22 Novembre 2026

Psicologia Punto One | Psicologia e Psicologi nel Web

Menu

Menu

SCHEDA SCIENTIFICA2026Metanalisi

Music-based interventions for acute and short-term anxiety in nursing students: a systematic review and meta-analysis.

Chen B, Wu S, Zou Z, Gao H

Frontiers in public health · 2026

Abstract originale

BACKGROUND: Nursing students encounter acute anxiety during examinations, simulations, clinical skills training, and other health-related student contexts. We evaluated music-based interventions, with acute or short-term self-reported anxiety as the principal outcome and other psychological, physiological, health, and educational outcomes as secondary domains. METHODS: Following peer review, we repeated searches in PubMed/MEDLINE, CINAHL Complete, Embase, APA PsycINFO, the Cochrane Central Register of Controlled Trials (CENTRAL), the Web of Science Core Collection, and ClinicalTrials.gov from their inception to 19 September 2026, without outcome, study design, publication type, or language limits. Two reviewers independently screened, extracted, and assessed the risk of bias. The principal meta-analysis used a uniform post-intervention Hedges' g from each eligible, individually randomized parallel-group trial. Random-effects models were estimated using Restricted Maximum Likelihood or Residual Maximum Likelihood (REML); Wald confidence intervals were the primary estimate, and Hartung-Knapp and prediction intervals were used to assess small-sample uncertainty. RESULTS: Of the 737 records identified, 337 duplicates were removed, and 400 records were screened. A total of 64 reports were sought; 39 unique complete documents were assessed; 26 reports representing 24 studies were included; and 13 full-text reports were excluded. Twenty studies were randomized, and four were non-randomized. Nine parallel-group trials (650 participants) contributed to the principal anxiety synthesis (g = -0.31, 95% confidence interval (CI): -0.58 to -0.04; p = 0.024; I 2 = 64%; τ 2 = 0.107). The Hartung-Knapp interval (from -0.63 to 0.01; p = 0.055) and the 95% prediction interval (from -1.15 to 0.53) crossed zero. All nine principal trials were at high overall risk of bias, and certainty was very low. CONCLUSION: Music-based interventions may reduce acute or short-term self-reported anxiety on average. However, the results are sensitive to the method used to quantify uncertainty, and the effects are likely to vary across settings. Evidence is insufficient to infer sustained psychological benefits, improved learning, or support for routine implementation. SYSTEMATIC REVIEW REGISTRATION: The review was registered in PROSPERO (CRD420251274752). Public record URL: https://www.crd.york.ac.uk/PROSPERO/view/CRD420251274752.

Parole chiave

music interventionmusic therapyanxietynursing studentsacute anxietyself-reported anxietynursing educationMusic Therapy and HealthDiverse Music Education InsightsMusic Education and AnalysisHumansAnxietyStudents, NursingMusic Therapyhealth workforcemeta-analysismusic listeningsystematic reviewMeta-analysisSystematic reviewNursing StudentsHealth WorkforceMusic Listening
Leggi il testo integrale originale

Testo estratto dalla pubblicazione: sono esclusi immagini, media, supplementi e parti soggette a diritti distinti.

Music-based interventions for acute and short-term anxiety in nursing students: a systematic review and meta-analysis

Background

Nursing students encounter acute anxiety during examinations, simulations, clinical skills training, and other health-related student contexts. We evaluated music-based interventions, with acute or short-term self-reported anxiety as the principal outcome and other psychological, physiological, health, and educational outcomes as secondary domains.

Methods

Following peer review, we repeated searches in PubMed/MEDLINE, CINAHL Complete, Embase, APA PsycINFO, the Cochrane Central Register of Controlled Trials (CENTRAL), the Web of Science Core Collection, and ClinicalTrials.gov from their inception to 19 September 2026, without outcome, study design, publication type, or language limits. Two reviewers independently screened, extracted, and assessed the risk of bias. The principal meta-analysis used a uniform post-intervention Hedges’ g from each eligible, individually randomized parallel-group trial. Random-effects models were estimated using Restricted Maximum Likelihood or Residual Maximum Likelihood (REML); Wald confidence intervals were the primary estimate, and Hartung–Knapp and prediction intervals were used to assess small-sample uncertainty.

Results

Of the 737 records identified, 337 duplicates were removed, and 400 records were screened. A total of 64 reports were sought; 39 unique complete documents were assessed; 26 reports representing 24 studies were included; and 13 full-text reports were excluded. Twenty studies were randomized, and four were non-randomized. Nine parallel-group trials (650 participants) contributed to the principal anxiety synthesis (g = −0.31, 95% confidence interval (CI): −0.58 to −0.04; p = 0.024; I2 = 64%; τ2 = 0.107). The Hartung–Knapp interval (from −0.63 to 0.01; p = 0.055) and the 95% prediction interval (from −1.15 to 0.53) crossed zero. All nine principal trials were at high overall risk of bias, and certainty was very low.

Conclusion

Music-based interventions may reduce acute or short-term self-reported anxiety on average. However, the results are sensitive to the method used to quantify uncertainty, and the effects are likely to vary across settings. Evidence is insufficient to infer sustained psychological benefits, improved learning, or support for routine implementation.

