The effectiveness of the Chinese Five-Element Music intervention on older adults with depression and anxiety disorder: A systematic review and meta-analysis

Abstract

Background:

Depression and anxiety are among the most prevalent mood disorders worldwide, affecting hundreds of millions of people and contributing to significant disability, particularly in older adults. Conventional treatments such as pharmacotherapy and psychotherapy are effective but often limited by side effects or accessibility. As a complementary approach, Chinese Five-Element Music (CFEM), rooted in Traditional Chinese Medicine, aims to restore emotional balance through the connection between musical tones, organ systems, and emotions. Recent studies suggest its potential to alleviate psychological symptoms, yet evidence in older populations remains limited.

Methods:

The systematic review identified and mapped available literature on CFEM interventions, guiding the selection and analysis of randomized controlled trials (RCTs) for meta-analysis. Databases including Web of Science, PubMed, CNKI, Wanfang, and CQVIP were searched in March 2025. Eighteen RCTs were included. Outcome measures assessed included the Hamilton Depression Rating Scale (HAMD), Hamilton Anxiety Rating Scale (HAMA), Zung Self-Rating Depression and Anxiety Scales (SDS and SAS), Pittsburgh Sleep Quality Index (PSQI), overall response rate (ORR), and Quality of Life (QOL). The risk of bias was assessed using the Cochrane Handbook criteria. Most studies reported adequate randomization, but blinding methods were often unclear.

Results:

A meta-analysis using a random-effects model was conducted on 18 RCTs. The results indicated that, compared to conventional care, the CFEM intervention significantly alleviated symptoms in older adults with depression and anxiety. Specifically, the outcomes were as follows: HAMD score (MD = −3.63, 95% CI: −4.97 to −2.3, P < .00001), HAMA score (MD = −3.89, 95% CI: −5.80 to −1.97, P < .0001), SAD score (MD = −6.69, 95% CI: −8.32 to −5.07, P < .00001), SAS score (MD = −6.84, 95% CI: −10.76 to −2.92, P = .0006), ORR score (95% CI: 1.61 to 4.89, P = .0003), QOL score (MD = −0.58, 95% CI: −15.65 to 14.48, P = .94), PSQI score (MD = −1.86, 95% CI: −2.54 to −1.17, P < .00001).

Conclusion:

CFEM may be an effective complementary intervention to alleviate depression, anxiety, and sleep problems in older adults. However, due to moderate study quality and lack of blinding, further high-quality trials are needed.

1. Introduction

According to the Global Health Data Exchange, in 2019, approximately 970 million people were reported to be suffering from some form of mood disorder,^[1] with anxiety and depression being the most prevalent. Of the 970 million, 301 and 280 million people suffered from anxiety disorders and depression, respectively. In 2020, the prevalence of anxiety and depression increased significantly due to the COVID-19 pandemic, with the number of individuals suffering from anxiety and major depression rising by 26% and 28%, respectively, within a year.^[2] Depression is a common mood disorder affecting an estimated 3.8% of the global population, with 5.0% of adults and 5.7% of those over 60 years of age affected.^[3] According to a 2019 study in China, more than 95 million Chinese individuals have experienced depression at some point in their lives, manifesting as a constant state of sadness, loss of appetite, insomnia, and, in severe cases, disruption of daily life or suicide.^[4]

Common treatments for depression and anxiety include pharmacological treatments, psychotherapy, and electroconvulsive therapy.^[5] Pharmacological interventions have been shown to be effective and are recommended in several clinical practice guidelines.^[6] However, they can cause unwanted side effects. Several nonpharmacological approaches, including music therapy, have demonstrated effectiveness in treating depression.^[7] In China, Chinese Five-Element Music (CFEM) is the most common form of music intervention.^[8] CFEM is based on traditional Chinese medicine (TCM) and the theory of the 5 elements – wood, fire, earth, metal, and water. These elements are linked to 5 tones (Gong, Shang, Jue, Zhi, and Yu), 5 organs (spleen, lung, liver, heart, and kidney).^[9] For example, the liver is linked with Jue and the emotion of anger, while the heart is linked with Zhi and joy. CFEM aims to balance the body by using music tones that correspond to specific organ-emotion systems. The music is thought to regulate the flow of qi and blood, promote relaxation, and improve emotional health.^[9] Recent studies have demonstrated the therapeutic potential of CFEM for psychological symptoms. Liu et al reported that CFEM reduced depression scores and improved emotional well-being in elderly patients with recurrent spontaneous abortion.^[10] Zhang et al found that CFEM outperformed western music in reducing anxiety, depression, and emotional distress among hemodialysis patients.^[11] Similarly, Liao et al showed that CFEM improved Quality of Life (QOL) and reduced symptom burden in advanced cancer patients.^[12] Although there have been some related systematic reviews, their scope has been limited to patients with specific cancers, diabetes, and individuals with perinatal psychological subhealth. This study aims to conduct a meta-analysis of the effectiveness of CFEM intervention on older adults with depression and anxiety.

