therefore, the complexity of the perioperative pathway needs to be considered in both the design of primary studies and the assessment of these studies via systematic review. Authors have recently questioned the status of RCTsin remaining the 'gold standard' design to inform perioperative decision-making. 8,9 Several authors have suggested that carefully designed before-andafter (observational) studies can be used to inform perioperative decision-making, with the benefit of being less resource-intensive, and more indicative of the feasibility of implementing interventions in actual practice. 8,9 However, well-conducted, RCTsoffer the highest level of scrutiny with the lowest level of bias, and therefore the greatest benefits to our patients, and remain the gold standard of experimental studies. 6 Nurse-led research in the perioperative setting The multidisciplinary nature of perioperative care can result in challenges for nurses when trying to implement evidence-based practice change, such as negotiating staff buy-in across large multidisciplinary groups. 10,11 Challenges also exist for perioperative nurses engaging in primary research that is pertinent to the discipline, such as funding. Potential sources of funding for specifically nurse-led research may also be even more scarce given the seemingly limited lack of financial backing for perioperative research both locally and internationally. 12 Yet, the importance of supporting perioperative nurses to undertake research is vital in both facilitating evidence-based change in this domain of care. Nurses must drive research priorities that are relevant to perioperative nursing care. 13 Although perioperative, nurse-led research may be increasing, the extent to which of these are nurseled perioperative RCTshas not been evaluated.

Methods Aim
The purpose of this scoping review is to identify in which domains of perioperative care nurses are leading experimental research.

Objectives
The main objectives of the scoping review were the following: • to identify in which domains of perioperative care nurse-led RCTshave been conducted • to analyse the issues impacting upon the quality of experimental research undertaken in the perioperative setting • to identify what, if any, gaps exist in nurse-led experimental research in the perioperative setting, thus identifying priorities for future research.

Design
This scoping review was conducted in reference to the methodology set out by the Joanna Briggs Institute (JBI), 14

Inclusion and exclusion criteria
Studies that met the following inclusion criteria were eligible for review: Population: participants receiving care during one or more phases of the perioperative pathway: preoperatively, intra-operatively or immediately post-operatively.
Concept (study designs): only nurseled randomised controlled study designs were included. To enable the identification of these particular trials, in-depth investigation of author names and qualifications were performed for those studies in which details were not listed on the abstract or full text. Other trials were included if known to be led by nursing academics but whose qualifications are not explicitly stated in the citation.
Context: studies focused on perioperative care including the pre-operative, intra-operative or immediate post-operative setting.

Screening and eligibility process
Four reviewers conducted screening of titles and abstracts to identify relevant papers for full-text retrieval (JM, NH, LD, SM). Full texts were then screened for eligibility against the inclusion criteria by the authorship team using a verification form developed for this purpose (see Supplement 1).

Data charting process
A flow chart was generated to indicate the papers included in the review at each stage, as per the PRISMA guidelines ( Figure 1). 16 A data charting form was developed to record and extract study characteristics and variables relevant to the review question (see Supplement 2). Pairs of reviewers undertook data extraction independently for each article and a third reviewer mediated where there was a lack of agreement.

Critical appraisal
Studies identified as relevant to the review were assessed for quality using the JBI Critical Appraisal Checklists for Randomised Controlled Trials. 21 While quality assessment is not considered mandatory in scoping reviews, undertaking this process assisted in identifying common issues that influenced or undermined the quality of RCTsin the perioperative setting. Pairs of reviewers also assessed each included study for quality, with disagreements resolved through discussion and consensus.
Where agreement was not resolved through this process, an independent third reviewer was used.

Synthesis
Following data extraction and quality assessment, key information from each study was tabulated to assist in determining country of origin, interventions, primary outcomes, surgical population, sample size and funding source (see Supplement 3

Results
Eighty-six studies were included in the final review ( Figure 1). The included studies were geographically widespread (   Issues impacting upon the quality of experimental research undertaken in the perioperative setting Issues impacting upon the quality of RCTs included in this review were related predominantly to the reporting of blinding techniques. Blinding of participants was unclear or not implemented in 79 per cent of included studies (n=68), blinding of those delivering the intervention was not used or was unclear in 80 per cent (n=69) of studies, and blinding of outcome assessors was not used or was unclear in 73 per cent (n=63) of included studies. Many studies did acknowledge the reasons for lack of blinding and most often this was related to the nature of the intervention under study; yet, most often, lack of blinding of one or more key groups was not discussed or acknowledged as a limitation.
In addition, a lack of, or unclear, randomisation was found in just over a quarter of included studies (35%, n=31). Similarly, a high number of included studies were assessed as having incomplete follow-up or there was inadequate analysis or description of differences between groups (32%, n =28). Duplication of study results was also found in one instance, where the same study was published in different journals with a different author order. 85,87