Systematic review registration

The review was registered in PROSPERO (CRD420251274752). Public record URL: https://www.crd.york.ac.uk/PROSPERO/view/CRD420251274752.

Introduction

Investing in nursing education is a health-workforce priority because nurses are central to health service delivery and progress toward universal health coverage (1). Nursing students encounter examinations, simulations, invasive-skills training, and early clinical placements while managing workloads, responsibilities, observations, and fears of making errors (2–5). Anxiety and related distress are common across nursing programs (6–9). Support for students’ mental health is therefore relevant to both their participation in training and the future nursing workforce.

In this review, music-based intervention is an umbrella term, not a claim of equivalence. We distinguish passive music listening or music medicine, active music making, and therapist-delivered music therapy. The last category requires a structured therapeutic intervention delivered by a music therapist; researcher- or educator-delivered listening is not labeled as formal music therapy (10, 11). We also identify trials in which music was added to a shared co-intervention, such as aromatherapy, because these comparisons estimate only the incremental contribution of music.

Music may redirect attention away from threat cues, support emotion regulation, increase perceived control when choice is offered, and alter autonomic arousal (12–17). The same attentional mechanism may also impose cognitive load or obscure task-relevant auditory information. Listening before a stressor may therefore differ from listening during simulation, skills performance, assessment, or clinical activity, where distraction could impair instruction, communication, alarms, or performance. These mechanisms were proposed more often than they were measured across the included studies.

Earlier reviews addressed test anxiety and performance in nursing students (18, 19). A 2026 meta-analysis across health sciences programs included 24 nursing-student reports (20). After conducting a corrective search and full-text reassessment, 15 of the 24 reports were included in the present review (resulting in a 62.5% overlap within each review, with a Jaccard overlap of 45.5% across 33 unique reports). The contribution of this review is therefore a nursing-specific, outcome-free update with design verification, separate analyses of randomized and non-randomized evidence, unit-of-analysis safeguards, an explicitly defined principal anxiety estimand, small-sample sensitivity analyses, and the Grading of Recommendations Assessment, Development and Evaluation (GRADE), rather than a largely independent evidence base.

The primary review question is as follows: among nursing students, do passive music listening, active music making, structured therapist-delivered music therapy, or music-based educational interventions, compared with no music, usual activity, an attention or sensory control, or a matched co-intervention, reduce acute or short-term self-reported anxiety at the first eligible post-intervention assessment? Other psychological, physiological, student health, and educational outcomes were secondary domains and were not interpreted as evidence that reduced anxiety improves competence or sustains mental health.

Methods

Protocol and reporting

The protocol was prospectively registered in PROSPERO (CRD420251274752), and the review was reported in accordance with the PRISMA 2020 guidelines (21). In response to peer review, we revised the electronic search by removing outcome and study-design restrictions, while retaining acute or short-term anxiety as the principal outcome. The amendment, search date, and outcome-selection hierarchy were documented before the corrective search was conducted.

Eligibility criteria and outcome domains

Participants were nursing students or student nurses at any stage of preregistration education. Eligible interventions included passive music listening or music medicine, active music making, structured therapist-delivered music therapy, and a music-based educational intervention. Music added to a co-intervention was eligible only when a comparator receiving the same co-intervention isolated the incremental music component. Studies using music solely as a metronomic or mnemonic cue were excluded if the intervention did not target psychological, physiological, student health, or broader educational outcomes. Comparators included no music, silence, usual education or activity, attention or sensory controls, and matched active controls.

The principal outcome was acute or short-term, self-reported state or situational anxiety at the first eligible post-intervention assessment. The context could occur during an examination, a clinical-skill simulation, an early clinical activity, or a student health event. Trait anxiety and test-specific anxiety were reported separately when they did not align with the selected acute construct. Stress, mood, burnout, pain, sleep, comfort, physiological measures, biomarkers, self-efficacy, knowledge, skills performance, and academic outcomes were secondary domains. Randomized and non-randomized controlled studies were eligible to be included; uncontrolled studies, reviews, editorials, case reports, and reports without an eligible controlled comparison were excluded.

Information sources and search strategy

We searched PubMed/MEDLINE, CINAHL Complete, Embase, APA PsycINFO, the Cochrane Central Register of Controlled Trials, the Web of Science Core Collection, and ClinicalTrials.gov from their inception through 19 September 2026. The corrective strategies combined only music-based interventions and nursing-student concepts. No outcome, study design, publication type, or language filters were applied. Complete platform-specific syntax, execution dates, and retrieval counts are provided in Supplementary Table S1.

The seven sources provided a total of 737 records: 114 from PubMed/MEDLINE, 191 from CINAHL Complete, 167 from Embase, 72 from CENTRAL, 36 from APA PsycINFO, 133 from the Web of Science Core Collection, and 24 from ClinicalTrials.gov. Database exports were preserved and deduplicated at the report level using DOI or registry identifiers, exact bibliographic matching, and a manual review of similar entries. Reference and companion-report checks supplemented electronic retrieval. The corrective search was a post-peer-review amendment and is reported accordingly.

Study selection and data extraction

Two reviewers independently screened 400 titles and abstracts and 39 unique full documents; they extracted study and outcome data, and resolved disagreements by discussion; a third reviewer was not required. Of the 64 reports sought, 23 could not be retrieved as complete documents, one was available only as a conference abstract, and one was an exact duplicate. Reports were linked to study families using authors, setting, sample, recruitment dates, intervention, outcomes, registry identifiers, and companion publication information. We did not impute unpublished numerical outcome data.