2. Methods

This study was reported in accordance with the PRISMA 2020 guidelines for the systematic review component, and followed standard practices for conducting and reporting meta-analyses.^[13] The completed PRISMA checklist is included in the Supplementary Material (Supplemental Digital Content, https://links.lww.com/MD/Q81). The protocol was preregistered in PROSPERO (CRD42023435684). Although the initial registration was based on the mid-night noon ebb-flow theory as a guiding framework for CFEM, the final analysis adopted a broader TCM-based perspective due to the limited number of studies directly applying that theory.

2.1. Search methods

A broad literature search was initially conducted to map the available evidence, which informed the subsequent systematic review and meta-analysis. Two independent reviewers searched for literature in the Web of Science, PubMed, CNKI, Wanfang, and CQVIP databases from inception to January 2025. To identify relevant literature, the following key search terms and MeSH headings were used (“depression*” OR “depressive symptom” OR “emotional depression”) OR (“anxiety*” OR “angst” OR “social anxieties” OR “hypervigilance” OR “nervousness” OR “anxiousness”) AND (“five-element music*” OR “five-element music therapy” OR “five line music” OR “five tune music” OR “five tone music” OR “five phase music” OR “Chinese music” OR “Chinese traditional music” OR “traditional Chinese medicine five-element music therapy.”).

2.2. Inclusion and exclusion

2.2.1. Studies types

This study collected all randomized controlled trials (RCTs) using five-element music to intervene in depression or anxiety in older adults aged 50 years and above, without restriction on publication language, and blinding.

2.2.2. Participant types

Participants were diagnosed with depression and anxiety disorders without limitations regarding race or region.

2.2.3. Intervention types

Patients in the intervention group received CFEM intervention, either as a standalone treatment or in combination with other therapies. The types, duration, and frequency of the music interventions were not restricted. The control group included patients who did not receive CFEM interventions, usual care, or no intervention.

2.2.4. Outcome types

Hamilton Depression Scale (HAMD); Hamilton Anxiety Scale (HAMA); Zung Self-Rating Depression Scale (SDS); Self-Rating Anxiety Scale (SAS); Pittsburgh Sleep Quality Index (PSQI); overall response rate (ORR); QOL. ORR was accepted only when it was clearly defined in the original study, typically based on a reported percentage of patients showing a predefined level of symptom improvement.

2.2.5. Exclusion criteria

Review articles or experimental design papers. Duplicate publications. Articles with unavailable full text or data. Non-RCTs. Studies with poorly-reported data, repeated reports, or incomplete and unusable fundamental data.

2.3. Data extraction and quality assessment

The literature and extracted data were independently screened and cross-checked based on the inclusion and exclusion criteria. The data extraction included the first author, publication year, country of publication, sample size, participant characteristics, intervention characteristics, and outcome measures. The risk of bias in eligible studies were independently assessed using the Cochrane Handbook for Systematic Reviews of Interventions.^[14] The assessment items included: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective outcome reporting, and other biases. Each bias risk was classified as low, unclear, or high risk. Any discrepancies were resolved through discussion or consultation with a third researcher.

2.4. Statistical analysis

This study’s data was statistically analyzed through the RevMan 5.3 software. When the variables in the included studies were continuous and assessed using different scales, standardized mean differences were calculated using Hedges’ g, which adjusts for small sample bias and is the default SMD metric in RevMan. Conversely, if the variables were continuous and measured using the same scales, the mean difference (MD) with a 95% confidence interval (95% CI) was employed. This study used chi-square tests and I² values to analyze the heterogeneity of included studies. Meta-analyses were conducted using a random-effects model, which accounts for potential clinical and methodological heterogeneity across studies. This decision was made a priori based on the variability in interventions (e.g., different combinations of CFEM and co-interventions), populations, and study settings. The I² statistic and its 95% CI were used to assess the degree of heterogeneity and guide interpretation of the results. As recommended by the Cochrane Handbook, the choice of statistical model was not determined by the I² value alone. Subgroup analysis was conducted to identify possible sources of heterogeneity if needed. Additionally, funnel plots were utilized to assess publication bias when more than 10 studies were included. The significance level for the meta-analysis was set at α = 0.05. Confidence intervals for I² statistics were calculated to assess the precision of heterogeneity estimates, in line with the Cochrane Handbook recommendations.

