Baseline predictors of treatment response to Chinese tuina and manual physical therapy in knee osteoarthritis: a secondary analysis of a randomized controlled trial

Abstract

Introduction Although manual therapy, including traditional Chinese tuina and manual physical therapy, is generally effective for knee osteoarthritis (KOA), individual responses vary considerably. This secondary analysis aimed to identify the quantifiable baseline characteristics that predict treatment response, thereby supporting personalized care. Methods Data from a randomized controlled trial comparing tuina and manual physical therapy for KOA were analyzed. At week 4, 59 of 127 patients (46.5%) were responders by Outcome Measures in Rheumatology-Osteoarthritis Research Society International (OMERACT-OARSI) criteria. Candidate variables were screened by univariate analysis (P < 0.25). Multivariate logistic regression identified independent predictors. Internal validation used 1000 bootstrap resamples, with sensitivity analyses including least absolute shrinkage and selection operator (LASSO) and stepwise regression based on Akaike information criterion or Bayesian information criterion. Results Univariate analysis identified disease duration, Kellgren-Lawrence grade, prior treatment satisfaction, all Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) subscales, numeric rating scale (NRS) pain score, 12-item short form health survey (SF-12) mental health score, functional tests, and treatment expectations as associated with response (P < 0.25). Multivariate logistic regression revealed two independent predictors: higher baseline WOMAC pain score (odds ratio [OR] = 1.56, 95% confidence interval [CI]: 1.31, 1.87, P < 0.001) and higher baseline SF-12 mental health score (OR = 1.06, 95% CI: 1.01, 1.12, P = 0.023). The model showed good discrimination (apparent area under the curve [AUC] = 0.838). Internal validation yielded an optimism-corrected AUC of 0.763 and Brier score of 0.169. Calibration and decision curve analyses indicated acceptable fit and net benefit. Conclusion Patients with more severe baseline pain and better mental health are more likely to respond favorably to either tuina or manual physical therapy. These findings bridge pain and mental health as predictors, providing evidence for precision manual therapy in KOA; external validation in larger cohorts is required.

Keywords

knee osteoarthritis / manual therapy / predictors / pain / mental health / logistic models

Acknowledgements

The authors thank all patients who participated in this study for their cooperation. We also acknowledge the tuina therapists and physical therapists for their professional delivery of the interventions, and the research assistants for their support in data collection and management.

Funding

This study was supported by National Natural Science Foundation of China (grant no.: 81803956) and Beijing Municipal Science & Technology Commission (grant no.: Z181100001718165).

Author Contribution Statement

Ying Wang: Software, formal analysis, writing - original draft, visualization, and writing - review & editing. Xi-you Wang: Methodology, visualization, funding acquisition, project administration, and writing - review & editing. Ya-nan Sun: Formal analysis and writing - review & editing. Chang-he Yu: Conceptualization, formal analysis, supervision, funding acquisition, project administration, and writing - review & editing.

Declaration of Competing Interest

The authors have no competing interests to declare that are relevant to the content of this article.

Data Availability

Additional data related to this paper may be obtained from the corresponding author upon reasonable request.

Use of AI Statement

None.

References

[1] Lv Y, Sui L, Lv H, Zheng J, Feng H, Jing F. Burden of knee osteoarthritis in China and globally from 1992 to 2021, and projections to 2030: a systematic analysis from the Global Burden of Disease Study 2021. Front Public Health. 2025; 13:1543180.

[2] Liu Z, Zhang Y, Liu J, et al. Effect of traditional Chinese non-pharmacological therapies on knee osteoarthritis: a narrative review of clinical application and mechanism. Orthop Res Rev. 2024; 16:21-33.

[3] Feng T, Wang X, Jin Z, et al. Effectiveness and safety of manual therapy for knee osteoarthritis: an overview of systematic reviews and meta-analyses. Front Public Health. 2023; 11:1081238.

[4] Zhu B, Ba H, Kong L, et al. The effects of manual therapy in pain and safety of patients with knee osteoarthritis: a systematic review and meta-analysis. Syst Rev. 2024; 13(1):91.

[5] Kolasinski SL, Neogi T, Hochberg MC, et al. 2019 American college of rheumatology/arthritis foundation guideline for the management of osteoarthritis of the hand, hip, and knee. Arthritis Rheumatol. 2020; 72(2):220-233.

