A systematic review of the literature was performed using the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines [27]. A comprehensive search was completed in January 2023 by 2 authors (SL and JY) using the following 5 electronic databases: PubMed, MEDLINE (OvidSP), CINAHL complete, PSYCINFO, and Embase.

The search keywords were selected from the PICO format (population: adult patients with procedures in the inpatient and outpatient settings; intervention: technology-based music intervention; comparison: standard care or usual care; outcome: pain and anxiety). These included keywords such as (“inpatient*” OR “hospitalization” OR “intensive care unit*” OR “emergency ward*” OR “general ward*” OR “patient*”) AND (“music*” OR “music intervention” OR “music therapy” OR “music medicine” OR “music listen*” OR “music-based” OR “music methods”) AND (“mobile application*” OR “smartphone” OR “telemedicine” OR “tablet*” OR “computer” OR “mhealth*” OR “ehealth* OR “technolog*” OR “cellphone*” OR “internet*” OR “internet-based” OR “mobile-based” OR “technology-based” OR “smartphone-based” OR “mhealth-based” OR “app*” OR “ipad”) AND (“anxiety” OR “pain”). Results were limited to adults (18 years or older), English text, and publication within 5 years (2018-2023).

This systematic review selected original empirical research studies on technology-based music interventions. The following articles were included: (1) randomized controlled trials (RCTs), (2) studies using interactive technology (eg, smartphones, mHealth, tablets, applications, and virtual reality), (3) empirical studies reporting pain and anxiety outcomes, and (4) English articles published from 2018 to 2023 (as of January 18, 2023). Articles were excluded if they were (1) not full-text articles (eg, conference abstracts and poster abstracts), (2) review articles, (3) study protocols, (4) studies that were not focused on music interventions (eg, therapist-focused), and (5) studies that targeted inpatients and outpatients who were younger than 18 years. We specifically selected the most recent articles published within the last 5 years to ensure the most up-to-date information on technology-based interventions and to improve upon previous systematic reviews [9,25,26,28,29].

Database searches were independently carried out by 2 authors (SL and JY) using electronic databases and cross-references in January 2023. Initially, relevant bibliographic details, including article titles, authors, journal names, publication years, keywords, and abstracts, were retrieved from each electronic database. The management of duplications was facilitated through the use of the EndNote program (Clarivate).

Following the deduplication process, the titles and abstracts underwent independent screening by the 2 authors (SL and JY). Any discrepancies encountered during this phase were systematically resolved through consensus-building between them. Upon the completion of this initial screening stage, the identified primary articles underwent a comprehensive full-text review.

Subsequently, data extraction from the selected studies was conducted with precision to effectively synthesize the study findings. A matrix table employing Excel (Microsoft Corp) spreadsheets was proficiently used throughout the review process to manage and consolidate the extracted data. The extracted information included a comprehensive array of elements, such as authors’ names, research objectives, baseline sample characteristics, study designs, intervention modalities related to music, control group specifications, and outcome variables, with a particular emphasis on elucidating findings pertinent to pain and anxiety.

The full-text screening was independently executed by the same 2 authors. In instances where discrepancies arose during this phase, the intervention of a third author (SP) was sought. The role of the third author entailed a meticulous review of the identified articles to ensure the accuracy and consistency of the selection process. Any disparities or ambiguities were meticulously addressed and resolved under the scrutiny of the third author.

Throughout the review process, adherence to academic standards and methodology was paramount. Any disagreements or discrepancies encountered at any stage were effectively addressed through consensus-building, thereby enhancing the reliability of the synthesized evidence.

The risk of bias was assessed using the Cochrane Risk of Bias version 2 tool [30]. This tool is used to evaluate the risk of bias for individual RCTs. There are five domains: (1) randomization process, (2) deviations from intended interventions, (3) missing outcome data, (4) measurement of the outcomes, and (5) selection of the reported results. The 3 researchers (SP, SL, and JY) assessed primary articles independently using the Risk of Bias version 2 tool. All disagreements and discrepancies were discussed and resolved through meetings until a consensus was achieved.

