This paper is in the following e-collection/theme issue:

Digital Health Reviews (1724) Psychiatry (95) Anxiety and Stress Disorders (1330) Methods and New Tools in Mental Health Research (558) Telepsychiatry (212) Prescribable Digital Interventions (Digital Therapeutics) (44) Digital Mental Health Interventions, e-Mental Health and Cyberpsychology (2111)

Published on in Vol 27 (2025)

Preprints (earlier versions) of this paper are available at https://preprints.jmir.org/preprint/74905, first published .
Magnitude of the Digital Placebo Effect and Its Moderators on Generalized Anxiety Symptoms: Systematic Review and Meta-Analysis

Magnitude of the Digital Placebo Effect and Its Moderators on Generalized Anxiety Symptoms: Systematic Review and Meta-Analysis

Magnitude of the Digital Placebo Effect and Its Moderators on Generalized Anxiety Symptoms: Systematic Review and Meta-Analysis

Authors of this article:

Takashi Hosono1, 2 Author Orcid Image ;   Rinka Tsutsumi1 Author Orcid Image ;   Yuki Niwa1 Author Orcid Image ;   Masuo Kondoh1, 3 Author Orcid Image
Article Authors Cited by Tweetations Metrics

    1Graduate School of Pharmaceutical Sciences, The University of Osaka, 1-6 Yamadaoka, Suita, Osaka, Japan

    2Clinical Research, R&D, NS Pharma Inc., Paramus, NJ, United States

    3Center for Infectious Disease Education and Research (CiDER), The University of Osaka, Osaka, Japan

    Corresponding Author:

    Masuo Kondoh, Prof, PhD


    Background : Digital therapeutics (DTx) have attracted attention as the substitutes or add-ons to conventional pharmacotherapy. The number of clinical trials for DTx has increased recently, and one of the main targets for DTx is psychiatric disorders. Generalized anxiety disorder (GAD) is one of the most common and notable psychiatric disorders, and it’s known that the magnitude of placebo effect in the pharmacotherapy is quite large. The randomized controlled trials (RCTs) with digital placebos are the most reliable clinical trials to evaluate the safety and efficacy of DTx. However, the magnitude of the digital placebo effect and its moderators on GAD have not been investigated, although they are critical to assess the true treatment effect of DTx.

    Objective : The objectives of this study were to identify RCTs with digital placebos as comparators that evaluated GAD assessment scores, to review the characteristics of the RCTs and of the digital placebos in the systematic review, and to investigate the magnitude and its moderators in the meta-analysis.

    Methods : The RCTs evaluating the GAD assessment scores by setting digital placebos as comparators were identified by searching the database of PubMed, Web of Science, and Scopus in July 2024. The characteristics of the RCTs and of the digital placebos were reviewed systematically. The meta-analysis, including subgroup analyses and meta-regressions, was conducted to investigate the magnitude and its moderators of the digital placebos.

    Results : A total of 54 RCTs were included in the systematic review and 32 RCTs with 3 GAD assessment scores were included in the meta-analysis with a total of 5311 participants. The magnitude of digital placebos for all the included studies was small to moderate (Hedges g=0.28, 95% CI 0.18-0.38). The subgroup analyses showed the significant difference in the magnitude among target population (P=.03), placebo approach (P=.02), and baseline values (P=.02). The meta-regressions also indicated that the primary psychiatric patients in the target population (P=.01), “Removed” type in placebo approach (P=.04) and high baseline values (P =.02) were moderators for the magnitude of digital placebos.

    Conclusions : This study showed the small-to-moderate and statistically significant digital placebo effect on GAD assessment scores. Target population, placebo approach, and baseline values were also identified as the moderators of the placebo effect. It would be effective to create the study protocols for the DTx trials with digital placebos by considering the moderators identified in this study.

    J Med Internet Res 2025;27:e74905

    doi:10.2196/74905

    Keywords

    digital health; digital intervention; digital therapeutics; randomized clinical trial; psychiatry; psychiatric; generalized anxiety disorder 


    Digital Therapeutics (DTx) are rapidly spreading as substitutes or add-ons to conventional pharmacotherapy recently. According to the definition by International Organization for Standardization, DTx are defined as health software intended to treat or alleviate a disease, disorder, condition, or injury by generating and delivering a medical intervention that has a demonstrable positive therapeutic impact on a patient [1]. The main difference between DTx and conventional wellness apps is that DTx are developed with clinical evidence usually generated from clinical trials [2].

    In many countries, DTx are regulated through the similar regulatory framework with medical devices as part of software as a medical device (SaMD), which is defined as software intended to be used for one or more medical purpose(s) that perform(s) these purposes without being part of hardware medical device by International Medical Device Regulators Forum [3]. Some regulatory authorities have recently implemented the policies that promote the development and the patient access of DTx and SaMD [4].

    With the implementation of regulatory policies, the number of clinical trials of DTx is also increasing in the world [5]. The designs of the confirmatory studies of DTx are basically comparable to those of drugs and medical devices except for some elements such as blinding and comparators. Traditionally, no intervention, waiting list, or treatment-as-usual was widely used as the comparators in the clinical trials of DTx due to the technical difficulties to keep blinding by using appropriate comparators [6].

    However, with the advancement of digital technology, digital placebos have been recently adopted in the clinical trials of DTx to evaluate the true effect of active interventions in blind manners appropriately [7]. Digital placebos are also called digital shams, sham apps, etc, in contexts, but the international standardized definition of digital placebos has never been fully discussed. It is critical to set appropriate digital placebos as comparators and manage them accordingly in the clinical trials to assess the true treatment effect of DTx, but there are still some hurdles related to the validation, the design variability, the patient engagement, the infrequent reports of the results, etc [8].

    The therapeutic areas of DTx are diverse, but psychiatric disorders are one of the major targets because of the benefit from cognitive behavioral interventions [9]. In the International Classification of Diseases 11, the clinical trials of DTx for mental, behavioral, and neurodevelopmental disorders have been mostly conducted recently [5]. In addition to depression, substance use disorders, posttraumatic stress disorders, etc, anxiety disorders are one of the main and notable psychiatric disorders, affecting 301 million people in the world in 2019 [10]. The World Health Organization also indicated that the COVID-19 pandemic triggered 25 % increase in prevalence of anxiety and depression worldwide [11].

    Anxiety disorders are classified into generalized anxiety disorder (GAD), social anxiety disorder (SAD), panic disorder, agoraphobia etc. GAD is defined as excessive anxiety and worry about a number of events or activities, occurring more days than not for at least 6 months [12]. Psychotherapy, such as cognitive behavioral therapy, as well as pharmacotherapy, is widely used as the first-line treatment for anxiety disorders, including GAD, in the various treatment guidelines [13-15]. The disadvantages of face-to-face psychotherapy are low accessibility and high cost. Alternatively, the digital interventions, including DTx, have gained attention for the treatment of psychiatric disorders to overcome the hurdles [9].

    The effect of digital interventions for anxiety, including GAD, has been investigated in the systematic review and the meta-analysis [16]. However, the magnitude and its moderators of digital placebos on GAD have not been investigated. The magnitude of placebo effect is quite large in GAD [17], and it is critical to identify the magnitude and its moderators of digital placebos and to reflect them to the protocols of clinical trials appropriately.

