Acupuncture for subthreshold depression: a randomized controlled trial Authors: Jin Xian, Ling Wang, Mi Sun, Xiao-Ming Zang, Hui-Juan Yu, Xin Zhang, Bin Cheng, Qi-Wen Tan Published: 2025 ABSTRACT AbstractBackgroundSubthreshold depression (SD), characterized by depressive symptoms not meeting major depressive disorder (MDD) criteria, poses a significant risk for MDD development. Acupuncture has shown promise for SD, though its efficacy beyond placebo remains unclear. Therefore, we designed this study to evaluate the efficacy of verum acupuncture (VA) compared to sham acupuncture (SA) for SD.MethodsA randomized clinical trial was conducted at the Affiliated Hospital of Shandong University of Traditional Chinese Medicine, China, from September 2021 to October 2023. Participants with SD were randomized 1:1 to receive either VA or SA for 8 weeks, with 12-week follow-up. The primary outcome was the change in Hamilton Depression Scale-17 (HAMD-17) scores from baseline to 8 weeks.ResultsOf 231 screened patients, 64 participants (71.87% female; mean age, 29.83 [SD 6.84] years) were included. Fifty-nine participants (92.2%) completed the study. The mean reduction in HAMD-17 scores from baseline to week 8 was -7.28 (95% CI, -8.26 to -6.3) in the VA group and -3.80 ( 95% CI, -4.42 to -2.77) in the SA group. The between-group difference at week 8 was -4.08 (95% CI, -5.15 to -3.02; P < .001). Effects persisted during the 12-week follow-up. No serious adverse events occurred.ConclusionsAcupuncture demonstrated significant therapeutic effects for SD compared to SA, with benefits extending beyond placebo. These findings support acupuncture as a specific and effective intervention for SD. However, the study's limitations, including its single-center design and relatively young sample, should be considered when generalizing the results.Trial registrationChiCTR2100049660; Registered August 7, 2021.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12906-025-05165-6.Keywords: Acupuncture, Subthreshold depression, Randomized clinical trial Background Subthreshold depression (SD) characterized by depressive symptoms that do not meet the criteria for major depressive disorder (MDD), may coexist with or transition into MDD over time [1]. The probability of patients with SD developing into MDD is approximately 17.62%, significantly higher than the 6.08% probability in non-depressed individuals [2]. The prevalence of SD is estimated to be two to three times that of MDD in Europe [3] and has been reported with rates ranging from 1.4% to 17.2% in community samples [4]. In China, similarly high rates have been observed, with studies among college students reporting prevalence around 30%–37% [5, 6]. Moreover, a longitudinal study during the early stage of the COVID-19 pandemic found that 47.8% of adults exhibited SD symptoms [7]. Although symptoms are typically less severe, SD remains linked with increased risks of suicide and mortality [8] and correlates positively with cardiovascular morbidities, including coronary artery disease and stroke [9]. Therefore, Prompt recognition, diagnosis, and management of SD are crucial for preventing its escalation to MDD, thereby mitigating its impact on health [10]. Current evidence suggests that for mild depressive symptoms, the efficacy of antidepressants is not significantly different from placebo [11]. Consequently, non-pharmacological treatments should be prioritized for managing SD [12]. Meta-analysis results indicate that psychotherapy, especially cognitive behavioral therapy (CBT), may be the most effective non-pharmacological intervention to improve SD in adults [13, 14], but it requires considerable time and financial investment. Additionally, a systematic review and individual participant data meta-analysis demonstrated that psychological interventions yield small to moderate benefits and can significantly reduce the risk of progression to MDD [15]. Acupuncture, a non-pharmacological therapy valued for its simplicity and cost-effectiveness, is widely accepted and currently extensively utilized in the treatment of MDD [16, 17] and its associated insomnia [18]. There is also evidence that acupuncture can improve symptoms of SD [19]. However, there are currently limited data directly comparing the efficacy