Abstract
Conservative treatment for chronic non-specific low back pain (CLBP) includes lumbar extension traction (LET) to re-align lumbar lordosis (LL). This study explores the use of machine learning (ML) models to predict post-treatment outcomes in patients with CLBP undergoing LET and how these predictions can support clinical decision-making. We utilized a retrospective database of 431 consecutive patients with uncomplicated CLBP. Post-treatment variables predicted included LL, NRS pain score, and Oswestry Disability Index (ODI). Input model variables included pre-treatment LL, sacral base angle (SBA), ratio of LL/SBA fit type, NRS, ODI, frequency, duration, LET compliance, and demographic variables of age and BMI. Initial variables were analyzed to predict post-treatment outcomes. Three ML models—Random Forest (RF), XGBoost, and Multilayer Perceptron (MLP)—were employed to handle both continuous and categorical variables, and performance was evaluated for predictive accuracy. Factors affecting outcomes were identified using Shapley Additive Explanations. Treatment was a multimodal spine rehabilitation program featuring LET applied 3–6 times per week, varied between 4 and 10 weeks, and follow-up was performed at the end of care. Improvements in LL, NRS, and ODI were − 11.5° to − 23.6°, 7.3/10 to 3.3/10, and 33.2% to 10.4%, respectively. Among the ML models, XGBoost demonstrated the highest predictive accuracy for lumbar lordotic angle (R2 = 0.728) and pain score (R2 = 0.648), while Random Forest slightly outperformed XGBoost for ODI (0.631 vs. 0.616). MLP performed poorly for ODI predictions (R2 = 0.201), indicating difficulty in capturing functional disability patterns. SHAP analysis identified fit type, compliance, traction frequency, pre-treatment lumbar curve, and BMI as the most influential predictors. These predictors offer actionable insight for clinical decision-making by allowing clinicians to stratify patients based on predicted responsiveness, tailor LET frequency and duration, and educate patients on the importance of compliance. This study demonstrates that ML models, particularly XGBoost and Random Forest, can effectively predict LET outcomes, supporting personalized treatment strategies for CLBP patients.
Data availability
Data is available upon request from the corresponding author.
References
Wu, A. et al. Global low back pain prevalence and years lived with disability from 1990 to 2017: estimates from the global burden of disease study 2017. Ann. Transl. Med. 8 (6). (2020).
Foster, N. E. et al. Prevention and treatment of low back pain: evidence, challenges, and promising directions. Lancet 391 (10137), 2368–2383 (2018).
Ketenci, A. & Zure, M. Pharmacological and non-pharmacological treatment approaches to chronic lumbar back pain. Turkish J. Phys. Med. Rehabilitation. 67 (1), 1 (2021).
Guo, Y. et al. Bibliometric analysis of research on manual therapy for low back pain from 2013 to 2023. Med. (Baltim). 104 (8), e41618. https://doi.org/10.1097/MD.0000000000041618 (2025).
Hurwitz, E. L., Haldeman, S. & Cedraschi, C. The global spine care initiative: applying evidence-based guidelines on the non-invasive management of back and neck pain to low- and middle-income communities. Eur. Spine J. 27 (Suppl 6), 851–860. https://doi.org/10.1007/s00586-017-5433-8 (2018). Epub 2018 Feb 19.
WHO guideline for non-surgical management of chronic primary low back pain in adults in primary and community care settings. ISBN 978-92-4-008178-9 (electronic version). (Accessed 13 March 2025). https://www.who.int/publications/i/item/9789240081789 (2023).
Harrison, D. E., Cailliet, R., Harrison, D. D., Janik, T. J. & Holland, B. Changes in sagittal lumbar configuration with a new method of extension traction: nonrandomized clinical controlled trial. Arch. Phys. Med. Rehabil. 83 (11), 1585–1591. https://doi.org/10.1053/apmr.2002.35485 (2002).
Diab, A. A. & Moustafa, I. M. The efficacy of lumbar extension traction for sagittal alignment in mechanical low back pain: a randomized trial. J. Back Musculoskelet. Rehabil. 26 (2), 213–220. https://doi.org/10.3233/BMR-130372 (2013).
Diab, A. A. & Moustafa, I. M. Lumbar lordosis rehabilitation for pain and lumbar segmental motion in chronic mechanical low back pain: a randomized trial. J. Manipulative Physiol. Ther. 35 (4), 246–253. https://doi.org/10.1016/j.jmpt.2012.04.021 (2012).
Lee, C. H., Heo, S. J., Park, S. H., Jeong, H. S. & Kim, S. Y. Functional changes in patients and morphological changes in the lumbar intervertebral disc after applying Lordotic Curve-Controlled traction: A Double-Blind randomized controlled study. Med. (Kaunas). 56 (1), 4. https://doi.org/10.3390/medicina56010004 (2019).
Moustafa, I. M. & Diab, A. A. Extension traction treatment for patients with discogenic lumbosacral radiculopathy: a randomized controlled trial. Clin. Rehabil. 27 (1), 51–62. https://doi.org/10.1177/0269215512446093 (2013).