    The objectives of this study were to identify the randomized controlled trials (RCTs), which used digital placebos as comparators, to review the characteristics of the RCTs and of the digital placebos systematically, and to investigate the magnitude and its moderators on GAD assessment scores in the meta-analysis. To our knowledge, this is the first research to investigate the magnitude and its moderators of digital placebos.


    Search Strategy

    This study followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) reporting guidelines [18] (see Checklist 1). In terms of the PICO (Population, Intervention, Control, and Outcome) framework, the target population of this study was adults aged 18 years and older, but was not limited to GAD diagnosed patients to generate a broader data pool. Instead, the impact of difference among the target population was investigated with the subgroup analysis in the meta-analysis. The intervention and the control were limited to DTx as active interventions and to digital placebos as controls, respectively. The study designs were also limited to RCTs. As for the outcome, GAD assessment scores were adopted in this study.

    The digital placebos were defined based on previous research in this study [8]. In brief, digital placebos are comparators designed to mimic the DTx (eg, with a similar design, components, and duration of treatment as the DTx), but the DTx active principle or component being removed or reduced in intensity. The control with a different delivery type from the intervention was not regarded as the digital placebo in this study. For example, delivering control via the web while providing the intervention through virtual reality was not considered as a digital placebo. We included the studies with DTx that were not intended to treat patients in our study if the GAD assessment scores were adopted because the target population in this study was not limited to GAD diagnosed patients.

    The papers reporting the results of clinical trials that used digital placebos as comparators were identified in the following way. The total of 3 databases, including PubMed, Web of Science, and Scopus, were systematically searched with the keywords: Anxiety AND (random* OR RCT) AND (blind OR blinded) AND (digital OR mhealth OR ehealth OR app OR apps OR application* OR smartphone OR mobile OR online OR computer-based OR web-based OR internet-based OR internet-delivered OR “virtual reality ” OR VR OR “ augmented reality ” OR AR OR wearable* OR game* OR gamifi*), in July 2024. Multimedia Appendix 1 shows the detailed search strategy . The search language was limited to English.

    After excluding the duplicates and the papers before 2000 automatically on the reference management software, the papers on nonclinical trials, drugs, and supplements, the ones without results, and the ones with hardware evaluation ( eg, transcranial direct current stimulation and acupuncture) were excluded by the review of the titles and the abstracts of each report.

    Next, the papers without digital placebos as comparators were excluded with the full-text review. The trials which adopted waiting list or treatment-as-usual without any digital interventions were excluded in this process. Subsequently, we excluded the papers from the duplicated studies and the ones for children or adolescents aged 18 years and younger. We also excluded the papers that did not set any GAD assessment scores as the primary or secondary endpoints. Those included the reports which adopted only the scores for SAD and the ones for children or adolescent anxiety. The studies without enough data for statistical analysis, the ones with wrong study designs to evaluate the effect of digital placebos, and the ones with less than 5 pre or post data were excluded. In order to control heterogeneity, the meta-analysis was conducted only with 3 GAD assessment scores, the reported number of which were the largest. The identification of the target papers in this study was conducted by TH and RT independently and the discrepancies were resolved through consensus meetings.

    Data Extraction

    The following information and variables were extracted from the selected papers for this systematic review and meta-analysis: (1) general information (author, clinical trial registration number, and publication year), (2) demographic characteristics (target population and age), (3) placebo device characteristics (placebo delivery type and placebo approach), (4) study design information (number of total patients, blindness, treatment period, number of group, and country), and (5) outcome information (primary and secondary endpoints, mean (SD) of pre and post intervention of comparators).

    The target population was classified into 3 groups in this study. The group of primary psychiatric disorders included the participants who had been diagnosed as psychiatric disorders or who met the criteria of specific psychiatric disorders based on clinical guidelines such as Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition at the timing of screening [12]. The group of other diseases included the patients with underlying conditions rather than psychiatric disorders such as cancers and pains. The others were classified into the group of nonpatients.

    The delivery type and approach of digital placebos was also set based on previous research [8]. The delivery type included “Web”, “Mobile”, “Computer”, “Virtual Reality (VR) / Augmented Reality (AR) ” and “Wearables”.The placebo approach was classified into the following 4 categories. “Replaced” is an approach which replaces the active component with an inactive or neutral component. “Removed” is an approach which simply removes active component. “Unrelated” is an approach that the active component is replaced by a different active component, which is unrelated. “Less” is an approach that is less intense version of the active component.

    Only 1 GAD assessment score was adopted from 1 study. The GAD assessment score in primary endpoint was prioritized if any. If no GAD assessment score was set in primary endpoint and some secondary endpoints for GAD were set in a study, we adopted the endpoint with the smallest placebo effect conservatively. The postintervention data at the end of treatment or the closest were adopted in each study although the long-term effect after the completion of the treatment was investigated in some selected studies.

    If the means or the SDs of the GAD assessment scores for meta-analysis were not reported in the papers, the corresponding author was contacted by email. If no reply was obtained, we imputed the data as recommended by the Cochrane group or extracted the data from the graphs in the papers, if possible [19]. The studies without enough information for meta-analysis, regardless of the work above, were excluded from this study. Sensitivity analysis was conducted by excluding the paper that required imputation for the means or the SDs. The data extraction from the identified papers was conducted by TH and double-checked by RT. The discrepancies were resolved through consensus meetings.

    Statistical Analysis

    The statistical analyses were performed by using R (version 4.4.2; R Foundation for Statistical Computing). The magnitude of placebo effect was computed based on the sample sizes, the means and the SDs of the baseline and the postintervention of the comparator group in each study. We adopted the Hedges g to estimate the magnitude of digital placebo effect because the studies with small sample size were included in this study. A random effects model was used by assuming high heterogeneity. The publication bias was evaluated by the Begg modified funnel plot, the Duval and Tweedie trim-and-fill procedure, and the Egger regression intercept.

    Subgroup analyses and meta-regressions were conducted to explore the reasons for heterogeneity and the potential moderators for the magnitude of digital placebo effect by using variables which were extracted from the selected studies. Variables were considered statistically significant when the P values were under .05 in both the subgroup analyses and the meta-regressions.

    Risk of bias assessment was performed by using the revised Cochrane risk-of-bias tool for randomized trials [19]. Sensitivity analysis was conducted by excluding the studies with high risk of bias.


    Study Selection for Systematic Review

    The PRISMA flowchart (see Figure 1) shows that 2718 records were identified as potentially relevant. Following screening with titles, abstracts, and methods, 374 papers were identified. Through the full-text reviews, 52 papers and 54 studies were identified for the systematic review [20-71]. 1 paper included 2 studies [21] and the population in another article was divided into 2 groups in accordance with the mental risk they defined. We dealt with those as independent studies [25].

    ‎
    Figure 1. PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) chart. The PRISMA chart for systematic review and meta-analysis is shown with the number of records included or excluded at each step. GAD: generalized anxiety disorder.

