This was a 6-week, parallel-group, unblinded, randomized controlled trial conducted at Hunan University of Technology and Business, Changsha City, Hunan Province, China, from September 3, 2024, to December 12, 2024. The study protocol was approved by the Institutional Ethics Committee of Hunan Technology and Business University (Approval No. 2025-HUTB-PSY-02006). All participants provided written informed consent before the study. The study is reported in accordance with the Consolidated Standards of Reporting Trials (CONSORT) checklist.

The study participants were senior college students enrolled in a Music Appreciation elective course. The inclusion criteria were as follows: (1) full-time senior college students, (2) age ≥18 years old, and (3) having registered for the elective course of Music Appreciation. The exclusion criteria were as follows: (1) students with severe mental illness and suicidal ideation or behaviors; (2) students who were absent due to sickness, vacation, studying abroad, or other reasons during the study period; and (3) students who had received psychotropic medications or other therapies (e.g., mindfulness interventions, music therapy, and cognitive-behavior therapy) during the past 3 months. The attrition criteria were as follows: (1) participants in the intervention group who completed fewer than 2 MBMT sessions, (2) participants who voluntarily dropped the course or withdrew from the study, and (3) students who did not complete the pre- and post-survey questionnaires.

The sample size was estimated using G-Power 3.1.9.7, based on two primary outcome indicators: depression, as assessed by the Zung Self-Rating Depression Scale (SDS), and anxiety, as assessed by the Zung Self-Rating Anxiety Scale (SAS). For this study, α was set at 0.05, the power was set at 0.95, assuming a medium effect size of 0.35 for both SAS and SDS. Considering an attrition rate of 15%, the required sample size was 200 for each group.

All participants were fully informed of the study’s purpose, procedure, benefits, and risks at the first introductory class of the Music Appreciation course. After providing electronic informed consent, participants were randomly assigned to either the intervention group or the control group. A centralized computer system was used to ensure allocation concealment by managing the random assignment of participants through a secure, independent, and automated process. The statistician (Lei Pan) used pre-specified Stata code to generate random numbers, ensuring a 1:1 allocation of participants (StataCorp and Banks, 2015). The list containing student IDs for participants and their assigned groups was returned to the researcher (Xiaoyan Zhang). A total of 400 students were registered in the Music Appreciation course and were recruited, who were randomized into two groups: an intervention group (n = 200) that received MBMT alongside the course, and a control group (n = 200) that received the standard course content without MBMT.

The intervention was described according to the Reporting Guidelines for Music-based Interventions checklist (Robb et al., 2025). The intervention was adapted from Lesiuk’s protocol, and the contents were based on four core mindfulness attitudes: non-judging, beginner’s mind, suspending judgment, and acceptance (Lesiuk, 2016). Non-judging refers to observing thoughts, feelings, and sensations without attaching labels or judgments to them. The Beginner’s mind is a concept that encourages viewing the world with an open, curious, and fresh perspective, as if for the first time, without being burdened by past experiences or judgments. Suspending judgment involves the willingness to try new things and listen without immediately judging or rejecting them, allowing for a deeper exploration and understanding of the experience. Acceptance is a mental practice that involves embracing the present moment without trying to change or resist it, focusing on what you can control instead of what you cannot. These four attitudes cultivate a state of openness and acceptance, which allows for a deeper connection to the present experience (Lesiuk, 2016). In addition, music can act as a tool to support these attitudes, ultimately helping individuals remain grounded in the moment (Lesiuk, 2016).

The interventionists included the instructor of the Music Appreciation course, a certified music therapist with 10 years of experience in music therapy, and a teacher assistant, a certified clinical psychologist with 5 years of experience in mental health counseling. The interventionists received standard uniform training on MBMT by attending a one-week workshop before the study. The intervention was delivered through the Music Appreciation course, and we designed six weekly MBMT sessions embedded in six music classes, with each session lasting approximately 40 min. Each MBMT session combined a variety of mindfulness practices (such as meditation, yoga, and mindful breathing) and music-related activities (such as listening to classical music, singing, imitating rhythms, and playing instruments). During these sessions, the participants were also guided to imagine themselves in specific scenarios, such as in a park or lying on the lawn, while concurrently focusing on their breathing. To enhance effectiveness, homework was assigned after each intervention to practice the session for 15 to 20 min daily. Further details of the themes and content of the intervention are shown in Table 1.

The control group received the standard course content without MBMT but could access other forms of therapy at their own discretion, the same way as the intervention group.

Students completed paper-based questionnaires to assess their mental health at the beginning and end of the Music Appreciation course. The questionnaires included demographic information such as age, gender, major, and residence, and standard scales to assess depression, anxiety, obsessive-compulsive symptoms, sleep quality, and resilience.

