The perspectives and experiences of physiotherapists and general practitioners in the use of the STarT Back Tool Tool: A review and meta- synthesis

Background: Chronic low back pain is associated with disability and work absence. Stratied primary care management of acute low back pain using a prognostic screening tool (STarT Back Tool, SBT) triages people with LBP into targeted treatment groups, matched to their level of risk of chronicity. The SBT was designed for use in the health service in the United Kingdom (UK) where it has been shown to improve clinical outcomes, patient satisfaction, and reduce treatment costs and time off work. Successful implementation of the SBT in health care outside the UK is dependent on the health practitioners who will use it and the healthcare system in which they work. Gaining health practitioners’ perspectives on the SBT is an important step in effective implementation. Methods: A computerised search of qualitative literature was conducted across seven databases in March 2021 using keywords to identify studies that investigate the perspectives of physiotherapists and general practitionners on the use of the SBT in primary health care. Study quality was assessed using the CASP tool. Data were coded and analysed using reexive thematic analysis. Results: Eight articles met the inclusion criteria and included the views of general practitioners and physiotherapists from primary health care settings from four countries. Three overarching themes were created from the data, the rst ‘making it work’, identies factors that may inuence implementation and continued use of the SBT in clinical practice. The second ‘will I do it?’, captured potential consequences of adopting the SBT and the third ‘it’s all about the patient’ emphasised how the SBT may affect patients and their potential reactions to risk stratication and matched treatments. Conclusions: Physiotherapists and general practitioners found using the SBT frequently enhanced their practice. from to to each healthcare environment.


Background
Low back pain (LBP) is a signi cant problem worldwide; prevention and management strategies have not reduced escalating prevalence rates (1)(2)(3)(4)(5)(6)(7)(8). A high proportion of those with LBP experience recurrent episodes over their lifetime (9)(10)(11)(12)(13). It is estimated that approximately 10% of people with acute LBP go on to develop chronic LBP which often results in disability and work absence (4,11,14,15). The impact on society is signi cant, in the personal cost of pain and suffering and the nancial cost of treatment and work absence. (6, 16-18) Traditionally a biomedical approach has been used to manage LBP (23). However there is evidence of an association between psychosocial factors and the propensity for acute LBP to become chronic (15,18,(24)(25)(26). This has led to the development of management strategies which consider psychosocial factors. Several screening tools have been developed for the management of LBP; the STarT Back Tool Screening Tool (SBT) (28) (http://www.keele.ac.uk/sbst/startbacktool/) was designed, developed and tested in the United Kingdom to support primary health care practitioners such as physiotherapists (PTs) and general practitioners (GPs). The evidence-based recommendation is for the SBT to be used at the rst contact for each new LBP episode (National Institute for Health and Care Excellence [NICE], 2016). The SBT is a 9-item questionnaire prognostic screening tool which identi es modi able psychosocial risk factors for developing chronic LBP (29). Patients are strati ed to receive targeted treatment, matched to their level of risk (30): Low risk patients require one treatment session which includes assessment and evidence based education on self-management; medium risk patients also receive education on self-management and an additional six sessions of evidence based physiotherapy; high risk patients receive the same as for medium risk and psychologically informed physiotherapy (15,(31)(32)(33). This strategy of screening and matched care, supports health practitioners in clinical decision-making (27,34,35). Strati cation of people with LBP using the SBT has been shown to reduce health care costs, individual suffering and productivity loss in the UK (4,15,33,36).
The SBT was initially validated for use in the National Health Service in the UK (31) and has since been used in a number of countries. However, the introduction of a strati cation tool into a primary care setting is dependent on those who will administer it and the constraints imposed on them by the health care system in which they work (32). Understanding practitioners' perspectives will help address barriers to successful implementation of the SBT (32,37).
This study is a systematic review of studies investigating the perspectives of physiotherapists and GPs on the use of the SBT in clinical practice.

Methods
The review was conducted using Preferred Reporting Items for Review and Meta-Analysis guidelines (PRISMA; (38).

Databases searched and inclusion and exclusion criteria
The databases searched were CINAHL, MEDLINE, Sport Discus, EBSCO, Scopus, Web of Science, Cochrane Review Database, and Cochrane Central Register of Controlled Trials published in English in peer reviewed journals from the inception of the SBT (2008) to March 2021. Articles were included if they were of qualitative design or included qualitative data reported separately from quantitative data. Participants in the study were health practitioners (general practitioners or physiotherapists) working in primary care. We included studies which had collected data during face to face interviews, focus groups, telephone interviews or workshops. The focus was perspectives of previous use or intended future use of the SBT. All retrieved articles were imported into bibliographic management software (Endnote X8). Duplicates were identi ed and removed.

