The Impact of Warm Needle Acupuncture Based on Traditional Chinese Medicine on Psychological Well-Being and Quality of Life in Osteoporotic Patients: A Retrospective Cohort Study Authors: Yunhui Qiu, Xugui Li, Zhen Zeng, Yanping Meng, Xiao Pan, Zhengyang Wang, Zheng Lü, Weidan Xiao Published: 2026 ABSTRACT AbstractObjectiveOsteoporosis is associated with psychological distress and diminished quality of life, yet conventional treatments primarily target bone metabolism. The effects of warm needle acupuncture on psychological well-being and quality of life in patients with osteoporosis remain underexplored.MethodsThis retrospective cohort study included 120 patients with osteoporosis (warm needle acupuncture group, n = 63; control group, n = 57). Groups were formed according to the treatment regimen received in routine clinical practice, without randomization. The control group received conventional anti-osteoporosis therapy (calcium and vitamin D supplementation, anti-osteoporosis medications as clinically indicated, and lifestyle guidance); the warm needle acupuncture group received the same conventional therapy plus warm needle acupuncture (three sessions per week for 8 weeks). Outcomes included the Self-Rating Depression Scale (SDS), Self-Rating Anxiety Scale (SAS), 36-Item Short Form Health Survey (SF-36), Visual Analog Scale (VAS), and lumbar bone mineral density (BMD). Post-treatment scores differed significantly between groups in five of eight SF-36 domains (PF, RP, BP, SF, and MH; all P < 0.05, all remaining significant after false discovery rate correction), and improvements in change scores were significantly greater in the warm needle acupuncture group across all eight domains (all adjusted P < 0.05). In an exploratory multiple regression analysis, warm needle acupuncture was the only covariate independently associated with mental component summary improvement (B = 7.53, 95% CI: 5.73–9.33, P < 0.001).ResultsThe warm needle acupuncture group showed significantly greater improvements in SDS, SAS, and VAS scores (all P < 0.001). BMD remained unchanged in both groups over the short 8-week observation period. Depression reduction was more strongly correlated with quality-of-life improvement (rs = −0.415, P < 0.001) than anxiety reduction (rs = −0.225, P = 0.014).ConclusionIn this retrospective, non-randomized cohort study, warm needle acupuncture combined with conventional therapy was associated with greater short-term improvements in psychological symptoms, pain, and multiple quality of life domains in patients with osteoporosis; these findings represent associations rather than confirmed causal effects. Depression improvement showed a stronger association with quality-of-life gains than anxiety improvement. Prospective trials are needed to confirm these associations.Keywords: osteoporosis, acupuncture therapy, moxibustion, depression, anxiety, quality of life Introduction Osteoporosis is a prevalent metabolic bone disorder characterized by progressive deterioration of bone microarchitecture and reduced bone mineral density (BMD), leading to increased fracture susceptibility. The global osteoporosis prevalence varied considerably across countries, ranging from 4.1% to 52.0%, with higher rates observed in women and older populations.1 Data from the Global Burden of Disease Study 2021 further demonstrated that low BMD, as a modifiable risk factor, accounted for a substantial and increasing proportion of disability-adjusted life years globally, with East Asia bearing a particularly heavy burden.2 Beyond skeletal fragility, patients with osteoporosis frequently experience worse mental health and quality of life, including heightened pain and psychological distress, compared with those without the condition.3 Recent investigation of the brain-bone axis has elucidated the bidirectional epidemiological and pathophysiological links between depression and musculoskeletal disorders, with shared mechanisms centered on the neuroimmune-inflammatory axis, as well as autonomic and endocrine dysregulation.4 Psychological factors such as kinesiophobia have also been shown to be negatively correlated with BMD in hospitalized older adults with osteoporosis, underscoring the interplay between fear-avoidance behavior, inactivity, and bone loss.5 Despite these findings, conventional pharmacological treatments for osteoporosis predominantly target bone metabolism while neglecting the psychological and functional dimensions of the disease. Acupuncture, as a key therapeutic modality in traditional Chinese medicine (TCM), has attracted increasing attention for its potential in managing osteoporosis. A meta-analysis involving 2758 patients confirmed that acupuncture as an adjuvant therapy significantly enhanced BMD and reduced pain in osteoporotic patients.6 A network meta-analysis comparing various acupuncture-related therapies for postmenopausal osteoporosis further demonstrated the favorable effects of combined needle-moxibustion approaches on clinical outcomes.7 Bibliometric analysis of research trends has also revealed that pain management, bone metabolism, and quality of life represent emerging hotspots in the acupuncture-osteoporosis field.8 Warm needle acupuncture, which integrates needle insertion with moxibustion combustion at the needle handle, delivers both mechanical stimulation and thermal effects simultaneously. Animal experiments have demonstrated that warm needle acupuncture modulates pain thresholds through regulation of analgesic-related neurotransmitters and protects tissue integrity via the Hippo-YAP signaling pathway.9,10 Regarding psychological regulation, acupuncture has been shown to elevate 5-hydroxytryptamine