2.5. Assessment of evidence quality

The quality of evidence was assessed using the GRADE (Grading of Recommendations Assessment, Development and Evaluation) approach, following the Cochrane Handbook for Systematic Reviews of Interventions. Five domains were evaluated: risk of bias, inconsistency, indirectness, imprecision, and publication bias. The certainty of evidence was assessed only for the primary outcomes and categorized as high, moderate, low, or very low. A summary of findings table presenting the GRADE ratings is provided in Supplementary Material (Supplemental Digital Content, https://links.lww.com/MD/Q81).

3. Results

3.1. Search process

The PRISMA flowchart illustrates the selection process. Initially, 2829 records were identified from the database, and 1970 records remained for screening after removing 531 duplicates. Following title and abstract screening, 245 reports were retrieved for further evaluation. In the eligibility stage, 169 reports were assessed for inclusion, with 23 excluded after full-text screening; 49 review articles were removed, and 79 studies that did not focus on CFEM intervention were excluded. Ultimately, 18 studies were included in the analysis. This flowchart clearly outlines the various stages of the study selection process and the exclusion criteria.

3.2. Study characteristics

Table 1 shows the characteristics of the study. The studies included in this meta-analysis varied in design and intervention details. The sample sizes for the experimental and control groups ranged from 20 to 60 participants per group. The participants’ ages were generally between 50 and 80 years, with a few studies specifying mean ages for both groups. The interventions in the experimental groups primarily involved the use of CFEM combined with other therapies, such as Progressive Muscle Relaxation Training, Baduanjin exercise, or TCM. In contrast, the control groups typically received conventional treatments, such as progressive muscle relaxation, conventional nursing care, or exercise without CFEM. Interventions were administered at varying frequencies, with most studies applying CFEM for about 20 to 30 mins daily for 4 to 8 weeks. Evaluation times were generally reported after the intervention was completed.

Table 1.

Basic characteristics of included studies.

