Study design:
This is a systematic review protocol for clinical trials prepared in accordance with the guidelines of the protocol Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA-P) [15]. The PRISMA-P checklist is included as an additional file (see additional file 1).
Study registration
In accordance with the guidelines, our systematic review protocol was registered with the International Prospective Registry of systematic reviews (PROSPERO) on 10/27/2020 (registration number CRD42020211302).
Available at: https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42020211302
Eligibility Criteria
Types of studies:
Randomized and non-randomized clinical trials will be included. Observational trials, systematic reviews and meta-analysis will be excluded, but the studies used in them will be analyzed.
However, controlled studies comparing different methods of auricular acupuncture will be excluded because these studies cannot produce and demonstrate the net effect of auricular acupuncture.
Types of participants:
Studies in which participants have any depressive disorder, diagnosed through standardized criteria such as the Diagnostic and Statistical Manual of Mental Disorders (DSM), the International Classification of Diseases (ICD) and other diagnostic criteria used internationally will be included. There will be no restrictions on age, gender, ethnicity, education or economic status.
Types of intervention/comparison:
Auricular acupuncture aims at establishing the balance of energy flow (Qi). The Qi flows in the body through channels that pass through specific points connected to each organ, so in cases of imbalance of the Qi, such points can be stimulated, which changes the flow of energy and regains the individual's health by restoring the body's balance. The stimulation can be from a needle, is the technique most used today, but can also be used seeds and magnets, and several other techniques.
Studies will be accepted that have as experimental intervention any type of auricular acupuncture (using needles, seeds, magnetic stones, lasers, ultrasound, bleeding or electrical treatment).
The control interventions that will be considered for the studies will be conventional medical treatment for each depressive disorder, no treatment, placebo and other active therapies.
Types of outcome measures
Primary results:
It is expected to observe a change in the level and symptoms of depression from the beginning to the end of treatment, being analyzed by qualified and validated forms of depression, such as the Beck-II inventory, Hamilton evaluation scale and Zung Depression Self-assessment scale.
Secondary results:
Evaluation of the change in quality of life or adverse reactions with the intervention.
Research methods to identify the studies:
Databases used will be: Pubmed Medline (see additional file 2), Scielo, Lilacs, Scopus, Web of Science and Embase. The search will not have restriction of language and period of publication. Unpublished studies will not be analyzed.
The keywords used will be auricular acupuncture (auriculotherapy, acupuncture therapy, among other synonyms) and the types of depressive disorders (major depressive disorder, bipolar disorder, postpartum depression (peripartum), seasonal affective disorder, psychotic depression, persistent depressive disorder (dysthymia), pre-menstrual dysphoric disorder, situational depression and atypical depression). The ENTREE TERM referring to these terms will be analyzed for the Embase and the DECS and their MESH terms will be used in the other databases.
The reference list of all included studies will be checked in order to identify potentially relevant studies.
Data collection and analysis
Selection of studies:
Two reviewers (KNS and MRJS) will independently review literature titles and abstracts mainly to exclude articles that do not meet the inclusion criteria, and then make an additional judgment on whether articles that may meet the inclusion criteria can be included. The disagreement of the two reviewers to include or exclude an article will be determined by a third reviewer (LPL). The entire process will be conducted in the Rayyan [16] application, a specific application to assist researchers in producing reviews.
Data extraction and management:
As well as the selection of articles, evaluators will independently extract (KNS and MRJS) data from the articles included in the study. The contents extracted include details of the intervention and comparator used, measured results and results according to the STRICTA p checklist (Revised standards for reporting interventions in acupuncture clinics) [17]. The details will be organized in excel tables. For articles published only in summary or articles where important information is missing, we will seek complete information on methods and results by contacting the authors. Disagreement on data extraction will be determined by a third reviewer (LPL).
Bias risk assessment:
The risk of bias in each study will be assessed by 2 independent reviewers (KNS and MRJS) using the Cochrane Rev Man "risk of bias tool" which analyzes 7 domains: random sequence generation, concealment of allocation, blinding of participants and staff, blinding of result evaluation, incomplete result data, selective result reports and other biases. The bias risk for each item will be classified as "low risk of bias", clear risk of bias or "high risk of bias". Any disagreement will be resolved through the third reviewer (LPL).
Measures of effect on treatment:
If the results are continuous, the data will be expressed with mean difference (MD) or standard difference (SMD) with 95% CI. And in the case of dichotomous results, data will be expressed as the relative risk (RR) with 95% CI.
Dealing with missing data:
Regarding missing data, it will be analyzed why the data are missing. Whenever possible, the authors will be contacted to request any inappropriate and missing data from the included studies. If it is not possible to obtain missing data, a case analysis will be performed, only on known data. And we will address the potential impact of missing data on the results of the review in the bias study.
Evaluation of heterogeneity:
The evaluation of heterogeneity will be conducted by the Review Manager (V.5.4). Chi-square test and forest value I2 to detect non-overlap ping CIs and investigating x2 (with P value > 0.10 indicating no heterogeneity), The I2 value is classified into 4 levels: little or no heterogeneity (0% -40%), moderate heterogeneity (30% -60%), substantial heterogeneity (50% -90%) and considerable heterogeneity (75% -100%). We will use a cutoff point of 50% for significant heterogeneity among the studies included. If I2 > 50%, a significant heterogeneity between trials will be considered and a meta-analysis will not be performed.
If the data found does not fit the quantitative analysis, the qualitative description will be performed. If the meta-analysis is adequate, according to the data of good performance in homogeneity, it will be applied to conduct a quantitative analysis.
Evaluation of report bias:
If the number of studies included in the review is more than 10, funnel graphs will be generated to analyze the possible presence of report biases and effects of small studies.
Data synthesis:
According to the heterogeneity of the interventions of the studies, the grouping of the studies that fit in the necessary criteria for the accomplishment of the meta-analysis will be made. If it is not possible, only a descriptive review of the studies that meet the inclusion criteria will be performed. The results will be considered statistically significant when p is greater than 0.05.
Subgroup analysis:
We intend to perform subgroup analysis if we identify substantial heterogeneity. The following factors will be analyzed: different types of depression, techniques used in auricular acupuncture and whether the studies are randomized or not.
Sensitivity analysis:
In order to ensure the robustness of the results, the sensitive analysis will be performed to remove the impact of low quality studies, if after the heterogeneity exists after the subgroup analysis. Thus the meta-analysis will be performed again after the elimination of the studies of inferior qualities. The two meta-analyses will be compared in order to make the decision on the effective elimination of low quality studies, according to the sample size, to forced evidence and missing data.
Classifying the quality of evidence:
The quality of the evidence will be evaluated by the Classification of Evaluation, Development and Evaluation of Recommendations (GRADE), this classification evaluates the quality of evidence based on 5 factors: limitations of the study, consistency of the effect, imprecision action bias, indirect and publication. The evaluation is categorized into 4 levels: high, moderate, low and very low quality.