The prevalence of chronic low back pain (LBP) worldwide is still on the rise, possibly because of mismanagement and inadequate treatment approaches that fail to recognise and manage the cognitive, psychological and social factors associated with non-specific chronic LBP (NSCLBP) (Bourne, Machado & Nagel 2014; Darlow et al. 2012).

Contemporary clinical practice guidelines for managing NSCLBP recommend a shift from the traditional biomedical model – focused on identifying and treating tissue pathology – towards a biopsychosocial orientation framework. This transition reflects growing recognition that biological factors alone cannot fully explain the persistence of pain and disability (Cowell et al. 2019; Meroni et al. 2021). Patients’ knowledge, attitudes and beliefs about pain, including misconceptions about injury, fear-avoidance behaviours and catastrophising, have been shown to influence symptom chronicity and treatment outcomes (Cowell et al. 2019; Meroni et al. 2021). Evidence indicates that limited pain knowledge and negative beliefs are associated with greater disability and poorer recovery (Tarimo & Diener 2017), while clinicians’ own biomedical-oriented attitudes can affect advice, activity recommendations and sick-leave practices (Gardner et al. 2017; Mukoka, Olivier & Ravat 2019; Lotia & Sheth 2021). Accordingly, a biopsychosocial approach is essential to address maladaptive cognitions and behaviours, enhance patient understanding and improve functional outcomes.

Recent studies have increasingly recommended the use of a systematic method of stratifying patients into clinically relevant sub-groups at low, medium and high-risk for transitioning to chronicity based on psychosocial factors and the tailoring of treatment according to their risk (Al Zoubi et al. 2019; Beneciuk et al. 2019). Reduced pain intensity, more targeted use of healthcare resources and a reduction in absenteeism were among the benefits found in a trial comparing stratified care with current best practice guidelines (Foster et al. 2014).

In African populations, contextual factors are widely recognised as key determinants of clinicians’ adherence to clinical practice guidelines, particularly those promoting biopsychosocial approaches to NSCLBP management (Ahenkorah et al. 2019). Among these contextual influences, the knowledge, attitudes and beliefs of healthcare providers themselves constitute a critical factor, as they directly shape clinicians’ readiness and ability to implement guideline-recommended psychosocial strategies in routine practice (Fitzgerald et al. 2018; Synnott et al. 2016). Limited training in recognising and addressing psychosocial contributors to NSCLBP continues to hinder effective implementation, highlighting the need for improved clinician education in pain science and biopsychosocial care (Fitzgerald et al. 2018; Synnott et al. 2016).

Strong evidence demonstrates that healthcare professionals’ knowledge, attitudes and beliefs directly influence clinical decision-making and treatment selection for LBP, with biomedical-oriented beliefs consistently associated with poorer adherence to evidence-based guidelines (Alshehri et al. 2020; Cowell et al. 2019; Gardner et al. 2017). Studies conducted in Europe, Asia, the Middle East and Australasia report persistent biomedical treatment orientations among physiotherapists and other healthcare professionals, despite growing endorsement of the biopsychosocial model (Alhowimel et al. 2021, 2022; Mikamo & Takasaki 2021; Rufa et al. 2022). For example, Alhowimel et al. (2021, 2022) identified moderate pain neurophysiology knowledge and a tendency towards biomedical beliefs among physiotherapists, which were associated with higher fear-avoidance messaging and less guideline-concordant care. Similarly, Mikamo and Takasaki (2021) reported that clinicians with stronger biomedical orientations were more likely to recommend rest, imaging and activity restriction, while Cowell et al. (2019) and Gardner et al. (2017) demonstrated that clinicians trained in a biopsychosocial framework achieved superior long-term reductions in pain and disability, even among patients who had previously failed conservative or surgical interventions. Collectively, these findings highlight substantial variability in clinician knowledge and treatment orientation across healthcare systems and cultural contexts, highlighting the importance of region-specific investigation. To date, however, no published studies have examined the knowledge, attitudes, beliefs or treatment orientations of clinicians managing LBP within the South African context. Given South Africa’s distinct healthcare structure, resource constraints and high musculoskeletal disease burden, locally derived evidence is necessary to support educational strategies and promote consistent, high-quality biopsychosocial care.