All data were analyzed via STATA statistics (version 17.0). Continuous variables were expressed as means and standard deviations (SDs), while categorical variables were presented as numbers and proportions. The analysis used an intention-to-treat (ITT) approach. Since no more than five individuals for each measure at baseline had a single entry missing, indicating missing at random, multiple imputation (MI) was adopted for ITT analysis. Independent sample t-tests and χ2 tests were used to compare the baseline characteristics between the intervention and control groups, as well as between the study sample and the dropout sample. The intervention effects were assessed using linear mixed-effects models (LMMs) with time (pre/post) as a within-subject factor and group (MBMT vs. control) as a between-subject factor. The LMM accounts for the non-independence of repeated measures taken from the same individual by including both fixed and random effects. Main effects of group, time, and group × time interaction effects were examined. Cohen’s d was calculated as the effect size between groups and between pre-post test measurements, based on the ITT rule. Effect sizes were categorized as small (d = 0.2), medium (d = 0.5), and large (d = 0.8). All statistical tests were two-sided, with a p-value < 0.05 indicating statistical significance. For multiple group comparisons, the Holm-Bonferroni correction was performed to control for type I error.

A total of 415 students were registered for the Music Appreciation course, among whom 15 were excluded due to sickness-related absences, resulting in 400 students who were eligible and completed informed consent. They were randomly assigned to the intervention group (n = 200) and the control group (n = 200). Among the 400 participants, all completed the baseline assessment; however, 22 were excluded from the final analysis due to dropping out of the course (n = 4), completing fewer than two MBMT sessions (n = 6), or failing to complete the post-intervention assessment (n = 12). Finally, 378 participants were included in the final analysis, comprising 181 in the intervention group and 197 in the control group (Figure 1). The intervention group had a significantly higher attrition rate than the control group (9.50% vs. 1.50%, p < 0.001).

Table 2 shows the sample characteristics at baseline. The participants had a mean age of 20.15 ± 0.73 years, and 59.65% (238/400) were Female. Most students were from urban areas (65.50%), and a large proportion of them were majoring in music (39.50%), pursuing the same music degree. Their mean scores for depression, anxiety, obsessive-compulsive symptoms, sleep quality, and resilience were 38.87 ± 7.80, 36.34 ± 5.54, 35.56 ± 8.70, 3.60 ± 0.97, and 84.69 ± 13.52, respectively. There were no significant differences in baseline characteristics between the intervention group and the control group, and between the completed sample and the missing sample.

As shown in Table 3, the LMM revealed statistically significant group effects (p < 0.001 for all) and group-by-time interaction effects (p < 0.001 for all) for both depression (SDS) and anxiety (SAS). After the intervention, the MBMT group showed more substantial decreases in SDS (β = −2.48, p < 0.001) and SAS scores (β = −2.41, p < 0.001) than the control group, with intergroup effect sizes being 0.40 (95% CI 0.27–0.81) and 0.30 (95% CI 0.25–0.78), respectively.

Table 3 also shows the LMM results for obsessive-compulsive symptoms (OSS), sleep quality (PSQI), and resilience (CD-RISC). Statistically significant group effects (p < 0.001 for all), time effects (p < 0.001 for all), and group-by-time interaction effects (p < 0.001 for all) were observed for both obsessive-compulsive symptoms and resilience. After the intervention, both groups showed significant changes in OSS (p < 0.001) and CD-RISC scores (p < 0.001). However, the MBMT group showed more substantial changes in OSS (β = −3.85, p < 0.001) and CD-RISC scores (β = 6.94, p < 0.001) than the control group, with intergroup effect sizes being 0.76 (0.28, 0.92) and 0.78 (0.47, 0.81), respectively. In addition, statistically significant group effect (p < 0.001) and group-by-time interaction effect (p < 0.001) were observed for sleep quality. The MBMT group showed a more substantial decrease in PSQI score (β = 0.65, p < 0.001) than the control group, with intergroup effect size being 0.63 (0.16, 0.83). Similar significant intervention effects were also observed in the subscales of OSS and CD-RISC, and the details are shown in Table 3.

Table 4 shows the intervention group’s feedback on the MBMT sessions as assessed by the AS. The participants expressed high levels of acceptability and satisfaction with the intervention, with the mean score of each item ranging from 4.84 ± 0.49 to 4.91 ± 0.43 (all ≥4).

Our study is the first of its kind to combine mindfulness and music therapy, innovatively integrating them into a selective course for senior students to enhance their mental health and overall well-being. Our results showed that although the intervention group and control group had comparable characteristics and mental health status at baseline, the intervention group demonstrated significant improvement in depression, anxiety, obsessive-compulsive symptoms, sleep quality, and resilience as compared to baseline and the control group. Our findings confirm the feasibility and effectiveness of MBMT in improving mental health among senior college students, which offers important implications for future mental health promotion programs on campus.