levels and improve clinical outcomes in patients with depression, while electroacupuncture ameliorates anxiety and depressive symptoms by modulating neurotransmitter pathways and the complement system.11,12 Moxibustion-based therapies have also been reported to regulate the hypothalamic-pituitary-adrenal axis in kidney-yang deficiency models, providing a TCM-specific mechanistic basis.13 From a theoretical perspective, warm needle acupuncture provides simultaneous mechanical and thermal stimulation, and the warming action of moxibustion is considered in traditional Chinese medicine theory to warm the meridians and tonify yang deficiency, a common pattern in older patients with osteoporosis; it may therefore offer incremental benefits over manual acupuncture or supportive care alone. However, most existing clinical studies on acupuncture for osteoporosis have focused primarily on BMD and pain outcomes, with limited attention to the impact on psychological well-being and quality of life. Moreover, symptom relief, quality-of-life improvement, and skeletal outcomes represent distinct therapeutic dimensions that may respond to treatment over different time scales, and few studies have evaluated them simultaneously. The present study aimed to evaluate the effects of warm needle acupuncture based on TCM theory on anxiety, depression, quality of life, and pain in patients with osteoporosis, thereby providing clinical evidence for the role of this integrative approach in the holistic management of osteoporosis. Materials and Methods Study Design This retrospective cohort study was conducted at The Affiliated Hospital of Wuhan Sports University. Patients with osteoporosis who received treatment between September 2022 and September 2024 were screened for eligibility. Group allocation was not randomized; it reflected routine clinical decision-making, in which warm needle acupuncture was offered as an optional add-on therapy and the final regimen was determined jointly by the treating physician’s recommendation and informed patient preference, including consideration of treatment cost and the required visit frequency. This study is reported in accordance with the STROBE guidelines for observational studies, and a completed STROBE checklist is provided as Supplementary Material. The study protocol was approved by the institutional ethics committee of the hospital (Approval Number: 672HREC20260330-L028), and the requirement for individual informed consent was waived due to the retrospective nature of the study. All procedures were performed in accordance with the ethical standards of the Declaration of Helsinki. Study Population The diagnosis of primary osteoporosis was established according to Chinese clinical guidelines.14 Osteoporosis was diagnosed when dual-energy X-ray absorptiometry (DXA) measurement at the lumbar spine (L1–L4), femoral neck, total hip, or distal one-third of the radius yielded a T-score ≤ −2.5. Patients who had sustained a hip or vertebral fragility fracture were also diagnosed with osteoporosis regardless of BMD values. Patients with a T-score between −2.5 and −1.0 who had experienced fragility fractures of the proximal humerus, pelvis, or distal forearm also met the diagnostic criteria. The inclusion criteria were as follows: (1) a confirmed diagnosis of primary osteoporosis according to the above criteria; (2) aged 50 years or older; (3) completion of at least 8 weeks of treatment; and (4) availability of complete pre- and post-treatment clinical data. The exclusion criteria were: (1) secondary osteoporosis caused by endocrine, metabolic, or other identifiable conditions; (2) severe cardiac, hepatic, or renal insufficiency; (3) pre-existing diagnosed psychiatric disorders; (4) history of acupuncture therapy within 6 months prior to treatment initiation; (5) concurrent use of glucocorticoids or other medications known to affect bone metabolism; (6) malignant tumors or severe systemic diseases; and (7) incomplete medical records. Treatment Groups Chinese clinical guidelinesBased on the treatment regimen received during the study period, patients were divided into two groups. The control group (n = 57) received standard anti-osteoporosis therapy in accordance with,14 including elemental calcium with a recommended total daily intake of 1000–1200 mg (supplemental dose of 600 mg/day), vitamin D supplementation (800 IU/day), and anti-osteoporosis medications as clinically indicated (primarily oral bisphosphonates such as alendronate sodium, with calcitonin or active vitamin D analogues reserved for patients with contraindications or intolerance), along with lifestyle guidance on weight-bearing exercise, fall prevention, and nutrition. Using standard needling techniques, inserted perpendicularly or slightly obliquely to a depth of 0.8–1.5 cun (approximately 20–40 mm) depending on the acupoint, until de qi sensation was achieved. A 2 cm segment of a standardized moxa roll (approximately 18 mm in diameter) was then attachedThe warm needle acupuncture group (n = 63) received the same therapy supplemented with additional warm needle acupuncture based on TCM theory. The acupoint prescription included Shenshu (BL23), Mingmen (GV4), Zusanli (ST36), Guanyuan (CV4), Dazhu (BL11), and Xuanzhong (GB39).15 After routine skin disinfection, filiform needles (0.30 mm × 40 mm) were inserted at the selected acupoints to the needle handle and ignited, allowing the thermal stimulation to transmit through the needle to the acupoint. Each session lasted approximately 30 min, and treatment was administered 3 times per week for a total duration of 8 weeks (24 sessions). All treatments were delivered by licensed acupuncturists with