Item #	Author, year	Sample size	Age (yr)	Intervention (experimental group)	Intervention (control group)	Music types	Frequency and duration	Evaluation time	Outcomes
1	Liao et al, 2018[12]	E = 29, C = 27	E:61.07 ± 14.24 C:62.57 ± 14.15	Progressive Muscle Relaxation Training + CFEM	Progressive Muscle Relaxation Training	CFEM	20 min/d, 8 wk	4, 8 wk after intervention	Depression: HADS
2	Chen and Xiong 2017[13]	E = 36, C = 34	E: 54. 3 ± 11.1 C:52. 3 ± 10. 9	Baduanjin Exercise + CFEM	Conventional Nursing	CFEM	30 min twice a day, 4 wk	4 wk after intervention	Depression: HAMD Depression: HAMD Depression: HAMD Anxiety: HAMA
3	Jiang et al, 2018[14]	E = 35, C = 35	60.90 ± 10.86	UC + PA + CFEM	UC	Zhi, Gong Tone Music	30 min once a day, 4 wk	4 wk after intervention	Depression: HAMD Anxiety: HAMA
4	Li et al, 2019[15]	E = 30, C = 30	E:61.2 C:59.7	Exercise + CFEM	Exercise + Chinese Medicine	CFEM	30 min once a day, 4 wk	1 wk, 2 wk, 4 wk after intervention	Anxiety: HAMA Kolcaba General Comfort Questionnaire
5	Li et al, 2016[16]	E = 40, C = 40	60 ± 11.49	CFEM + UC	UC	CFEM	30 min twice a day, 1 wk	1 wk after intervention	Depression: SDS
6	Liu and Liu 2019[17]	E = 44, C = 44	E:58.1 C:58.7	Conventional Nursing + CFEM	Conventional Nursing	CFEM	30 min once a day, 4 wk	4 wk after intervention	Depression: HAMD; SDS Depression: HAMD; SDS Sleep quality: PSQI
7	Liu and Liu 2019[18]	E = 39, C = 40	E:52.3 ± 4.6 C:53.5 ± 5.1	Progressive Rehabilitation Nursing + CFEM	Progressive Rehabilitation Nursing	Yu Tone Music	30 min twice a day, 4 wk	4 wk after intervention	Depression: SDS Anxiety: SAS
8	Pan et al, 2017[19]	E = 48, C = 47	-	Conventional Nursing + CFEM	Conventional Nursing	CFEM	20–30 min twice a day, 4 wk	4 wk after intervention	Depression: HAMD
9	Tang et al, 2017[20]	E = 63, C = 61	-	Conventional Nursing + CFEM Combined with Tai Chi	Conventional Nursing	CFEM	30 min twice a day, 4 wk	4 wk after intervention	Depression: HADS Anxiety: HADS-A Distress Management (DT)
10	Wang et al, 2011[21]	E = 20, C = 25	-	CFEM + Conventional Therapy	Western Music + Conventional Therapy	CFEM	30 min twice a day, 9 d	9 d after intervention	Depression: SDS Anxiety: SAS
11	Wang et al, 2015[22]	E = 36, C = 36	E:54.97 ± 9.36 C:58.31 ± 6.96	Conventional Therapy + CFEM + Ear Acupressure	Conventional Therapy	Gong, Shang, Jue, Gong Tone Music	30 min twice a day, 10 d	10 d after intervention	Depression: SDS Anxiety: SAS Quality of life: QLQ-C30
12	Wang 2017[23]	E = 50, C = 50	E:51.4 ± 8.3 C:53.6 ± 9.5	Progressive Muscle Relaxation Training + CFEM	Progressive Muscle Relaxation Training	CFEM	30 min once a day, 8 wk	8 wk after intervention	Quality of Life: Short Form-36 Item
13	Wen et al, 2016[24]	E = 35, C = 35	E:51. 56 ± 11.45 C:50. 89 ± 10.58	Conventional Therapy + CFEM	Conventional Therapy	CFEM	20–30 min twice a day, 3 d	3 d after intervention	Depression: SDS Anxiety: HAMA; SAS
14	Wen et al, 2017[25]	E = 20, C = 20	64.5 ± 3.4	CFEM + TCM	TCM	CFEM	24 min once a day, 3 wk	3 wk after intervention	Depression: HAMD
15	Chen et al, 2019[27]	E = 60, C = 60	E:87.28 ± 3.12 C:80.21 ± 5.48	UC + CFEM	UC	CFEM	30 min twice a day, 4 wk	4 wk after intervention	Depression: HAMD; SDS
16	Zhai and Liu 2017[28]	E = 53, C = 53	58.3 ± 10.2	UC + CFEM	UC	CFEM	30 min once a day, 4 wk	4 wk after intervention	Depression: HAMD
17	Xu et al, 2019[29]	E = 60, C = 60	E:58. 9 ± 6.7 C:58.7 ± 6.9	Conventional Nursing + CFEM	Conventional Nursing	CFEM	30 min once a day, 45 d	45 d after intervention	Depression: SDS Anxiety: SAS Sleep quality: PSQI
18	Yang et al, 2016[25]	E = 47, C = 21	E:75.15 ± 7.15 C:76.48 ± 733	CFEM + Conventional Therapy	Conventional Therapy	CFEM	40 min twice a day, 4 wk	4 wk after intervention	Depression: HAMD; SDS

This summary presents the characteristics of studies involving Chinese Five-Element Music (CFEM) interventions. It includes authors, sample sizes, participant ages, intervention types, music used, frequency/duration, evaluation periods, and measured outcomes. Summary of findings table (GRADE Assessment): Summary of findings based on GRADE assessment includes number of studies, total sample size, effect estimates (MD or OR) with 95% CI, and certainty of evidence. Downgrade reasons reflect methodological limitations.

C = control group, E = experimental group, GRADE = grading of recommendations assessment, development and evaluation, MD = mean difference.

3.3. Study quality

The risk of bias assessment for the included studies is presented in Figures 1 and 2. Regarding random sequence generation, 9 studies^[15–23] were judged as low risk, while 2 studies^[24,25] were rated as high risk due to grouping patients according to the order of admission. The remaining studies were classified as having an unclear risk. Six studies^{[15–17,19,20,22]} mentioned allocation concealment using random numbers, but the remaining studies showed an unclear risk of bias. In terms of participant and personnel blinding, all studies were judged as having an unclear risk, as no detailed descriptions were provided. Nine studies^{[15,18,19,24–26]} demonstrated a low risk of bias for outcome data and assessment, indicating that dropouts and withdrawals were appropriately managed. The remaining studies exhibited an unclear risk.

Figure 1.