Our study revealed a high adherence rate to interventions, with 90.50% of participants completing at least two sessions and both pre- and post-assessments. Additionally, all students who attended the sessions completed their homework, achieving a 100% completion rate. Homework is essential for helping students apply new mindfulness and music strategies to manage stress and anxiety in real-world settings. Integrating homework, such as mindful music listening and mindful instrument practice, into the MBMT intervention can reinforce skills learned in sessions, foster independent practice, and extend the therapeutic effects into daily life. The high adherence rate confirms that embedding MBMT sessions within selective college courses is a viable approach for promoting intervention compliance, with potential long-term sustainability. Integrating the intervention within the college curriculum offers a feasible and practical approach as it reduces barriers to access and promotes participation for college students who often juggle multiple academic and extracurricular demands. Further post-intervention assessments showed that participants expressed high levels of acceptance and satisfaction with the MBMT sessions. Such positive feedback provides further evidence of the practical application and effectiveness of the MBMT intervention. This finding suggests that MBMT may be scalable to a larger student population if effectively integrated into more courses or programs, thereby benefiting a greater number of students and improving their mental health and overall well-being.

However, it should also be noted that the intervention group had a significantly higher attrition rate than the control group (9.5% vs. 1.5%), also known as differential attrition (Bell et al., 2013). Multiple factors may affect participants’ adherence to the intervention, including academic overload, daily practice demands, unmet expectations, lack of perceived benefits, participant-investigator relationships, and other logistical challenges (Collins et al., 2022). For instance, the intervention may have imposed a higher academic burden than the control group’s standard activities by requiring additional time, effort, or lifestyle changes that were unsustainable for some students, leading to higher dropout rates. This finding suggests the need for proactive protocols to address known causes of attrition and minimize the risk of differential attrition. These strategies include maintaining clear communication, enhancing participant engagement, minimizing participant burden, providing transparent information, offering appropriate incentives, and collecting follow-up data for all participants (Elfeky et al., 2020).

Our study showed that MBMT significantly improved depression and anxiety symptoms of senior college students. This finding aligns with previous research demonstrating the effectiveness of mindfulness-based interventions in alleviating psychological distress across various clinical settings, indicating their potential for broader application (Chan et al., 2023; Knoerl et al., 2022; Yildirim and Ciris Yildiz, 2022). Previous research suggests that mindfulness-based interventions are more effective in reducing symptoms of depression and anxiety than non-evidence-based treatments, such as relaxation training, health education, supportive psychotherapy, and imagery or suppression techniques (Khoury et al., 2013; Hofmann and Gomez, 2017). These beneficial effects remained stable even after the follow-up period (Khoury et al., 2013; Hofmann and Gomez, 2017).

Furthermore, mindfulness-based interventions have demonstrated comparable efficacy with cognitive behavioral therapy (CBT), a well-established gold standard treatment for anxiety and depression (Khoury et al., 2013; Hofmann and Gomez, 2017). Previous meta-analyses showed that mindfulness-based interventions and CBT were equally effective in treating depression (Sverre et al., 2023) and anxiety (Li et al., 2021) at both post-intervention and follow-up. While MBMT and CBT share many similar characteristics and are both effective in managing depression and anxiety symptoms, they offer different approaches and aims. MBMT emphasizes acceptance and observation of thoughts to prevent relapses and reduce their negative impact, while CBT focuses on changing or restructuring negative thoughts and behaviors (Sverre et al., 2023; Li et al., 2021). Our findings will extend the available treatment options for depression and anxiety beyond the currently recommended psychotherapy, such as CBT, suggesting MBMT as a cost-effective alternative to address mental health problems (Sverre et al., 2023; Li et al., 2021). Our study has significant clinical implications, offering additional treatment approaches and empowering participants to choose the approach that best fits their needs, which aligns with the principle of personalized approaches to mental healthcare (Deif and Salama, 2021).

Apart from emotional distress, MBMT has also been shown to decrease obsessive-compulsive symptoms and improve sleep quality and resilience. Obsessive-compulsive symptoms are characterized by persistent, unwanted thoughts (obsessions) that cause significant anxiety, often accompanied by repetitive behaviors (compulsions) aimed at reducing the anxiety (Woody et al., 2019). A meta-analysis showed that mindfulness-based interventions were effective in reducing obsessive-compulsive symptoms, with a small to medium pooled effect (SMD = −0.35) (Chien et al., 2022). MBMT can be an effective tool in managing obsessive-compulsive symptoms by fostering non-judgmental awareness of intrusive thoughts and reducing their perceived significance (Chien et al., 2022).