    Characteristics of the Included Studies for Systematic Review

    Table 1 lists the studies included in this systematic review. Multimedia Appendix 2 summarizes the characteristics of the studies. All the selected studies were randomized, blinded, parallel group comparison studies, except for one in which cross-over design was adopted. Almost 65 % (35/54) of the studies were published in 2019 and after. Target population was almost evenly divided into 3 groups. The placebo delivery type in most included studies was mobile or web. Most of the placebo approach was that the active component was replaced with inactive or neutral components. The total numbers of participants in the selected studies were relatively large and approximately 15% (8/54) of the selected studies included over 501 participants. Almost half of the studies were double-blind studies. The most frequent treatment period and the number of groups was >31 days and <90 days and 2 groups, respectively. The multicountry clinical trials were conducted only in 3 studies and none of them were multiregional clinical trials (MRCTs).

    Table 1. The studies included in the systematic review.
    Study, yearTarget populationAge (years), mean (SD)Placebo deliveryPlacebo approachTotal patients, NBlindnessTreatment (days)Groups, nCountriesEndpoint
    Carlbring et al [20], 2012 Primary38.0 (12.0)WebReplaced79Double282SingleBAI a
    Sharpe et al Study 1 [21], 2012Other diseases 40.6 (15.8)WebReplaced34Double12SingleDASS-Ab
    Sharpe et al Study 2 [21], 2012 Other diseases 45.6 (14.5)WebReplaced34Double282SingleDASS-Ab
    Boettcher et al [22], 2013 Primary 42.8 (11.5)WebReplaced129Double146SingleBAIa
    Glozier et al [23], 2013 Other diseases 58.4 (6.6)WebReplaced562Double842SingleGAD-7c
    Enock et al [24], 2014 Nonpatients 34.8 (11.4) Mobile Replaced 429 Double 28 3 Single PSWQd
    Musiat et al High [25], 2014 Nonpatients 21eWeb Unrelated 181 Single-Pr84 2 Single GAD-7c
    Musiat et al Low [25], 2014 Nonpatients 21eWeb Unrelated 859 Single-P 84 2 Single GAD-7c
    Salemink et al [26], 2014 Primary 38.6 (10.8) Web Replaced 47 Double 11 2 Single STAI-STg
    Taylor-Rodgers and Batterham [27], 2014 Nonpatients 21.9 (1.9) Web Unrelated 67 Single-P 21 2 Single GAD-7c
    Green et al [28], 2014 Other diseases 59.6 (13.4) Web Replaced 200 Single-P 1 2 Single STAI-STg
    Buntrock et al [29], 2016 Nonpatients 44.4 (11.8) Web Replaced 406 Single-Ih 3 ‐ 6 weeks 2 Single HADS-Ai
    Carleton et al [30], 2017 Primary 39.7 (12.5) Web Replaced 113 Double 28 2 Single STAI-TRj
    Dennis-Tiwary et al [31], 2017 Nonpatients 31.1 (6.2) Mobile Replaced 33 Double 28 2 Single DASS-Ab
    McAllister et al [32], 2017 Primary 39.7 (16.0) Web Replaced 31 Double 28 2 Single STAIk
    Peters et al [33], 2017 Nonpatients 24.9 (6.3) Web Replaced 111 Double 42 2 Single STAI-STg
    Baker et al [34], 2018 Other diseases 42.0 (15.4) Web Replaced 39 Single-P 56 2 Single BAIa
    Bennell et al [35], 2018 Other diseases 61.3 (7.1) Web Replaced 144 Double 56 2 Single DASS-Ab
    Kyrios et al [36], 2018 Primary 34.2 (9.9) Web Replaced 179 Double 84 2 Single HAM-Al
    Amir et al [37], 2019 Primary 35.3 (12.0) Computer Removed 169 Double 84 4 Single HAM-Al
    Clarke et al [38], 2019 Other diseases 57.7 (10.0) Web Removed 780 Double 84 2 Single GAD-7c
    Glozier et al [39], 2019 Primary 58.1 (6.1) Web Replaced 87 Double 84 2 Single HAM-Al
    Ahorsu et al [40], 2020 Other diseases 38.0 (9.9) Mobile Replaced 320 Double 42 2 Single HADS-Ai
    Fuller-Tyszkiewicz et al [41], 2020 Nonpatients 39.2 (5.9) Mobile Removed 183 Single-P 35 2 Single DASS-Ab
    Majd et al [42], 2020 Primary 35.3 (5.8) Mobile Replaced 312 Single-I 42 3 Single HADS-Ai
    Andersson et al [43], 2021 Primary 43.0 (14.0) Web Replaced 311 Single-P 70 2 Single PSWQd
    Bove et al [44], 2021 Other diseases 49.2 (10.9) Mobile Replaced 44 Double 42 2 Single STAI-TRj
    Fiol-DeRoque et al [45], 2021 Nonpatients 40.6 (9.6) Mobile Replaced 482 Single-P 14 2 Single DASS-Ab
    Hirsch et al [46], 2021 Primary 35.7 (11.5) Web Replaced 230 Single-P 28 2 Single PSWQd
    Morriss et al [47], 2021 Nonpatients 38.4 (14.3) Web Replaced 790 Single-I 42 2 Single GAD-7c
    Ong et al [48], 2021 Other diseases 58.0 (16.0) Mobile Replaced 182 Single-I 365 2 Single HADS-Ai
    Aganov et al [49], 2022 Nonpatients 39.1 (9.4) VRmReplaced 94 Single-I 7 2 Single STAI-STg
    Cuijpers et al [50], 2022 Primary 27.1 (8.1) Web Replaced 680 Single-P 56 2 Single GAD-7c
    Haas et al [51], 2022 Primary 36.9 (13.2) Web Replaced 161 Single-I 35 3 Single BAIa
    Rodrigues et al [52], 2022 Other diseases 48.5 (16.9) VR Replaced 44 Double 1 2 Single HADSn
    Teles et al [53], 2022 Nonpatients 58.8 (12.4) Web Replaced 42 Single-I 84 2 Single HADS-Ai
    Torok et al [54], 2022 Nonpatients 21.7 (2.2) Mobile Replaced 455 Double 42 2 Single GAD-7c
    Ditton et al [55], 2023 Nonpatients 25.4 (5.8) Mobile Less 108 Double 35 3 Single DASS-Ab
    Eek et al [56], 2023 Primary 53.7 (10.7) Web Less 166 Single-P 84 3 Single GAD-7c
    Ehlers et al [57], 2023 Primary 35.8 (11.5) Web Unrelated 217 Single-I 84 3 Single GAD-7c
    Fatori et al [58], 2023 Primary 32.3 (4.9) Mobile Replaced 81 Double 56 2 Single GAD-7c
    Hoffmann et al [59], 2023 Nonpatients 47.0 (13.0) Web Removed 89 Double 30 2 Multiple HADS-Ai
    Karlsson-Good et al [60], 2023 Primary 39.3 (14.1) Web Less 273 Single-P 84 2 Single BAIa
    Nijman et al [61], 2023 Primary 39.7 (12.4) VR Unrelated 81 Single-I 56 2 Single BAIa
    Sharpe et al [62], 2023 Other diseases 49.9 (13.9) Web Replaced 288 Double 28 4 Single DASS-Ab
    van Gelder et al [63], 2023 Nonpatients 35.1 (NA) Web Replaced 198 Single-P 90 2 Single HADS-Ai
    Zion et al [64], 2023 Other diseases 52.5 (11.4) Mobile Replaced 449 Double 84 2 Single PROMIS-Ao
    Kleinau et al [65], 2024 Nonpatients 24eMobile Replaced 1493 Single-I 56 2 Single GAD-7c
    Romano et al [66], 2024 Other diseases 34.0 (10.2) Web Replaced 62 Single-P 56 2 Multiple DASS-Ab
    Sandhu et al [67], 2024 Nonpatients 30.3 (5.1) Web Removed 306 Single-P 365 2 Single DASS-Ab
    Scazufca et al [68], 2024 Primary 60pMobile Replaced 603 Single-I 42 2 Single GAD-7c
    Thompson et al [69], 2024 Primary 33.0 (1.4) Web Unrelated 133 Single-P 42 2 Multiple DASS-Ab
    Vereschagin et al [70], 2024 Nonpatients 20eMobile Removed 1489 Single-I 30 2 Single GAD-7c
    Zainal et al [71], 2024 Primary 21.8 (3.4) Mobile Removed 191 Single-I 14 2 Single GADQq

    a BAI: Beck Anxiety Inventory.