Search strategy
Search terms were developed to identify studies relating to the perspectives on the use of the SBT and were adapted for different databases (see Table 1). Four key term clusters were used, which identi ed the group of participants of interest; health care practitioners (physiotherapists and general practitioners), the intervention; strati ed care (SBT), the condition of interest; low back pain, and the phenomenon of interest; and perspectives of those administering the tool.

Appraisal of included papers
Title and abstracts of articles were reviewed against inclusion criteria to identify relevant articles by two reviewers (FT and JH). Full texts of articles where the title and abstract t the inclusion criteria were reviewed using the same eligibility criteria by FT. JH independently reviewed the identi ed eligible articles. Where disagreement existed, NS acted as arbitrator. Reference lists of all included articles were searched for additional relevant articles and two authors, who have published widely in the area (Hill, JC and Caeiro, C) were contacted and invited to view the nal article list to ensure all relevant literature had been included. "health* practitioner*" OR "health care pract*" OR "health* professional*" OR "health person*" OR physiotherap* OR "physical therap*" OR "primary care*" OR "general pract*" OR gp OR doctor* OR physician* OR nurse* OR pt "low* back" OR "lumbar spine" OR "lumbar pain" OR lbp OR "non speci c back pain" OR "nonspeci c back pain" AND Perspectives (beliefs and experiences) opinion* OR thought* OR perspective* OR experience* OR impression* OR view* OR training OR support* OR administer* OR implement* OR attitude*

Quality Assessment
The Critical Appraisals Skills Programme (CASP; Critical Appraisal Skills Programme, 2018) tool was applied independently by two reviewers (F.T. and G.A.) to assess the quality of included articles. No article was excluded on methodological quality however the Cochrane Collaboration recommend this tool to assess con dence in the quality of a review (40). Any discrepancies of opinion were resolved using an additional reviewer (JH) to make the nal decision.

Data Extraction
The following data were extracted from the included articles: study design, participants, country, and method of data collection. Two reviewers (F.T and G.A) extracted data independently from the results section of each paper in the form of direct quotes and summaries of participant interviews.

Data Analysis
Re exive thematic analysis with an experiential orientation was used to gather ndings and generate themes (41). This method captured the experience of the participants by focussing on the language used (42). Two reviewers (F.T. and G.A.) familiarised themselves with the data by reviewing and repeatedly reading the selected papers. Each article was coded line-by-line and the codes were grouped and categorised into candidate themes, (sub-themes) ensuring that the context of the original article was considered. The data were then reviewed by four reviewers (F.T., G.A., J. H. & N.S.) and the candidate themes were grouped under overarching themes which went beyond the content of the original studies.
Themes were nalised following multiple reviews of candidate themes and subsequently de ned and named.
Representative quotes were selected to support the themes. The aim of this was to develop themes which are more directly relevant to clinical practice.

Literature Search
The process for study selection can be found in Fig. 1.
Six hundred and three journal articles were identi ed, 300 remained after removal of duplicates. Results from eight studies undertaken between 2011 and 2020 were included in the nal review and meta-synthesis. Authors were contacted to request full publications of two abstracts that appeared relevant to this review; one was not published (43) and one author contacted twice did not reply (44). In the eight articles retrieved, 76 physiotherapists and 65 GPs were interviewed or surveyed about their perspectives on the use of the SBT.

Study Characteristics
A summary of the articles included in this review can be found in Table 2. Studies were completed in four countries: four in the UK, two in Germany and one each in the USA and Portugal between 2011 and 2020. Four studies interviewed participants following exposure and use for several months (36, [45][46][47]. The remaining four studies reported on interviewed participants who had completed workshops where potential for the use of the SBT use was being investigated (4,44,48,49).