of acupuncture versus SA in treating SD. Therefore, we designed this study. Methods Study design This study was a randomized, double-blinded clinical trial with 8-week treatment and 12-week follow-up for patients with SD. The protocol has been previously published [20], was approved by the Institutional Review Board of the Affiliated Hospital of Shandong University of Traditional Chinese Medicine, and this study was registered in the Chinese Clinical Trial Registry (ChiCTR2100049660). This study adhered to the Consolidated Standards of Reporting Trials (CONSORT) guidelines [21] and the Standards for Reporting Interventions in Clinical Trials of Acupuncture (STRICTA) guidelines [22]. All participants signed informed consent forms before randomization. Participants Participants were included if they met the following criteria: diagnosed with SD using the Mini International Neuropsychiatric Interview (MINI, version 5.0), presenting with at least two depressive symptoms lasting for more than two weeks and some degree of functional impairment, but not fulfilling the diagnostic criteria for MDD according to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5); aged 18–60 years; HAMD-17 score between 7 and 16; no depression-related treatment had been received in the previous two weeks; and had not taken antibiotics, probiotics, yogurt, or other fermented dairy products prior to sample collection. Exclusion criteria were: meeting the criteria for MDD, bipolar disorder, or psychotic disorder according to DSM-5; severe liver or kidney damage, severe arrhythmia, or cardiac insufficiency; pregnancy; depression secondary to psychoactive or non-addictive substances or brain diseases; a history of suicide attempt or suicidal behavior; inability to tolerate acupuncture; or difficulty communicating or cooperating with researchers. Randomization and blinding The R 4.0.3 software package was used to generate random numbers and employ dynamic block randomization (block sizes 4, 6, and 8), allocating participants in a 1:1 ratio to the VA group or the SA group. The randomization sequence was concealed in opaque, sealed envelopes and managed by a team member who was not involved in recruitment, assessment, intervention, or statistical analysis. Participant blinding was maintained, and communication between participants was minimized using independent treatment rooms and an appointment system. Outcome assessments were conducted by researchers blinded to group assignments. Interventions Both VA and SA were performed by certified acupuncturists with at least 5 years of clinical experience. All acupuncturists received standardized training prior to the trial, covering acupoint and non-acupoint locations, manipulation techniques, and minimal acupuncture stimulation protocols. Subjects in both groups received 24 sessions (3 times a week for 8 weeks; 30 min per session). Baihui (GV20), Yintang (GV29), Shenmen (HT7), Taichong (LR3), Tianshu (ST25), and Zusanli (ST36) were used in the VA group. The selection of acupoints was based on the Evidence-Based Guidelines of Clinical Practice with Acupuncture and Moxibustion for Depression (Revised) [23], and was further adapted according to the clinical experience of our research team. Sterile disposable Huatuo needles (0.30 × 40 mm) were inserted at the acupoints. The twirling and lifting technique was applied for 30 s at each point to elicit Deqi sensations (soreness, numbness, distension, and heaviness). In the SA group, acupuncture was applied at non-acupoints (see eTable 1 in Supplement 1) using sterile needles (0.30 × 25 mm). The needles were inserted 2 mm deep without any manual stimulation, and participants were not expected to experience Deqi (no sensation after insertion). No participants in either group received any form of antidepressant treatment during the intervention period. Table 1Baseline CharacteristicsCharacteristicsVA group(N = 32)SA group(N = 32)Sex, n (%) Male7 (21.9%)11 (34.4%) Female25 (78.1%)21 (65.6%)Age, mean (SD)29.6 (7.20)30.1 (6.57)BMI, mean (SD)22.1 (3.40)23.2 (3.01)Years of Educationa M [Q25, Q75]16.0 [15.8,19.2]16.0 [16.0,19.0]Employment Status, n (%) Full-time19 (59.4%)16 (50.0%) Student11 (34.4%)10 (31.2%) Freelancer1 (3.12%)4 (12.5%) Unemployed1 (3.12%)2 (6.25%)Relationship