Discussion
To our knowledge, this is the first scoping review to investigate the range of nurse-led randomised controlled trials conducted in the perioperative setting. Geographically, this review has revealed that North America contributed the highest number of studies to this review, with the United States of America (USA) the most prolific individual country in terms of conducting nurse-led perioperative RCTs in the last five years. This contrasts with a recent scoping review of RCTs and quasiexperimental studies published in nursing journals, whereby Taiwanese nursing researchers were found to have published the most frequently in nursing journals. 110 However, our review also included studies that, although nurse-led, were published in journals that were not specifically nursing-focused, and only focused on RCTs which was appropriate to address the review question. Similarly, though, our review also found no African studies for inclusion. 110 This may be unsurprising given that a 2015 scoping review of clinical nursing and midwifery research in African countries found that, at the time of the review, most included research was qualitative, and focused on primary or secondary prevention of cancer. 111 Additional obstacles to conduct and publication of nursing research in this region include a lack of resources (including funding, library access, equipment and collaborators) and political and civil unrest. 112 This review of 86 studies revealed that there are six clearly identifiable areas in which nurses are leading experimental research (specifically RCTs) relevant to perioperative care. The most common primary outcome across included studies was the prevention of anxiety and this was investigated using a range of supportive interventions. Given how commonly pre-operative anxiety is experienced, and the detrimental patient outcomes associated with anxiety, 54,93 this may be justified despite anxiety prevention not being a stated priority by professional associations. The investigation of supportive or complementary therapies may be reflective of the growing interest in complementary therapies in health care more broadly.
The quality issues noted in this review, in which a large proportion of studies assessed the effectiveness of supportive therapies, indicate that nursing researchers are utilising facets of the randomised controlled study design adaptively (and creatively). Given the expense and resources required to conduct RCTs, it is imperative for nurses to ensure that these resources are well spent on trials that are well conducted and provide useful findings. At this stage, it may be pertinent for the focus on anxiety prevention to shift from primary research to translation into practice.
Almost half of the included studies (47%) assessed interventions that were delivered during the preoperative phase. A moderate number (n=13, 15%) delivered interventions during the intra-operative phase but due to the nature of the interventions and outcomes under study -for example, the focus on anxiety reduction which would be difficult to assess intra-operatively due to anaesthesia -few studies assessed outcomes during the intra-operative phase of care (n=4, 5%). This gap in the literature is an opportunity for nurses to design experimental studies that measure the outcomes of interventions and outcomes related to intra-operative or procedural nursing care. Despite anxiety prevention being the most common outcome in the included studies, one did highlight that further investigation with teens or adolescents is worthy of future study. 54 While some regions and countries have established perioperative research priorities, [113][114][115] an international consensus is not evident. The lack of consensus may be influenced by the diverse and differing needs between developed and under-developed regions, but also reflects the variation in the processes used to determine the published perioperative priorities (including the variation in stakeholder involvement). The perioperative pathway is complex, multi-staged and involves numerous health professions in the delivery of care. Therefore, it is logical that any work to establish areas of perioperative care that requires a stronger evidence base needs to ensure multidisciplinary input -as well as ensuring that health care consumers also have input.
In the United Kingdom (UK), the National Institute of Academic Anaesthesia and James Lind Alliance (JLA) Research Priority Setting Partnership's agreed on ten anaesthetic and perioperative care priorities include a range of issues. These range from the study of the term effects of anaesthesia, to establishing 'success' measures for perioperative care. 113 The authors determined that specific care and physiological questions were ranked more highly by clinicians, whereas lay stakeholders ranked communication and long-term outcomes of anaesthesia more highly. 113 Similarly, Biccard et al's Delphi study of perioperative investigators in South Africa, while recognising the need for a co-ordinated perioperative research agenda, established national priorities that focused on a e-57 wide range of quite specific clinical care aspects although lay input into this process was not evident. 115 The failure to investigate outcomes that matter to patients within pragmatic trials is not unique to perioperative care. 6 Nonetheless, the primary outcomes of anxiety prevention and knowledge generation identified in this review align more closely with lay stakeholder-identified priorities related to communication, 26 which may be unsurprising given that patient advocacy is a key nursing role.
This review also found that safety outcomes received minimal attention in the nurse-led trial research included in this review. It has also been argued that safety outcomes, having also been neglected, should also be reported in pragmatic trials in the perioperative setting. 6 Within the perioperative nursing field, Steelman's top ten patient safety priority areas, established by perioperative nurses in the USA, identify only one of the primary outcomes of interest found in the included studies in this review as a safety concern (perioperative hypothermia prevention). 116 However, many of these safety concerns may not lend themselves as a focus of experimental research due to being rare events (for example, wrongsite surgery, prevention of retained surgical items, surgical fires) while others are less so (medication errors, pressure injuries). 116 118 This report highlights that, for some areas, meeting the key performance indicators has been problematic. For example, in 2017 there was an increased incidence of perioperative hypothermia reported. 118 Therefore, it can be argued that the continued focus on developing strategies to manage this condition is warranted.
All health care professionals leading experimental perioperative research need to ensure that the populations upon which research is focused are reflective of the needs of the surgical populations. As mentioned, no studies specifically focused on the needs of older adults were found in this review. Studies of younger, fitter populations may not be truly reflective of surgical populations outside of trial settings; thus, the practical application of research findings is reduced, and the interests of the older adults receiving surgical care may not be met. This need has been evident over the last ten years. In 2010, a large multicentre, prospective observational study of older adults undergoing surgery in Australia and New Zealand highlighted that complications and mortality among this cohort were prevalent, and strategies were urgently needed to address these issues. 119 However, nurse-led RCTs in the perioperative setting do not reflect the trend of focusing on older adults, and patients with cancer, which were reported more broadly in nurse-led experimental research across clinical settings. 110 This review has also revealed that common quality indicators are problematic in the conduct of RCTs in this setting. Unclear randomisation was evident across the majority of studies, despite the inclusion criteria only specifying randomised controlled designs. There was a lack of blinding in the included studies.
In the studies where blinding was implemented, the method of blinding varied considerably. Successful blinding may have occurred for the participant, those delivering interventions and/or the outcome assessors. While a number of studies acknowledged and provided an explanation for a lack of blinding, many other studies either reported but did not explain, or did not acknowledge the lack of blinding at all. Where acknowledged, most often blinding was not achieved due to the nature of the intervention. This is perhaps unsurprising, given that most of the interventions were delivered and/or outcomes assessed at time points of care where patients were awake. It is acknowledged that interventions such as the use of forced air warming, or some complementary therapies, are extremely problematic when trying to include effective blinding techniques for participants. 99 Nonetheless, bias related to lack of participant blinding may be offset by the assessment of objective outcome measures and the use of outcome assessor blinding, where possible. 120