more than five years of clinical experience who had been trained on a standardized operating protocol, and the same practitioner treated each patient whenever feasible. Outcome Measures The primary outcome measures included psychological status and health-related quality of life. Depressive symptoms were assessed using the Self-Rating Depression Scale (SDS), a 20-item self-report questionnaire with raw scores ranging from 20 to 80, where higher scores indicate more severe depressive symptoms.16 Anxiety symptoms were evaluated using the Zung Self-Rating Anxiety Scale (SAS), which adopts a similar structure and scoring methodology, with higher scores reflecting greater anxiety severity.17 Health-related quality of life was measured using the 36-Item Short Form Health Survey (SF-36), which assesses 8 domains: physical functioning (PF), role-physical (RP), bodily pain (BP), general health (GH), vitality (VT), social functioning (SF), role-emotional (RE), and mental health (MH), each scored from 0 to 100 with higher scores indicating better quality of life. Two component summary scores were further derived: the Physical Component Summary (PCS) and the Mental Component Summary (MCS), approximated as the unweighted mean of the four mental health–related domain scores (VT, SF, RE, and MH) rather than computed using population-specific norm-based algorithms; analyses involving MCS were therefore considered exploratory.18 The secondary outcome measures were pain intensity and BMD. Pain was assessed using the Visual Analog Scale (VAS), in which patients rate their pain intensity on a 100 mm horizontal line, with 0 mm representing no pain and 100 mm representing the worst imaginable pain.19 Lumbar spine BMD (L1–L4) was measured by DXA.20 Because an 8-week interval is shorter than the recommended dual-energy X-ray absorptiometry monitoring interval of one to two years,21 BMD was assessed to document skeletal stability over the treatment period rather than to detect treatment-induced changes in bone mass. All outcome measures were assessed at two time points: at baseline before treatment initiation and at the end of the 8-week treatment period. Statistical Analysis Baseline between-group comparability of all pre-treatment outcome scores was formally tested using the same distributional approach. Between-group comparisons were additionally performed on change scores (post-treatment minus pre-treatment), and effect sizes were calculated as Hedges’ g for parametric comparisons and rank-biserial correlation (r) for non-parametric comparisons. To account for multiple comparisons across outcomes, P values from between-group comparisons were adjusted using the Benjamini–Hochberg false discovery rate procedure. Model fit was assessed using the coefficient of determination (R2), adjusted R2, and the overall F-test; multicollinearity was evaluated using variance inflation factors; and residual normality was examined using the Shapiro–Wilk test. Statistical analyses were performed using SPSS version 26.0 (IBM Corp., Armonk, NY, USA). Continuous variables were assessed for normality using the Shapiro–Wilk test. Normally distributed data were expressed as mean ± standard deviation, and non-normally distributed data were expressed as median (interquartile range). Between-group comparisons of continuous variables were conducted using the independent samples t-test for normally distributed data and the Mann–Whitney U-test for non-normally distributed data. Within-group comparisons between pre- and post-treatment values were performed using paired t-tests for normally distributed data and Wilcoxon signed-rank tests for non-normally distributed data. Categorical variables were presented as frequencies and percentages and compared using the chi-square test. To identify independent factors associated with improvement in quality of life, multiple linear regression analysis was performed with the change in SF-36 MCS as the dependent variable, incorporating age, sex, disease duration, baseline VAS score, baseline SDS score, baseline SAS score, and treatment group as covariates. Pearson correlation analysis was used for normally distributed variables and Spearman rank correlation analysis for non-normally distributed variables to examine the relationship between changes in psychological scores (SDS and SAS) and changes in SF-36 MCS. A two-tailed P value of less than 0.05 was considered statistically significant. No a priori sample size calculation was performed, because the study included all eligible patients treated during the study period. A post-hoc power analysis based on the primary psychological outcome (SDS change score) indicated that the sample of 63 and 57 patients per group provided greater than 99% power to detect the observed between-group difference (Hedges’ g = 1.14) at a two-sided α of 0.05. Results Patient Enrollment and Baseline Characteristics A total of 187 patients with osteoporosis were initially screened from the hospital records. After applying the inclusion and exclusion criteria, 67 patients were excluded, and 120 patients were enrolled in the final analysis, including 63 in the WNA group and 57 in the control group (Figure 1). Baseline demographic and clinical characteristics are summarized in Table 1. No statistically significant differences were observed between the two groups in age, sex, BMI, disease duration, osteoporosis type, education level, fragility fracture history, comorbidities, smoking status, alcohol consumption, or menopause duration (all P > 0.05), Formal between-group testing of pre-treatment outcome scores showed no significant differences in SDS (P = 0.058), SAS (P = 0.659), VAS (P = 0.592), lumbar BMD (P = 0.117), or seven of the eight SF-36 domains (all P > 0.05). However, baseline social functioning scores were significantly higher in the WNA group than in the control group (57.78 ± 13.49 vs 50.84 ± 18.29, P = 0.019); this imbalance was addressed through change-score comparisons, as described below (Table 2 and Table 3). Flow diagram of the study. Table 1Baseline Demographic and Clinical Characteristics of the Two GroupsVariableWarm Needle Acupuncture Group (n = 63)Control Group (n = 57)P ValueAge (years)63.17 ± 7.3663.91 ± 7.060.577aSexMale25 (39.68%)31 (54.39%)0.153bFemale38 (60.32%)26 (45.61%)BMI (kg/m2)21.60 (20.00–23.85)22.00 (20.10–24.00)0.971cDisease duration (years)3.20 (1.70–5.70)3.40 (1.50–4.80)0.528 cOP typePostmenopausal25 (39.68%)21 (36.84%)0.895bSenile38 (60.32%)36 (63.16%)Education levelJunior high school or below18 (28.57%)20 (35.09%)0.720bSenior high school23 (36.51%)20 (35.09%)College or above22 (34.92%)17 (29.82%)Fragility fracture historyYes20 (31.75%)18 (31.58%)1.000bNo43 (68.25%)39 (68.42%)ComorbiditiesHypertension23 (36.51%)17 (29.82%)0.561bDiabetes mellitus11 (17.46%)14 (24.56%)0.465bCoronary heart disease9 (14.29%)8 (14.04%)1.000bSmokingYes12 (19.05%)14 (24.56%)0.610bNo51 (80.95%)43 (75.44%)Alcohol consumptionYes10 (15.87%)10 (17.54%)1.000bNo53 (84.13%)47 (82.46%)Menopause duration (year) †12.39 ± 7.5712.27 ± 6.440.953aNotes: Data are presented as mean ± SD or median (IQR) for continuous variables, and n (%) for categorical variables. †Postmenopausal women only (WNA group n = 25, control group n = 21). a independent samples t-test; b chi-square test; c Mann–Whitney U-test.Abbreviations: WNA, warm needle acupuncture; BMI, body mass index; OP, osteoporosis; DM, diabetes mellitus; CHD, coronary heart disease; SD, standard deviation; IQR, interquartile range. Baseline Demographic and Clinical Characteristics of the Two Groups Notes: Data are presented as mean ± SD or median (IQR) for continuous variables, and n (%) for categorical variables. †Postmenopausal women only (WNA group n = 25, control group n = 21). a independent samples t-test; b chi-square test; c Mann–Whitney U-test. Abbreviations: WNA, warm needle acupuncture; BMI, body mass index; OP, osteoporosis; DM, diabetes mellitus; CHD, coronary heart disease; SD, standard deviation; IQR, interquartile range. Table 2Comparison of SF-36 Domain Scores and Change Scores Between the Two GroupsSF-36 DomainWNA Group (n = 63)Control Group (n = 57)P (Between)P (Baseline)Change Score (Post − Pre)PrePostP (Within)PrePostP (Within)WNAControlP (Between)PF56.43 ± 16.2169.21 ± 18.59<0.001a54.11 ± 13.6858.42 ± 14.04<0.001a<0.001b0.401b11.00 (7.00–18.00)4.00 (0.00–8.00)<0.001eRP48.13 ± 17.5762.49 ± 18.22<0.001a49.26 ± 19.6354.46 ± 18.47<0.001a0.018b0.739b14.37 ± 9.385.19 ± 6.51<0.001fBP47.67 ± 16.9460.95 ± 18.05<0.001a45.51 ± 17.3451.47 ± 16.74<0.001a0.004b0.492b13.29 ± 9.155.96 ± 7.60<0.001fGH52.59 ± 15.4760.52 ± 16.20<0.001a54.04 ± 16.2156.14 ± 16.540.002a0.145b0.618b7.00 (3.00–11.50)1.00 (−1.00–5.00)<0.001eVT49.05 ± 15.3059.00 (47.50–70.00)<0.001c51.09 ± 15.6955.04 ± 16.80<0.001a0.163d0.473b11.60 ± 8.033.95 ± 5.49<0.001fSF57.78 ± 13.4969.75 ± 16.59<0.001a50.84 ± 18.2955.86 ± 19.21<0.001a<0.001b0.019b11.00 (6.00–18.50)3.00 (−1.00–10.00)<0.001eRE42.67 ± 18.4955.02 ± 21.21<0.001a45.09 ± 21.9651.07 ± 24.59<0.001a0.347b0.514b12.35 ± 10.345.98 ± 9.07<0.001fMH50.11 ± 14.5962.76 ± 15.29<0.001a50.18 ± 14.9154.54 ± 16.02<0.001a0.005b0.981b12.65 ± 7.864.37 ± 6.48<0.001fNotes: Data are presented as mean ± SD or median (IQR). apaired t-test; bindependent samples t-test; cWilcoxon signed-rank test; dMann–Whitney U-test. eMann–Whitney U-test on change scores; findependent samples t-test on change scores. Between-group comparisons of post-treatment scores remained significant after Benjamini–Hochberg false discovery rate correction (adjusted q: PF = 0.002, RP = 0.027, BP = 0.009, SF < 0.001, MH = 0.010; GH, VT, and RE non-significant), and all change-score comparisons also remained significant after correction (all q < 0.05).Abbreviations: WNA, warm needle acupuncture; SF-36, 36-Item Short Form Health Survey; PF, physical functioning; RP, role-physical; BP, bodily pain; GH, general health; VT, vitality; SF, social functioning; RE, role-emotional; MH, mental health; SD, standard deviation; IQR, interquartile range; Pre, pre-treatment; Post, post-treatment. Comparison of SF-36 Domain Scores and Change Scores Between the Two Groups Notes: Data are presented as mean ± SD or median (IQR). apaired t-test; bindependent samples t-test; cWilcoxon signed-rank test; dMann–Whitney U-test. eMann–Whitney U-test on change scores; findependent samples t-test on change scores. Between-group comparisons of post-treatment scores remained significant after Benjamini–Hochberg false discovery rate correction (adjusted q: PF = 0.002, RP = 0.027, BP = 0.009, SF < 0.001, MH = 0.010; GH, VT, and RE non-significant), and all change-score comparisons also remained significant after correction (all q < 0.05). Abbreviations: WNA, warm needle acupuncture; SF-36, 36-Item Short Form Health Survey; PF, physical functioning; RP, role-physical; BP, bodily pain; GH, general health; VT, vitality; SF, social functioning; RE, role-emotional; MH, mental health; SD, standard deviation; IQR, interquartile range; Pre, pre-treatment; Post, post-treatment. Table 3Comparison of VAS, Lumbar BMD, and Their Change Scores Between the Two GroupsVariableWNA Group (n = 63)Control Group (n = 57)P (Between)P (Baseline)Change Score (Post − Pre)PrePostP (Within)PrePostP (Within)WNAControlP (Between)VAS (mm)59.06 ± 13.9537.89 ± 15.51<0.001a57.77 ± 12.2248.60 ± 13.43<0.001a<0.001b0.592b−21.17 ± 9.26−9.18 ± 7.08<0.001fLumbar