Quality of Articles
All studies met at least seven of the ten criteria in the CASP tool, with the relationship between the researcher and included participants most frequently not addressed. (see Table 3).
Relationship between research and participants considered Additionally, physiotherapists identi ed speci c areas of education they needed to manage psychosocial factors. They reported feeling under prepared by their tertiary education to undertake cognitive behavioural therapy, exercise prescription and the level of skilled communication required (4,44). Some physiotherapists hoped that the SBT would increase the trust GPs had in physiotherapists' ability to manage patients with LBP (4).
[…] our everyday reality is that there is little con dence in our capabilities […] this approach enhances our own con dence as well as the physicians' trust in us, that the patients receive appropriate physiotherapy treatment. [...] the GP is the rst person the patient sees, if they could be better educated with regards to back pain and then that ltered down and everyone trying to do a fairly similar thing, not similar for every single problem but approaching problems in similar ways. (Physiotherapist; Sanders et al., 2014) System policies refers to how speci c health system policies affected the motivation of participants to use the SBT appropriately. GPs reported a fear of being penalised by insurance companies for overloading services, reduced their motivation to refer patients to physiotherapists (47,48).
[...] I wasn't going to get my wrists slapped, because you do sometimes get your wrists slapped [...] we get messages saying what the wait is for physio and you know, please don't overload us and we're often being told that so I try and manage. (GP; Sanders et al., 2011) Physiotherapists also expressed concerns about restrictions placed on them from insurance companies. These included lack of freedom in decision making, prespeci ed time for consultations, and a restriction in the number of follow-up appointments. These restrictions could limit the appropriate use of the SBT (4). Physiotherapists reported that closer working relationships with GPs reduced waiting times (44) and the change in frequency and duration of consultations resulted in a reduced case load (46).
We may be seeing them over a longer period of time, but we're not seeing them as often. So that's working better for us from a case load point-of-view. This captured participants' con icting perceptions of whether it was worth undertaking, and potential consequences of adopting it. There were ve subthemes: 'is there time?', 'ability to change', 'recognise me', 'I know better', and 'changing me as a health professional'.
Is there time? examines whether there was time available in participants' sessions to administer the SBT l. Most GPs indicated that they lacked time to administer the SBT assessment and treatment recommendations (44,(47)(48)(49), with some GPs having as few as nine minutes per consultation (48). Some physiotherapists agreed that the SBT was time consuming for GPs and both professions suggested ways to reduce the time taken. For example, education being delivered remotely via media instead of face-to-face (4, 45, 46, 48). Ability to change, considers the factors affecting GP's and physiotherapist's ability to make the change required to implement the SBT. A factor that was pertinent to the success of SBT implementation was ensuring the local availability of recommended referral services (49).
Whatever decision you come out with, the care option that it's recommending must be embedded within the healthcare system that you're working in. So there's no point of having these treatment options in a tool if it's not going to be available in your practice area. (Physiotherapist; Saunders et al., 2016) Another in uencing factor was the reported stress associated with a change to routine care practises. Some participants forgot to use the tool in consults (44,45,47). Recognise me refers to the factors that affected physiotherapists' motivation to upskill and treat more challenging patients. Due to the signi cant role they would play in the management of LBP patients using the SBT, physiotherapists advocated for increased recognition from health industry colleagues (4, 46).
[…] that would indeed be a rst agship. Meaning, that in medicine or even politics they would say: 'look how important physiotherapy is!' (Physiotherapist; Karstens et al., 2018) The reduction in the number of treatment sessions required for low-risk patients, with the consequent reduction in income was of concern to physiotherapists (4). This concern, combined with an increase in complexity of treatment required for high-risk patients, meant physiotherapists expressed a desire for remuneration to re ect the change (4).
I think reimbursement is rather cause of frustration for all of us therapists [laughter], we are convinced that we don't get paid to an extent we think we are quali ed [...]. With these tasks [described for the STarT-Approach], with these additional quali cations, physiotherapy is gaining more, dramatically more importance and thus deserves a higher reimbursement. A few GPs confessed they were not using the SBT, not only because LBP was not a priority for them, but that it was a clinically uninteresting condition that often did not require a GP visit (45,47). Others felt conditions such as heart disease and diabetes took precedence over LBP (47). Physiotherapists also noticed when they referred patients back to GPs due to unsuccessful treatment, GPs were reluctant to engage with patients (46). Changing me as a health professional, illustrates how the SBT affected the professional attributes of GPs and physiotherapists. Both professions expressed that the SBT improved their con dence when dealing with the psychosocial factors that often accompany chronic LBP (45)(46)(47).