Status, n (%) Single19 (59.4%)12 (37.5%) In a Relationship2 (6.25%)5 (15.6%) Married11 (34.4%)12 (37.5%) Divorced0 (0.00%)3 (9.38%)Exercise Habits, n (%) None12 (37.5%)14 (43.8%) Occasionally (1–2 times per week)16 (50.0%)14 (43.8%) Frequently (≥ 3 times per week)4 (12.5%)4 (12.5%)Eating Habits, n (%) Irregular time and quantity5 (15.6%)13 (40.6%) Regular quantity, irregular time4 (12.5%)5 (15.6%) Regular time, irregular quantity9 (28.1%)6 (18.8%) Regular time and quantity14 (43.8%)8 (25.0%)Smoking, n (%) Non-smoker32 (100%)28 (87.5%) Occasionally (≤ 5 days per month)0 (0.00%)2 (6.25%) Frequently (> 6 days per month)0 (0.00%)2 (6.25%)Drinkingb, n (%) Non-drinker21 (65.6%)19 (59.4%) Occasionally (≤ 5 days per month)11 (34.4%)13 (40.6%)HAMD-17, mean (SD)c11.5 (2.63)12.3 (2.40)PHQ9, mean (SD)d9.91 (1.97)10.7 (2.22)GAD7, mean (SD)e6.84 (3.31)6.59 (2.82)SF-12v2 PCS, mean (SD)f51.3 (5.20)52.9 (5.43)SF-12v2 MCS, mean (SD)g32.8 (7.41)34.1 (8.53)Abbreviations: VA group Verum Acupuncture group, SA group Sham Acupuncture group, HAMD-17 Hamilton Depression Rating Scale-17, PHQ-9 Patient Health Questionnaire-9, GAD-7 Generalized Anxiety Disorder Scale-7, SF12v2-PCS 12-item Short Form Health Survey, Version 2 Physical Component Summary, SF12v2-MCS 12-item Short Form Health Survey, Version 2 Mental Component SummaryaNonparametric test (Mann–Whitney U test) was usedbChi-square test was usedcScores range, 0–52; higher scores indicate worse depressiondScores range, 0–27; higher scores indicate worse depressioneScores range, 0–21; higher scores indicate worse anxietyfScores ranges, 0–100; higher scores indicate better physical healthgScores ranges, 0–100; higher scores indicate better mental health Baseline Characteristics Abbreviations: VA group Verum Acupuncture group, SA group Sham Acupuncture group, HAMD-17 Hamilton Depression Rating Scale-17, PHQ-9 Patient Health Questionnaire-9, GAD-7 Generalized Anxiety Disorder Scale-7, SF12v2-PCS 12-item Short Form Health Survey, Version 2 Physical Component Summary, SF12v2-MCS 12-item Short Form Health Survey, Version 2 Mental Component Summary aNonparametric test (Mann–Whitney U test) was used bChi-square test was used cScores range, 0–52; higher scores indicate worse depression dScores range, 0–27; higher scores indicate worse depression eScores range, 0–21; higher scores indicate worse anxiety fScores ranges, 0–100; higher scores indicate better physical health gScores ranges, 0–100; higher scores indicate better mental health Outcomes The primary outcome was the mean change in HAMD-17 scores from baseline to week 8. HAMD-17 assessments were conducted at baseline, week 4, week 8 (end of treatment), and week 20 (follow-up). The Hamilton Depression scale [24], developed by Professor Max Hamilton in 1960, is a widely used instrument for evaluating depression severity. Secondary outcomes included changes in the 9-item Patient Health Questionnaire (PHQ-9) [25], the 7-item Generalized Anxiety Disorder (GAD-7) scale [26], and the SF-12v2 Health Survey (SF-12v2) Mental Component Summary (MCS) and Physical Component Summary (PCS) [27]. These were also assessed at baseline, week 4, week 8, and week 20. Expectations regarding acupuncture were evaluated at baseline. Acupuncture tolerability was assessed after the first and the final treatment sessions. Blinding success was assessed at week 8, immediately after the last treatment session. Adverse events, including broken or missing needles, post-acupuncture dizziness, unbearable pain (VAS ≥ 8), prolonged pain lasting more than 2 h (VAS ≥ 4), local hematoma, infection, or abscess, as well as other unforeseen events, were monitored and recorded throughout the entire study period. Statistical analysis The sample size was calculated based on the reduction in HAMD-17scores from baseline to week 8 in the pilot study. The mean (SD) reduction in HAMD scores was −6.1 (2.38) in the VA group and −3.7 (2.21) in the SA group. With 25 patients in each group, the study has 95% power to detect a significant difference at a 2-sided significance level of 0.05. Accounting for a 20% dropout rate, the total sample size was set at 64 (32 per group). Statistical analyses were performed using the R 4.3.1 software package, with a significance level of P < 0.05 (two-sided test) for