Cochrane risk of bias. Summary of risk of bias across all included studies using the Cochrane Risk of Bias tool. The chart shows the proportion of studies rated as low risk (green), unclear risk (yellow), and high risk (red) for each bias domain: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting, and other bias.

Figure 2.

Cochrane Risk of Bias Summary. Risk of bias assessment for each included randomized controlled trial. Each row represents a study and each column represents a specific bias domain. Green circles indicate low risk, yellow circles indicate unclear risk, and red circles indicate high risk.

3.4. Efficacy analysis

3.4.1. Depression assessment results

The forest plot (Figs. 3 and 4) presents the results of a meta-analysis comparing the experimental and control groups. Each study reported the mean, standard deviation, and sample size for both groups, along with the calculated MD and its 95% CI. Eight RCTs^{[15,17,23,25–29]} used the HAMD scale to compare 657 older adult patients with depression and anxiety disorders. The meta-analysis results from the random-effects model showed that the HAMD scale of CFEM had a significant effect on older adult patients with depression (MD = −3.63, 95% CL: −4.97 to −2.3, P < .00001).

Figure 3.

HAMD Scale. Forest plot comparing the HAMD outcomes between experimental and control groups across 8 RCTs. CFEM intervention significantly reduced HAMD scores compared to controls (MD = −3.63, 95% CI: −4.97 to −2.3, P < .00001). CFEM = Chinese Five-Element Music, CI = confidence interval, HAMD = Hamilton Depression Rating Scale, MD = mean difference, RCTs = randomized controlled trials.

Figure 4.

SDS Scale. Forest plot summarizing the effect of CFEM on SDS scores across 9 RCTs. Results showed a significant reduction in SDS scores for the experimental group (MD = −6.69, 95% CI: −8.32 to −5.07, P < .00001). CFEM = Chinese Five-Element Music, CI = confidence interval, MD = mean difference, RCTs = randomized controlled trials, SAS = self‐rating anxiety scale, SDS = self‐rating depression scale.

Nine RCTs^{[16–19,21–23,25,30]} used the SAS scale to compare 702 older adult patients with depression and anxiety disorders. The meta-analysis results from the random-effects model indicated that CFEM significantly reduced SDS scores in older adults with depression and anxiety (MD = −6.69, 95% CI: −8.32 to −5.07, P < .00001).

Visual inspection of the funnel plots did not reveal clear asymmetry. However, due to the small number of studies and the lack of variation in study precision, the risk of publication bias remains inconclusive, which as presented in Figures 5 and 6.

Figure 5.

Funnel plot of HAMD scale. Funnel plot evaluating potential publication bias in studies reporting HAMD outcomes. The distribution appears approximately symmetrical, suggesting a low risk of publication bias. HAMD = Hamilton Depression Rating Scale.

Figure 6.

Funnel plot of SDS scale. Funnel plot for studies reporting SDS outcomes. The plot shows no clear asymmetry, indicating minimal likelihood of publication bias. SDS = self‐rating depression scale.

3.4.2. Anxiety assessment results

Four RCTs^{[15,23,24,27]} used the HAMA scale to evaluate 230 older adult patients with depression and anxiety disorders. The meta-analysis results using a random-effects model indicated that CFEM significantly reduced HAMA scores (MD = −3.89, 95% CI: −5.80 to −1.97, P < .0001; Fig. 7).

Figure 7.

HAMA Scale. Forest plot summarizing the effects of CFEM on Hamilton Anxiety Rating Scale (HAMA) outcomes across 4 studies. A significant reduction in HAMA scores was observed (MD = −3.89, 95% CI: −5.80 to −1.97, P < .0001). CFEM = Chinese Five-Element Music, CI = confidence interval, HAMA = Hamilton Anxiety Rating Scale, MD = mean difference.

Five RCTs^{[18,19,21,22,30]} used the SAS scale to assess a total of 346 older adult patients with depression and anxiety disorders. The results from the random-effects model revealed that CFEM significantly reduced SAS scores compared to the control group (MD = −6.84, 95% CI: −10.76 to −2.92, P = .0006). However, substantial heterogeneity was observed among the included studies (I² = 90%, 95% CI: 75% to 96%), indicating considerable variability in the effect sizes (Fig. 8).

Figure 8.

SAS scale. Forest plot analyzing SAS scores from 5 studies. The meta-analysis showed a significant effect in favor of CFEM (MD = −6.84, 95% CI: −10.76 to −2.92, P = .0006), although with high heterogeneity. CFEM = Chinese Five-Element Music, CI = confidence interval, MD = mean difference, SAS = self‐rating anxiety scale.