    b DASS-A: Depression Anxiety Stress Scale-Anxiety.

    c GAD-7: Generalized Anxiety Disorder-7.

    d PSWQ: Penn State Worry Questionnaire.

    e Median (no information for IQR value was described in the paper).

    fSingle-P: single blinding against patients.

    g STAI-ST: State Trait Anxiety Inventory-State.

    hSingle-I: single blinding against investigators.

    i HADS-A: Hospital Anxiety and Depression Scale-Anxiety.

    j STAI-TR: State Trait Anxiety Inventory-Trait.

    k STAI: State Trait Anxiety Inventory.

    l HAM-A: Hamilton Anxiety Rating Scale.

    mVR: virtual reality.

    n HADS: Hospital Anxiety and Depression Scale.

    o PROMIS-A: Patient-Reported Outcomes Measurement Information System-Anxiety.

    p Target age 60-70 years.

    q GADQ: Generalized Anxiety Disorder Questionnaire.

    Study Selection for Meta-Analysis

    The 3 most frequently adopted GAD assessment scores in the selected studies for this meta-analysis were Generalized Anxiety Disorder-7 (GAD-7), Depression Anxiety Stress Scales-Anxiety (DASS-A), and Hospital Anxiety and Depression Scale-Anxiety (HADS-A; see Multimedia Appendix 2) . These 3 assessment scores covered 74.9% (5311/7092) of all cases included in the systematic review. The brief descriptions for each assessment score are as follows.

    GAD-7 is a 7-item self-reported scale to assess anxiety symptoms over the past 2 weeks. Scores range from 0 (not to all) to 3 (nearly every day), with a maximum score of 21 [72]. DASS-A is a 21-item self-reported scale to assess depression, anxiety, and stress. Items are rated on 4-point scale ranging from 0 (did not apply to me at all) to 3 (applied to me very much). DASS-A is the anxiety component of DASS [73]. HADS-A is a 14-item self-reported scale that assesses anxiety (7 items) and depression (7 items). Scores of each item range from 0 to 3 with higher score representing higher levels of anxiety and depression. HADS-A is the anxiety component of the Hospital Anxiety and Depression Scale [74]. All of them are self-reported scales that assess anxiety and several studies reported high correlation among GAD-7, DASS-A, and HADS-A in specific diseases and population [75,76].

    Magnitude of Digital Placebo Effect

    A total of 32 studies, which adopted GAD-7, DASS-A, or HADS-A as assessment scores were included in the meta-analysis with the total number of 5311 [21,23,25,27,29,31,35,38,40-42,45,47,48,50,53-59,62,63,65-70]. The pooled effect size for all included studies was Hedges g=0.28 (95% CI 0.18-0.38) with an overall I2 = 76%. The Egger test did not indicate publication bias (P =.47). We also investigated the risk of publication bias with a funnel plot visually and applied the trim-and-fill method to investigate any asymmetry of the funnel plot. The funnel plot did not show any substantial publication bias and the trim-and-fill analysis showed the presence of 2 unpublished studies (see Multimedia Appendix 3). Considering these studies in the pooled analysis, the magnitude of placebo effect was adjusted to Hedges g=0.24 (95% CI 0.13-0.35). Figure 2 shows the summary the results of risk of bias and Multimedia Appendix 4 shows the detailed results in each study . The placebo effect remained when considering only studies with a low risk of bias (Hedges g=0.26, 95% CI 0.15-0.36). The sensitivity analysis by excluding the studies that required imputation of means or SDs did not show any significant changes on the outcome (Hedges g =0.30, 95% CI 0.20-0.40).

    ‎
    Figure 2. Results of risk of bias assessment of included studies for meta-analysis using the revised Cochrane risk-of-bias tool for randomized trials [19].

    Results of Subgroup Analyses and Meta-Regressions

    The subgroup analysis across the 3 GAD assessment scores adopted in this study did not show the significant difference in the magnitude of placebo effect (P=.16). On the other hand, the subgroup analysis showed the significant difference in the magnitude of placebo effect across the target population (P=.03) although the placebo effect of all the subgroups was statistically significant (see Figure 3). The placebo effect of the subgroup for primary psychiatric disorders was larger than the one for the other 2 groups. The subgroup analysis also showed the significant difference in the magnitude among placebo approach (P=.02) and baseline value (P=.02). The placebo effect of the “Removed” type in placebo approach was smaller than other approaches. Also, the placebo effect was larger in the higher baseline groups in each GAD assessment score than the lower baseline groups.

    ‎
    Figure 3. Forest plot of digital placebo effect from the subgroup analysis across primary psychiatric patients, patients with other diseases rather than psychiatric disorders and nonpatients. IV: inverse variance method; Other diseases: patients with other diseases rather than psychiatric disorders; Primary: primary psychiatric patients; Random: random effects model. Std: standardized [21,23,25,27,29,31,35,38,40,41,42,45,47,48,50,53,54,55,56,57,58,59,62,63,65,66,67,68,69,70].

    We investigated the factors contributing to the high heterogeneity through meta-regression and identified that target population, placebo approach, baseline values, and GAD assessment scores as main contributing factors (R2=58.5%). The meta-regressions in this study also showed that the primary psychiatric patients in target population (P=.01), “Removed” type in placebo approach (P=.04) and high baseline values (P=.02) were statistically significant (see Table 2). There were no strong correlations between the 3 variables. The correlation coefficient between target population and baseline values was 0.15. The one between target population and placebo approach was 0.21. The one between baseline values and placebo approach was 0.09.