Chun, S. W., Lim, C. Y., Kim, K., Hwang, J. & Chung, S. G. The relationships between low back pain and lumbar lordosis: a systematic review and meta-analysis. Spine J. 17 (8), 1180–1191. https://doi.org/10.1016/j.spinee.2017.04.034 (2017).
Sadler, S. G., Spink, M. J., Ho, A., De Jonge, X. J. & Chuter, V. H. Restriction in lateral bending range of motion, lumbar lordosis, and hamstring flexibility predicts the development of low back pain: a systematic review of prospective cohort studies. BMC Musculoskelet. Disord. 18 (1), 179. https://doi.org/10.1186/s12891-017-1534-0 (2017).
Cardoso, L. et al. Computational modeling of posteroanterior lumbar traction by an automated massage bed: predicting intervertebral disc stresses and deformation. Front. Rehabil Sci. 3, 931274. https://doi.org/10.3389/fresc.2022.931274 (2022).
Lee, C. H., Heo, S. J. & Park, S. H. The real time geometric effect of a Lordotic Curve-Controlled spinal traction device: A randomized cross over study. Healthc. (Basel). 9 (2), 125. https://doi.org/10.3390/healthcare9020125 (2021).
de Andrada Pereira, B. et al. Influence of lumbar lordosis on posterior rod strain in long-segment construct during Biomechanical loading: a cadaveric study. Neurospine 18 (3), 635 (2021).
Moustafa, I. M. et al. Utilizing machine learning to predict post-treatment outcomes in chronic non-specific neck pain patients undergoing cervical extension traction. Sci. Rep. 14 (1), 11781 (2024).
Tangsrivimol, J. A. et al. Artificial intelligence in neurosurgery: a state-of-the-art review from past to future. Diagnostics (Basel) 13 (14), 2429. https://doi.org/10.3390/diagnostics13142429 (2023).
Javaid, M., Haleem, A., Pratap Singh, R., Suman, R. & Rab, S. Significance of machine learning in healthcare: Features, pillars and applications. Int. J. Intell. Networks. 3, 58–73. https://doi.org/10.1016/J.IJIN.2022.05.002 (2022).
Tschuggnall, M. et al. Machine learning approaches to predict rehabilitation success based on clinical and patient-reported outcome measures. Inf. Med. Unlocked 24 https://doi.org/10.1016/J.IMU.2021.100598 (2021).
Tagliaferri, S. D. et al. Artificial intelligence to improve back pain outcomes and lessons learnt from clinical classification approaches: three systematic reviews. Npj Digit. Med. 3, 1–16. https://doi.org/10.1038/s41746-020-0303-x (2020).
Shanahan, C., Ward, A. R. & Robertson, V. J. Comparison of the analgesic efficacy of interferential therapy and transcutaneous electrical nerve stimulation. Physiotherapy 92, 247–253. https://doi.org/10.1016/J.PHYSIO.2006.05.008 (2006).
Robertson, V., Ward, A., Low, J. & Reed, A. Electrotherapy explained, Principles and Practice 4th edn (Heinemann, 2006).
Kisner, C. & Colby, L. A. Therapeutic Exercise: Foundation and Techniques 5th edn (F.A.Davis Company, 2007).
Horton, W. C. et al. Is there an optimal patient stance for obtaining a lateral 36 radiograph? A critical comparison of three techniques. Spine (Phila Pa. 1976). 30 (4), 427–433. https://doi.org/10.1097/01.brs.0000153698.94091.f8 (2005).
Betz, J. W. et al. Reliability of the Biomechanical assessment of the sagittal lumbar spine and pelvis on radiographs used in clinical practice: A systematic review of the literature. J. Clin. Med. 13 (16), 4650. https://doi.org/10.3390/jcm13164650 (2024).
Harrison, D. E., Haas, J. W., Moustafa, I. M., Betz, J. W. & Oakley, P. A. Can the mismatch of measured pelvic morphology vs. Lumbar lordosis predict chronic low back pain patients? J. Clin. Med. 13 (8), 2178. https://doi.org/10.3390/jcm13082178 (2024).
Kobayashi, T. M. D., Atsuta, Y. M. D., Matsuno, T. M. D. & Takeda Naoki MD†. A longitudinal study of congruent sagittal spinal alignment in an adult cohort. Spine 29 (6), 671–676. https://doi.org/10.1097/01.BRS.0000115127.51758.A2 (2004).
Ostelo, R. W. et al. Interpreting change scores for pain and functional status in low back pain: towards international consensus regarding minimal important change. Spine (Phila Pa. 1976). 33 (1), 90–94. https://doi.org/10.1097/BRS.0b013e31815e3a10 (2008).
Fritz, J. M. & Irrgang, J. J. A comparison of a modified Oswestry low back pain disability questionnaire and the Quebec back pain disability scale. Phys. Ther. 81 (2), 776–788. https://doi.org/10.1093/ptj/81.2.776 (2001).
Raptis, S., Ilioudis, C. & Theodorou, K. From pixels to prognosis: unveiling radiomics models with SHAP and LIME for enhanced interpretability. Biomedical Phys. Eng. Express. 10 (3), 035016 (2024).