all tests. All analyses were conducted using the intention-to-treat (ITT) approach, with missing outcome data imputed using the Last Observation Carried Forward (LOCF) method [28]. The normality of continuous variables was assessed using the Shapiro–Wilk test, followed by either independent samples t-test or Mann–Whitney U test; categorical data were analyzed using the chi-square test or Fisher's exact test. Repeated measures data were analyzed using linear mixed-effects models (LMM) fitted with the lmerTest package [29], with baseline values as covariates, and time points, groups, and their interactions as fixed effects, and individual participants as random effects. Age, sex, and BMI were additionally tested as covariates, and the optimal model was selected using P-values, Akaike Information Criterion (AIC), and model parsimony. Results A total of 231 patients were screened, resulting in the inclusion of 64 participants (46 females [71.87%]; mean age, 29.83 [SD 6.84] years) in the intention-to-treat analysis. Both groups were randomized with 32 participants each. Of these, 59 (92.2%) completed the intervention and the subsequent 12-week follow-up (Fig. 1). There were no differences between the study groups in the number of patients lost to follow-up (eTable 2 in Supplement 1). All clinical and baseline demographic characteristics, including expectations of acupuncture, were balanced between the two groups (Table 1, eTable 3 in Supplement 1). Study Flowchart for Enrollment, Allocation, Follow-up, and Analysis Table 2HAMD-17 score among study participantsOutcomeMean change in HAMD-17 score from baseline (95% CI)aDifferenceLsmean (95%CI)P valuebVA groupSA groupWeek 4−5.31 (−6.21 to −4.42)−2.56 (−3.38 to −1.75)−3.14 (−4.21 to −2.08) < 0.001Week 8−7.28 (−8.26 to −6.3)−3.8 (−4.42 to −2.77)−4.08 (−5.15 to −3.02) < 0.001Week 20−6.86 (−8.03 to −6.22)−2.45 (−3.31 to −1.32)−5.21 (−6.27 to −4.14) < 0.001Abbreviations: VA group Verum Acupuncture group, SA group Sham Acupuncture group, HAMD-17 Hamilton Depression Rating Scale-17, LSmean Least Squares MeanaScores range, 0–52; higher scores indicate worse depressionbCalculated using Bonferroni correction HAMD-17 score among study participants Abbreviations: VA group Verum Acupuncture group, SA group Sham Acupuncture group, HAMD-17 Hamilton Depression Rating Scale-17, LSmean Least Squares Mean aScores range, 0–52; higher scores indicate worse depression bCalculated using Bonferroni correction Table 3Second outcomes among study participantsOutcomesMean change from baseline (95% CI)DifferenceLsmean (95% CI)P valueeVA groupSA groupPHQ-9a Week 4−5.06 (−6.04 to −4.09)−2.16 (−2.96 to −1.35)−3.5 (−4.74 to −2.26) < 0.001 Week 8−6.53 (−7.47 to −5.6)−2.97 (−4.11 to −1.83)−4.15 (−5.39 to −2.91) < 0.001 Week 20−6.5 (−7.59 to −5.41)−1.75 (−2.79 to −0.71)−5.34 (−6.58 to −4.1) < 0.001GAD-7b Week 4−3.19 (−3.81 to −2.56)−2.03 (−3.2 to −0.86)−1.01 (−2.1 to 0.08)0.07 Week 8−4.53 (−5.74 to −3.32)−2.56 (−3.65 to −1.47)−1.82 (−2.91 to −0.73)0.0012 Week 20−4.59 (−5.72 to −3.47)−1.53 (−2.45 to −0.62)−2.92 (−4.01 to −1.82) < 0.001SF12v2-PCSc Week 43.21 (1.03 to 5.39)0.8 (−1.25 to 2.86)1.21 (−0.84 to 3.25)0.25 Week 83.79 (1.67 to 5.91)1.3 (−0.29 to 2.9)1.29 (−0.76 to 3.33)0.22 Week 203.8 (1.47 to 6.12)1.04 (−0.55 to 2.64)1.56 (−0.49 to 3.6)0.14SF12v2-MCSd Week 412.4 (9.57 to 15.23)4.22 (1.04 to 7.39)7.32 (3.8 to 10.83) < 0.001 Week 816.7 (13.52 to 19.88)7.04 (3.69 to 10.39)8.79 (5.28 to 12.3) < 0.001 Week 2019.04 (16.27 to 21.81)3.58 (0.71 to 6.45)14.59 (11.08 to 18.1) < 0.001Abbreviations: VA group Verum Acupuncture group, SA group Sham Acupuncture group, PHQ-9 Patient Health Questionnaire-9, GAD-7 Generalized Anxiety Disorder Scale-7,SF12v2-PCS 12-item Short Form Health Survey, Version 2 Physical Component Summary, SF12v2-MCS 12-item Short Form Health Survey, Version 2 Mental Component SummaryaScores range, 0–27; higher scores indicate worse depressionbScores range, 0–21; higher scores indicate worse anxietycScores ranges, 0–100; higher scores indicate better physical healthdScores ranges, 0–100; higher scores indicate better mental healtheCalculated using Bonferroni correction Second outcomes among study participants Abbreviations: VA group Verum Acupuncture group, SA group Sham Acupuncture group, PHQ-9 Patient