    Table 2. Results of meta-regressions.
    Variables Level N Estimate SE P value Lower limit Upper limit
    Publication year N/Aa32 0.015 0.014 .28 − 0.012 0.041
    Age N/A 29 −0.0004 0.004 .93 −0.083 0.0076
    Total number of patients N/A 32 0.0000 0.0001 .89 −0.0003 0.0002
    Treatment period N/A 31 −0.001 0.001 .11 −0.002 0.0002
    Number of groups N/A 32 0.080 0.737 .46 −0.133 0.293
    Target population Nonpatients 32 Reference
    Primary —a 0.308 0.113 .01 0.087 0.530
    Other diseases — −0.005 0.108 .97 −0.216 0.206
    Placebo delivery type Mobile 32 Reference
    Web — 0.003 0.103 .97 −0.198 0.205
    Placebo approach Less 32 Reference
    Removed — −0.048 0.240 .04 −0.954 - 0.013
    Replaced — −0.300 0.223 .18 −0.737 0.136
    Unrelated — −0.328 0.248 .19 −0.814 0.158
    Blindness Double 32 Reference
    Single-investigators — 0.032 0.118 .79 −0.200 0.264
    Single-patients — −0.161 0.118 .17 −0.392 0.070
    Number of countries Multiple 32 Reference
    Single — −0.06 0.207 .77 −0.466 0.345
    GADb assessment scores DASS-Ac32 Reference
    GAD-7d— 0.172 0.114 .13 −0.050 0.395
    HADS-Ae— 0.005 0.138 .97 −0.266 0.276
    Baseline values High 32 Reference
    Low — −0.222 0.094 .02 −0.406 −0.038

    aN/A: not applicable.

    bGAD: generalized anxiety disorder.

    cDASS-A: Depression Anxiety Stress Scales-Anxiety.

    dGAD-7: Generalized Anxiety Disorder-7.

    eHADS-A: Hospital Anxiety and Depression Scale-Anxiety.


    Principal Findings

    The systematic review in this study revealed that digital placebos have been actively adopted as comparators in recent DTx clinical trials and most of the study designs were parallel-group, blind RCTs conducted in a single country. The meta-analysis in our study revealed the small-to-moderate and statistically significant digital placebo effect on GAD assessment scores. The findings were not changed after the adjustment by trim-and-fill method, by risk of bias assessment, or by sensitivity analysis. The subgroup analyses also revealed the significant difference on the placebo effect across the target population, placebo approach, and baseline values. According to the meta-regressions, primary psychiatric disorders in target population, “Removed” type in placebo approach, and higher baseline values of GAD assessment scores were associated with effect of digital placebos. The correlation of each variable was not high and the 3 variables were considered to be independent moderators.

    Comparison With Previous Work

    The finding that digital placebos have been actively adopted as comparators in clinical trials recently is consistent with the recent research [7]. The placebo approach adopted mostly was “replaced with inactive/neutral” in this study, which was consistent with the previous research that covered all the psychiatric disorders [8]. The placebo delivery type adopted mostly was web followed by mobile, and only 1 study adopted computer in this study, which was different from the previous research that computer, wearable, and virtual reality were adopted more [8]. With the advancement of digital technology, the ratio and variety of digital placebos used in clinical trials are expected to change in the future.

    The study designs of the selected studies in this study were very similar to the ones with conventional pharmacotherapy. Most of them were parallel-group, blind RCTs. However, most of the clinical trials selected in this study were conducted in a single country, that was very different from the clinical trials of pharmacotherapy that MRCTs were quite common. The lack of internationally harmonized regulatory review and authorization system to facilitate the MRCTs is considered as one of the reasons in addition to the hurdles of language and sociocultural factors [77]..

    To our knowledge, this study is the first research to investigate the magnitude of placebo effect and its moderators with meta-analysis. The result that the magnitude of the digital placebo effect in the group of primary psychiatric disorders was larger than the other 2 groups is consistent with a previous report that has shown the large placebo effect among GAD patients [17]. Further comprehensive research is needed to investigate the reasons the digital placebo effect in the group of primary psychiatric disorders was higher, but expectancy effects for digital devices and neurological changes observed in pharmacotherapy may occur even with digital placebos [78,79]. At the same time, the results in the group of primary psychiatric disorders in this study indicated that the magnitude of digital placebo effect might be smaller than the placebo effect in pharmacotherapy or neurostimulation [17,80]. Further research, which will compare the magnitude between digital placebos and conventional placebos with the same target population directly is needed to address this question.

    In this study, “Removed” type in placebo approach was identified as a moderators of digital placebo effect. The result might be reasonable because the other 3 approaches still include less, unrelated, or neutral components of active interventions. The finding of higher baseline groups as a moderator might be also reasonable given the wide range of fluctuations. However, more research is needed to clarify whether these variables affect the magnitude of digital placebo effect.

    Limitations

    There are several limitations in this study. First, we defined digital placebos based on previous research, under which no universal definition of digital placebos exists. The systematic review and meta-analysis with different definitions of digital placebos might lead to different conclusions. For example, the conclusion of meta-analysis may change if we include the cases where the delivery types between interventions and the controls were different. Second, this study used only 3 reliable academic databases used in medical and scientific research for the keyword search. The Egger test and funnel plot in this study suggested that publication bias was minimal. However, the risk of bias or incomplete evidence can not be ruled out. Third, this study included only adults aged 18 years and older. Other assessment scores need to be evaluated to investigate the digital placebo effect for children and adolescents. Fourth, we included only 3 GAD assessment scores, the reported number of which were the largest in this meta-analysis to control heterogeneity. These 3 assessment scores covered 74.9% (5311/7092) of all cases included in the systematic review, but more comprehensive studies are needed to discuss the generalizability of the findings in this study. Also, further studies on other diseases are needed to investigate the generalizability of the findings in this study beyond GAD. Fifth, we did not take the long-term effects of digital placebos into consideration. We adopted the postintervention data at the end of treatment or the closest to standardize the conditions although the long-term effect after the treatment was investigated in some selected studies. Finally, we did not consider the psychological or sociocultural factors that may impact on the effect of digital placebos.

    Conclusions

    This study showed the small-to-moderate and statistically significant digital placebo effect on GAD assessment scores. Target population, the digital placebo approach, and high baseline values were also identified as moderators of the digital placebo effect in this study. It would be effective to create the study protocols for the DTx trials with digital placebos by considering the factors identified in this study.

    Authors' Contributions

    TH contributed to the conception and design of the study, analysis and interpretation of the study data, and drafting and revision of the paper. RT independently screened the articles and double-checked the extracted data. YN contributed to the revision of the paper. MK contributed to the conception of the study, interpretation of data, and drafting and revision of the paper. TH and MK are co-corresponding authors.

    Conflicts of Interest

    TH is an employee of NS Pharma, Inc, New Jersey, The United States, which does not offer digital health solutions. This research was conducted independently from the company with no financial or institutional influence.

    Multimedia Appendix 1

    Search strategy.

    DOCX File, 15 KB
    Multimedia Appendix 2

    Characteristics of studies included in the systematic review.

    DOCX File, 18 KB
    Multimedia Appendix 3

    Funnel plot and trim-and-fill funnel plot.

    DOCX File, 53 KB
    Multimedia Appendix 4

    Results of risk of bias in each study.

    DOCX File, 40 KB
    Checklist 1

    PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) 2020 checklist.