And I think now I've got more con dence in how to deal with some of the things they're saying to me. Before if I'd done that and I wasn't con dent I think I would've reverted back. (Physiotherapist; Sanders et al., 2014) Physiotherapists found the SBT training was useful as it helped them re ect on their current practice and gave a fresh perspective on the treatment for LBP. Additionally, they reported a shift in focus from a biomedical approach to a biopsychosocial one (45,46).
It made me re ect on things and it's helped me see, identify some habits that I probably have got into over the years that I need to look at changing. Conversely, a prominent concern of GPs was that using the SBT would reduce clinical autonomy and would undermine GPs' professional identity (49). Physiotherapists were also fearful that addressing more psychosocial aspects and using fewer traditional hands-on treatments could decrease patient respect (4,46). Karstens and colleagues (4) also reported that some physiotherapists were afraid that selected areas of traditional treatment would become obsolete.
[…] the more you as a physiotherapist address these psycho-social aspects of medicine, don't you risk rationalising yourself out of the picture? (Physiotherapist; Karstens et al., 2018) Theme 3: It's all about the patient The theme 'it's all about the patient' emphasised how the SBT may affect patients and their potential reactions to risk strati cation and matched treatments. Both professions identi ed three areas to be considered prior to implementation: patient individuality, the in uence of patient demand on decision making, and a patient's motivation to improve. Participants felt they were pigeonholing patients by stratifying them into one of only three groups which they felt fails to consider patient individuality (4,47).
[…] I'm one of these that thinks that there's an art to general practice and it's more a sort of conversation and a feeling between two people. Now, you can't put feelings into a questionnaire [reference to the tool] so what's right for one, is completely wrong for another and unless you had a questionnaire that was a thousand questions long, you're just not going to capture that, are you? (GP; Sanders et al., 2011) Both professions agreed that patient demand could signi cantly in uence decision-making surrounding referrals (36, 46, 47). These participants recognised that patients have rights to request speci c treatment and therefore they felt the need to consider the requests to keep generating business.
Whereas, with a patient it's more of, I don't know how to put it really, it's not like a bartering but it is a bit like that you know. They want physio, […] and you're thinking well you know, should they have it or not and they may only score one but if they really want physio, they going to get physio because we're here to meet the patient expectations and Participants mentioned that some patients prefer passive treatments and lack motivation to recover (4, 46, 47).
If he is really motivated and wants to recover quickly, then things will go faster […], for those people who enjoy being given a sick note and spending a week or two at home … (Physiotherapist; Karstens et al., 2018) On the other hand, some physiotherapists anticipated that this method of treatment may be embraced by patients due to previous unsuccessful treatment. Karstens and colleagues (4) reported that patients who received a preliminary education session prior to physiotherapy presented with an improved attitude towards treatment. Additionally, physiotherapists found the SBT encouraged patient engagement in treatment (46).
[…] it means you've got a totally different starting point with the patient, […] the patient is pro-active, and we can get away from never-ending treatment sessions. (Physiotherapist; Karstens et al., 2018) When considering patient-clinician interactions and how patient-centred care was affected, both professions highlighted that the SBT enhanced communication by enabling conversations with patients regarding different aspects of their LBP (36, 45,46).
What was really useful to me was the discussions with the [trainer]… That really changed my focus when talking with patients about back pain, really letting them know that no harm will come to them from being active and how to prepare them appropriately for what physical therapy could offer. (GP; Hsu et al., 2019) Discussion Implementation of a tool to help manage a clinical problem is dependent on the willingness of health practitioners to accept new knowledge and utilise new concepts to drive a change in behaviour. The tool needs to make sense and have meaning to intended users to increase the likelihood of positive impact and successful implementation (46). This review captures the perspectives on the use of the SBT and how health practitioners feel it may in uence their practice.

