This protocol will be performed in accordance with the guidelines of Preferred Reporting Items for Systematic Review and Meta- Analysis Protocols (PRISMA- P) 2015 [54] (Additional file 1).
Inclusion criteria
Types of studies
Only randomized controlled trials (RCTs) of music intervention for COVID-19 with mental disorders will be included. Quasi-randomized controlled trials, reviews and animal experiments will be excluded. Only articles that were published in English or Chinese will be included.
Types of participants
We will only include patients diagnosed as COVID-19 with mental disorders, such as anxiety, depression, or sleep disorder. No limitations relating to age, gender, nationality, ethnicity, and education level.
Types of interventions and controls
The treatment group will be treated with music intervention, including music therapy or music medicine, regardless of duration and frequency. The control group will receive a placebo treatment, conventional western medicine, no music intervention, traditional Chinese medicine, or even no treatment. The curative effects of the two groups will then be compared.
Types of outcomes
As this study aims to systematically evaluate the effects of music intervention on COVID-19 patients with mental disorders such as anxiety, depression, and sleep disorders. The primary outcomes of our study are the scores of questionnaires relating to anxiety, depression, and sleep disorder, and reflect the level of anxiety, depression and sleep quality, from the related scales. The anxiety score will be measured by the SAS and the Hamilton HAMA. The score for depression will be assessed by the SDS and the HAMD. The score for sleep quality will be assessed by the PSQI. The secondary outcomes will be the quality of life and safety. Quality of life will be measured by the MOS Item Short from Health Survey (SF-36), while safety will be measured by the incidence of adverse events.
Data collection and analysis
Search strategy
We plan to search a range of electronic databases from inception to the May 2021, including PubMed, Embase, Cochrane Library, Web of Science, China National Knowledge Infrastructure, Wanfang Database, Chinese Biomedical Literature Database, and Chinese Science and Technology Periodical Database (VIP). We will include all RCTs that were written in English or Chinese that are associated with music intervention for COVID-19 patients with mental disorders. We will also search the reference lists of the all selected articles to identify relevant trails and reviews, and manually search the gray literature, such as trail registries. Further details of the PubMed search strategy are shown in Table 1.
Table 1
Search strategy for PubMed
|
Number
|
Search terms
|
#1
|
“COVID-19”[Mesh Terms]
|
#2
|
“2019-nCoV”[Title/Abstract]OR“coronavirus”[Title/Abstract] OR“COVID19”[Title/Abstract]OR“COVID-19”[Title/Abstract] OR“SARS-CoV-2”[Title/Abstract]OR“coronavirus covid-19”[Title/Abstract]OR“SARSCoV-2”[Title/Abstract]OR“COVID-19 pneumonia”[Title/Abstract]OR“novel coronavirus”[Title/Abstract] OR“new coronavirus”[Title/Abstract]OR“coronavirus covid-19”[Title/Abstract]OR“nCoV-2019”[Title/Abstract]OR“SARS-CoV”[Title/Abstract]OR“novel coronavirus pneumonia”[Title/Abstract] OR“novel coronavirus 2019”[Title/Abstract]
|
#3
|
#1OR#2
|
#4
|
“Anxiety”[Mesh Terms] OR“anxieties”[Title/Abstract] OR“anxious symptom”[Title/Abstract]OR“nervousness”[Title/Abstract]
|
#5
|
“Depression”[Mesh Terms] OR“depressions”[Title/Abstract] OR“depressive symptom”[Title/Abstract] OR“emotional depression”[Title/Abstract]
|
#6
|
“Sleep disorder”[Mesh Terms]OR“sleep disorders”[Title/Abstract] OR“sleep disturbance”[Title/Abstract] OR“sleep disturbances”[Title/Abstract] OR“insomnia”[Title/Abstract] OR“somnipathy”[Title/Abstract] OR“sleeplessness”[Title/Abstract]
|
#7
|
#4OR#5OR#6
|
#8
|
“Music intervention”[Mesh Terms] OR“music therapy”[Title/Abstract] OR“music medicine”[Title/Abstract]OR“music”[Title/Abstract] OR“music listening”[Title/Abstract]OR“singing”[Title/Abstract] OR“song”[Title/Abstract]
|
#9
|
“Randomised controlled trial”[Publication Type] OR“randomised”[Title/Abstract] OR“placebo”[Title/Abstract]
|
#10
|
#3AND#7AND#8AND#9
|
Study selection
Endnote V.X9 software will be conducted to manage literature. After reading the titles and abstracts, two independent investigators (SP and ZX) will search and screen for appropriate studies. Any differences will be submitted to a third researcher (X-YZ). The protocol to be used is shown in Additional file 2.