    DOCX File, 280 KB
    1. ISO/TR 11147, health informatics-personalized digital health-digital therapeutics health software systems. International Organization for Standardization. URL: https://www.iso.org/obp/ui/#iso:std:iso:tr:11147:ed-1:v1:en [Accessed 2025-03-19]
    2. Wang C, Lee C, Shin H. Digital therapeutics from bench to bedside. NPJ Digit Med. Mar 10, 2023;6(1):38. [CrossRef] [Medline]
    3. Software as a medical device (SaMD) key definitions. International Medical Device Regulators Forum URL: https:/​/www.​imdrf.org/​sites/​default/​files/​docs/​imdrf/​final/​technical/​imdrf-tech-131209-samd-key-definitions-140901.​pdf [Accessed 2025-03-19]
    4. DTx by Country. Digital Therapeutics Alliance. URL: https://dtxalliance.org/understanding-dtx/dtx-by-country/ [Accessed 2025-03-19]
    5. Masanneck L, Stern AD. Tracing digital therapeutics research across medical specialties: evidence from ClinicalTrials.gov. Clin Pharma and Therapeutics. Jul 2024;116(1):177-185. URL: https://ascpt.onlinelibrary.wiley.com/toc/15326535/116/1 [CrossRef] [Medline]
    6. Huh KY, Oh J, Lee S, Yu KS. Clinical evaluation of digital therapeutics: present and future. Healthc Inform Res. Jul 2022;28(3):188-197. [CrossRef] [Medline]
    7. Lutz J, Offidani E, Taraboanta L, Lakhan SE, Campellone TR. Appropriate controls for digital therapeutic clinical trials: A narrative review of control conditions in clinical trials of digital therapeutics (DTx) deploying psychosocial, cognitive, or behavioral content. Front Digit Health. 2022;4:823977. [CrossRef] [Medline]
    8. Bos E, Preller KH, Kaur G, Malhotra P, Kharawala S, Motti D. Challenges with the use of digital sham: systematic review and recommendations. J Med Internet Res. Oct 24, 2023;25:e44764. [CrossRef] [Medline]
    9. Dang A, Arora D, Rane P. Role of digital therapeutics and the changing future of healthcare. J Family Med Prim Care. May 2020;9(5):2207-2213. [CrossRef] [Medline]
    10. Anxiety disorders. World Health Organization. URL: https://www.who.int/news-room/fact-sheets/detail/anxiety-disorders [Accessed 2025-03-19]
    11. COVID-19 pandemic triggers 25% increases in prevalence of anxiety and depression worldwide. World Health Organization. URL: https:/​/www.​who.int/​news/​item/​02-03-2022-covid-19-pandemic-triggers-25-increase-in-prevalence-of-anxiety-and-depression-worldwide [Accessed 2025-03-19]
    12. Diagnostic and Statistical Manual of Mental Disorders (Text Revision) (DSM-5-TR). American Psychiatric Association; 2022. URL: https://www.psychiatry.org/psychiatrists/practice/dsm [Accessed 2025-03-19]
    13. Generalized anxiety disorder and panic disorder in adults: management. National Institute for Health and Care Excellence. 2020. URL: https://www.nice.org.uk/guidance/cg113?unlid=25209853520162212188 [Accessed 2025-03-19]
    14. Generalized anxiety disorder and panic disorder in adults. American Academy of Family Physician. 2015. URL: https://www.aafp.org/pubs/afp/issues/2022/0800/generalized-anxiety-disorder-panic-disorder.html [Accessed 2025-03-19]
    15. Katzman MA, Bleau P, Blier P, et al. Canadian clinical practice guidelines for the management of anxiety, posttraumatic stress and obsessive-compulsive disorders. BMC Psychiatry. 2014;14 Suppl 1(Suppl 1):S1. [CrossRef] [Medline]
    16. Pauley D, Cuijpers P, Papola D, Miguel C, Karyotaki E. Two decades of digital interventions for anxiety disorders: a systematic review and meta-analysis of treatment effectiveness. Psychol Med. Jan 2023;53(2):567-579. [CrossRef] [Medline]
    17. Bschor T, Nagel L, Unger J, Schwarzer G, Baethge C. Differential outcomes of placebo treatment across 9 psychiatric disorders: a systematic review and meta-analysis. JAMA Psychiatry. Aug 1, 2024;81(8):757-768. [CrossRef] [Medline]
    18. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. Mar 29, 2021;372:n71. [CrossRef] [Medline]
    19. Higgins JPT, Thomas J, Chandler J, et al, editors. Cochrane Handbook for Systematic Reviews of Interventions Version. Cochrane; 2024. URL: https://www.cochrane.org/authors/handbooks-and-manuals/handbook [Accessed 2025-03-19]
    20. Carlbring P, Apelstrand M, Sehlin H, et al. Internet-delivered attention bias modification training in individuals with social anxiety disorder--a double blind randomized controlled trial. BMC Psychiatry. Jun 25, 2012;12:66. [CrossRef] [Medline]
    21. Sharpe L, Ianiello M, Dear BF, Perry KN, Refshauge K, Nicholas MK. Is there a potential role for attention bias modification in pain patients? Results of 2 randomised, controlled trials. Pain. Mar 2012;153(3):722-731. [CrossRef] [Medline]
    22. Boettcher J, Leek L, Matson L, et al. Internet-based attention bias modification for social anxiety: a randomised controlled comparison of training towards negative and training towards positive cues. PLoS One. 2013;8(9): e71760 . [CrossRef] [Medline]
    23. Glozier N, Christensen H, Naismith S, et al. Internet-delivered cognitive behavioural therapy for adults with mild to moderate depression and high cardiovascular disease risks: a randomised attention-controlled trial. PLoS ONE. 2013;8(3):e59139. [CrossRef] [Medline]
    24. Enock PM, Hofmann SG, McNally RJ. Attention bias modification training via smartphone to reduce social anxiety: a randomized, controlled multi-session experiment. Cogn Ther Res. Apr 2014;38(2):200-216. [CrossRef]
    25. Musiat P, Conrod P, Treasure J, Tylee A, Williams C, Schmidt U. Targeted prevention of common mental health disorders in university students: randomised controlled trial of a transdiagnostic trait-focused web-based intervention. PLoS ONE. 2014;9(4):e93621. [CrossRef] [Medline]
    26. Salemink E, Kindt M, Rienties H, van den Hout M. Internet-based cognitive bias modification of interpretations in patients with anxiety disorders: a randomised controlled trial. J Behav Ther Exp Psychiatry. Mar 2014;45(1):186-195. [CrossRef] [Medline]
    27. Taylor-Rodgers E, Batterham PJ. Evaluation of an online psychoeducation intervention to promote mental health help seeking attitudes and intentions among young adults: randomised controlled trial. J Affect Disord. Oct 2014;168:65-71. [CrossRef] [Medline]
    28. Green MJ, Schubart JR, Whitehead MM, Farace E, Lehman E, Levi BH. Advance care planning does not adversely affect hope or anxiety among patients with advanced cancer. J Pain Symptom Manage. Jun 2015;49(6):1088-1096. [CrossRef] [Medline]
    29. Buntrock C, Ebert DD, Lehr D, et al. Effect of a web-based guided self-help intervention for prevention of major depression in adults with subthreshold depression: a randomized clinical trial. JAMA. May 3, 2016;315(17):1854-1863. [CrossRef] [Medline]
    30. Carleton RN, Teale Sapach MJN, Oriet C, LeBouthillier DM. Online attention modification for social anxiety disorder: replication of a randomized controlled trial. Cogn Behav Ther. Jan 2017;46(1):44-59. [CrossRef] [Medline]