[6] Arden NK, Perry TA, Bannuru RR, et al. Non-surgical management of knee osteoarthritis: comparison of ESCEO and OARSI 2019 guidelines. Nat Rev Rheumatol. 2021; 17(1):59-66.

[7] Ma P, Liu L, Li S, et al. Comparative effectiveness of tuina therapy versus manual physical therapy for knee osteoarthritis: a randomized controlled trial. BMC Complementary Med Ther. 2025; 25(1):128.

[8] Testa M, Rossettini G. Enhance placebo, avoid nocebo: how contextual factors affect physiotherapy outcomes. Man Ther. 2016; 24:65-74.

[9] Ezzatvar Y, Dueñas L, Balasch-Bernat M, Lluch-Girbés E, Rossettini G. Which portion of physiotherapy treatments’ effect is not attributable to the specific effects in people with musculoskeletal pain? A meta-analysis of randomized placebo-controlled trials. J Orthop Sports Phys Ther. 2024; 54(6):391-399.

[10] Ghesquiere A, Wyka K, Smith M, Kutner JS. Associations between psychological symptoms and treatment outcomes of a massage therapy intervention: secondary analyses of a randomized controlled trial. Complementary Ther Med. 2019; 46:116-122.

[11] Fang L, Fang M, Jiang S, Chen H. Optimization of parameters of Yi Zhi Chan tuina manipulation promotes peripheral circulation. J Tradit Chin Med. 2015; 35(5):558-563.

[12] Perlman A I, Ali A, Njike V Y, et al. Massage therapy for osteoarthritis of the knee: a randomized dose-finding trial. PLoS One. 2012; 7(2):e30248.

[13] Fan XR, Li DD, Wang SS, Guo Y, Yu CH. A qualitative study on the factors influencing the treatment of musculoskeletal pain by tuina manipulation. Chin Gen Pract. 2023; 26(2):219-224.

[14] Sun YN, Yu CH. Theoretical exploration of intervention component analysis (ICA) identifying the critical features of traditional Chinese medicine complex interventions-application of ICA on tuina treatment for knee osteoarthritis as an example. World Sci Technol Mod Tradit Chin Med Mater Med. 2023; 25(8):2851-2857.

[15] Zhou YJ, Wang SS, Wang XY, et al. Construction of an influencing factor framework for pain relief with tuina therapy based on Delphi method. Int J Trad Chin Med. 2023; 45(4):391-396.

[16] Chinese Medical Association. Chinese guideline for diagnosis and treatment of osteoarthritis (2021 edition). Chin J Orthop. 2021; 41(18):1291-1314.

[17] Altman R, Asch E, Bloch D, et al. Development of criteria for the classification and reporting of osteoarthritis: classification of osteoarthritis of the knee. Arthritis Rheum. 1986; 29(8):1039-1049.

[18] Kellgren JH, Lawrence JS. Radiological assessment of osteo-arthrosis. Ann Rheum Dis. 1957; 16(4):494-502.

[19] Lemeshow S, Sturdivant RX, Hosmer DW. Applied Logistic Regression. 3rd ed. Hoboken, USA: Wiley; 2013.

[20] Bursac Z, Gauss CH, Williams DK, Hosmer DW. Purposeful selection of variables in logistic regression. Source Code Biol Med. 2008; 3(1):17.

[21] van der Esch M, van der Leeden M, Roorda LD, Lems WF, Dekker J. Predictors of self-reported knee instability among patients with knee osteoarthritis: results of the Amsterdam osteoarthritis cohort. Clin Rheumatol. 2016; 35(12):3007-3013.

[22] Chen LM, Guo SJ, Xu GR, et al. Analysis of TCM syndrome characteristics in 28763 patients with knee osteoarthritis. Chin J Tradit Chin Med Pharm. 2022; 37(2):1061-1064.

[23] Yi ZP, Zhu ZS, Mao M, et al. Diagnosis and treatment strategy of treating both sinews and bones in knee osteoarthritis from the perspective of meridian sinew theory. Fujian J Tradit Chin Med. 2024; 55(1):29-31.

[24] Wang XY. Current status and challenges of clinical and mechanistic research on tuina for knee osteoarthritis. Beijing J Tradit Chin Med. 2025; 44(3):262-268.

[25] Liu ZF, Wang HR, Yu TY, Zhang YQ, Jiao Y, Wang XY. Tuina for peripherally-induced neuropathic pain: a review of analgesic mechanism. Front Neurosci. 2022; 16:1096734.

