Background
Schizophrenia is one of the most serious psychiatric diseases worldwide. It causes significant dysfunction in multi-dimensions of personal function of the patients, with a wide range of clinical symptoms. In general, the symptoms of schizophrenia can be classified as positive symptoms, negative symptoms, and general psychopathology [
1]. In addition, patients with schizophrenia also often have comorbid mood symptoms, either in the form of a manic mood or depression. Furthermore, risk factors associated with schizophrenia such as violence and suicide have been found to be correlated with the specific symptoms of schizophrenia. Among them, the risk of violence has been reported to be correlated with positive symptoms [
2], and the risk of suicide has been reported to be correlated with negative symptoms and mood symptoms [
3,
4].
Music therapy has been used in the treatment of schizophrenia for decades, and it is often used as adjunct therapy to medication. It can take the format of individual therapy [
5], large group therapy [
6], or a combination of individual and group therapy [
7] through either “passive listening” [
8‐
10] or “active participation” [
6,
7]. Every form of music therapy has an effect in schizophrenic patients to a certain extent. However, most studies investigating the treatment effect of music therapy have focused on the effect on negative and mood symptoms only [
6,
7,
9‐
13].
Recently, there is one huge review article discussing the clinical application of music therapy in different disease entity, including neoplasm, disease of central nervous system, psychiatric illness, and so on [
14]. In this article, the authors reviewed the evidences of music therapy in the schizophrenic patients mainly according to one recent meta-analysis conducted by the Mössler and the colleagues (2011) [
15]. In this meta-analysis conducted by Mössler and the colleagues, the authors suggested that the application of music therapy in schizophrenic patients would be beneficial with regards to negative symptoms and mood symptoms. On the other hand, because of a lack of evidence, the role of music therapy in the treatment of positive symptoms in schizophrenic patients is inconclusive. However, positive symptoms have been proven to be highly associated with the risk of violence [
2], which contributes to the stigmatization of patients with schizophrenia [
16]. Furthermore, the report by Mössler et al. (2011) did not provide further information about the association between the treatment effect of music therapy and the specific characteristics of music therapy applied in schizophrenic patients.
In order to explore (1) the treatment effect of music therapy in patients with schizophrenia, and (2) the role of specific characteristics of music therapy in the treatment effect, we conducted this meta-analysis to evaluate the treatment effect of music therapy on the specific subscale symptoms of schizophrenia, including the overall disease severity of schizophrenia, positive symptoms, negative symptoms, general psychopathology, and mood symptoms. We performed meta-regression analysis to evaluate the association between the treatment effect of music therapy and the clinical variables or specific characteristics of music therapy, including frequency, duration of each session, total number of sessions of music therapy, and total duration of music therapy.
Discussion
The results of this meta-analysis indicated that adjunct music therapy to standard treatment was associated with a significantly better treatment effect in schizophrenic patients than control patients, not only in negative or mood symptoms but also in positive symptoms. In addition, this treatment effect was significantly positively associated with the whole duration of illness, and significantly inversely associated with the number of previous hospitalizations.