Visual inspection of the funnel plot for the Hamilton Anxiety Scale and Self-Rating Anxiety Scale outcomes revealed a roughly symmetrical distribution, suggesting a low risk of publication bias (Figs. 9 and 10).

Figure 9.

Funnel plot of HAMA Scale. Funnel plot assessing publication bias among studies evaluating the HAMA outcomes. The studies are distributed symmetrically along the vertical axis, indicating a low risk of publication bias. HAMA = Hamilton Anxiety Rating Scale.

Figure 10.

Funnel plot of SAS scale. Funnel plot illustrating potential publication bias for studies reporting SAS outcomes. The distribution appears roughly symmetrical, suggesting minimal likelihood of publication bias. SAS = self-rating anxiety scale.

3.4.3. Overall response rate (ORR)

The forest plot (Fig. 11) presents the results of a meta-analysis of 4 RCTs^{[19,23,27,29]} and compares the clinical effectiveness of CFEM intervention versus control, measured as a dichotomous outcome. The data was analyzed using a fixed-effects model, and the pooled ORR was 2.80 (95% CI: 1.61 to 4.89, P = .0003), indicating that the intervention significantly increased the likelihood of clinical improvement compared to the control group. The total number of effective cases was 95 out of 138 in the experimental group and 48 out of 112 in the control group. The heterogeneity was recorded as low (I² = 0%, 95% CI: 0% to 72%), suggesting consistency across the included studies.

Figure 11.

Overall response rate. Forest plot showing the odds ratios for overall clinical response in 4 randomized controlled trials. The pooled result indicates that CFEM significantly increased the overall response rate compared to control interventions (OR = 2.80, 95% CI: 1.61 to 4.89, P = .0003). No heterogeneity was observed (I² = 0%). CFEM = Chinese Five-Element Music, CI = confidence interval, MD = mean difference.

3.4.4. Prognostic assessment

The forest plot (Fig. 12) presents the results of a subgroup meta-analysis examining the effects of CFEM on sleep quality using the PSQI and QOL. Two RCTs^[17,22] assessed PSQI outcomes in participants. The pooled analysis using a random-effects model showed a significant improvement in PSQI scores for the experimental group compared to the control group (MD = −1.86, 95% CI: −2.54 to −1.17, P < .00001), with moderate heterogeneity (I² = 48%, 95% CI: 0% to 86%).

Figure 12.

PSQI and QOL scale. Forest plot depicting the pooled effects of CFEM on sleep quality (PSQI) and QOL. The intervention significantly improved PSQI scores (MD = −1.86, 95% CI: −2.54 to −1.17, P < .00001), while no significant effect was observed on QOL (MD = −0.58, 95% CI: −15.65 to 14.48, P = .94). Substantial heterogeneity was present in the QOL subgroup. CFEM = Chinese Five-Element Music, CI = confidence interval, MD = mean difference, PSQI = Pittsburgh Sleep Quality Assessment Scale, QOL = quality of life.

Two additional RCTs^[19,20] evaluated QOL in 172 participants. The results indicated no significant difference between the groups (MD = −0.58, 95% CI: −15.65 to 14.48, P = .94), and substantial heterogeneity was observed (I² = 89%, 95% CI: 30% to 98%).

The overall pooled analysis across all 4 studies^{[17,19,20,22]} showed a statistically significant benefit of the intervention (MD = −2.71, 95% CI: −4.55 to −0.86, P = .004), although heterogeneity was high (I² = 85%, 95% CI: 53% to 96%). Subgroup analysis did not reveal a statistically significant difference between PSQI and QOL outcomes (P = .87).

4. Discussion

The aim of this systematic review and meta-analysis is to investigate the effects of CFEM interventions for individuals suffering from depression and anxiety disorders to gather evidence of their effectiveness. It also provides an overview of the use of CFEM as an intervention. The results showed significant improvements in HAMD, HAMA, SAS, and SDS scores in the intervention groups compared to control groups. This meta-analysis focuses on the application of CFEM specifically for older adults with depression and anxiety disorders. While previous meta-analyses have primarily focused on populations such as perinatal women or cancer patients with comorbid depression and anxiety, this study uniquely addresses the mental health challenges faced by older adults with mood disorders. Given the increasing prevalence of depression and anxiety among older adults and the growing interest in nonpharmacological interventions, this study provides novel insights into the potential benefits of CFEM within this demographic.