Figure 1 shows the flow diagram of the literature search and selection process. A total of 292 articles were identified through 5 electronic database searches. Initially, 87 duplicate articles were eliminated. Subsequently, 205 articles were screened based on titles and abstracts. Among these, 153 articles were excluded due to either irrelevance (n=152) or unavailability of full text (n=1). Following thorough full-text reviews, 31 articles were further excluded for various reasons, including being out of focus (n=9), being therapist-focused (n=7), having a non-RCT design (n=4), and using an unclear technology device (n=11). Consequently, 21 articles met the inclusion criteria for this study.

The characteristics of the 21 articles included in this study are shown in Table 1. The purpose of all the studies was to examine the effects of technology-based music interventions on patient outcomes, including anxiety and pain. Two studies aimed to examine particularly interactive technology-based interventions [31,32]. Drzymalski et al [31] focused on the impact of self-selected or preselected music, and Anglin et al [32] focused on the effects of a patient’s choice. Most studies included male and female participants; however, 6 studies targeted only women [15,20,21,24,31,33]. The sample size ranged from 18 to 330. Most studies had 2 groups (ie, experimental group and control group), except 2 studies [31,33]. The interventions were performed for patients who underwent surgery or invasive procedures. Surgery included gynecologic surgery [20], cardiothoracic surgery [34], cataract surgery [14], orthopedic surgery [35,36], nasal bone fracture reduction [37], and cesarean delivery [21,31]. The procedures included radiation therapy [15], catheter insertion [38], colonoscopy [18], biopsy [24], bronchoscopy [39], steroid injection [40], transvaginal ultrasound-guided oocyte retrieval [33], wound care [16], eye procedures [41], pleural procedures [42], and urologic procedures [32]. The studies were conducted in many countries, including the United States [15,24,31,32,40], France [14,20,38], Hong Kong [18,33], Australia [42], Germany [21], Taiwan [34], India [35,41], Turkey [43], Spain [44], Chile [37], Malaysia [36], Iran [39], and Brazil [16].

The quality assessment of the selected studies was conducted using the Risk of Bias 2 criteria [30]. These criteria provide a more nuanced approach compared to the previous version of the tool, allowing for a more detailed evaluation of bias across different domains of study conduct. This structured assessment helps researchers and readers of systematic reviews to better understand and interpret the quality of evidence presented in the studies. “Low risk” indicates that there are sufficient measures in place within the study design and conduct to minimize bias, thus providing confidence in the validity of the study results. Studies categorized as having low risk of bias are considered to have minimal risk of distorting the intervention effect estimates. “Some concerns” indicates that there are certain aspects of the study design or conduct that raise concerns about the potential for bias. “High risk” indicates that there are significant issues in the design, conduct, or reporting that substantially increase the risk of bias. Studies categorized as high risk of bias are deemed to have limitations that seriously compromise the validity of the findings.

Of the 21 studies, 11 (52%) were classified as having low risk of bias and 10 (48%) were identified as having some concerns. More specifically, in terms of the randomization process, 76% (16/21) of studies were deemed to have low risk and 24% (5/21) were flagged as having some concerns. Regarding deviation from the intended intervention, 76% (16/21) of studies were categorized as having low risk, 19% (4/21) were identified as having some concerns, and 5% (1/21) were categorized as having high risk. In terms of the measurement of outcomes, 71% (15/21) of studies were classified as having low risk and 29% (6/21) were identified as having some concerns. All 21 studies (100%) were rated as low risk for missing outcome data and the selection of the reported results (Figure 2). The evaluation was conducted independently by 3 authors, and any disparities were resolved through discussion.

