PATIENT SELF-ASSESSMENT TOOL (BACK-ON-LINE™) FOR EARLY TARGETED SELF-MANAGEMENT OF LOW BACK PAIN IN THE WORKPLACE:CONSTRUCT AND DISCRIMINATE VALIDATION STUDY

M. Chen¹, V. Sparkes², L. Sheeran¹

¹Cardiff University, School of Healthcare Sciences, Cardiff, United Kingdom, ²Cardiff University, Biomechanics and Bioengineering Research Centre Versus Arthritis, Caridiff, United Kingdom

Background: The high prevalence of low back pain (LBP), combined with an unprecedented waiting time to access physiotherapy, has spurred research to seek a better assessment of pain subtypes to identify poor prognostic indicators for targeted treatment.

Purpose: This study objective was to validate a novel, online patient self-assessment tool (BACK-ON-LINE™; BOL), designed for the early identification of pain mechanisms (nociceptive [NC] and nociplastic [NP]) likely underlying a person’s LBP to target early self-management.

Methods: Between December 2019 and June 2021, 211 people with LBP were recruited from two work sectors (healthcare and transport), typically associated with a high prevalence of LBP, to complete an online BOL self-assessment with 4 subdomains (Pain behaviour, Pain perception, Impact of LBP on work and life). Demographics, pain intensity and duration, days of work absence, Rolland Morris Disability Questionnaire (RMDQ) and STarT Back were collected. The construct validity was determined using Spearman’s correlation test. The discriminant ability and predictive performance were analysed using BOL subgroup characteristics across reference standards (RMDQ≥7, VAS ≥7, Time off work ≥4 weeks, Pain Duration ≥6 months) with STarT Back as a comparator.

Results: 150 participants (71.1%) aged 20–61 years completed the survey. The proportions of patients allocated to NC and NP subgroups by BOL were 70% and 30%, respectively. There were significant differences between NC and NP subgroups in RMDQ, STarT Back scores and Patient Reported Outcome Measures (PROMs). Significant subgroup differences were also observed in BOL total and all subdomain scores. BOL total and most of the subdomain scores correlated positively with RMDQ and STarT Back except for the ‘Impact of LBP on work’ domain. The highest correlation coefficient of 0.66 was achieved with STarT Back indicating a good correlation with high statistical significance (p<0.0001). Receiver operating characteristic (ROC) curves indicated good discriminant ability and predictive performance of BOL subgroups across reference standards. BOL and STarT Back instruments discriminated similarly well across all selected reference standards including disability (RMDQ), absenteeism and pain intensity. When using pain duration as the reference standard, BOL scored better to STarT Back (Area under the curve (AUCs): 0.75 and 0.56, p<0.0001).

Conclusions: BOL self-assessment tool was successful in subgrouping the working population with LBP based on their pain type. The tool demonstrated high construct validity and discriminate validity equal to STarT Back across most of the reference standards and was better at discriminating patients with LBP over 6 months.

Implications: Our findings suggest BOL may contribute to the improvement of LBP assessment and potentially support early targeted self-management.

Funding acknowledgements: This study was co-funded by Cardiff University and China Scholarship Council (CSC).

Keywords:
Low back pain
Self-assessment
Validity

Topics:
Musculoskeletal

Did this work require ethics approval? Yes

Institution: Cardiff University

Committee: Health Research Authority and Health and Care Research Wales Ethics Committee

Ethics number: 19/HCRW/0035