When several anxiety measures were eligible, the hierarchy was as follows: the study-designated primary acute anxiety outcome; a validated state or situational measure; or a test- or event-specific anxiety measure. We selected the first postintervention assessment and did not count any participant twice. Trait anxiety and stress were not used as substitutes for the principal anxiety construct (22, 23). We selected the first postintervention assessment and did not count the same participants twice. Accordingly, trait anxiety and stress were not used as substitutes for the principal anxiety construct. In multi-arm trials, the comparison that isolated music was selected; for Öner et al. and Son et al., aromatherapy plus music was compared with aromatherapy alone.

Risk of bias assessment

Two reviewers independently assessed each target result using Risk of Bias 2 (RoB2), including the cluster and crossover variants where applicable, or Risk of Bias in Non-Randomized Studies of Interventions (ROBINS-I) (24–26). The target effect was assignment to the music-based intervention for the selected outcome. Prespecified confounders for non-randomized studies included baseline outcome severity, age, and sex; academic level or prior performance; musical experience or preference; and concurrent stress-management activities. Domain judgments and rationales are reported in Supplementary Tables S4, S5.

Reporting bias assessment

Funnel plots and regression tests were performed only when at least 10 sufficiently comparable studies were available (27). No synthesis met this threshold. Therefore, reporting bias was judged as undetermined and was not downgraded solely because formal testing was unavailable.

Certainty of evidence

We used GRADE to assess the risk of bias, inconsistency, indirectness, imprecision, and publication bias (28). Publication bias was downgraded only when independent evidence supported a concern, not merely because fewer than 10 studies were available.

Data synthesis

The principal synthesis included one effect from each individually randomized parallel-group trial that reported usable postintervention arm summaries. Cluster-randomized, crossover, and non-randomized studies were not pooled in this model unless valid design-adjusted effects and standard errors (SEs) were available. Hedges’ g standardized across anxiety instruments (29). For two independent groups, g = Jd, where J = 1–3/(4df − 1); its variance was (n1 + n0)/(n1n0) + g2/(2df). When only a symmetric 95% confidence interval was available, SE was reconstructed as (upper − lower)/(2 × 1.959964).

Random-effects models used inverse-variance weights wi = 1/(SEi2 + τ2), with τ2 estimated by restricted maximum likelihood. The pooled estimate was Σwiyi/Σwi. The primary 95% confidence interval used the normal quantile. An unmodified Hartung–Knapp sensitivity interval used q* = Σwi(yi − μ)2/(k − 1), SEHK = √(q*/Σwi), and a t distribution with k − 1 degrees of freedom. Heterogeneity was described with Q, I2, τ2, and a 95% prediction interval was also calculated (29, 30). The analyses used Python 3.12.13, NumPy 2.3.5, SciPy 1.17.0, and Matplotlib 3.10.8 (31).

The principal estimand was deliberately uniform: the unadjusted postintervention standardized mean difference across all included trials. As a sensitivity analysis, post-minus-pre–change-score effects were calculated for eight trials with compatible baseline and follow-up data. When a change-score SD was not reported, SDchange = √(SDpre2 + SDpost2 − 2r·SDpre·SDpost), using r = 0.50 primarily and r = 0.25 and 0.75 in sensitivity analyses. We also examined passive listening, the exclusion of add-on interventions, standalone passive listening, the exclusion of the high-risk trial from the randomization domain, and leave-one-out analyses.

The crossover trial by Lai et al. lacked paired variance or first-period data. The two cluster-randomized trials lacked sufficient cluster-adjusted information. These studies were not included in the principal pool. Stress, dysmenorrhea-related pain, physiological outcomes across different stressors, and educational constructs were not pooled because the number of comparable studies was small and clinical or methodological heterogeneity was substantial. Non-randomized effects were presented separately.

Results

Study selection

The corrective search identified 737 records. After 337 duplicates were removed, 400 titles and abstracts were screened, and 336 of these were excluded. A total of 64 reports were sought; 23 could not be retrieved. Of the 41 retrieved records, one was an abstract-only report, and one was an exact duplicate, leaving 39 unique, complete documents for full-text assessment. A total of 13 were excluded, and 26 reports representing 24 independent studies were included. Nine individually randomized parallel-group trials contributed to the principal anxiety meta-analysis (Figure 1; Supplementary Tables S6, S7).

Study and intervention characteristics

The 24 studies comprised 20 randomized studies—17 individually randomized parallel-group trials, two cluster-randomized trials, and one crossover trial—and four non-randomized controlled studies (32–57). A total of 20 articles evaluated passive or background listening, two evaluated active music-making interventions, one structured group intervention delivered by a music therapist, and one music-based auditory-training intervention. Comparators included silence, no music or usual activity, sensory or attention controls, and matched co-interventions. Intervention selection, provider, delivery, dose, tempo, and context are detailed in Supplementary Tables S2, S3.