Many patients undergoing surgical procedures to manage illnesses often endure physical discomfort and psychological strain, which can significantly impact medical expenses and clinical outcomes [45]. Considering these aspects, this systematic review investigated the efficacy of music interventions for alleviating anxiety and pain, drawing insights from recent clinical trials. Specifically, it delves into various methodological approaches, such as employing music-dispensing devices and considering individual preferences regarding music selection and genre. The advent of technological advancements has facilitated the integration of diverse devices for music interventions within clinical settings. These innovations have not only enhanced the accessibility of music interventions in such environments but also empowered patients to personalize their music experiences according to their preferences. This signifies the potential of music interventions in not only mitigating anxiety and pain but also fostering meaningful patient-clinician interactions during the intervention process. Thus, our findings underscore the significance of methodological nuances in music intervention studies, highlighting the pivotal role in yielding favorable outcomes.

A total of 21 studies met the inclusion criteria for this review. Among these 21 studies, 17 addressed anxiety, 11 addressed pain, and 6 addressed both. These studies aimed to identify the effectiveness of music interventions for pain and anxiety among hospital patients. Among the 17 studies on anxiety, 13 reported a significant reduction in anxiety in the experimental group compared with the control group, while 4 studies showed inconsistent results. Similarly, among the 11 studies on pain, 7 reported a significant decrease in pain in the experimental group compared with the control group, while 4 studies showed inconsistent results. These findings support earlier literature that music interventions can be used to manage physical and psychological symptoms [46,47] and can significantly benefit improvements in pain [46-49] and anxiety [46,47,49,50], but they indicated the need for repeated studies.

Previous reviews have underscored the value of music interventions across various patient populations and treatments [46,51], and they have been recommended to decrease anxiety, stress, and postoperative pain [46,47,51]. In this review, music interventions using technology were implemented across a range of medical settings, with experimental groups using diverse music playback devices or software, including web-based music applications (eg, QR code access), smartphones, tablets, computers, CD players, MP3 players, projectors, iPods, and personal or portable devices. However, the findings regarding which device delivered the music intervention better were unclear, and the difference in the effect of the medium on anxiety and pain was not directly discussed. Therefore, these findings indicate that selecting an appropriate device is an important factor in music interventions [52], and the effectiveness of the medium of the music intervention should be further assessed in future studies.

More than 50% (13/21, 62%) of studies included in this review allowed participants to select themes themselves, and of these, more than 70% (10/13, 77%) reported positive effects. Previous studies reported that anxiety and pain reductions were more effective when self-selected music was played [28,29,52,53]. The researchers interpreted that participants’ familiarity with self-selected music evoked a sense of pleasure and control that could act to reduce pain [29]. On this basis, this review suggests that the method of the selection of music is also an important factor in a music intervention for reducing anxiety and pain. Some studies required participants to select music from playlists available in music applications, which included pop-rock music, romantic music, religious music, natural sounds, popular songs, slow-tempo music, relaxing music, folk music, piano sonatas, and local folk music. Music content or genre and tempo are important for stimulation and relaxation [51]. Pop, film soundtrack, jazz, classical, folk, and instrumental music are widely used music genres [54]. Moreover, slow-tempo music (60-80 beats/min) supports relaxation, whereas fast-tempo music can cause discomfort [55]. Hatwar and Gawande [56] reported that consulting an expert while selecting music is beneficial. Yangöz and Özer [47] reported that a music genre should be chosen after consulting with an expert, and the patient or researcher should then select a music playlist. However, in our review, it was unclear whether the music was chosen after expert consultation. Therefore, consultation with experts can be considered in the process of selecting music playlists, and more studies are required to examine the difference between the effects of music selected by oneself and those of music selected by researchers [51]. Although music interventions can be used with individualized interaction, group interaction, or individual listening [47], most of the reviewed literature involved individual listening. Thus, identifying the effect of the music intervention in the context of group interaction in future studies will help implement appropriate music interventions.