Effective management of LBP within a biopsychosocial framework requires that physiotherapists possess both a robust understanding of pain neurophysiology and the ability to apply risk stratification and biopsychosocial approaches in clinical practice. Translating this knowledge into routine care remains challenging because of barriers such as limited access to structured training, time constraints and the resource-intensive nature of traditional postgraduate education (Beneciuk & George 2015; Beneciuk et al. 2019). Online education offers a promising solution, providing scalable, flexible and cost-effective access to standardised, evidence-based content tailored to individual learning needs (Cook, Levinson & Garside 2010). Asynchronous formats guided by Cognitive Load Theory (CLT) enable complex material to be delivered in segmented, self-paced modules, reducing cognitive overload and enhancing retention (Ahmed 2024). Despite advances in contemporary pain science, a persistent knowledge-practice gap remains within physiotherapy, with evidence indicating that many clinicians continue to endorse inaccurate pain concepts and encounter significant barriers to implementing biopsychosocial care despite formal training (Smart 2023; Van Dijk et al. 2023; Wilson et al. 2025). Accordingly, our study investigated the effectiveness of a brief webinar series in enhancing physiotherapists’ pain-related knowledge and promoting a shift from predominantly biomedical orientations towards a patient-centred biopsychosocial model of care.

A total of 536 physiotherapists completed both the demographic questionnaire and the PABS-PT questionnaire prior to the educational intervention (Figure 1). Only 216 participants (40.3%) returned the completed pre-intervention NPQ, as this requirement was introduced midway through data collection. In contrast, participants were asked to complete the post-webinar NPQ from the start of data collection, resulting in a larger number of post-intervention NPQ responses. Following completion of the webinar series, 521 participants (97.2%) submitted both a completed PABS-PT and NPQ. Consequently, pre- and post-intervention comparisons were conducted using 521 paired PABS-PT datasets and 201 paired NPQ datasets.

Of the 536 participants, most were aged 31–40 years (36.8%), followed by < 30 years (27.1%) (Table 2). The majority were female (86.8%). Most participants obtained their undergraduate qualification between 2000 and 2009 (35.0%). The highest qualification held by most respondents was a bachelor’s degree (84.5%), with smaller proportions holding master’s (11.8%) or doctoral degrees (1.1%).

A high number of participants (66.8%) had undergone some form of post-graduate training where pain neuroscience was a component of the material taught. Common fields of clinical practice included musculoskeletal outpatient care (91.6%), sport (46.5%) and in-hospital settings (39.7%), with fewer working in neurology (20.0%), paediatrics (16.0%), public health (2.1%) or other areas (14.2%).

The results from the PABS-PT and NPQ revealed significant mean score changes in the PABS-PT and NPQ outcome measures (Figure 2 and Table 3).

Analysis of pre- and post-intervention scores revealed significant changes in physiotherapists’ attitudes, beliefs and pain knowledge. On the PABS-PT biomedical subscale, participants’ scores decreased by 8.8% from a mean of 53.74 ± 12.80 (95% CI: 52.53–54.74) before the intervention to 43.04 ± 11.63 (95% CI: 42.04–44.04) after the intervention, representing a mean difference of 10.60 ± 10.44 (95% CI: 9.70–11.50, t = 23.16, p < 0.001). A reduction in the biomedical subscale score reflects an improvement, indicating a shift away from a purely biomedical orientation towards a more contemporary, biopsychosocial approach. Conversely, the PABS-PT biopsychosocial subscale scores demonstrated a 7.6% increase from 35.52 ± 4.65 (95% CI: 35.14–35.94) to 39.62 ± 4.70 (95% CI: 39.22–40.03), with a mean difference of −4.09 ± 5.09 (95% CI: −4.53 to −3.65, t = −18.32, p < 0.001). An increase on this subscale reflects improvement, indicating stronger endorsement of biopsychosocial principles. Similarly, NPQ scores improved by 6.6% from 10.19 ± 1.78 (95% CI: 9.92–10.41) to 11.04 ± 1.64 (95% CI: 10.79–11.25), with a mean difference of −0.86 ± 1.65 (95% CI: −1.09 to −0.63, t = −7.35, p < 0.001), demonstrating a significant increase in pain neurophysiology knowledge.