Twelve studies were conducted in Türkiye; the remaining studies were conducted in the United States, Iran, Taiwan, the Philippines, Austria, India, and South Korea. Sex reporting was incomplete, but women formed the large majority in studies with usable data. No included study was conducted in Africa, Latin America, or Oceania. The evidence does not support reliable sex, gender, cultural, geographic, or equity analyses.

Risk of bias

Among the 20 randomized studies, 18 were judged to be at a high overall risk of bias, and two had some concerns; none was at a low overall risk. All four non-randomized studies were at a serious overall risk of bias. Common concerns were inadequate allocation or cluster handling, missing data, unblinded self-reported outcomes, selective reporting, and uncontrolled class, period, expectancy, or attention effects (Figures 2, 3; Supplementary Tables S4, S5, Figure S1).

Principal short-term anxiety outcome

Nine individually randomized parallel-group trials involving 650 participants contributed to the principal postintervention analysis. The REML estimate favored music-based intervention (Hedges’ g = −0.31, 95% CI: −0.58 to −0.04; p = 0.024; I2 = 64%; τ2 = 0.107; Q = 22.38, df = 8, p = 0.004). However, the Hartung–Knapp interval included no effect (−0.63 to 0.01; p = 0.055), as did the 95% prediction interval (−1.15 to 0.53) (Figure 4). All nine trials were at a high overall risk of bias.

Excluding the structured active music-therapy trial yielded g = −0.36 (95% CI: −0.65 to −0.08; Hartung–Knapp CI − 0.71 to −0.02; k = 8). Excluding the two add-on contrasts gave g = −0.35 (−0.65 to −0.05; Hartung–Knapp CI − 0.72 to 0.02; k = 7). Standalone passive listening gave g = −0.43 (−0.75 to −0.12; Hartung–Knapp CI − 0.84 to −0.02; k = 6). Excluding the trial at high risk in the randomization domain gave g = −0.24 (−0.51 to 0.02; Hartung–Knapp CI − 0.56 to 0.08; k = 8). Leave-one-out Wald intervals crossed zero after the omission of six of the nine trials, indicating sensitivity to individual studies (Supplementary Tables S8–S9, Figures S2–S3).

Change-score analyses were available in eight trials. The estimates remained in the direction of lower anxiety at r = 0.25, 0.50, and 0.75, but their confidence intervals included no effect. These analyses used inferred correlations across several studies and excluded Gosselin et al. 2024 because compatible baseline summaries were unavailable. The principal effect corresponds to approximately 0.31 within-study standard deviations, but no validated, cross-instrument minimally important difference was identified for nursing students.

Stress and student health outcomes

Stress results were not pooled. The randomized evidence came from menstrual stress and clinical skills contexts with markedly different magnitudes; the non-randomized lecture study was kept separate. These estimates do not support a common stress effect. Pain was not pooled either: three randomized studies evaluated primary dysmenorrhea using different intervention doses, timing, and analytical approaches. Study-specific effects and their sources are reported in Supplementary Table S10.

Other student health outcomes, including depressed mood, burnout, mood disturbance, comfort, and sleep-related outcomes, were sparse and heterogeneous. No common effect was estimated, and these outcomes were not used to support conclusions about acute anxiety, learning, or sustained mental health.

Physiological outcomes

Physiological outcomes were reported across examination, simulation, clinical skills, and self-measurement settings. Due to differences in the measurement timing and stressors, no common effect was estimated. Directly verified arm-level effects are reported in Supplementary Table S10. Numerical values that could not be independently reproduced from the report-level data were summarized narratively and were not retained as quantitative effects.

Educational outcomes

Educational outcomes included knowledge, self-efficacy, simulation or clinical skills performance, self-confidence, academic performance, and auditory diagnostic skills. Directly verified between-group results are presented in Supplementary Tables S10, S11. The majority of the estimates were imprecise or vulnerable to outcome-measure limitations. Gosselin et al. reported higher post-intervention self-efficacy (g = 0.73, 95% CI: 0.33–1.13) and higher instructor-rated performance in a selected subset of 20 participants (g = 1.71, 0.67–2.75); the subset result was particularly susceptible to selection bias. Su et al. found no clear difference in the PIVC checklist score (MD: 0.46, −2.35 to 3.27). Other constructs and time points were too dissimilar to pool.

Within-group improvement was not treated as evidence of efficacy. Some studies reported task-specific educational effects, including auditory discrimination, but measurement validity, assessor blinding, and delayed retention were inconsistently reported. The evidence does not establish that lowering self-reported anxiety improves knowledge, competence, grades, or retention.

Reporting bias and certainty

No synthesis included at least 10 sufficiently comparable studies; reporting bias could therefore not be determined and was not used as a standalone downgrade. The certainty of the principal anxiety estimate was very low because all contributing trials were at a high risk of bias, heterogeneity was substantial, the contexts and intervention forms varied, and the Hartung–Knapp and prediction intervals included no effect. Secondary domains were also highly uncertain due to sparse, heterogeneous, or non-randomized evidence (Table 1).

Table 1Summary of findings and certainty of evidence.

Outcome Evidence Effect I2/τ2 HK interval Prediction interval Key limitations Reporting bias Certainty Interpretation

Principal acute/short-term anxiety 9 parallel-group RCTs; 650 participants g = −0.31 (95% CI: −0.58 to −0.04) 64% / 0.107 −0.63 to 0.01 −1.15 to 0.53 All nine trials at high RoB; varied interventions, controls, and contexts Undetermined; not downgraded Very low A small average reduction is possible, but robust uncertainty and the expected range across settings include no effect.