BMD (g/cm2)0.703 (0.663–0.738)0.697 ± 0.0640.685c0.677 ± 0.0700.678 ± 0.0700.727a0.118d0.117d+0.000 ± 0.010+0.001 ± 0.0130.905fNotes: Data are presented as mean ± SD or median (IQR). apaired t-test; bindependent samples t-test; cWilcoxon signed-rank test. dMann–Whitney U-test; eMann–Whitney U-test on change scores; findependent samples t-test on change scores. The mean change in lumbar BMD was +0.000 ± 0.010 g/cm2 (95% CI: −0.002 to 0.003) in the WNA group and +0.001 ± 0.013 g/cm2 (95% CI: −0.003 to 0.004) in the control group, confirming the expected absence of meaningful short-term BMD change.Abbreviations: WNA, warm needle acupuncture; VAS, Visual Analog Scale; BMD, bone mineral density; SD, standard deviation; IQR, interquartile range; Pre, pre-treatment; Post, post-treatment. Comparison of VAS, Lumbar BMD, and Their Change Scores Between the Two Groups Notes: Data are presented as mean ± SD or median (IQR). apaired t-test; bindependent samples t-test; cWilcoxon signed-rank test. dMann–Whitney U-test; eMann–Whitney U-test on change scores; findependent samples t-test on change scores. The mean change in lumbar BMD was +0.000 ± 0.010 g/cm2 (95% CI: −0.002 to 0.003) in the WNA group and +0.001 ± 0.013 g/cm2 (95% CI: −0.003 to 0.004) in the control group, confirming the expected absence of meaningful short-term BMD change. Abbreviations: WNA, warm needle acupuncture; VAS, Visual Analog Scale; BMD, bone mineral density; SD, standard deviation; IQR, interquartile range; Pre, pre-treatment; Post, post-treatment. Psychological Outcomes The reduction in SDS scores was significantly greater in the WNA group than in the control group (−10.56 ± 5.96 vs −4.40 ± 4.65; between-group comparison of change scores, P < 0.001, FDR-adjusted q < 0.001; Hedges’ g = 1.14) (Figure 2A). Changes in SDS and SAS scores before and after treatment are presented in Figure 2. Both groups showed significant reductions in SDS scores after 8 weeks of treatment (both P < 0.001 for within-group comparisons). A similar pattern was observed for SAS scores: both groups improved significantly from baseline (both P < 0.001), with the WNA group showing a significantly greater improvement than the control group in change scores (P < 0.001, FDR-adjusted q < 0.001; rank-biserial r = 0.52) and significantly lower post-treatment scores (P = 0.011) (Figure 2B). SDS and SAS scores before and after treatment. (A) SDS scores. (B) SAS scores. Error bars indicate SEM. *P < 0.05, ***P < 0.001. Abbreviations: SDS, Self-Rating Depression Scale; SAS, Self-Rating Anxiety Scale; WNA, warm needle acupuncture; SEM, standard error of the mean. Health-Related Quality of Life SF-36 domain scores before and after treatment are presented in Table 2. Both groups showed significant within-group improvements across all eight domains after treatment (all P < 0.05). For between-group comparisons of post-treatment scores, the WNA group scored significantly higher than the control group in five of the eight domains: PF, RP, BP, SF, and MH (all P < 0.05). No statistically significant between-group differences were observed for GH (P = 0.145), VT (P = 0.163), or RE (P = 0.347). All five significant between-group differences in post-treatment scores remained significant after Benjamini–Hochberg correction (adjusted q: PF < 0.01, RP = 0.027, BP < 0.01, SF < 0.001, MH = 0.010). Between-group comparisons of change scores showed significantly greater improvements in the WNA group across all eight domains (all P < 0.001; all FDR-adjusted q < 0.05), with large effect sizes for RP (Hedges’ g = 1.12), VT (g = 1.10), and MH (g = 1.14), and moderate-to-large effects for PF (rank-biserial r = 0.61), BP (g = 0.86), GH (r = 0.55), SF (r = 0.45), and RE (g = 0.65) (Table 2). Pain and Bone Mineral Density And the WNA group had significantly lower post-treatment VAS scores than the control group (P < 0.001); the reduction in VAS scores was significantly greater in the WNA group (−21.17 ± 9.26 mm vs −9.18 ± 7.08 mm; P < 0.001, FDR-adjusted q < 0.001; Hedges’ g = 1.44). VAS scores and lumbar BMD values before and after treatment are presented in Table 3. Both groups showed significant reductions in VAS scores after treatment (both P < 0.001 for within-group comparisons), In contrast, no significant changes in lumbar BMD were observed within either group or between the two groups after the 8-week treatment period (all P > 0.05); the mean change in BMD was negligible in both groups (WNA: +0.000 ± 0.010 g/cm2, 95% CI: −0.002 to 0.003; control: +0.001 ± 0.013 g/cm2, 95% CI: −0.003 to 0.004; between-group P = 0.905), consistent with the expected stability of DXA-measured BMD over short intervals. Factors Associated with Quality of Life Improvement Multiple linear regression analysis was performed to identify factors independently associated with improvement in SF-36 MCS (Figure 3 and Table 4). The model explained 39.4% of the variance in MCS change (R2 = 0.394, adjusted R2 = 0.356; F(7112) = 10.39, P < 0.001), with no evidence of multicollinearity (all variance inflation factors ≤ 1.31) and normally distributed residuals (Shapiro–Wilk P = 0.413). Among the covariates examined, WNA treatment was the only covariate independently associated with MCS improvement (B = 7.53, 95% CI: 5.73–9.33, P < 0.001); full regression results are presented in Table 4. Given the retrospective design and the approximated MCS scoring, this analysis should be considered exploratory. Age, sex, disease duration, baseline VAS score, baseline SDS score, and baseline SAS score were not significantly associated with MCS change (all P > 0.05). Forest plot of multiple linear regression analysis for factors associated with SF-36 MCS