    31. Dennis-Tiwary TA, Denefrio S, Gelber S. Salutary effects of an attention bias modification mobile application on biobehavioral measures of stress and anxiety during pregnancy. Biol Psychol. Jul 2017;127:148-156. [CrossRef] [Medline]
    32. McAllister J, Gascoine S, Carroll A, et al. Cognitive bias modification for social anxiety in adults who stutter: a feasibility study of a randomised controlled trial. BMJ Open. Oct 22, 2017;7(10): e015601 . [CrossRef] [Medline]
    33. Peters SE, Lumsden J, Peh OH, Penton-Voak IS, Munafò MR, Robinson OJ. Cognitive bias modification for facial interpretation: a randomized controlled trial of transfer to self-report and cognitive measures in a healthy sample. R Soc Open Sci. Dec 2017;4(12):170681. [CrossRef] [Medline]
    34. Baker KS, Georgiou-Karistianis N, Lampit A, Valenzuela M, Gibson SJ, Giummarra MJ. Computerised training improves cognitive performance in chronic pain: a participant-blinded randomised active-controlled trial with remote supervision. Pain. Apr 2018;159(4):644-655. [CrossRef] [Medline]
    35. Bennell KL, Nelligan RK, Rini C, et al. Effects of internet-based pain coping skills training before home exercise for individuals with hip osteoarthritis (HOPE trial): a randomised controlled trial. Pain. Sep 2018;159(9):1833-1842. [CrossRef] [Medline]
    36. Kyrios M, Ahern C, Fassnacht DB, Nedeljkovic M, Moulding R, Meyer D. Therapist-assisted internet-based cognitive behavioral therapy versus progressive relaxation in obsessive-compulsive disorder: randomized controlled trial. J Med Internet Res. Aug 8, 2018;20(8):e242. [CrossRef] [Medline]
    37. Amir N, Taboas W, Montero M. Feasibility and dissemination of a computerized home-based treatment for generalized anxiety disorder: a randomized clinical trial. Behav Res Ther. Sep 2019;120:103446. [CrossRef] [Medline]
    38. Clarke J, Sanatkar S, Baldwin PA, et al. A web-based cognitive behavior therapy intervention to improve social and occupational functioning in adults with type 2 diabetes (the SpringboarD trial): randomized controlled trial. J Med Internet Res. May 21, 2019;21(5):e12246. [CrossRef] [Medline]
    39. Glozier N, Christensen H, Griffiths KM, et al. Adjunctive internet-delivered cognitive behavioural therapy for insomnia in men with depression: a randomised controlled trial. Aust N Z J Psychiatry. Apr 2019;53(4):350-360. [CrossRef] [Medline]
    40. Ahorsu DK, Lin CY, Imani V, et al. Testing an app-based intervention to improve insomnia in patients with epilepsy: a randomized controlled trial. Epilepsy Behav. Nov 2020;112:107371. [CrossRef] [Medline]
    41. Fuller-Tyszkiewicz M, Richardson B, Little K, et al. Efficacy of a smartphone app intervention for reducing caregiver stress: randomized controlled trial. JMIR Ment Health. Jul 24, 2020;7(7):e17541. [CrossRef] [Medline]
    42. Rajabi Majd N, Broström A, Ulander M, et al. Efficacy of a theory-based cognitive behavioral technique app-based intervention for patients with insomnia: randomized controlled trial. J Med Internet Res. Apr 1, 2020;22(4):e15841. [CrossRef] [Medline]
    43. Andersson E, Ljótsson B, Hedman-Lagerlöf M, et al. Targeting excessive worry with internet-based extinction therapy: a randomised controlled trial with mediation analysis and economical evaluation. Psychol Med. Sep 2021;51(12):2023-2033. [CrossRef] [Medline]
    44. Bove R, Rowles W, Zhao C, et al. A novel in-home digital treatment to improve processing speed in people with multiple sclerosis: a pilot study. Mult Scler. Apr 2021;27(5):778-789. [CrossRef] [Medline]
    45. Fiol-DeRoque MA, Serrano-Ripoll MJ, Jiménez R, et al. A mobile phone-based intervention to reduce mental health problems in health care workers during the COVID-19 pandemic (PsyCovidApp): randomized controlled trial. JMIR Mhealth Uhealth. May 18, 2021;9(5):e27039. [CrossRef] [Medline]
    46. Hirsch CR, Krahé C, Whyte J, et al. Internet-delivered interpretation training reduces worry and anxiety in individuals with generalized anxiety disorder: a randomized controlled experiment. J Consult Clin Psychol. Jul 2021;89(7):575-589. [CrossRef] [Medline]
    47. Morriss R, Kaylor-Hughes C, Rawsthorne M, et al. A direct-to-public peer support program (Big White Wall) versus web-based information to aid the self-management of depression and anxiety: results and challenges of an automated randomized controlled trial. J Med Internet Res. Apr 23, 2021;23(4):e23487. [CrossRef] [Medline]
    48. Ong SW, Jassal SV, Porter EC, et al. Digital applications targeting medication safety in ambulatory high-risk CKD patients: randomized controlled clinical trial. CJASN. 2021;16(4):532-542. [CrossRef] [Medline]
    49. Aganov S, Nayshtetik E, Nagibin V, Lebed Y. Pure purr virtual reality technology: measuring heart rate variability and anxiety levels in healthy volunteers affected by moderate stress. Arch Med Sci. 2022;18(2):336-343. [CrossRef] [Medline]
    50. Cuijpers P, Heim E, Ramia JA, et al. Guided digital health intervention for depression in Lebanon: randomised trial. Evid Based Ment Health. Dec 2022;25(e1):e34-e40. [CrossRef] [Medline]
    51. Haas M, Jakubovski E, Kunert K, et al. ONLINE-TICS: internet-delivered behavioral treatment for patients with chronic tic disorders. JCM. Jan 4, 2022;11(1):250. [CrossRef] [Medline]
    52. Rodrigues IM, Lima AG, Santos AD, et al. A single session of virtual reality improved tiredness, shortness of breath, anxiety, depression and well-being in hospitalized individuals with COVID-19: a randomized clinical trial. JPM. May 19, 2022;12(5):829. [CrossRef] [Medline]
    53. Teles S, Ferreira A, Paúl C. Feasibility of an online training and support program for dementia carers: results from a mixed-methods pilot randomized controlled trial. BMC Geriatr. Mar 1, 2022;22(1):173. [CrossRef] [Medline]
    54. Torok M, Han J, McGillivray L, et al. The effect of a therapeutic smartphone application on suicidal ideation in young adults: findings from a randomized controlled trial in Australia. PLoS Med. May 2022;19(5):e1003978. [CrossRef] [Medline]
    55. Ditton E, Knott B, Hodyl N, et al. Evaluation of an app-delivered psychological flexibility skill training intervention for medical student burnout and well-being: randomized controlled trial. JMIR Ment Health. Feb 6, 2023;10:e42566. [CrossRef] [Medline]
    56. Eék N, Sundström C, Kraepelien M, Lundgren J, Kaldo V, Berman AH. High- versus low-intensity internet interventions for alcohol use disorders (AUD): a two-year follow-up of a single-blind randomized controlled trial. Internet Interv. Sep 2023;33:100630. [CrossRef] [Medline]
    57. Ehlers A, Wild J, Warnock-Parkes E, et al. Therapist-assisted online psychological therapies differing in trauma focus for post-traumatic stress disorder (STOP-PTSD): a UK-based, single-blind, randomised controlled trial. Lancet Psychiatry. Aug 2023;10(8):608-622. [CrossRef] [Medline]