Stress 2 RCTs plus 1 non-randomized study; not pooled Study-specific estimates only Not pooled Not applicable Not applicable Different constructs and contexts; self-report; serious/high RoB Undetermined Very low No common stress effect can be inferred.

Dysmenorrhea pain 3 RCTs; not pooled Study-specific mean differences Not pooled Not applicable Not applicable Different timing, dose, and analytic basis; unblinded self-report Undetermined Very low Pain findings are separate from the principal anxiety inference.

Physiological outcomes 7 heart-rate/pulse studies; 6 blood-pressure studies Mixed study-specific effects Not pooled Not applicable Not applicable Different stressors and timing; some legacy reconstructed effects Undetermined Very low No consistent physiological effect is established.

Educational outcomes Construct-specific evidence; not pooled Mixed knowledge, self-efficacy, skill, and performance findings Not pooled Not applicable Not applicable Non-equivalent constructs; assessor blinding, validity, and retention often unclear Undetermined Very low Reduced anxiety cannot be assumed to improve learning or competence.

CI, confidence interval; g, Hedges’ standardized mean difference; HK, Hartung–Knapp; PI, prediction interval; RCT, randomized controlled trial; RoB, risk of bias. Publication bias was not downgraded solely because fewer than 10 comparable trials were available, as this precluded formal testing.

Discussion

The average post-intervention estimate suggests a small reduction in acute or short-term self-reported anxiety, but this inference is fragile. The conventional Wald interval excluded zero, whereas the Hartung–Knapp interval narrowly crossed zero. The prediction interval spanned benefit, no benefit or harm, and the majority of the leave-one-out Wald intervals crossed zero. Together with a uniformly high risk of bias across the nine principal trials, these results support the possibility of efficiency rather than a definitive claim of efficacy.

The study-level comparison with the 2026 review shows meaningful overlap: 15 of the 24 reports were shared, while 9 were unique to each review. The present review differs in that it applies an outcome- and design-unrestricted corrective search, restricts the population to nursing students, verifies study design, separates non-randomized and design-invalid estimates, uses a uniform postintervention principal estimand, and quantifies small-sample uncertainty. Differences in effect size and certainty should therefore be attributed to differences in eligibility criteria, the inclusion of newly available studies, contrast selection, design handling, and analytical methods rather than to an independent evidence base (20). A detailed comparison is provided in Supplementary Table S12.

The observed reduction could reflect music-specific emotion regulation, attentional redirection, perceived control, or autonomic effects. Expectancy, demand characteristics, participant preference, and researcher attention are also plausible explanations for unblinded self-reported outcomes. The choice of comparator changes the causal question: silence and usual activity estimate the effect of adding music, sensory controls address a more specific contrast, and aromatherapy-only controls estimate the incremental contribution of music.

Reduced anxiety does not necessarily improve learning. Moderate arousal may support engagement, whereas excessive relaxation or auditory distraction may impair attention to instructions, communication, alarms, or task-relevant cues. Secondary psychological, physiological, student health, and educational findings were too heterogeneous for a common pooled interpretation. No included study formally evaluated cost, implementation fidelity, reach, equity, or scalability.

Intervention heterogeneity also limits generalization. The majority of the studies evaluated brief passive or background listening; active music making, therapist-delivered group work, and music-based auditory training involve different mechanisms, staff competencies, participant engagement, and dosage. Preference, familiarity, tempo, volume, delivery, duration, and prior musical experience were reported inconsistently, rendering them unsuitable for credible subgroup analysis or meta-regression.

Limitations include 23 reports for which a complete document could not be retrieved, potential selective non-publication, sparse evidence within intervention categories, and incomplete reporting on allocation, missing data, assessor blinding, adherence, and intervention fidelity. Although the corrective search used no language limit, non-English and grey literature may remain underrepresented. The principal outcome hierarchy was amended following peer review. To prevent the reuse of uncertain data, 38 ancillary estimates that could not be reproduced from arm-level report data and that were available only as rounded values from earlier forest plots were excluded from the quantitative dataset. Evidence was geographically concentrated and predominantly female.

Future trials should conceal allocation, use blinded performance assessors, prespecify a single validated acute anxiety outcome, separate from educational outcomes, and report arm-level data at every time point. Reports should distinguish passive listening, active music making, therapist-delivered therapy, and add-on designs; they should document preferences, familiarity, expectancy, distraction, adverse effects, fidelity, and access; and use adequate cluster or crossover analyses (58). Educational studies should use validated measures and delayed retention tests. Future updates should maintain outcome- and design-unrestricted searching.

Conclusion

Music-based interventions may, on average, reduce acute or short-term self-reported anxiety among nursing students, but very-low-certainty in the evidence and sensitivity to uncertainty methods preclude definitive efficacy claims. The effects on learning, physiology, student health, and sustained wellbeing remain uncertain, and current evidence does not support the routine implementation of music-based interventions.

References

1.World Health O. State of the World’s Nursing 2025: Investing in Education, Jobs, Leadership and Service Delivery . Geneva: World Health Organization (2025 ).