improvement. The dependent variable was the change in SF-36 MCS score. Dots represent regression coefficients and horizontal lines represent 95% confidence intervals. Blue indicates statistically significant factors (P < 0.05) and gray indicates non-significant factors. Abbreviations: SF-36, 36-Item Short Form Health Survey; MCS, Mental Component Summary; VAS, Visual Analog Scale; SDS, Self-Rating Depression Scale; SAS, Self-Rating Anxiety Scale; WNA, warm needle acupuncture; CI, confidence interval. Table 4Multiple Linear Regression Analysis of Factors Associated with Improvement in SF-36 MCSVariableBSEtP value95% CIVIFIntercept0.945.890.160.874−10.73 to 12.61—Age (years)0.030.060.510.609−0.09 to 0.161.02Sex−0.840.92−0.920.361−2.66 to 0.981.06Disease duration (years)−0.200.13−1.460.148−0.46 to 0.071.07Baseline VAS score0.040.041.240.219−0.03 to 0.111.03Baseline SDS score0.020.050.460.650−0.08 to 0.131.30Baseline SAS score−0.000.04−0.030.976−0.09 to 0.091.25WNA treatment7.530.928.20<0.0015.73 to 9.331.07Notes: The dependent variable was the change in SF-36 MCS score (post-treatment minus pre-treatment); MCS was approximated as the unweighted mean of the VT, SF, RE, and MH domain scores. Model fit: R2 = 0.394, adjusted R2 = 0.356, F(7112) = 10.39, P < 0.001; residuals were normally distributed (Shapiro–Wilk P = 0.413).Abbreviations: B, unstandardized regression coefficient; SE, standard error; CI, confidence interval; VIF, variance inflation factor; VAS, Visual Analog Scale; SDS, Self-Rating Depression Scale; SAS, Self-Rating Anxiety Scale; WNA, warm needle acupuncture; SF-36, 36-Item Short Form Health Survey; MCS, Mental Component Summary. Multiple Linear Regression Analysis of Factors Associated with Improvement in SF-36 MCS Notes: The dependent variable was the change in SF-36 MCS score (post-treatment minus pre-treatment); MCS was approximated as the unweighted mean of the VT, SF, RE, and MH domain scores. Model fit: R2 = 0.394, adjusted R2 = 0.356, F(7112) = 10.39, P < 0.001; residuals were normally distributed (Shapiro–Wilk P = 0.413). Abbreviations: B, unstandardized regression coefficient; SE, standard error; CI, confidence interval; VIF, variance inflation factor; VAS, Visual Analog Scale; SDS, Self-Rating Depression Scale; SAS, Self-Rating Anxiety Scale; WNA, warm needle acupuncture; SF-36, 36-Item Short Form Health Survey; MCS, Mental Component Summary. Correlation Between Psychological Improvement and Quality of Life Improvement Spearman rank correlation analysis revealed a significant negative correlation between ΔSDS and ΔSF-36 MCS (rs = −0.415, P < 0.001) (Figure 4A), indicating that greater reductions in depressive symptoms were associated with greater improvements in mental health-related quality of life. A weaker but statistically significant negative correlation was also observed between ΔSAS and ΔSF-36 MCS (rs = −0.225, P = 0.014) (Figure 4B). Scatter plots showing the correlation between changes in psychological scores and changes in SF-36 MCS. (A) Correlation between ΔSDS and ΔSF-36 MCS. (B) Correlation between ΔSAS and ΔSF-36 MCS. Circles represent individual patients in the WNA group and squares represent patients in the control group. The solid line represents the overall linear regression fit. Spearman correlation coefficients (r) and P values are shown. Negative correlations indicate that greater reductions in depression/anxiety scores were associated with greater improvements in mental health-related quality of life. Δ, change from baseline. Abbreviations: SDS, Self-Rating Depression Scale; SAS, Self-Rating Anxiety Scale; SF-36, 36-Item Short Form Health Survey; MCS, Mental Component Summary; WNA, warm needle acupuncture. Discussion In this retrospective cohort study, warm needle acupuncture combined with conventional anti-osteoporosis therapy was associated with greater improvements in depressive symptoms, anxiety symptoms, pain intensity, and five of the eight SF-36 domains compared with conventional therapy alone. BMD remained unchanged over the 8-week observation period. Exploratory regression analysis identified warm needle acupuncture as the only covariate independently associated with mental component summary improvement, and depressive symptom reduction was more strongly correlated with quality-of-life improvement than anxiety reduction. Given that depression affects approximately one-third of older adults with osteoporosis,22 these psychological benefits may carry particular clinical relevance for the management of this condition. It should be emphasized that none of these mechanistic markers were measured in the present cohort; these pathways are cited as biologically plausible hypotheses rather than as explanations directly supported by our data. The psychological benefits of warm needle acupuncture are consistent with the known serotonergic, neuroendocrine, and neuroimmune mechanisms through which acupuncture modulates mood disorders.11–13 A recent review provided additional evidence that acupuncture regulates the hypothalamic-pituitary-adrenal axis through effects on corticotropin-releasing hormone, adrenocorticotropic hormone, and cortisol at the level of the paraventricular nucleus, hippocampus, and amygdala.23 The thermal component of warm needle acupuncture may amplify these effects by enhancing local blood circulation and providing sustained analgesic stimulation, thereby disrupting the cycle of pain, inactivity, and mood deterioration. An overview of warm needle acupuncture for osteoarthritis reported that nine of fourteen included reviews reached positive conclusions, although the total effective rate used as the primary endpoint in that literature is a poorly standardized