Consistently in previous literature [57], the interventions were performed before, during, and after the surgery or procedure, and the duration of playing music varied from 10 to 60 minutes. Music duration is a factor in the intervention’s effect on pain and anxiety [52]. Yangöz and Özer [47] reported that music duration ranged from 15 to 180 minutes, and they recommended the duration be restricted to less than 20 minutes because listeners could get bored if the duration is too long. Gillen et al [58] reported that the most common duration for listening to music was 15 to 30 minutes, and Pittman and Kridli [51] found that a duration of 15 to 20 minutes was effective in music interventions. The appropriate playing duration for a music intervention could vary according to the treatment environment (eg, the required time for a surgery or procedure differs). However, studies that determine the appropriate duration for music interventions are limited [51,52].

The risk of bias assessment is an established approach to evaluate the credibility of results at the systematic review level [59]. Nevertheless, the results of this review need to be interpreted with caution. One of the studies included in this review had a high risk of bias in deviations from intended interventions since we could not clarify the interventionist. Given the nature of a music intervention as a therapy, a professional may be required for intervention delivery. However, if researchers perform the intervention, they could be aware of the potential influence on the results. Thus, researchers should design the approach by clearly considering the importance of study quality.

Future studies should examine the effect of advanced technology-based music interventions in the context of interaction using various smart devices and software. Advanced technologies, such as artificial intelligence, robotics, and smartphone apps, are continually evolving for better patient outcomes. More interactive and patient-tailored music interventions using these advanced technologies can be developed and tested to reduce anxiety and pain by allowing to control the type of music and music delivery method. It is also necessary to review the effectiveness of in-patient mobile communication interventions between medical staff and patients, and to promote the development of different field application methods using this technology.

While our systematic review offers valuable insights, it is important to acknowledge its limitations. Despite our diligent efforts to include comprehensive studies meeting the eligibility criteria, there remains the possibility of incomplete retrieval owing to the search keywords used. Considering the significant evolution in music delivery technology, we decided to include articles published within the past 5 years. Moreover, our search was confined to 5 databases and included only studies written in English, potentially leading to an incomplete list of included studies.

Despite its limitations, our review has significant value as it has conducted an in-depth analysis of the current literature and has provided a lucid summary of intervention protocols and outcomes concerning anxiety and pain. Emphasizing technology use and its outcomes, our findings underscore the critical importance of selecting suitable devices in music interventions, presenting a notable challenge for future intervention development. Notably, our review distinguishes itself with its comprehensive summary of primary articles, intervention methodologies, and outcomes related to anxiety and pain. By highlighting technology use and its outcomes, we offer a global perspective on intervention protocols and the efficacy of technology-based music interventions for alleviating pain and anxiety among hospitalized patients.

While acknowledging the potential impact of a meta-analysis, our review prioritizes elucidating intervention methodologies specific to technology-based music interventions. The diverse array of technologies employed in the included studies can pose challenges in clearly discerning the effectiveness of the used devices. By focusing on methodologies related to technology use, our review underscores significant challenges associated with selecting appropriate devices in intervention protocols aimed at mitigating pain and anxiety in hospitalized patients. This approach can help to lay the foundation for validating the effectiveness of technology-based device use in future studies.

Music interventions are valuable for various patients and treatments and have been recommended to decrease anxiety, stress, and postoperative pain. A music intervention is a nonpharmacological complementary approach and shows a positive effect on patients, and up to 70% of studies reported a positive effect when patients were allowed to select the music themselves. This systematic review provided an in-depth review of the current literature on technology-based music interventions for patients undergoing various procedures. Future studies are needed to examine the effectiveness of interactive technology-based music interventions for reducing anxiety and pain among hospitalized patients undergoing procedures. This review contributes to the research on technology-based music interventions and can help to select a practical methodology for the interventions. Moreover, further meta-analyses should be conducted to enhance statistical power through the combination of studies.