In general, the results of this study are consistent with the previous meta-analysis conducted by Mössler et al. (2011) [
15], which report moderate to large effects in general mental state, negative symptoms, depression, and anxiety. In addition, the current meta-analysis provides additional information to the current knowledge about the benefits of adjunct music therapy for patients with schizophrenia, which provided the evidences of the significant treatment effect of adjunctive music therapy on the positive symptoms, negative symptoms, and mood symptoms. In the meta-analysis conducted by Mössler et al. (2011), the treatment effect of music therapy on positive symptoms was inconclusive because of limited evidence. In recent years, several new trials and studies have been published [
11,
23,
24]. Among them, most studies discussed the significantly better treatment effect of adjunct music therapy on the positive symptoms in schizophrenic patients [
23,
24]. The significantly better treatment effect in newer published studies rather than previous reports might be derived from the different intervention design (group therapy versus individual therapy) [
5,
23], different frequency of music therapy [
24,
25], or different total duration of music therapy [
5,
24]. Thus, it is necessary to conduct a further meta-analysis based on these newer evidences. The clinical implication that adjunct music therapy can alleviate positive symptoms in schizophrenic patients more effectively than in those who do not receive adjunct music therapy is important, in particular because positive symptoms are thought to be highly associated with the risk of violence [
2]. In addition, Courtright P and the colleague (1990) reported that the use of music therapy could help in reducing aggressive behavior in schizophrenic patients [
29]. Among the positive symptoms, the hallucination and delusion had been thought to be more prominent in clinical practice. Furthermore, the relationship between the auditory hallucination and music therapy was also one of the interesting topics about the treatment effect of adjunct music therapy in schizophrenic patients. However, we could not perform such meta-analysis of these topics because of the limited reports providing these data [
7,
24]. Among these articles, there was only one discussing the differences of severity of hallucination or delusion after treatment of adjunct music therapy [
7] and only one comparing the changes of severity of hallucination or delusion between experiment group (patients with adjunct music therapy) and control group (patients with standard treatment only) [
24]. When we investigated the results of these two studies, we found significantly lower severity of hallucination and delusion after treatment of adjunct music therapy and significantly better improvement in hallucination in patients with adjunct music therapy than those with standard treatment only but not in delusion.
We also performed meta-regression analysis to evaluate the possible confounding effect of clinical variables. The results showed that the treatment effect of music therapy was positively associated with the “whole duration of illness”. This means that the application of music therapy for schizophrenic patients could alleviate disease severity, especially in those with chronic illness. Several authors have investigated the effect of music therapy on cognitive function in patients with chronic schizophrenia, and reported significant benefits [
30]. In addition, the usage of specific music therapy has been reported to improve attention and motor performance in chronic schizophrenic patients [
8,
31,
32]. Taken together, supplemental music therapy can provide significant benefits for schizophrenic patients with a chronic course with regards to disease severity and also cognitive function.
In this study, we found a significantly inverse association between the treatment effect of music therapy and the number of previous hospitalizations. This seems to be in contrast to our other finding of a positive association between the treatment effect of music therapy and the whole duration of illness. A possible explanation may be the specific characteristics of the patients with more previous hospitalizations. In previous reports, schizophrenic patients with more previous hospitalizations have been reported to have worse treatment adherence, poorer treatment attitude, easier relapse, and poor support systems [
33‐
35]. Therefore, the adherence or attendance rate to music therapy in schizophrenic patients with more previous hospitalizations may be poorer than those with fewer previous hospitalizations. However, we could not investigate the association between adherence/attendance rate and the treatment effect of music therapy because of a lack of data.
With regards to safety profile, the dropout rate was not significantly different between the schizophrenic patients who did and not receive adjunct music therapy. In addition, when we investigated the included studies individually, no severe adverse effect profile was reported with the use of music therapy.
With regards to the cost-effectiveness of music therapy in the treatment of schizophrenic patients, we found no significant associations between the treatment effect of music therapy and the frequency of music therapy, the total duration of music therapy, and the duration of each session of music therapy in meta-regression analysis. Furthermore, we investigated whether there was any difference in the treatment effect of adjunct music therapy in subgroup analysis by different frequency of music therapy or different total duration of music therapy. The results showed that the treatment effect was significantly better compared to those who did not receive adjunct music therapy, both in a total duration of music therapy shorter than 3 months and 3 months or longer, and a frequency of less than 4 times per week and 4 times a week or more. This indicates that adjunct music therapy in schizophrenic patients can result in a better treatment effect compared to those who do not receive adjunct music therapy when introduced into a treatment program regardless of the total duration, frequency, or duration of each session of music therapy. Because music therapy is costly and complex, clinicians may be reluctant to include it in clinical practice. However, our findings show that the effect of music therapy is beneficial regardless of the frequency, length, and duration. Therefore, we suggest that clinicians should consider including adjunct music therapy into their clinical practice to provide their patients with the best outcomes possible.