Our study found a weak but significant negative relationship between changes in PABS-PT biomedical and biopsychosocial subscales scores (r = −0.2326, p = 0.0000). Significant improvements in biopsychosocial attitudes were positively associated with increases in NPQ scores (r = 0.1519, p = 0.0314), while reductions in biomedical attitudes were inversely correlated with NPQ improvements (r = −0.1215, p = 0.0766), suggesting that shifts in physiotherapists’ beliefs are closely linked to their understanding of pain neurophysiology.

The mean scores per item on the PABS-PT before and after the webinar are reflected in Figure 3. Before the intervention, physiotherapists’ responses on the PABS-PT reflected an apparent biomedical orientation towards LBP. Several biomedical subscale items demonstrated relatively high scores, indicating moderate to strong agreement with statements emphasising caution, activity limitation and the presence of tissue damage (e.g. ‘Reduction of daily physical exertion is a significant factor in treating back pain’; ‘Not enough effort is made to find the underlying organic causes of back pain’). Conversely, biopsychosocial subscale items generally yielded lower scores, reflecting limited endorsement of biopsychosocial perspectives, particularly regarding the influence of psychosocial factors and patient perceptions on pain and functional outcomes (e.g. ‘A patient suffering from severe back pain will benefit from physical exercise’; ‘Therapy may have been successful even if pain remains’).

Conversely, biopsychosocial subscale scores increased, indicating enhanced recognition of psychosocial, cognitive and behavioural contributions to pain. Notably, agreement with ‘Mental stress can cause back pain even in the absence of tissue damage’ increased from 5.20 ± 1.04 to 5.54 ± 0.96, and ‘The way patients view their pain influences the progress of the symptoms’ increased from 5.46 ± 0.85 to 5.67 ± 0.73. Internal consistency was high for the biomedical subscale, both pre- and post-intervention (Cronbach’s α = 0.865–0.879), whereas the biopsychosocial subscale demonstrated moderate reliability (Cronbach’s α = 0.524 pre-intervention; 0.540 post-intervention), consistent with prior validations of the PABS-PT.

Following the intervention, physiotherapists’ beliefs regarding back pain exhibited a discernible shift from predominantly biomedical orientations towards greater acknowledgement of biopsychosocial factors. On the biomedical subscale, mean scores decreased across many items, indicating a reduction in distinct biomedical perspectives. For instance, agreement with statements such as ‘Back pain sufferers should refrain from all physical activity to avoid injury’ declined from 1.53 ± 0.88 pre-intervention to 1.29 ± 0.78 post-intervention, while endorsement of ‘Pain is a nociceptive stimulus, indicating tissue damage’ decreased from 2.68 ± 1.50 to 1.88 ± 1.30.

The percentage of correct answers per item on the NPQ before and after the webinar are reflected in Figure 4.

At baseline, participants’ knowledge was limited for several items, notably Item 2 (‘When part of your body is injured, special pain receptors convey the pain message to your brain’), which was correctly answered by only 33.8% of participants. Moderate performance was observed for Items 1 (59.7%) and 8 (69.0%), indicating gaps in understanding of pain perception and neural mechanisms.

Following the intervention, correct responses on the NPQ increased across most items, with the largest improvement observed for those with low baseline scores. Item 2 improved from 33.8% to 53.9%, Item 8 from 69.0% to 85.0% and Item 11 from 77.3% to 92.5%. Items with high baseline scores (Items 12 and 13) exhibited minimal change, consistent with ceiling effects. These findings indicate that the intervention substantially enhanced participants’ understanding of neurophysiological pain mechanisms, particularly in areas of prior low knowledge. Item-level internal consistency, as measured by Cronbach’s alpha, ranged from 0.50 to 0.57 before the intervention and 0.51 to 0.58 after, with overall NPQ reliability increasing slightly from 0.559 to 0.567.