composite outcome.24 However, those studies focused primarily on pain and physical function outcomes. The present findings extend this evidence by demonstrating that the benefits of warm needle acupuncture also encompass the psychological and mental quality of life domains in patients with osteoporosis. The observation that depressive symptom reduction correlated more strongly with quality-of-life improvement than did anxiety reduction may reflect the greater representation of somatic items in the SDS that overlap with pain-related complaints, rendering depression scores more responsive to interventions that concurrently relieve pain. Social functioning also reached significance; however, this result should be interpreted with caution because baseline social functioning scores differed significantly between groups (P = 0.019), with the WNA group starting approximately 7 points higher. Importantly, the between-group comparison of change scores, which is less sensitive to baseline imbalance, also showed a significantly greater improvement in the WNA group (P < 0.001, FDR-adjusted q < 0.05), supporting the robustness of this finding. Although post-treatment scores in general health, vitality, and role-emotional did not differ significantly between groups, change-score comparisons indicated significantly greater improvements in the WNA group for these domains as well, albeit with smaller between-group effects than in the physical domains. General health and vitality reflect global self-assessments of health status and energy that are typically shaped by long-standing health beliefs and physical deconditioning, and an 8-week intervention may be insufficient to shift these perceptions to the level of absolute between-group differences aligns with the regression finding that warm needle acupuncture was independently associated with mental component summary improvementThe between-group differences in physical functioning, role-physical, and bodily pain are consistent with the observed pain reduction and likely reflect improved physical capacity. The significance of mental health directly supports the psychological benefits captured by the SDS and SAS and. The smaller post-treatment between-group difference in role-emotional, despite clear improvements on depression and anxiety scales, does not contradict the overall pattern of psychological benefit. The role-emotional scale uses a dichotomous response format that produces only a limited number of distinct score levels, making it susceptible to floor and ceiling effects and less sensitive to incremental change than continuously scored domains such as mental health.25 The high standard deviations observed for this domain in both groups further support this interpretation. A multicenter randomized trial reported that individualized acupuncture for chronic neck pain produced sustained improvements in pain and disability over 24 weeks, with physical function and bodily pain domains showing larger effect sizes than general health perceptions, a pattern broadly similar to the domain-specific findings in the present study.26 The pain reduction in the warm needle acupuncture group may represent one pathway linking the intervention to the observed psychological and quality of life improvements. However, this mediation hypothesis is inferred from post-hoc correlational analyses and was not formally tested in a causal framework; it should therefore be regarded as hypothesis-generating rather than established. A path analysis study in menopausal women with osteoporosis demonstrated that pain and physical function were significant predictors of depressive symptoms, with the model explaining 36% of variance in depression and 72% in physical quality of life.27 Pain relief may therefore facilitate increased physical activity and social participation, which in turn alleviates depressive symptoms and improves overall well-being. This interpretation is supported by the significant improvement in the bodily pain domain of the SF-36. The analgesic mechanisms of acupuncture include local effects through adenosine release, segmental spinal inhibition, and supraspinal release of endogenous opioid peptides.28 Experimental evidence has further demonstrated that electroacupuncture recruits beta-endorphin-containing neutrophils to the site of peripheral inflammation, and that this analgesic effect is abolished by beta-endorphin neutralizing antibodies,29 providing evidence for the opioidergic basis of acupuncture analgesia. Guidelines recommendLumbar BMD remained unchanged in both groups after 8-week treatment, which is consistent with the slow biological timeline of bone remodeling and the short observation window. Follow-up DXA intervals of typically one to two years after treatment initiation, as shorter intervals are unlikely to yield changes exceeding the least significant change of the instrument.21 A multicenter randomized trial of acupuncture for osteopenia demonstrated significant reductions in bone turnover markers, including beta-C-terminal telopeptide and N-osteocalcin, after three months of treatment, suggesting that bone metabolic effects may precede detectable BMD changes.30 Previous meta-analyses have also confirmed that acupuncture can improve BMD when treatment extends beyond three months.6,7 The absence of bone turnover marker measurements in the present study prevents determination of whether warm needle acupuncture influenced bone formation or resorption pathways during the observation period. The exploratory regression analysis identified warm needle acupuncture as the only covariate independently associated with mental component summary improvement, whereas baseline demographic