Health Questionnaire-9, GAD-7 Generalized Anxiety Disorder Scale-7, SF12v2-PCS 12-item Short Form Health Survey, Version 2 Physical Component Summary, SF12v2-MCS 12-item Short Form Health Survey, Version 2 Mental Component Summary aScores range, 0–27; higher scores indicate worse depression bScores range, 0–21; higher scores indicate worse anxiety cScores ranges, 0–100; higher scores indicate better physical health dScores ranges, 0–100; higher scores indicate better mental health eCalculated using Bonferroni correction Changes in HAMD-17 scores over time were analyzed using LMM. The results demonstrated significant main effects of group (F = 104.05, P < 0.001), time (F = 11.99, P < 0.001), and group × time interaction (F = 5.68, P < 0.01), indicating that symptom trajectories differed between groups. At the 8-week posttreatment assessment (primary endpoint), the VA group experienced a mean reduction of 7.28 points from baseline (95% CI −8.26 to −6.3). A significant between-group difference was observed between the VA and SA groups (−4.08, 95% CI, −5.15 to −3.02; P < 0.001). At week 4, midway through the treatment phase, the VA group showed a significantly greater improvement in HAMD-17 scores compared to the SA group (−3.14, 95% CI −4.21 to −2.08, P < 0.001). This therapeutic advantage was sustained during the follow-up period: at week 20, the VA group maintained significantly lower HAMD-17 scores than the SA group (−5.21, 95% CI −6.27 to −4.14, P < 0.001) (Table 2; Fig. 2A). Results of the per-protocol analysis were similar (eTable 4 in Supplement 1). Therapeutic Effects of Acupuncture A, HAMD-17; B, PHQ-9; C, GAD-7 D, PCS; E, MCS HAMD-17, Hamilton Depression Rating Scale 17; PHQ-9, the 9-item Patient Health Questionnaire; GAD-7, the 7-item Generalized Anxiety Disorder (GAD-7); PCS, physical health compared; MCS, mental health compared Secondary outcomes also supported the superiority of VA. At week 4, the VA group demonstrated a significantly greater improvement in PHQ-9 scores (−3.5, 95% CI −4.74 to −2.26) and SF-12v2 MCS (7.32, 95% CI 3.8 to 10.83) compared to SA (P < 0.001 for both). No significant differences were observed in GAD-7 (−1.01, 95% CI −2.1 to 0.08, P = 0.07) or SF-12v2 PCS (1.21, 95% CI −0.84 to 3.25, P = 0.25). At week 8, VA led to greater improvements in PHQ-9 (−4.15, 95% CI −5.39 to −2.91, P < 0.001), GAD-7 (−1.82, 95% CI −2.91 to −0.73, P < 0.01), and SF-12v2 MCS (8.79, 95% CI 5.28 to 12.3, P < 0.001) compared to SA. Differences in SF-12v2 PCS remained non-significant (1.29, 95% CI −0.76 to 3.33, P = 0.22). At the 20-week follow-up, the VA group continued to show superior outcomes: PHQ-9 (−5.34, 95% CI −6.58 to −4.1), GAD-7 (−2.92, 95% CI −4.01 to −1.82), and SF-12v2 MCS (14.59, 95% CI 11.08 to 18.1) were all significantly better than in the SA group (P < 0.001 for all), while SF-12v2 PCS remained statistically comparable (1.56, 95% CI −0.49 to 3.6, P = 0.14) (Table 3; Fig. 2B–F). After 8 weeks of treatment, all patients found acupuncture acceptable, with no statistically significant differences (eTable 5 in Supplement 1). Seven mild adverse events were reported during the study, and no serious adverse events were observed (eTable 6 in Supplement 1). There was no significant difference among the groups in the proportion of patients who guessed they had received acupuncture immediately after the 24th session for the blinding assessment (eTable 7 in Supplement 1). Disscusion In line with previous studies showing the superiority of acupuncture over SA for SD [30], our trial confirmed similar effects. Moreover, we extended prior work by employing a wider range of outcome measures and assessing effects across multiple timepoints, including follow-up. Our findings revealed that by week 8, both SA and VA groups showed reductions in HAMD-17 scores. However, the VA group demonstrated a significantly greater reduction in these scores, highlighting its superior efficacy. A notable observation was the sustained effect of acupuncture beyond the treatment period. At week 20, the HAMD-17 scores in the VA group remained stable, whereas there was a marked increase in the SA group, suggesting that the benefits of acupuncture persist long after the cessation of treatment, unlike SA which showed a transient effect. This