The findings of our study indicate that participation in a 5-h pain neuroscience webinar significantly enhanced participants’ understanding of pain mechanisms. Moreover, the intervention promoted a discernible shift in treatment orientation, moving from a predominantly biomedical approach towards a more comprehensive biopsychosocial model of care. These outcomes align with international evidence indicating that targeted educational interventions can effectively modify clinicians’ conceptualisation of pain and promote alignment with contemporary, evidence-based guidelines (Bareiss, Nare & McBee 2019; Leysen et al. 2021; Springer, Gleicher & Hababou 2018).

The integration of psychological and social elements within the assessment and management of LBP is fundamental to the biopsychosocial model of care, which has been associated with enhanced patient outcomes and reduced healthcare utilisation (Childs et al. 2015; Foster et al. 2018). Clinicians’ attitudes and beliefs impose a substantial influence on patients’ perceptions, recovery expectations, and functional outcomes (Darlow et al. 2012; Gardner et al. 2017). Practitioners exhibiting stronger biomedical orientations are more likely to advise activity restriction and utilise passive modalities that are incongruent with current guideline recommendations (Alshehri et al. 2020; Meroni et al. 2021). Conversely, those demonstrating biopsychosocial orientations are more inclined to employ active strategies such as graded exercise, patient education and cognitive-behavioural principles (Oliveira et al. 2018).

Despite substantive advances in conceptualising pain as a complex experience influenced by nervous system plasticity and biopsychosocial factors, a persistent knowledge–practice gap exists in physiotherapy.

Cross–sectional evidence indicates that while physiotherapists report high perceived competence in pain science education (PSE), a significant proportion endorse inaccurate pain science concepts, despite formal training, and report multiple implementation barriers, including patient expectations, time constraints and clinician–specific challenges in addressing maladaptive beliefs and behaviours (Wilson et al. 2025). Systematic reviews further demonstrate that although physiotherapists may be generally aware of biopsychosocial principles, knowledge, skills, attitudes, environmental context, confidence and patient expectations continue to limit effective integration of these approaches in chronic pain care (Van Dijk et al. 2023).

These challenges are exacerbated by systemic educational gaps that leave many graduates underprepared for complex pain management, as pain neuroscience and biopsychosocial frameworks are inconsistently integrated into curricula (Smart 2023). The evidence emphasises that improving pain management in physiotherapy requires the dissemination of contemporary pain science that aligns clinician beliefs and behaviours with best–practice biopsychosocial care.

Our study sample comprised 536 physiotherapists, representing a robust cohort relative to similar international investigations (Alhowimel et al. 2021; Mikamo & Takasaki 2021; Petit et al. 2019). While the sample size is a strength, the use of a convenience sampling strategy, a non-probability method, introduces inherent limitations. Participants were recruited based on their accessibility and voluntary election to attend the webinars (Andrade 2021). Although this approach is a pragmatic necessity in educational research, it carries an increased risk of selection bias. Clinicians who self-select into PNE may differ systematically from the broader professional population regarding baseline knowledge, motivation or existing professional interests (Andrade 2021). Consequently, the representativeness of this sample may be compromised, potentially limiting the external validity and generalisability of the findings to the wider physiotherapy community.

Baseline PABS-PT and NPQ scores indicated a relatively favourable starting point for this cohort, likely reflecting prior postgraduate exposure or musculoskeletal specialisation. However, the statistically significant post-intervention shifts suggest that structured online education can further enhance alignment with biopsychosocial management strategies, even among clinicians with moderate-to-high baseline knowledge.

This observation is consistent with literature demonstrating that focused interventions can influence practice orientation regardless of prior experience levels (Domenech et al. 2011; Overmeer et al. 2011; Rufa et al. 2022).

The magnitude of change observed in our study is consistent with outcomes reported in comparable international interventions. For instance, Saracoglu et al. (2021) reported a 6% – 7% increase in biopsychosocial orientation and concomitant reductions in biomedical beliefs among physiotherapists following a structured online module, despite relatively high baseline scores. Similarly, Bareiss et al. (2019) demonstrated significant improvements in PABS-PT and NPQ scores among physiotherapy students after short-format, interactive pain science workshops. Zimney et al. (2018) reported that brief online PNE interventions could yield significant improvements in clinician knowledge, attitudes and confidence in managing NSLBP. The knowledge gains observed in this study (6.6%) further support the efficacy of short-format interventions for practising clinicians (Cox, Louw & Puentedura 2017; Wassinger 2021). These results suggest that online education serves to reinforce and consolidate pre-existing knowledge, bridging the gap between undergraduate training and advanced clinical reasoning (Mikamo & Takasaki 2021; Fourré et al. 2022).