2.Labrague LJ McEnroe-Petitte DM Gloe D Thomas L Papathanasiou IV Tsaras K . A literature review on stress and coping strategies in nursing students . J Ment Health . (2017 ) 26 :471 –80 . doi: 10.1080/09638237.2016.1244721 , 27960598

3.Pulido-Martos M Augusto-Landa JM Lopez-Zafra E . Sources of stress in nursing students: a systematic review of quantitative studies . Int Nurs Rev . (2012 ) 59 :15 –25 . doi: 10.1111/j.1466-7657.2011.00939.x

4.Jimenez C Navia-Osorio PM Diaz CV . Stress and health in novice and experienced nursing students . J Adv Nurs . (2010 ) 66 :442 –55 . doi: 10.1111/j.1365-2648.2009.05183.x , 20423427

5.Ribeiro FMSS Mussi FC Pires CGS Silva RM Macedo TTS Santos CAST . Stress level among undergraduate nursing students related to the training phase and sociodemographic factors . Rev Lat Am Enfermagem . (2020 ) 28 :e3209 . doi: 10.1590/1518-8345.3036.3209 , 32321037 PMC7164923

6.Aloufi MA Jarden RJ Gerdtz MF Kapp S . Reducing stress, anxiety and depression in undergraduate nursing students: systematic review . Nurse Educ Today . (2021 ) 102 :104877 . doi: 10.1016/j.nedt.2021.104877 , 33905898

7.McCarthy B Trace A O'Donovan M Brady-Nevin C Murphy M O'Shea M . Nursing and midwifery students' stress and coping during their undergraduate education programmes: an integrative review . Nurse Educ Today . (2018 ) 61 :197 –209 . doi: 10.1016/j.nedt.2017.11.029 , 29227889

8.Efstathiou M Kakaidi V Tsitsas G Mantzoukas S Gouva M Dragioti E . The prevalence of mental health issues among nursing students: an umbrella review synthesis of meta-analytic evidence . Int J Nurs Stud . (2025 ) 163 :104993 . doi: 10.1016/j.ijnurstu.2025.104993 , 39809132

9.Savitsky B Findling Y Ereli A Hendel T . Anxiety and coping strategies among nursing students during the COVID-19 pandemic . Nurse Educ Pract . (2020 ) 46 :102809 . doi: 10.1016/j.nepr.2020.102809 , 32679465 PMC7264940

10.American Music Therapy A. What is music Therapy? Silver Spring, MD: American Music Therapy Association (2024 ).

11.World Federation of Music T. About WFMT and music Therapy . Raleigh, NC: World Federation of Music Therapy (2011 ).

12.Chanda ML Levitin DJ . The neurochemistry of music . Trends Cogn Sci . (2013 ) 17 :179 –93 . doi: 10.1016/j.tics.2013.02.007 , 23541122

13.Koelsch S . Brain correlates of music-evoked emotions . Nat Rev Neurosci . (2014 ) 15 :170 –80 . doi: 10.1038/nrn3666 , 24552785

14.Bradt J Dileo C Potvin N . Music for stress and anxiety reduction in coronary heart disease patients . Cochrane Database Syst Rev . (2013 ) 2013 :CD006577 . doi: 10.1002/14651858.CD006577.pub3 , 24374731 PMC8454043

15.de Witte M Spruit A van Hooren S Moonen X Stams GJ . Effects of music interventions on stress-related outcomes: a systematic review and two meta-analyses . Health Psychol Rev . (2020 ) 14 :294 –324 . doi: 10.1080/17437199.2019.1627897 , 31167611

16.Linnemann A Ditzen B Strahler J Doerr JM Nater UM . Music listening as a means of stress reduction in daily life . Psychoneuroendocrinology . (2015 ) 60 :82 –90 . doi: 10.1016/j.psyneuen.2015.06.008 , 26142566

17.Thoma MV La Marca R Brönnimann R Finkel L Ehlert U Nater UM . The effect of music on the human stress response . PLoS One . (2013 ) 8 :e70156 . doi: 10.1371/journal.pone.0070156 , 23940541 PMC3734071

18.Khaira MK Raja Gopal RL Mohamed Saini S Md Isa Z . Interventional strategies to reduce test anxiety among nursing students: a systematic review . Int J Environ Res Public Health . (2023 ) 20 :1233 . doi: 10.3390/ijerph20021233 PMC9858718 36673999

19.Martin RD Naziruddin Z . Systematic review of student anxiety and performance during objective structured clinical examinations . Curr Pharm Teach Learn . (2020 ) 12 :1491 –7 . doi: 10.1016/j.cptl.2020.07.007 , 33092780

20.Erbay Dalli Ö Pehlivan S . The effectiveness of music interventions on psychological, physiological and academic outcomes in health sciences students: a meta-analysis . BMC Med Educ . (2026 ) 26 :357 . doi: 10.1186/s12909-026-08693-y PMC12869969 41495795

21.Page MJ McKenzie JE Bossuyt PM Boutron I Hoffmann TC Mulrow CD . The PRISMA 2020 statement: an updated guideline for reporting systematic reviews . BMJ . (2020 ) 372 :n71 . doi: 10.1136/bmj.n71 , 33782057 PMC8005924

22.Lovibond SH Lovibond PF . Manual for the Depression Anxiety Stress Scales . 2nd ed. Sydney: Psychology Foundation (1995 ).