Hartvigsen, J. et al. Lancet low back pain series working Group. What low back pain is and why we need to pay attention. Lancet 391, 2356–2367 (2018).
Buchbinder, R. et al. Lancet low back pain series working Group. low back pain: a call for action. Lancet 391, 2384–2388 (2018).
Nuckols, T. K. et al. Rigorous development does not ensure that guidelines are acceptable to a panel of knowledgeable providers. J. Gen. Intern. Med. 23, 37–44 (2008).
Oakley, P. A., Ehsani, N. N., Moustafa, I. M. & Harrison, D. E. Restoring lumbar lordosis: a systematic review of controlled trials utilizing chiropractic bio Physics® (CBP®) non-surgical approach to increasing lumbar lordosis in the treatment of low back disorders. J. Phys. Ther. Sci. 32 (9), 601–610. https://doi.org/10.1589/jpts.32.601 (2020).
Haas, M., Vavrek, D., Peterson, D., Polissar, N. & Neradilek, M. B. Dose-response and efficacy of spinal manipulation for care of chronic low back pain: a randomized controlled trial. Spine J. 14 (7), 1106-16. https://doi.org/10.1016/j.spinee.2013.07.468 (2014).
Haas, M., Groupp, E. & Kraemer, D. F. Dose-response for chiropractic care of chronic low back pain. Spine J. 4 (5), 574–583. https://doi.org/10.1016/j.spinee.2004.02.008 (2004).
Legaye, J., Duval-Beaupère, G., Hecquet, J. & Marty, C. Pelvic incidence: a fundamental pelvic parameter for three-dimensional regulation of spinal sagittal curves. Eur. Spine J. 7 (2), 99–103. https://doi.org/10.1007/s005860050038 (1998).
Mendoza-Lattes, S., Ries, Z., Gao, Y. & Weinstein, S. L. Natural history of spinopelvic alignment differs from symptomatic deformity of the spine. Spine (Phila Pa. 1976). 35 (16), E792–E798. https://doi.org/10.1097/BRS.0b013e3181d35ca9 (2010).
Maiers, M. J., Albertson, A. K., Major, C., Mendenhall, H. & Petrie, C. P. The association between individual radiographic findings and improvement after chiropractic spinal manipulation and home exercise among older adults with back-related disability: a secondary analysis. Chiropr. Man. Th. 33 (1), 2. https://doi.org/10.1186/s12998-024-00566-9 (2025).
Haslam-Larmer, L. et al. Gleaning a lot from the history and physical exam, and reasonably confident without imaging: a qualitative study of primary care clinicians’ management of patients with low back pain. BMC Prim. Care. 26 (1), 26. https://doi.org/10.1186/s12875-025-02726-z (2025).
Williams, B., Gichard, L., Johnson, D. & Louis, M. An investigation into the chiropractic practice and communication of routine, repetitive radiographic imaging for the location of postural misalignments. J. Clin. Imaging Sci. 14, 28. https://doi.org/10.25259/JCIS_68_2024 (2024).
Noshchenko, A. et al. Spinopelvic parameters in asymptomatic subjects without spine disease and deformity: A systematic review with Meta-Analysis. Clin. Spine Surg. 30, 392–403 (2017).
Fujishiro, T. et al. European spine study Group, ESSG. Decision-making factors in the treatment of adult spinal deformity. Eur. Spine J. 27 (9), 2312–2321 (2018).
Banno, T. et al. The cohort study for the determination of reference values for spinopelvic parameters (T1 pelvic angle and global tilt) in elderly volunteers. Eur. Spine J. 25, 3687–3693 (2016).
Tominaga, R. et al. Dose-response relationship between spino-pelvic alignment determined by sagittal modifiers and back pain-specific quality of life. Eur. Spine J. 30 (10), 3019–3027 (2021). https://doi.org/10.1007/s00586-021-06965-3
Acknowledgements
Partial funding for this project was received from The NCMIC Foundation and CBP NonProfit for funding of open access fees if accepted for publication.
Author information
Authors and Affiliations
Contributions
Authors I.M.M., D.U.O., M.T.M., S.Z., I.K., P.A.O., and D.E.H. all participated in the research idea and participated in its design. I.M.M., D.U.O., M.T.M., and D.E.H. contributed to the statistical analysis. I.M.M., D.U.O., M.T.M., S.Z., I.K. participated in data collection and supervision. I.M.M., D.U.O., M.T.M., S.Z., I.K., P.A.O., and D.E.H. All contributed to the interpretation of the results and wrote the drafts. All authors have read and agreed to the published version of the manuscript.
Corresponding author
Ethics declarations
Competing interests
PAO is a paid consultant for CBP NonProfit, Inc. DEH teaches rehabilitation methods and is the CEO of a company that distributes spine rehabilitation equipment to physicians in the U.S.A. as used in this manuscript. All the other authors declare that they have no competing interests.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Moustafa, I.M., Ozsahin, D.U., Mustapha, M.T. et al. Machine learning models for predicting treatment outcomes in chronic non-specific back pain patients undergoing lumbar extension traction. Sci Rep (2026). https://doi.org/10.1038/s41598-026-38059-9
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-026-38059-9