Music therapy has a long history in Chinese medicine, with the theory of 5 sounds for healing being proposed in the Inner Canon of Huangdi. The 5 elements of TCM music are said to influence 5 organs (liver, heart, spleen, lung, and kidney) and the 5 spirits of patients to prevent, treat, and cure illnesses.^[31] In ancient Chinese music, the 5 tones of Gong, Shang, Jue, Zhi, and Yu correspond to the attributes of the 5 organs. The relationship between emotional and mental activities and the 5 organs is as follows: the liver is associated with anger, the heart with joy, the spleen with thought, the lung with worry, and the kidney with fear.^[10] The 5 tones are used to regulate the body’s meridians and maintain the normal functioning of qi and blood, achieving a balanced state of yin and yang. The tone of Jue, which is connected to the liver, has the effect of draining the liver and tonifying qi, as well as pacifying the liver and strengthening yang. The tone of Shang, associated with the lung, promotes the internal collection of qi throughout the body, regulating the propagation and purification of lung qi, and has the effect of protecting the kidney and suppressing the liver. The tone of Yu, connected to the kidney, promotes the subduction of qi throughout the body, strengthens the function of the spirit, and helps the liver’s yin.^[32] When the thoughts and emotions expressed in music resonate with a person’s emotions, the effects of music therapy are further amplified, allowing patients to empathize with the emotions conveyed in the music, thereby alleviating anxiety, depression, and other symptomatic manifestations.^[33]

CFEM therapy operates through sound waves on nerve cells in the cerebral cortex, affecting all parts of the body through coordinated rhythmic, melodic, and forceful acoustic vibrations, causing harmonious resonance of the body’s organs, muscles, and brain waves. Consequently, the dysfunctional state of various organs is improved. The effects of five-element music include pain relief, reduced attention to undesirable stimuli, distraction from sad moods and worries, and feelings of relaxation. This is the foundational theoretical basis for the effectiveness of Five Elements Music Therapy.^[34] CFEM is easily accepted by patients because it is simple, convenient, economical, and safe. Additionally, it avoids the adverse effects associated with acupuncture and medication.^[35] Currently, CFEM has been widely used in the treatment and care of various diseases and has proven effective in specifically treating mental illnesses, depression, insomnia, and subhealth.^[11]

We found that although all included studies used CFEM as a music intervention method, the tuning is different. Almost all of the music interventions in the studies included the tone of Jue music. The Jue tone in the 5 elements of music belongs to the wood element of the 5 elements, which is connected to the liver and depicts the revival of the earth in spring.^[36] The tone of Jue music can improve the reticular structure of the limbic system and cortex of the human brain, regulate the functions of the immune system and neuroendocrine system, and weaken the sympathetic nerves. It can be effective in enhancing the secretion of parasympathetic nerves and endorphins, reducing the secretion of catecholamines in the body, thereby reducing anxiety and depression symptoms and improving sleep quality.^[37] Jue music is soothing, optimistic, cheerful, and positive. Listening to Jue music can destress the liver, tonify the heart and spleen, and regulate negative psychology, which is helpful in relieving patients’ depression and anxiety symptoms and can guide patients’ emotional state, allowing them to improve their psychological and physical well-being, thereby improving their QOL.^[38] Besides, 1 study selected 4 tones of CFEM music (Gong, Shang, Jue, Yu), another selected 2 tones (Zhi, Gong), and another only selected the Yu tone, which was related to the actual situation of the subjects involved in the study.

Additionally, some of the included studies used CFEM as an independent therapy in the intervention group, while others combined it other therapies (such as progressive muscle relaxation, Baduanjin exercise, or TCM). Although the combined therapies varied, the application of CFEM remained the core component, supporting its comprehensive effectiveness in the meta-analysis. Most interventions followed a similar protocol, with sessions lasting approximately 20 to 30 minutes, administered once or twice daily for 1 to 8 weeks. These consistent parameters suggested a common practical framework for CFEM in clinical settings. Subgroup analyses further demonstrated that CFEM had positive effects on sleep quality as measured by the PSQI scale. However, while there was some evidence of improved QOL, the effect was not statistically significant, and substantial heterogeneity was noted. This variability may be attributed to differences in intervention duration, participant characteristics, or concurrent treatments.