23.Spielberger CD Gorsuch RL Lushene RE Vagg PR Jacobs GA . Manual for the State-Trait Anxiety Inventory . Palo Alto, CA: Consulting Psychologists Press (1983 ).

24.Higgins J Green S . Cochrane Handbook for Systematic Reviews of Interventions . London: Cochrane (2024 ).

25.Sterne JAC Hernan MA Reeves BC Savovic J Berkman ND Viswanathan M . ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions . BMJ . (2016 ) 355 :i4919 . doi: 10.1136/bmj.i4919 , 27733354 PMC5062054

26.Sterne JAC Savovic J Page MJ Elbers RG Blencowe NS Boutron I . RoB 2: a revised tool for assessing risk of bias in randomised trials . BMJ . (2019 ) 366 :l4898 . doi: 10.1136/bmj.l4898 31462531

27.Egger M Davey Smith G Schneider M Minder C . Bias in meta-analysis detected by a simple, graphical test . BMJ . (1997 ) 315 :629 –34 . doi: 10.1136/bmj.315.7109.629 , 9310563 PMC2127453

28.Guyatt GH Oxman AD Vist GE Kunz R Falck-Ytter Y Alonso-Coello P . GRADE: an emerging consensus on rating quality of evidence and strength of recommendations . BMJ . (2008 ) 336 :924 –6 . doi: 10.1136/bmj.39489.470347.AD , 18436948 PMC2335261

29.Borenstein M Hedges LV Higgins JPT Rothstein HR . Introduction to Meta-Analysis . Chichester: Wiley (2009 ).

30.Higgins JPT Thompson SG Deeks JJ Altman DG . Measuring inconsistency in meta-analyses . BMJ . (2003 ) 327 :557 –60 . doi: 10.1136/bmj.327.7414.557 , 12958120 PMC192859

31.Virtanen P Gommers R Oliphant TE Haberland M Reddy T Cournapeau D . SciPy 1.0: fundamental algorithms for scientific computing in Python . Nat Methods . (2020 ) 17 :261 –72 . doi: 10.1038/s41592-019-0686-2 , 32015543 PMC7056644

32.Aksoy B Öztürk L . A randomized controlled trial on the effect of music and white noise listening on anxiety and vital signs during intramuscular injection skill learning . Teach Learn Nurs . (2024 ) 19 :e52 –8 . doi: 10.1016/j.teln.2023.08.020

33.Bittman BB Snyder C Bruhn KT Liebfreid F Stevens CK Westengard J . Recreational music-making: an integrative group intervention for reducing burnout and improving mood states in first year associate degree nursing students: insights and economic impact . Int J Nurs Educ Scholarsh . (2004 ) 1 :Article 12 . doi: 10.2202/1548-923X.1044 , 16646877

34.Borzoo T Tafazoli M Ebrahimzadeh M Kazemi-Arpanahi H Tabahfar R Kamyari N . The effect of aromatherapy and music therapy on blood pressure and heart rate of nursing students before taking the exam . Sci Rep . (2025 ) 15 :12783 . doi: 10.1038/s41598-025-97199-6 , 40229405 PMC11997094

35.Çetin S Kaya K . Comparative effectiveness of Hegu point ice massage versus music for dysmenorrhea-related pain and comfort levels: a randomized controlled trial . Altern Ther Health Med . (2024 ) 30 :27 –35 . 39480682

36.Chen CJ Sung HC Lee MS Chang CY . The effects of Chinese five-element music therapy on nursing students with depressed mood . Int J Nurs Pract . (2015 ) 21 :192 –9 . doi: 10.1111/ijn.12236 , 24593291

37.Contreras JA Malecha A Freysteinson W Cesario S . Music in academia for undergraduate nursing students: a randomized controlled study . Nurs Educ Perspect . (2025 ) 46 :185 –7 . doi: 10.1097/01.NEP.0000000000001287 , 38819215

38.Ergin E Günay İsmailoğlu E Özdemir B Gündoğdu Ö Çınar Yücel Ş . Effects of music on anxiety and comfort levels of nursing students during their first simulation experience: a randomized controlled study . Altern Ther Health Med . (2025 ) 31 :25 –31 .40293846

39.Erzincanlı S Yüksel A . Effect of listening to music on anxiety levels of nursing students during their first experience with intravenous catheterization: a quasi-experimental study . J Educ Res Nurs . (2025 ) 22 :78 –82 . doi: 10.14744/jern.2025.12316

40.Evangelista K Macabasag RLA Capili B Castro T Danque M Evangelista H . Effects of classical background music on stress, anxiety, and knowledge of Filipino baccalaureate nursing students . Int J Nurs Educ Scholarsh . (2017 ) 14 :1 –7 . doi: 10.1515/ijnes-2016-0076 , 29095699

41.Eyüboğlu G Göçmen Baykara Z Çalışkan N Eyikara E Doğan N Aydoğan S . Effect of music therapy on nursing students' first objective structured clinical exams, anxiety levels and vital signs: a randomized controlled study . Nurse Educ Today . (2021 ) 97 :104687 . doi: 10.1016/j.nedt.2020.104687 , 33310698