The beneficial effects of MBMT on sleep quality were congruent with a meta-analysis of 61 RCTs, which showed that mindfulness-based interventions were effective in improving sleep quality, with a large pooled effect (SMD = −0.794) (Yang et al., 2022). MBMT can reduce stress and anxiety that disrupt sleep, promote relaxation and easy transition to sleep, enhance body awareness, and release physical tension that interferes with sleep, all contributing to better sleep quality (Yang et al., 2022). Finally, the finding that MBMT significantly improved participants’ resilience also aligns with a meta-analysis of 57 RCTs, which showed a medium pooled effect size of mindfulness-based interventions in improving resilience (O’Connor et al., 2023). MBMT can help individuals develop a deeper understanding of their own mental and emotional processes, which is a core component of resilience (Christopher et al., 2018). In addition, MBMT encourages individuals to focus on the present and be flexible in their thinking and behavior, which is crucial for adapting to changing circumstances and building resilience (Christopher et al., 2018).

Notably, our study also showed that the control group experienced significant improvement in OSS and resilience, despite not participating in the MBMT intervention, demonstrating a control group effect. This finding highlights the potent indirect benefits of the Music Appreciation class through musical stimulation and social interaction. While the control group did not engage in active MBMT, passive engagement, such as listening to and learning about music, can still have a positive effect on brain and cognitive function, leading to improved resilience and decreased OSS (Thompson et al., 2023). Additionally, the social aspect of the Music Appreciation class can foster a sense of community, enhance interpersonal skills, and promote empathetic development, all of which contribute to improved mental health (Thompson et al., 2023). The improvement in the control group may complicate the interpretation of results and challenge the assumed effectiveness of the intervention. To address this issue, future research should incorporate more rigorous control conditions, such as enhanced blinding, repeated measurements, and waitlist control, to ensure that research findings are valid and reliable (Aday et al., 2022; Jensen et al., 2017).

This study has several limitations. First, the sample was drawn from a single university and from a specific course (Music Appreciation), which may limit the external validity of our findings. The conclusions may not be generalizable to students at other institutions with different academic cultures, pedagogical approaches, and student demographics. To enhance the generalizability and robustness of these findings, future research should replicate this study across a diverse range of institutions, including public, private, and community colleges, as well as with students from various disciplines and socioeconomic backgrounds. Second, the intervention was delivered through a selective course of Music Appreciation, which may potentially serve as an informal music therapy-related intervention. Therefore, the control group should be framed as an active comparator rather than a true placebo or inactive control, and the effect sizes represent a comparison between the two, not a comparison of a treatment versus no treatment at all. Although the MBMT sessions differ from the music course content, combining MBMT and music courses can generate a synergistic effect that enhances both. Future studies may consider a multi-arm trial that includes an accurate placebo/no-intervention control to compare the effectiveness of music courses with or without MBMT, as compared to the control group, to gain a better understanding of the intervention effects. Third, we only collected data at pre- and post-intervention, and the absence of follow-up assessments is a key limitation. This can lead to unknown long-term efficacy, inability to identify delayed effects, risk of bias, limited validity, and a lack of insight into maintenance. Future RCTs with multiple assessments and longer follow-ups are needed to assess the long-term effectiveness of MBMT.

Fourth, this RCT was not blinded due to the nature of behavioral interventions. The lack of assessor blinding and the reliance on self-reported outcomes can introduce significant risks of bias, including observer bias, participant bias, social desirability bias, and recall bias, potentially leading to an overestimation of treatment effects. Future studies can implement strategies to mitigate bias, including blinding outcome assessors, using objective behavioral measures, and conducting sensitivity analyses. Finally, our study was conducted within a specific Chinese cultural context, and our results may not apply to other cultural groups, which could affect their validity and generalizability. Future studies should incorporate a cross-cultural perspective by comparing findings across different cultural groups to identify universal patterns versus culturally specific phenomena.

Our study carries significant practical implications for implementing MBMT in academic environments, which should include comprehensive facilitator training and adapting the format to fit the institutional context, such as hybrid delivery. Facilitators should have an established personal mindfulness practice (lasting one to 2 years) with a deep, personal understanding of mindfulness to guide others more effectively. Facilitators should complete certification programs that blend online learning, in-person intensives, and individualized guidance with a focus on teaching skills, curriculum design, inclusivity, and trauma-informed care. Ongoing professional development is also necessary to ensure that facilitators remain current with research and maintain their professional practice. In addition, a hybrid format, which combines in-person and online instruction, can make MBMT more accessible and flexible for a diverse academic community by increasing accessibility, accommodating diverse schedules, and enhancing learning opportunities.

Apart from practical implications, our study also highlights the theoretical contribution of MBMT as a conceptual framework that moves beyond simply demonstrating its effectiveness as an intervention. It presents a multimodal and synergistic conceptual framework. It proposes a new, integrated model for combining music and mindfulness to produce a synergic effect that is more powerful than the sum of each component. This framework has important implications for both research and practice. Future research based on the framework should focus on the synergistic interactions between modalities, which can better inform and optimize MBMT interventions.