disparity can be attributed to the fact that acupuncture likely combines specific and non-specific effects, whereas SA predominantly yields non-specific placebo effect. The PHQ-9 mirrored the HAMD-17 results, reinforcing the consistency of our findings across different assessment tools. Beyond statistical significance, the minimal clinically important difference (MCID) is crucial for understanding the clinical relevance of observed changes [31]. For the HAMD-17, MCID estimates typically range between 3 and 8 points [32]. In our study, the anchor-based MCID was determined as 6 points. At the end of the treatment phase, 75% of participants in the acupuncture group reached or exceeded this threshold, whereas only 21.9% in the SA group achieved it. This highlights the superiority of acupuncture in achieving clinically meaningful improvements in depressive symptoms. Interestingly, the SA group also demonstrated a reduction in HAMD-17 scores at week 8, which may reflect non-specific placebo responses or the natural tendency of remission in SD. Regarding the choice of sham control, superficial needling at non-acupoints was selected because it is widely used in acupuncture trials [33, 34], and it enables effective participant blinding while minimizing specific therapeutic effects. Moreover, given that individuals with SD are often more sensitive and skeptical, we were concerned that non-penetrating placebo needles might compromise blinding credibility. Therefore, superficial needling was adopted to maximize the success of blinding while maintaining methodological rigor. In contrast to depression outcomes, differences in anxiety symptoms did not emerge until week 8, suggesting that acupuncture may require cumulative stimulation to achieve anxiolytic effects and highlighting the potential for a dose–response relationship. Regarding quality of life, as measured by the SF-12v2 PCS and MCS components, acupuncture did not markedly influence PCS but did show a positive impact on MCS. Specifically, acupuncture led to greater improvements in MCS at weeks 4 and 8, with these benefits persisting up to week 20. Although the SA group also showed initial improvements in MCS, the benefits waned rapidly after the conclusion of treatment. No serious adverse events were reported during the trial, and only a few mild adverse events occurred, confirming the favorable safety profile of acupuncture. This supports its acceptability and feasibility as a treatment for SD. The blinding assessment showed a random distribution of guesses between groups. In this trial, only two options (“VA” or “SA”) were provided to encourage participants to make a definite choice based on their perceptions, a method that has also been adopted in previous acupuncture trials [35, 36]. Nonetheless, we acknowledge that including a “don’t know” option could provide additional information on uncertainty, and we will consider this refinement in future studies. Our acupuncture protocol included acupoints such as Zusanli (ST36) and Tianshu (ST25), which are well recognized for their roles in gastrointestinal regulation and have been widely applied in acupuncture treatment for depression [37, 38]. While previous studies have often emphasized points along the Governor Vessel (Du Mai) or those associated with the heart and mind, our approach integrates gastrointestinal-targeted points based on the well-established correlation between gut function and emotional well-being. This reflects a preventive and holistic perspective consistent with traditional Chinese medicine theory. SD is associated with alterations in brain function and gut microbiota composition compared to healthy individuals [39]. Acupuncture has been shown to modulate brain function in depressed patients [40], leading us to hypothesize that its therapeutic mechanism for SD might involve similar neurobiological pathways. Additionally, preliminary data suggest that acupuncture’s antidepressant effects could be mediated through modulation of the gut microbiome [41, 42]. To further explore this hypothesis, stool and plasma samples were collected both at baseline and week 8 as part of the study design, with the aim of conducting microbiome and metabolomic analyses to investigate potential mechanisms underlying acupuncture’s