These convergent findings suggest that even short, well-structured educational interventions can produce meaningful cognitive and attitudinal change among practising clinicians, particularly when designed to engage learners actively and leverage principles such as Cognitive Load Theory (Ahmed 2024). A critical advantage of online education is its potential to overcome traditional barriers to widespread pain neuroscience training.

Conventional postgraduate courses are often resource-intensive, requiring travel, extended time commitments and high financial costs, which restricts participation and limits equitable access (Beneciuk & George 2015; Beneciuk et al. 2019). In contrast, online modules can be delivered flexibly, allowing self-paced learning without substantial incremental cost. This is particularly relevant in low- and middle-income settings, where access to formal postgraduate education may be limited, and for clinicians balancing clinical responsibilities with professional development. Additional advantages include accessibility for geographically dispersed clinicians, opportunities for repeated review and reinforcement, and the potential to integrate interactive elements, such as quizzes and case-based scenarios to strengthen application in clinical reasoning (Hung et al. 2024). Compared with synchronous approaches, asynchronous learning maintains comparable outcomes while increasing learner autonomy and engagement (Hung et al. 2024). By leveraging these advantages, online postgraduate pain education represents a feasible and effective strategy to strengthen physiotherapists’ biopsychosocial competencies, directly addressing the objectives of our study to evaluate the impact of an online, modular PNE intervention on physiotherapists’ knowledge, attitudes and beliefs in LBP management.

Nevertheless, challenges remain for large-scale implementation. Engagement and completion rates can vary, and ensuring interactivity and application to clinical reasoning is critical for translating knowledge gains into practice (Cook et al. 2010; Hung et al. 2024). Furthermore, sustained reinforcement, through repeated modules, mentorship or integrated clinical practice exercises, may be necessary to consolidate learning and support behavioural change. Cost-effectiveness analyses of online interventions remain sparse, but preliminary evidence suggests that the reduced logistical and personnel costs associated with asynchronous delivery could make online education a viable strategy for broad-scale dissemination of contemporary pain science (Ahmed 2024; Hung et al. 2024).

Correlation analyses revealed a weak negative association between biomedical and biopsychosocial subscales, suggesting that a reduction in biomedical beliefs was accompanied by a corresponding increase in biopsychosocial orientation. Moreover, a measurable positive correlation emerged between enhanced pain knowledge and biopsychosocial beliefs, aligning with previous findings indicating that improved pain knowledge and comprehension are linked to greater adherence to evidence-based, patient-centred care (Christe, Darlow & Pichonnaz 2021; Fitzgerald et al. 2018; Mikamo & Takasaki 2021). Collectively, these findings reinforce the premise that conceptual understanding of pain mechanisms is integral to the adoption of contemporary management paradigms for CLBP. Whether this conceptual shift leads to meaningful behavioural changes remains uncertain.

The positive effects observed in our study likely derive from pedagogical design, which engaged clinicians in the conceptual foundations of contemporary pain science. The inclusion of content addressing nervous system plasticity, cognitive and psychosocial contributors to pain, and the limitations of strictly biomedical approaches appears to have facilitated meaningful conceptual shifts. These mechanisms are in line with prior research indicating that comprehension of pain mechanisms is linked to stronger endorsement of biopsychosocial principles and evidence-based treatment behaviours (Christe et al. 2021; Fitzgerald et al. 2018).

Our study demonstrates that a brief online education programme can effectively enhance physiotherapists’ knowledge and foster a shift from biomedical to biopsychosocial treatment orientations for chronic LBP.

These findings highlight the modifiable nature of clinicians’ beliefs and underscore the potential of targeted, accessible education to promote evidence-based, person-centred physiotherapy practice. Broader implementation of such interventions may strengthen clinical reasoning, improve adherence to guidelines and ultimately optimise patient outcomes in chronic pain management.