Limitations
There is potential that some nurseled RCTs meeting the inclusion criteria have been inadvertently missed, despite our extensive and thorough search process. The process e-58 of identifying nurse-led studies was complex during the search phase of this review. Not all studies clearly identified the professional background of authors. This meant that additional searches of the primary author's name were, in some instances, needed to identify whether or not studies were nurse-led.
This review also only provides a picture of randomised controlled studies conducted by nurses in the last five years. Quasi-experimental, observational and qualitative studies were not included, nor were secondary analyses such as systematic reviews and metaanalyses. Therefore, this review cannot provide an indication of the non-experimental or synthesised body of evidence generated by nurses in this clinical setting. We also only included studies published in English. Future studies may seek to investigate the body of nurse-led research conducted using these study designs to gain a more inclusive snapshot of research in this clinical setting.

Conclusions
This scoping review has identified clear areas of perioperative care that have been the focus of nurse-led randomised controlled trials. The emphasis has been on supportive care of both patients, and caregivers.
Most conducted research has involved multiple phases of care, across the perioperative pathway. Significant issues affecting the quality of experimental nurse-led research conducted in the perioperative setting have also been identified, mainly relating to blinding and randomisation. Acknowledging these issues provides opportunities for maximising research quality in nurseled experimental research. Gaps in perioperative nursing research exist in focused assessment of intra-operative or procedural aspects of care, patient safety outcomes and care of vulnerable groups. Opportunities also exist for nurses to contribute to multidisciplinary research priority setting in the perioperative field and focus on the translation of evidence to practice in areas such as anxiety prevention where further extensive experimental research may not be warranted. Priority settings must also include patients and caregivers as stakeholders to ensure that we are meeting their needs.

Ethical considerations
This review did not involve primary research and therefore ethical approval was not required. However, a potential conflict of interest relating to one of the primary review authors also being the author of one of the included randomised controlled trials was noted. In this instance, the review author was not involved with the critical appraisal of this study.

Acknowledgments
This review is one of a series of scoping reviews currently being conducted by researchers within the acute and critical care research group at Queensland University of Technology (QUT). They aim to identify current nurse-led research activities in acute and critical care settings (including perioperative care) and nursing research priorities. This collaborative group includes a number of university-based researchers and clinician researchers working in acute and critical care settings to ensure that the review outcomes are clearly linked to clinical practice. Within this group, we wish to acknowledge the input of Dr Petra Lawrence for assistance in critical appraisal and data extraction.

Funding
There is no funding to report.

Disclosure
A potential conflict of interest related to one of the primary review authors also being the author of one of the included randomised controlled trials was noted. However, in this instance, the review author was not involved with the critical appraisal of this study. SK reports that her employer (QUT) has received monies on her behalf from BD Medical for educational consultancies, outside the submitted work. The authors report no other possible conflicts of interest in this work.

Timing of intervention (I) and timing of outcome (O) Funding
Al-Azawy,