Data extraction and management
Two reviewers (SP and ZX) will use a predefined extraction template to extract data independently, including: (1) general information [the first author, journal, year, country, and funding information]; (2) patient characteristics [sample size, average age, gender, anxiety score, depression score, sleep quality, and comorbidity]; (3) treatment-protocol for music interventions [types, duration and frequency] and protocols for comparators [types, duration and frequency]; (4) study design [random sequence generation, allocation concealment, blinding, and follow-up]; (5) outcomes [primary, secondary and other outcomes, including scores for anxiety, depression, and sleep quality; adverse events]. The extracted information will be cross-checked by SP and ZX, and any differences will be discussed and resolved with a third reviewer (X-YZ). When necessary, the author will be contacted for more relevant information.
Assessment of risk of bias
Two investigators (CX and GX) will assess the risk of bias independently according to the Cochrane Collaboration’s Risk of Bias tool [55]. The following items will be examined: random sequence generation, allocation concealment, incomplete data, blinding, selective reporting, and other bias. The results will be evaluated systematically and any disagreements will be resolved by the third reviewer (X-YZ). The grades derived from these evaluations will be given as ‘low’, ‘high’, or ‘unclear’ risk of bias.
Measures of treatment effect
We will calculate the risk ratio (RR) for dichotomous data with 95% confidence intervals (CIs) and the mean difference (MD) will be included in our study for continuous data. As all outcomes (scores for anxiety, depression, and sleep quality; safety) are continuous variables, the MD and 95% CIs will be calculated by two authors (CX and GX) independently.
Dealing with missing data
If important data is missing in the selected article, or the reported details are insufficient, we will contact the author via various means to supplement and complete the content. If the information is not available, then sensitivity analyses will be performed to address missing data.
Assessment of heterogeneity
Clinical heterogeneity refers to the variation caused by interventions, different participants, and different end-point indicators. The heterogeneity will be assessed by calculating the I2 value. If I2≤50% for a given study, then the study will not show statistically significant heterogeneity and will be considered in the study. Studies with an I2>50%, will be deemed to show statistically significant heterogeneity and will not be included in our meta-analysis.
Assessment of publication bias
When the meta-analysis contains 10 or more RCTs, we will use funnel plots and Egger’s test to evaluate any publication bias.
Data synthesis
We will use Review Manage software V.5.3.5 to analyze all data. Since all of the outcomes arising from this meta-analysis will be continuous variables, we will calculate MD and 95% CIs. If the included studies are sufficiently homogeneous, we will use the fixed-effect model to process the data. While subgroup analyses or sensitivity analyses will be used in studies with significant heterogeneity.
Subgroup analysis
To explore the potential sources of heterogeneity in a given study, subgroup analysis will be used according to the following factors:
- Different age groups.
- Gender.
- The type of mental disorders (anxiety, depression, or sleep disorders).
- The type of music intervention.
- Differences in duration and frequency.
- The types of control group (placebo, conventional western medicine, no music intervention, traditional Chinese medicine, no treatment).
Sensitivity analysis
Sensitivity analysis is an important method that can be used to assess the robustness and reliability of the combined results in meta-analysis. Sensitivity analysis will determine whether there is any change in the results by including or excluding a specific study. If the results are unstable, we can then remove research studies with a high risk of bias or check the missing data.
Evaluating the evidence
The quality of evidence will be evaluated according to the Grading of Recommendations Assessment, Development and Evaluation (GRADE) guideline. We will evaluate the five factors of risk of bias, heterogeneity, inaccuracy, indirectness, and publication bias, and divide the results into ‘high’, ‘moderate’, ‘low’, and ‘very low’ levels of quality.