Qualitative study exploring the views of patients and healthcare providers on current rehabilitation practices after lumbar fusion surgery


Low back pain (LBP) is one of the most common, costly and debilitating musculoskeletal disorders worldwide.1 Lumbar fusion surgery (LFS) is an operative technique that joins two or more vertebrae and is increasingly used to alleviate persistent LBP due to degenerative lumbar spine conditions.2 LFS is thought to mediate ‘biomechanically unstable or malaligned’ spine segments, despite the lack of high-quality clinical trials on its efficacy and poorly defined biomechanical criteria.3 4 Further but less debated, LFS is commonly performed as an adjunct to nerve root decompression surgery when the decompression inevitably compromises anatomical integrity.5 6

No consensus exists on the optimal content and timing of preoperative, perioperative and postoperative rehabilitation in LFS, despite the fact that its importance has been widely acknowledged.7 8 This lack of evidence on what can be considered ‘good clinical practice’ rehabilitation could contribute to unfavourable outcomes after LFS, including poor functional status, ongoing pain and psychosocial burden.9 For other musculoskeletal disorders such as total knee and total hip arthroplasty,10 11 evidence-based rehabilitation care pathways have been shown to improve the clinical outcome and to reduce healthcare costs after surgery. However, such care pathways focusing on rehabilitation for LFS are lacking.4 5 A crucial step in the initial development phase of a care pathway is to understand patients’ and experts’ opinions on the current and preferred management of a specific condition, here LFS. These opinions are important to detect existing research gaps as well as to identify potential barriers and facilitators for implementing the aimed intervention.12

Qualitative methods are particularly suitable for exploring these opinions with a low risk of narrowing down complexity.13 14 As an important first step in the development of an evidence-based rehabilitation care pathway for patients following LFS, this qualitative study was conducted to fuel further pathway development steps, in particular a Delphi process to reach consensus on key interventions. Therefore, the primary aim of this study was to gain insight into patients’ and healthcare providers’ (HCPs) opinions regarding the current management of LFS. The secondary aim was to obtain these stakeholders’ opinions on the preferred rehabilitation care pathway for patients undergoing LFS.


Study design

A qualitative descriptive interpretive design was applied. The Consolidated criteria for Reporting Qualitative research15 provided guidance in reporting this study.16 The current study is part of the Rehabilitation for lumbar Arthrodesis Controlled Trialproject.


The study was approved by the Ethics Committee Research UZ/KU Leuven (S60109) and registered at clinicaltrials.gov (NCT03427294). All participants provided written informed consent.

Patient and public involvement

As no specific patient society with trained patient research partners for LFS exists, we involved one patient with LFS recruited in our clinic in the design of our study.


A total of 36 participants were selected based on their experience with rehabilitation of LFS, either as HCP (n=31) or patient (n=5). Two researchers (TT and MB) recruited and purposefully selected the patients out of two departments in an academic hospital (orthopaedic surgery and neurosurgery). All interviews with the five patients took place between 1 week and 1 year postoperative. Participation was voluntary. There were 4 HCPs from non-academic hospitals and 27 HCPs from an academic hospital. The purposeful sampling method12 was used to capture heterogeneity in the expertise and work settings (academic and non-academic hospital setting) of HCPs and in the demographics and clinical outcomes of patients. Details on participants are summarised in table 1.

Table 1

Participants’ demographic details (n=36)

Interview procedure

Face-to-face, in-depth interviews were conducted by TT. TT is a musculoskeletal physiotherapist with 18 years of experience in treating patients with LBP. She was trained in conducting semistructured interviews, analysing data in Nvivo and developing clinical care pathways. To increase credibility,16 the results of the interviews were analysed by two independent researchers (TT and MB). The epistemological background of our research group was dominantly ‘constructionism’. MB is a specialist in physical and rehabilitation medicine and trained in conducting interviews and analysing data in Nvivo; he has a special interest in spine pathology. In between, two poor and two rich interviews were also analysed by two independent researchers VVA and TS for verification with the same goal to increase credibility but also reflexivity (ie, critical self-reflection about oneself as researcher). Interviews lasted approximately 45 min and took place at the office of the HCPs in the hospital (except for one patient at home). Data were collected, coded and analysed from 2018 to 2022. Additional interviews were performed until data saturation. All interviews were audio-recorded and transcribed verbatim, except for two participants (one patient and one HCP) who did prefer to conduct the interview without audio-recording.

A topic guide was developed upfront based on a thorough review of the literature7 and iteratively refined during the interview process. Interview topics and example questions are provided in table 2. The topic guide included questions targeting the (1) participant’s preoperative and postoperative perceptions and experiences, (2) underlying attitudes and beliefs towards the LFS procedure, (3) outcome expectations after LFS, (4) facilitators and barriers to recovery after LFS, (5) ideal content of physiotherapy after LFS and (6) experience with rehabilitation and return to prior functioning after LFS.