In this study, the funnel plot was symmetrically distributed, indicating that there was no publication bias in the included literature. However, this study included 2 pieces of literature with high bias risks. The rigor of the relevant research on CFEM music intervention for anxiety and depression in older adults needs further improvement. Moreover, the heterogeneity observed in some outcomes suggested that methodological inconsistencies and variations in intervention implementation could have influenced the results. Most of the included trials were conducted in China and published in Chinese, which may limit the generalizability of the findings to other cultural or healthcare contexts. Additionally, the absence of long-term follow-up data prevents conclusions about the sustained benefits of CFEM. Despite these limitations, the findings of this systematic review and meta-analysis support the integration of CFEM as a complementary nonpharmacological approach for managing depression and anxiety in older adults. Future research should aim to standardize intervention protocols, improve study design rigor, and explore the long-term effects of CFEM across diverse populations.

5. Limitation

The authors found several limitations that were common among current studies. First, these studies generally lacked long-term efficacy observation and follow-up. Few studies included a long-term follow-up after the completion of music therapy, leading to no data on the probability of disease recurrence. Second, most studies included patients with 1 type of specific illness, the generalization of the study results may be limited. And the samples for these studies were generally small. Sample sizes should be expanded where possible to obtain more accurate data. Besides, the clinical treatment lacks standardization, and there are no standardized criteria either for the selection of music in the music intervention process or the timing of interventions. At the same time, musical forms are monotonous and lack variety. Furthermore, patients are often unaware of music therapy. In most of the studies, most of the five-element music intervention trials were conducted in a ward where the volume effect can be influenced by the surrounding environment, and it is recommended that a dedicated music therapy treatment room be set up where available to reduce interference from environmental factors.

In addition, this study has several limitations that should be acknowledged. First, the exclusion of gray literature, such as dissertations, conference proceedings, or unpublished trial data. Although our database search was extensive, the omission of gray sources may contribute to publication bias and limit the comprehensiveness of the findings. Future reviews could benefit from incorporating strategies to identify and include gray literature. Second, some eligible reports could not be retrieved during the full-text screening stage. Although we made attempts to contact corresponding authors and use institutional channels, we were unable to obtain access to all full texts. This may have affected the comprehensiveness of the final dataset. Besides, since the music intervention methods used in the studies included in this review are unique to China, most of the literature included in the analysis comes from CNKI and Wanfang database. This regional concentration may introduce bias and limit the international applicability of the results.

Another important limitation lies in the quality of the included studies. The risk-of-bias assessment revealed a large number of items rated as “unclear,” particularly in domains such as allocation concealment, blinding, and selective reporting. These unclear risks reflect either a lack of transparency in reporting or insufficient methodological rigor, both of which diminish the overall confidence in the pooled results. Future studies should prioritize detailed reporting following CONSORT or similar guidelines to improve evaluability and credibility. Furthermore, the ability to assess publication bias was limited. Although funnel plots were generated, the number of included studies was small and the variation in precision was minimal. As such, the funnel plots do not provide conclusive evidence regarding publication bias.

The CFEM interventions included in this review shared a common theoretical framework rooted in TCM, but the exact implementation protocols varied across studies. Similarly, the comparison groups included standard care, psychological therapy, and other nonpharmacological interventions. This methodological heterogeneity may limit the interpretability of pooled effect sizes, as the interventions and controls are not entirely uniform. To address this, we used a random-effects model to reflect the diversity across studies. Nevertheless, the results should be interpreted with caution, and the pooled estimates are better understood as indicating general trends rather than precise effect sizes. Future trials with more consistent designs and standardized intervention protocols will help improve the reliability of meta-analytic syntheses in this field.

6. Conclusion

This systematic review and meta-analysis demonstrates that CFEM has significant positive effects on reducing symptoms of depression and anxiety among older adults. The intervention also beneficially impacted sleep quality, though its influence on overall QOL remains inconclusive. As a low-cost, low-side effect, and easy-to-implement therapy, CFEM deserves to be promoted in clinical practice. Despite certain methodological limitations and heterogeneity across studies, the findings highlight CFEM as a promising complementary approach to conventional treatments. Future high-quality RCTs with standardized protocols and long-term follow-up are needed to further validate its effectiveness and support its integration into clinical practice.

Author contributions

Conceptualization: Jiayin Li.

Data curation: Jiayin Li.

Formal analysis: Jiayin Li.

Methodology: Jiayin Li.

Software: Jiayin Li.

Supervision: I Ta Wang, Wen Fen Beh.

Validation: Jiayin Li, Kanghao Chen, Wen Fen Beh.

Visualization: Jiayin Li, Wen Fen Beh.

Writing – original draft: Jiayin Li, Kanghao Chen.

Writing – review & editing: Wen Fen Beh.

Supplementary Material

medi-104-e44764-s001.docx ^{(44KB, docx)}