42.Gebhart V Buchberger W Klotz I Neururer S Rungg C Tucek G . Distraction-focused interventions on examination stress in nursing students: effects on psychological stress and biomarker levels. A randomized controlled trial . Int J Nurs Pract . (2020 ) 26 :e12788 . doi: 10.1111/ijn.12788 , 31724291

43.Gosselin K Oliver H Rose D George J Hanson A Strout K . Efficacy of a 15-minute music intervention on nursing students' anxiety, self-efficacy, and performance in simulation testing: a randomized study . Clin Simul Nurs . (2024 ) 96 :101625 . doi: 10.1016/j.ecns.2024.101625

44.İnangil D Irmak Vural P Doğan S Körpe G . Effectiveness of music therapy and emotional freedom technique on test anxiety in Turkish nursing students: a randomised controlled trial . Eur J Integr Med . (2020 ) 33 :101041 . doi: 10.1016/j.eujim.2019.101041

45.İnangil D Şendir M Kabuk A Türkoğlu İ . The effect of music therapy and progressive muscle relaxation exercise on anxiety before the first clinical practice in nursing students: a randomized controlled study . Florence Nightingale J Nurs . (2020 ) 28 :341 –9 . doi: 10.5152/FNJN.2020.19075 , 34263213 PMC8134016

46.Ince S Çevik K . The effect of music listening on the anxiety of nursing students during their first blood draw experience . Nurse Educ Today . (2017 ) 52 :10 –4 . doi: 10.1016/j.nedt.2017.02.009 , 28214664

47.Indira A Pydimarry PV Katari K Hemanathan R Pal R Ghosh A . Effectiveness of music therapy on academic performance of nursing students . Int J Acad Med . (2018 ) 4 :278 –83 . doi: 10.4103/IJAM.IJAM_32_18

48.Küçükkaya B Başgöl Ş . The effect of listening to music and drawing on coping with dysmenorrhea complaints in nursing students: randomized controlled trial . BMC Womens Health . (2024 ) 24 :571 . doi: 10.1186/s12905-024-03398-0 , 39455971 PMC11515444

49.Lai HL Chen PW Chen CJ Chang HK Peng TC Chang FM . Randomized crossover trial studying the effect of music on examination anxiety . Nurse Educ Today . (2008 ) 28 :909 –16 . doi: 10.1016/j.nedt.2008.05.011 , 18597899

50.Öner U Çevik Aktura S Özbay H Erol A . Effects of aromatherapy and music on stress and self-esteem in nursing students . J Nurs Educ . (2026 ) 65 :293 –301 . doi: 10.3928/01484834-20260129-02 , 41711537

51.Pellico LH Duffy TC Fennie KP Swan KA . Looking is not seeing and listening is not hearing: effect of an intervention to enhance auditory skills of graduate-entry nursing students . Nurs Educ Perspect . (2012 ) 33 :234 –9 . doi: 10.5480/1536-5026-33.4.234 , 22916626

52.Karakuş Selçuk A Baysal E . The effect of dark chocolate and music on pain and anxiety in young women with primary dysmenorrhea: randomized controlled trial . Eur J Integr Med . (2022 ) 56 :102192 . doi: 10.1016/j.eujim.2022.102192

53.Son HK So WY Kim M . Effects of aromatherapy combined with music therapy on anxiety, stress, and fundamental nursing skills in nursing students: a randomized controlled trial . Int J Environ Res Public Health . (2019 ) 16 :4185 . doi: 10.3390/ijerph16214185 , 31671873 PMC6861884

54.Su S Ceylan B Özcan FY Güney A . The effect of music on peripheral intravenous catheter placement skills, anxiety, self-confidence perception, and vital signs in nursing students: randomized controlled trial . Mediterr Nurs Midwifery . (2026 ) 6 :225 –30 . doi: 10.4274/MNM.2025.25400

55.Summers S Hoffman J Neff JA Hanson S Pierce K . The effects of 60 beats per minute music on test taking anxiety among nursing students . J Nurs Educ . (1990 ) 29 :66 –70 . doi: 10.3928/0148-4834-19900201-06 , 2154565

56.Contreras JA . The Effects of a Music Intervention on Stress, Anxiety, and Academic Performance among New Undergraduate Nursing Students. [doctoral dissertation] National Library of Medicine (US). Denton (TX): Texas Woman’s University (2022 ). Available online at: https://clinicaltrials.gov/study/NCT05590910

57.ClinicalTrials.gov. Music Therapy and EFT: Integrative Medicine on Exam Anxiety in Turkish Nursing Students. Identifier NCT03645811. Updated 11 March 2020. Available at: https://clinicaltrials.gov/study/NCT03645811 (accessed 19 September 2026).

58.Hopewell S Chan AW Collins GS Hróbjartsson A Moher D Schulz KF . CONSORT 2025 statement: updated guideline for reporting randomised trials . BMJ . (2025 ) 389 :e081123 . doi: 10.1136/bmj-2024-081123 , 40228833 PMC11995449

Come citare questo studio

Chen B, Wu S, Zou Z, Gao H. (2026). Music-based interventions for acute and short-term anxiety in nursing students: a systematic review and meta-analysis.. Frontiers in public health. DOI: 10.3389/fpubh.2026.1974164

Consulta la pubblicazione originale ↗