effects in SD. Since these analyses are still in progress, the present manuscript focuses on reporting the clinical efficacy outcomes, while mechanistic findings will be explored once the analyses are completed. In terms of cost-effectiveness and practicality, acupuncture presents an attractive alternative to psychotherapy due to its lower costs and practical feasibility. However, a thorough economic analysis comparing acupuncture to psychotherapy would provide more definitive insights into their relative cost-effectiveness. Despite these promising findings, our study acknowledges several limitations. The participant pool was predominantly young adults, leaving room for future investigations into older adults and adolescents. In addition, anxiety symptom evaluation relied solely on self-reported scales without the inclusion of clinician-administered assessments, which may limit the objectivity of findings Moreover, the absence of a waiting-list control group precludes definitive conclusions about the natural course of SD without intervention. As a single-center trial with a relatively small sample size, the generalizability of our findings is limited. Multicenter studies with larger cohorts are warranted. Finally, although our follow-up extended to 20 weeks, the persistence of effects beyond this period remains to be investigated. Conclusions Acupuncture demonstrated significant therapeutic effects for SD compared to SA, with benefits extending beyond compared placebo. These findings support acupuncture as a specific and effective intervention for SD. However, the study's limitations, including its single-center design and relatively young sample, should be considered when generalizing the results. Supplementary Information Supplementary material 1. Acknowledgements We sincerely thank all the participants involved in this study. We also extend our special gratitude to Dr. Yuning Xie, Dr. Chunxiao Zhao, and Dr. Jian Wang for their valuable guidance on study design and methodology. Role of the Funder/Sponsor The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. Abbreviations Subthreshold depression Major depressive disorder Cognitive behavioral therapy Verum acupuncture Sham acupuncture The 17-item Hamilton depression scale The 9-item patient health questionnaire The SF-12v2 health survey The 7-item generalized anxiety disorder scale The SF-12v2 mental health compared score The SF-12v2 physical health compared score Intention-to-treat Body mass index Linear mixed-effects model The minimal clinically important difference Authors' contributions Concept and design: Jin Xian, Ling Wang, Qi-wen Tan. Acquisition, analysis, or interpretation of data: Mi Sun, Xiao-ming Zang, Qi-wen Tan. Drafting of the manuscript: Jin Xian, Ling Wang. Critical revision of the manuscript for important intellectual content: Hui-juan Yu, Xin Zhang, Bin Cheng, Qi-wen Tan. Statistical analysis: Mi Sun, Xiao-ming Zang. Obtained funding: Jin Xian, Qi-wen Tan. Administrative, technical, or material support: Xiao-ming Zang, Hui-juan Yu, Xin Zhang. Supervision: Jin Xian, Ling Wang. Funding This study was funded by support by the scientific research and innovation team of ZhiWeiBing of Shandong University of Traditional Chinese Medicine, Natural Science Foundation of Shandong Province (ZR2022QH197), Traditional Chinese Medicine Science and Technology Project of Shandong Province (2021Q110 & Q-2023088). Data availability The datasets used and/or analyzed in this study are available from the corresponding author upon reasonable request. Declarations Ethics approval and consent to participate The study protocol was approved by the Institutional Review Board of the Affiliated Hospital of Shandong University of Traditional Chinese Medicine ((2021) Ethical review No.(055)–KY) and registered in the Chinese Clinical Trial Registry (ChiCTR2100049660). The study was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants prior to randomization. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Jin Xian and Ling Wang contributed equally to this work. References Associated Data Supplementary Materials Supplementary material 1. Data Availability Statement The datasets used and/or analyzed in this study are available from the corresponding author upon reasonable request.