Table 2

Topic list of experiences, attitudes and/or expectations with example questions

Strict confidentiality was assured to the participants. Only the two interviewers had access to the original non-anonymised data. The interviews were transcribed verbatim by TT and two research assistants including the non-verbal signals (observed by TT during the interviews). Additionally, a short report about the interviewee and contextual characteristics of the interview was made, supporting data extraction afterwards. All audio-recordings were destroyed after transcription.

Data analysis

The data analysis was guided by the Qualitative Analysis Guide of Leuven.17 Accordingly, an elaborate preparation of the coding process was performed before the actual coding with the aid of computer-assisted software (Nvivo, developed by QSR international, V.12). Iterative processes of going deep in the data and repetitive moving between the various stages of the process were performed.17 18

First, the data were coded independently and analysed into meaningful concepts by two researchers (TT and MB). Next, the finalised code list was applied to all transcripts. A conceptual scheme was made from each interview, highlighting and summarising the emerging themes. Through a process of repetitive interpretation, synthesising and theorising, themes were identified. Transcripts were re-read several times, and the selected themes were finalised into a hierarchical conceptual framework based on consensus discussion between the two principal researchers.


To the best of our knowledge, this qualitative interview study is the first to explore the perspectives of both HCPs and patients regarding the preoperative, perioperative and postoperative rehabilitation of LFS. Starting from their opinions on the current rehabilitation practices, suggestions for a more optimal rehabilitation pathway for patients undergoing LFS pathway emerged.

Although the interviews adopted a neutral stance towards status quo versus change in current practice, participants predominantly focused on discussing the ideal care pathway for LFS. In this way, critique on the contemporary pathway could be positively and safely formulated, and the central theme of this work, namely, the need to change current LFS care, received maximal focus. This early finding immediately stressed the role of contextual factors and the complexity of implementing care pathways, which motivated us to summarise study findings into an implementation science framework.21 The 2022 updated CFIR19 20 provides an evidence-based menu of constructs that are associated with effective implementation and fit our findings best (figure 1). CFIR comprises five major domains: the ‘innovation domain’ with factors related to the implemented pathway itself, the ‘inner domain’ and ‘outer domain’ referring to the micro to macro setting in which the innovation is implemented, the individual’s domain’ including roles and characteristics of all stakeholders including patients and the ‘implementation process domain’ pointing to the activities and strategies used to implement the innovation. As participants’ opinions touched all and often multiple domains (online supplemental appendix A), the multidimensional nature of CFIR could be confirmed with a larger role for concepts related to the innovation, inner and individual’s domains. Below key emergent concepts are discussed.

Our results highlight that the precise role and competence of each member of the interdisciplinary team needs to be clearly defined, respected and discussed. A possible barrier could be an excessive hierarchy between different disciplines of the team.22 23 Some participants perceived this hierarchy and expressed this by non-verbal communication during the interviews rather than by specific quotes. This finding aligns with the plea for care pathway development with bottom-up co-creation involving all healthcare and research community stakeholders equally.5 24–27 Similarly, all participants in this study emphasised the crucial role of uniform communication. An example given was the mixed messages by different HCPs towards patients on when activities were allowed postoperatively.

A central point of contact, mostly assigned to an extended scope physiotherapist, was considered useful by interviewees to ensure this ‘one voice’ expert communication, given the large number of HCPs involved in the rehabilitation of LFS. Oestergaard et al
28 did however not find any added benefit of preoperatively started case management by a social medicine doctor, an occupational therapist or a social worker with late postoperative (telephone) support on clinical and work outcomes. We speculate that the professional background of the case manager and the exact timing or time investment may indeed affect results.

Somewhat related, participants believed that establishing a strong therapeutic alliance is essential for a positive clinical outcome following LFS rehabilitation. Indeed, a positive therapeutic alliance was associated with improvements in key clinical outcomes such as pain and disability in the broader literature of LBP research, but data in LFS are still lacking.29 Therapeutic alliance is known to be nurtured by agreement on treatment goals, shared decision-making and assignment of treatments, as well as by an adequate emotional bond between patient and HCP.30 Correspondingly, it was stressed by the participants that patients should be encouraged to express their expectations, preferences and views in a shared decision-making process during the rehabilitation after LFS.31 Participants therefore considered the preoperative period as the perfect ‘window of opportunity’ to initiate the rehabilitation pathway for LFS, ideally with a thorough assessment by a physiotherapist once there is an indication set for surgery. Participants felt this period is ideal to build a therapeutic alliance, streamline reassuring communication, discuss expectations and goals and educate patients to take informed shared decisions.32–34 Prehabilitation of lumbar fusion is still in its infancy, certainly compared with prehabilitation in other orthopaedic interventions such as knee and hip replacement.35 Nevertheless, preoperative physiotherapy and psychological therapy improved pain after LFS, in the study of Nielsen et al
36 and Reichart et al