    58. Fatori D, Zuccolo P, Xavier MO, Matijasevich A, Polanczyk GV. Smartphone-assisted online brief cognitive behavioral therapy to treat maternal depression: findings of a randomized controlled trial. Braz J Psychiatry. Mar 13, 2023;45(1):50-53. [CrossRef] [Medline]
    59. Hoffmann M, Jeitziner MM, Riedl R, et al. Effects of an online information tool on post-traumatic stress disorder in relatives of intensive care unit patients: a multicenter double-blind, randomized, placebo-controlled trial (ICU-Families-Study). Intensive Care Med. Nov 2023;49(11):1317-1326. [CrossRef] [Medline]
    60. Karlsson-Good M, Kaldo V, Lundberg L, Kraepelien M, Anthony SA, Holländare F. Increasing the accessibility to internet-based cognitive behavioural therapy for depression: a single-blind randomized controlled trial of condensed versus full-text versions. Internet Interv. Dec 2023;34:100678. [CrossRef] [Medline]
    61. Nijman SA, Pijnenborg GHM, Vermeer RR, et al. Dynamic Interactive Social Cognition Training in Virtual Reality (DiSCoVR) versus Virtual Reality Relaxation (VRelax) for people with a psychotic disorder: a single-blind multicenter randomized controlled trial. Schizophr Bull. Mar 15, 2023;49(2):518-530. [CrossRef] [Medline]
    62. Sharpe L, Jones EB, Pradhan P, Todd J, Colagiuri B. A double-blind phase II randomized controlled trial of an online cognitive bias modification for interpretation program with and without psychoeducation for people with chronic pain. Pain. Apr 1, 2023;164(4):e217-e227. [CrossRef] [Medline]
    63. van Gelder NE, Ligthart SA, van Rosmalen-Nooijens KA, Prins JB, Oertelt-Prigione S. Effectiveness of the SAFE eHealth intervention for women experiencing intimate partner violence and abuse: randomized controlled trial, quantitative process evaluation, and open feasibility study. J Med Internet Res. Jun 27, 2023;25:e42641. [CrossRef] [Medline]
    64. Zion SR, Taub CJ, Heathcote LC, et al. Effects of a cognitive behavioral digital therapeutic on anxiety and depression symptoms in patients with cancer: a randomized controlled trial. JCO Oncol Pract. Dec 2023;19(12):1179-1189. [CrossRef] [Medline]
    65. Kleinau E, Lamba T, Jaskiewicz W, et al. Effectiveness of a chatbot in improving the mental wellbeing of health workers in Malawi during the COVID-19 pandemic: a randomized, controlled trial. PLoS ONE. 2024;19(5):e0303370. [CrossRef] [Medline]
    66. Romano D, Chesterman S, Fuller-Tyszkiewicz M, et al. Feasibility, acceptability, and preliminary efficacy of acceptance commitment therapy for adults living with inflammatory bowel disease and distress. Inflamm Bowel Dis. Jun 3, 2024;30(6):911-921. [CrossRef] [Medline]
    67. Sandhu S, Hickey M, Koye DN, et al. Eggsurance? A randomized controlled trial of a decision aid for elective egg freezing. Hum Reprod. Aug 1, 2024;39(8):1724-1734. [CrossRef] [Medline]
    68. Scazufca M, Nakamura CA, Seward N, et al. Self-help mobile messaging intervention for depression among older adults in resource-limited settings: a randomized controlled trial. Nat Med. Apr 2024;30(4):1127-1133. [CrossRef] [Medline]
    69. Thompson EM, Albertella L, Viskovich S, Pakenham KI, Fontenelle LF. Internet-based acceptance and commitment therapy for obsessive-compulsive symptoms: a randomized controlled trial. Behav Res Ther. Sep 2024;180:104595. [CrossRef] [Medline]
    70. Vereschagin M, Wang AY, Richardson CG, et al. Effectiveness of the minder mobile mental health and substance use intervention for university students: randomized controlled trial. J Med Internet Res. Mar 27, 2024;26:e54287. [CrossRef] [Medline]
    71. Zainal NH, Tan HH, Hong RYS, Newman MG. Testing the efficacy of a brief, self-guided mindfulness ecological momentary intervention on emotion regulation and self-compassion in social anxiety disorder: randomized controlled trial. JMIR Ment Health. Apr 19, 2024;11:e53712. [CrossRef] [Medline]
    72. Spitzer RL, Kroenke K, Williams JBW, Löwe B. A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch Intern Med. May 22, 2006;166(10):1092-1097. [CrossRef] [Medline]
    73. Lovibond PF, Lovibond SH. The structure of negative emotional states: comparison of the Depression Anxiety Stress Scales (DASS) with the Beck Depression and Anxiety Inventories. Behav Res Ther. Mar 1995;33(3):335-343. [CrossRef]
    74. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand. Jun 1983;67(6):361-370. [CrossRef] [Medline]
    75. Peters L, Peters A, Andreopoulos E, Pollock N, Pande RL, Mochari-Greenberger H. Comparison of DASS-21, PHQ-8, and GAD-7 in a virtual behavioral health care setting. Heliyon. Mar 2021;7(3):e06473. [CrossRef] [Medline]
    76. Huang XJ, Ma HY, Wang XM, Zhong J, Sheng DF, Xu MZ. Equating the PHQ-9 and GAD-7 to the HADS depression and anxiety subscales in patients with major depressive disorder. J Affect Disord. Aug 15, 2022;311:327-335. [CrossRef] [Medline]
    77. Hosono T, Niwa Y, Kondoh M. Comparison of product features and clinical trial designs for the DTx products with the indication of insomnia authorized by regulatory authorities. Ther Innov Regul Sci. Nov 2024;58(6):1138-1147. [CrossRef] [Medline]
    78. Torous J, Firth J. The digital placebo effect: mobile mental health meets clinical psychiatry. Lancet Psychiatry. Feb 2016;3(2):100-102. [CrossRef] [Medline]
    79. Niazi SK. Placebo effects: neurological mechanisms inducing physiological, organic, and belief responses-a prospective analysis. Healthcare (Basel). Nov 20, 2024;12(22):2314. [CrossRef] [Medline]
    80. Huneke NTM, Amin J, Baldwin DS, et al. Placebo effects in randomized trials of pharmacological and neurostimulation interventions for mental disorders: an umbrella review. Mol Psychiatry. Dec 2024;29(12):3915-3925. [CrossRef] [Medline]

    ‎
    DASS-A: Depression Anxiety Stress Scales-Anxiety
    DTx: digital therapeutics
    GAD: generalized anxiety disorder
    GAD-7: Generalized Anxiety Disorder-7
    HADS-A: Hospital Anxiety and Depression Scale-Anxiety
    MRCTs: multiregional clinical trials
    PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses
    RCT: randomized controlled trial
    SAD: social anxiety disorder
    SaMD: software as a medical device

    Edited by Javad Sarvestan; submitted 24.03.25; peer-reviewed by Mohammad Eghbal Heidari, Shukun Yang; final revised version received 17.06.25; accepted 18.06.25; published 31.07.25.

    Copyright

    © Takashi Hosono, Rinka Tsutsumi, Yuki Niwa, Masuo Kondoh. Originally published in the Journal of Medical Internet Research (https://www.jmir.org) , 31.7.2025.

    This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research (ISSN 1438-8871), is properly cited. The complete bibliographic information, a link to the original publication on https://www.jmir.org/, as well as this copyright and license information must be included.