characteristics, pain severity, and psychological scores were not significantly associated. Given that psychological distress has been independently linked to osteoporosis diagnosis in population-based cohorts,31 the potential applicability of warm needle acupuncture across baseline psychological severity levels may be clinically meaningful, although this hypothesis requires confirmation in prospective studies before routine incorporation into osteoporosis management can be recommended. The Spearman correlation analysis further revealed that the association between depressive symptom reduction and mental quality of life improvement was approximately twice the magnitude of the corresponding anxiety association, suggesting that depression improvement may serve as a more important pathway to quality-of-life enhancement in this population. Screening for depressive symptoms using validated instruments may therefore be clinically useful in identifying patients with osteoporosis who may benefit most from the addition of warm needle acupuncture to standard therapy. Several limitations should be acknowledged. First, the retrospective, non-randomized design is the principal limitation: group assignment was determined by routine clinical decision-making rather than random allocation, so selection bias and confounding by indication cannot be excluded. Although measured baseline characteristics and pre-treatment outcome scores were largely comparable between groups (with the exception of social functioning), unmeasured factors—such as patient motivation, health literacy, socioeconomic status, and willingness to pay for acupuncture—may have differed between groups and correlated with the outcomes. Second, neither patients nor assessors were blinded, and all primary outcomes (SDS, SAS, SF-36, and VAS) were self-reported; expectation and placebo effects may therefore have contributed substantially to the observed between-group differences, particularly for psychological symptom scales, and the absence of a sham acupuncture control precludes separating specific from non-specific treatment effects. Third, the mental component summary was approximated from domain-level scores rather than calculated using population-specific norm-based algorithms; accordingly, the regression and correlation analyses based on this measure (Figures 3 and 4) should be considered exploratory. Fourth, patients with pre-existing diagnosed psychiatric disorders were excluded, which limits generalizability to patients with more severe psychological burden and may have restricted the observable range of psychological improvement. Fifth, no a priori sample size calculation was performed; although post-hoc power for the primary psychological outcome exceeded 99%, the study may have been underpowered for smaller effects in secondary outcomes. Sixth, multiple outcomes were compared; although false discovery rate correction was applied and the principal findings remained significant, the possibility of chance findings cannot be entirely excluded. Seventh, as a single-center study, generalizability is limited. Eighth, the 8-week observation period was insufficient for detecting BMD changes and may have been too short to observe improvements in general health, vitality, and role-emotional domains; bone turnover markers were not measured, and no long-term follow-up was conducted. Ninth, the intervention required three clinic visits per week for eight weeks, which may be difficult to implement widely among older patients with osteoporosis and may limit real-world feasibility. Future research should employ prospective, multicenter, randomized, sham-controlled trial designs with extended follow-up of at least six months, incorporating bone turnover markers and inflammatory cytokines to provide mechanistic insight. Formal mediation analysis of the pain-psychology-quality of life pathway and subgroup analyses by osteoporosis type would further refine clinical recommendations. In this retrospective cohort study, warm needle acupuncture combined with conventional anti-osteoporosis therapy was associated with greater short-term improvements in depressive symptoms, anxiety symptoms, pain, and quality of life compared with conventional therapy alone, whereas BMD remained unchanged over the 8-week observation period, as expected. Given the retrospective, non-randomized, and unblinded design, these findings demonstrate association rather than confirmed causal efficacy, and they support warm needle acupuncture only as a potential adjunct for short-term symptom and quality-of-life management in patients with osteoporosis. Prospective, randomized, sham-controlled trials with longer follow-up are needed before the intervention can be recommended with confidence as an evidence-based adjunct. Acknowledgment Yunhui Qiu and Xugui Li are the co-first authors, contributed equally to this work. Funding Statement This study was supported by Hubei Province Natural Science Foundation - Joint Funding Project for Sports Innovation and Development (Grant No: 2025AFD649) and the Shizhen Talent Program of Hubei Province for Scientific Research (Grant No: Hubei Health Document [2024] No. 256.) The funders had no role in study design, data collection and analysis, interpretation of the results, or the decision to publish. Ethics Approval The study protocol was approved by the Ethics Committee of The Affiliated Hospital of Wuhan Sports University (approval number: [672HREC20260330-L028]), and the requirement for informed consent was waived due to the retrospective nature of the study and the use of anonymized data. Disclosure The authors declare that they have no competing interests. References