There was no doubt among interviewees that rehabilitation for LFS should include a biopsychosocial approach to assessment and treatment, which is consistent with current guidelines on LBP management.5 37 Abbott et al
38 linked the content of interviews with patients to the well-known International Classification of Functioning, Disability and Health framework. This confirmed the broad impact of LFS seen on the levels of body structures and functions (35.5%), activities and participations of daily life (53.6%) and personal and environmental domains (10.9%). Several qualitative studies in LFS demonstrated longstanding and unacceptable preoperative pain and disability trajectories in patients, with a surgical decision bringing hope for relief that may bring unrealistic expectations about the outcome of LFS.13 14 39 Also, reports on the heterogenous postoperative outcome after LFS and the struggle to handle surgery-related pain as well as to regain a societal role illustrated ongoing stress.13 40 Jakobsson et al
41 and Abbott et al
42 found that pain catastrophising, preoperative control over pain and self-efficacy for exercise predict (although to a small extent) postoperative disability in patients with LFS. Not surprisingly, participants in our study felt that a referral to a psychologist should be facilitated, when necessary, because physiotherapists and other HCPs may lack competence in addressing specific or complex psychological barriers. Physiotherapists indeed formulated this need for psychological guidance to handle contextual factors in the broader literature on LBP management.43 However, evidence to show that trained HCPs can effectively deliver psychologically informed treatments (eg, based on acceptance and commitment therapy or exposure principles) is accumulating, but research on the value of such transdisciplinary HCP roles in the field of LFS is needed.44–47

All participants in this study acknowledged the importance of early mobilisation after LFS. This is in accordance with recent literature on surgery for degenerative spinal conditions for which benefits of this approach were seen.48 This early postoperative mobilisation contrasts with a still common practice of postoperative bracing following spinal surgery. A recent study showed that Belgian spinal surgeons prescribed postoperative lumbar bracing in more than half of LFS procedures.49 Interestingly, this decision to brace was mainly guided by the surgeons’ beliefs about pain relief and improved fusion rate. Similarly, in the qualitative work from Greenwood et al,14 patients reported to restrict their movement due to fear of surgical implant failure due to the absence of received information here or clear instructions not to move the first 3 months after surgery by the treating surgeon. This was confirmed by patient participants in this study who insisted on reassuring education on tissue integrity and recent qualitative work by Rushton et al
8 who stated that sufficient and coherent education in comprehensive management is needed to avoid and reduce threat associated with surgical procedure. Apart from this psychological point of view, sufficient movement is essential to promote bone healing in the vicinity of the LFS implants. Unfortunately, our study also revealed contradicting perspectives from different HCP disciplines that may hamper early mobilisation. Surgeons were much more stringent in postoperative restrictions than physiotherapists, who even felt that no restrictions on activities should be given. Also, adequate analgesia by a PC(I)A-pump was deemed necessary by some surgeons, whereas physiotherapists and occupational therapists perceived this as an important barrier to early mobilisation.

It was remarkable that some participant subgroups (physiotherapists, psychologists and patients) focused on return to work (RTW), while other stakeholders in this study did not discuss this topic. Rolving et al
50 found that only 42% of the patients returned to work, 1 year postoperatively. Oestergaard et al
28 indicated a stagnation of RTW rates after LFS from 6 months to 1 year postop. Further research on RTW is needed, because there are no clear guidelines on RTW for patients undergoing LFS and successful interventions are currently lacking.28

On the one hand, HCPs and patients in this study expressed views on LFS care pathways that broadly corroborated with the needs of patients found in the LFS literature and broader LBP treatment guidelines.5 8 14 While several of the HCPs also had a managing role in their hospital, it was nevertheless remarkable that opinions on cost-effectiveness of interventions, impact of legislation, staffing cost or hospital culture were largely absent. To gain insight into managerial aspects of lumbar fusion care, future studies should consider including more managers as participants.

On the other hand, discrepancies between team members on the exact rehabilitation contents exist. Gaining consensus on the rehabilitation content is an important challenge to be addressed by future research. This qualitative study lays the foundation for further developing an optimal rehabilitation pathway for LFS that encounters the needs of both patients and HCPs.7 35 49

This study has some limitations. First, the results of this qualitative study may be specific to the studied population and setting, that is, tertiary (non)-academic hospital care. However, transferability of findings was promoted by the recruitment of a large number and variety of participants and a thick description of their diverse backgrounds and opinions. In addition, each relevant discipline was represented by one or more participants, and four experts from non-university hospitals were purposefully included. Furthermore, the inclusion of five patients strengthens this study’s trustworthiness, since the insights of patients are essential to provide the best possible care.51 Second, it is possible that the reported perspectives may have been influenced by social desirability or power imbalances, particularly as the participants were aware of the role of the investigators.52 To limit this, we reassured confidentiality, allowed off-the-record statements and avoided the use of focus groups. Also, to increase reflexivity in our research team, interview coders from two disciplines (physiotherapy and physical medicine and rehabilitation medicine) were deliberately paired and supported by a methodological expert and independent researcher to promote self-criticism and credibility of findings.

This post was originally published on https://bmjopen.bmj.com