Summary of Results
The main themes identi ed health practitioner perspectives that in uenced the use of the SBT. Participants felt the SBT enhanced their practice though many found it constrained the use of their diagnostic skills. Time constraints were of concern, but many felt that if they put time in to making it habitual practice, using the SBT could become e cient. Involving patients in decision making on their treatment was thought to be one of the bene ts of using the SBT, though some GPs felt it had the potential to limit treatment options.
Physiotherapists and GPs felt the psychosocial education during the SBT training added to their skills and ability to treat more complex patients. Many would have valued additional support with this after they'd had a chance to use it in practice.
The wide range of pathologies GPs encounter meant they often placed LBP as a low priority meaning they discharged patients with chronic LBP sooner than physiotherapists believed was optimal. However, using the SBT gave physiotherapists and GPs a better understanding of each other's management of LBP resulting in an improved interprofessional relationship.
Physiotherapists had concerns about the nancial implications of using the SBT and the ability to translate it into their speci c healthcare system. Many felt that once pro cient in the use of the SBT, their skills should be recognised nancially.

Findings in relation to existing literature and guidance
The STarT Back Tool strati es patients with LBP to receive appropriate matched treatment at an optimal time point (32). Most participants felt that the SBT served its purpose as a strati cation tool and enhanced their practice. However, a contrasting view from some physiotherapists and GPs was that -the SBT prevented them from exercising autonomy.
Additionally, the inclusion of the SBT in some clinical guidelines gives the impression of compulsion for some participants, preventing the use of the diagnostic skills they have developed through clinical experience. The ability to diagnose often involves pattern recognition, which many GPs view as a component of 'the art of practice' (50). However reliance on pattern recognition can result in a longer pathway to effective treatment and using the SBT has been shown to enhance rather than detract from effective practice (15). The training focuses on the key messages, that the SBT is designed to: aid, not take over clinical decision making; improve the e ciency of LBP management; and help manage high and challenging caseloads (34,51).
A nding from this review was that some GPs perceived LBP to be a low priority, a nding corroborated by Poitras and colleagues (52) who reported on barriers to the use of LBP management guidelines by health practitioners.
Physiotherapists reported that patients with chronic LBP were discharged from their GP prematurely, with GPs expressing that their clinical expertise was better spent on major chronic conditions. GPs reported that guidelines were of low impact and that LBP represented a limited, clinically uninteresting portion of their caseload. This perception may stem from their requirement to manage a wide range of health conditions; in Australia, LBP only accounts for 1.8% of GPs caseload despite being the eighth most common condition they manage (51). Many GPs also express a lack of con dence and understanding of chronic LBP, with authors reporting that the more complex a patient is, the less willing GPs are to address the problem in its entirety (53,54). The belief that LBP is not a serious condition may explain why some patients are dismissed as malingerers. Frequently such patients have underlying psychosocial factors inhibiting their recovery and if not addressed early, there is an increased risk they will experience fear-avoidance behaviour (55) and may enter a cycle of pain, suffering and disability.
A promising nding from the current review was that physiotherapists and GPs expressed increased engagement in their role following the SBT training, due to increased con dence in addressing psychosocial factors (52,56,57). This facilitated a shift from a biomedical to a biopsychosocial approach (20), adding a new dimension to routine care.
Synnott and colleagues (57), reported on physiotherapy perspectives of managing psychosocial dimensions of LBP after intensive training. Participants described increased con dence and an improved level of skill in the management of patients with chronic LBP.
Some participants in this review believed that allowing patients a choice of treatment options made them the centre of decision-making. GPs expressed concern that strati ed care could limit options and result in a loss of patient choice and independence. Loss of patient autonomy has been shown to negatively affect patient biopsychosocial wellbeing (58-60). However, it is a matter of debate how much clinicians should allow patients to dictate their treatment. Health practitioners often deal with the internal con ict between a paternalistic approach of doing what is best for the patient and cultivating patient autonomy (58, 60). An example of this is whether imaging should be undertaken for LBP in the absence of red ags (61). In some cases imaging may reassure the patient that there are no abnormal ndings, but early imaging for acute LBP has been correlated with increased sick leave and likelihood of surgery, and an overall poorer prognosis (62). Consequently, when patients express a strong treatment preference, GPs and physiotherapists can choose to grant patient requests, to maintain therapeutic relationships or deny requests with the intention of doing what is best for the patient (60, 63).

Strengths and limitations of the study
Strengths of this study include the use of PRISMA guidelines and the CASP tool. The data analysis was subjected to a rigorous peer-review process; FT and GA presented their initial coding to NS and JH and there was discussion and recoding to ensure that each code re ected the participant data accurately. The process of writing the results yielded several iterations and re nements to the themes which were agreed upon by all authors. Studies were included from four countries, and participants had a range of experience in their professions. There was a dearth of studies undertaken outside the UK, despite undertaking a sensitive search of seven data bases and contacting key authors in the area. This may be due to the relatively recent implementation of the SBT in countries other than the UK. One limitation was that only four studies had participants who had used the SBT (36, [45][46][47], the remainder had been educated and were contemplating it's use (4,44,48,49). Although this pre implementation data is of interest, it is not experiential.

Clinical and research implications
In the United Kingdom (UK) healthcare is primarily publicly funded by the National Health Service (NHS) (64). The STarT Back Tool was developed in the UK and hence was speci cally designed to t their funding model. Research in other countries is needed to gain an understanding of how the SBT can be translated and adapted to t health practitioners working in other health funding models. Consultation and collaboration with health practitionners and other stakeholders is required to facilitate development and effective translation of the SBT into their environment.
Understanding how different healthcare systems implement strati ed care will help to develop an effective system for managing people with LBP. Furthermore, whilst evidence-based practice has a signi cant role in the translation and implementation of the SBT, patient expectation and satisfaction are equally important for successful implementation.
Further research needs to focus on patient self-reported clinical outcomes using the SBT approach and their satisfaction with the care they received. The translation of a clinical tool designed for a speci c health system to a different context necessitates consideration of many factors and consultation with stake holders. This will facilitate adaptation of the SBT approach for each country's healthcare system. GA and FT analysed the ndings. All authors contributed to identi cation of the themes and analysis of the themes and drafted the manuscript. Study selection process