[26] Xu H, Zhou YF, Shi MW, et al. Clinical exploration of tuina in prevention and treatment of central sensitization in patients with knee osteoarthritis. Acta Chin Med. 2024; 39(2):286-289.

[27] Liu ZF, Yu CH, Wang XY, et al. Therapeutic effects and mechanisms of “tendons instead of bone” acupoint-pressing manipulation on knee osteoarthritis rats with chronic pain and depression. Beijing J Tradit Chin Med. 2025; 44(3):283-288.

[28] Wang XB, Hou MJ, Gou YY, Xu Y, Lin ZL, Tao J. Application of the rehabilitation principle of “gu zheng (bones integrity) and jin he (muscles harmony), xing shen gong tiao (union of body and mind)” in chronic musculoskeletal dysfunction. Rehabil Med. 2023; 33(1):14-23.

[29] Gao XY, Tian GP, Wu N, et al. Clinical study on the treatment of knee osteoarthritis with figure and spirit co-adjustment acupuncture method. Chin J Tradit Med Traumatol Orthop. 2025; 33(2):14-18.

[30] Zhang XM, Wang HD, Wang TT, et al. Acupotomy diagnosis and treatment for knee osteoarthritis based on the theory of form qi spirit. Rheumat Arthritis. 2022; 11(11):60-62.

[31] Yan Z, Wang Z, Liang Q, Liu J, Wu S, Luan X. Longitudinal trajectories of depressive symptoms among patients with knee osteoarthritis: the role of pain intensity. Pain Manag Nurs. 2023; 24(2):151-156.

[32] Yang W, Ma G, Li J, et al. Anxiety and depression as potential risk factors for limited pain management in patients with elderly knee osteoarthritis: a cross-lagged study. BMC Musculoskelet Disord. 2024; 25(1):995.

[33] Imamura M, Pacheco-Barrios K, de Melo PS, Marduy A, Battistella L, Fregni F. Maladaptive compensatory neural mechanisms associated with activity-related osteoarthritis pain: dissociation of psychological and activity-related neural mechanisms of WOMAC pain and VAS pain. J Clin Med. 2025; 14(11):3633.

[34] Wheeler CHB, de C Williams AC, Morley SJ. Meta-analysis of the psychometric properties of the pain catastrophizing scale and associations with participant characteristics. Pain. 2019; 160(9):1946-1953.

[35] Karasawa Y, Yamada K, Iseki M, et al. Association between change in self-efficacy and reduction in disability among patients with chronic pain. PLoS One. 2019; 14(4):e0215404.

[36] Yu CH, Lu HL, Shao H, et al. Social psychological mechanism, implementation difficulties, coping and strategies of the manual therapy placebo scheme design. China Med Herald. 2025; 22(15):186-190.

[37] Bin Kwon S, Ku Y, Han HS, Lee MC, Kim HC, Ro DH. A machine learning-based diagnostic model associated with knee osteoarthritis severity. Sci Rep. 2020; 10:15743.

[38] Atukorala I, Pathmeswaran A, Makovey J, et al. Can pain flares in knee osteoarthritis be predicted? Scand J Rheumatol. 2021; 50(3):198-205.

[39] Paz-González R, Balboa-Barreiro V, Lourido L, et al. Prognostic model to predict the incidence of radiographic knee osteoarthritis. Ann Rheum Dis. 2024; 83(5):661-668.

[40] Liu J, Jin H, Yon DK, et al. Risk factors for depression in patients with knee osteoarthritis: a systematic review and meta-analysis. Orthopedics. 2024; 47(5):e225-e232.

[41] Royuela A, Kovacs FM, Campillo C, Casamitjana M, Muriel A, Abraira V. Predicting outcomes of neuroreflexotherapy in patients with subacute or chronic neck or low back pain. Spine J. 2014; 14(8):1588-1600.

[42] Barroso J, Wakaizumi K, Reckziegel D, et al. Prognostics for pain in osteoarthritis: Do clinical measures predict pain after total joint replacement? PLoS One. 2020; 15(1):e0222370.

[43] Urquhart DM, Phyomaung PP, Dubowitz J, et al. Are cognitive and behavioural factors associated with knee pain? A systematic review. Semin Arthritis Rheum. 2015; 44(4):445-455.