Limitation
There are some limitations to the current meta-analysis. First, most of the included studies conducted adjunct music therapy as a combination of passive listening and active participation and most of them were performed in forms of group therapy. This may lead to the problem that, for most of the studies that involved playing music, the therapy group would likely consist of patients who knew how to play an instrument. However, in the real world, this situation is unlikely to occur. Further studies are needed to investigate whether “passive listening” or “individual therapy” can provide benefits for schizophrenic patients, either in disease severity or with other symptoms. Second, as mentioned in the Discussion, we could not further analyze the association between the treatment effect of music therapy and the attendance rate of music therapy because of a lack of data. However, the attendance rate may be a confounding effect on the treatment effect of music therapy. Third, in order to perform the meta-analysis for the treatment effect of music therapy on the general aspect of disease severity, positive symptoms, negative symptoms, general psychopathology, and mood symptoms, we pooled different studies with different rating scales including the PANSS, BPRS, SANS, and SDS. This may have increased the heterogeneity in the meta-analysis, which achieved significance. However, in previous reports, disease severity was shown to be associated between different rating scales to some degree [
27,
28]. In addition, there were no absolute contraindications to pool these different rating scales together. Therefore, in order thoroughly meta-analyze the subscales of specific symptoms, we chose to pool those rating scales together. Besides, in current study, we could not perform detailed meta-analysis of treatment effect of adjunct music therapy for the specific symptoms, such as hallucination or delusion, because of the limited reports providing such information. Finally, we tried to investigate the treatment effect of music therapy on schizophrenic patients with different disease severity. However, we could only perform the meta-analysis on those with mild disease severity but not those with moderate or severe disease severity because of a lack of sufficient trials.
Conclusion
Our meta-analysis provides strong evidence to prove the role of adjunct music therapy in the treatment of schizophrenic patients. In these patients, adjunct music therapy improved the effect of treatment in negative and mood symptoms, and also positive symptoms compared to those who did not receive music therapy regardless of the total duration, frequency, total number of sessions, or the duration of each session of music therapy. The effect of treatment was especially pronounced for those with a chronic disease course. However, in order to extensively apply adjunct music therapy in a clinical setting, further trials are needed to evaluate the treatment effect of music therapy with “passive listening,” the possible confounding effect of attendance rate, and the treatment effect of music therapy on those with moderate or severe disease severity.
Availability and requirements
Competing interests
The authors stated that there are no any competing interests in current literature.
Author contributions
PTT, the corresponding author and one of the first authors, takes the responsibility of not only writing down the section of discussion and part of section of introduction, but also collecting and revising all the information and manuscripts from all the other authors. YWC, who is a junior neurologist and works equally as the other one of the first author, takes the responsibility of writing down the section of the methods and results. PYL, one of the senior psychiatrists and specialist of meta-analysis, helps in not only the procedure of meta-analysis, but also complete on the section of introduction of this manuscript and revision of part of the method/result. KYT, one of the junior psychiatrists, takes the responsibility of both literature searching and help in the literatures summarization. HYW, one of the junior psychiatrists, takes the responsibility of both literature searching and help in the revision of section of discussion. YSC, one of the junior psychiatrists, takes the responsibility of both final judgment of literature searching and revision of the section of introduction. YCC, one of the senior psychiatrists, raises the subgroup meta-analysis of treatment effect of music therapy in different total duration of music therapy. CHC, one of the senior psychiatrists, raises the subgroup meta-analysis of treatment effect of music therapy in different frequency of session of music therapy. WC, .one of the senior psychiatrists, raises the concept of meta-regression analysis of treatment effect of music therapy and clinical variables. CKW, the other one of the senior psychiatrists, suggest to make further comparison of treatment effect of music therapy in different total numbers of sessions of music therapy and help in study design. All authors reviewed the manuscript. All read and approved the final manuscript.