Body image encompasses an individual’s cognition, emotional attitudes, and behavioral regulation regarding their body, and is a multidimensional concept [1]. Dissatisfaction with one’s body’s appearance, including its shape, weight, and other features, is known as body dissatisfaction [2]. Body dissatisfaction has a pervasive hold on children and emerging adults, and the phenomenon has been described as “normative discontent” [3]. Studies showed the prevalence of body dissatisfaction ranged from 42.2% to 80.9% among children and emerging adults across the world [4-7]. Females, those with higher BMI, and those from minority groups were reported to experience greater body dissatisfaction [8]. Body dissatisfaction can lead to a range of physiological issues, such as low self-esteem [9], anxiety [10], and depression [11], as well as behavioral burdens, such as eating disorders [12], extreme weight loss behaviors [13], alcohol, drug abuse [14], and excessive pursuit of cosmetic procedures [15]. The direct economic costs of body dissatisfaction were estimated to be US $84 billion in the United States [16].

Existing face-to-face interventions, such as cognitive behavioral therapy (CBT) and behavior interventions and supports (BIS), demonstrated small to moderate efficacy in reducing body dissatisfaction. For instance, CBT helps reframe maladaptive thoughts about body image [17], while dissonance-based interventions like the EVERYbody Project reduced thin-ideal internalization and eating disorder symptoms in college students [18]. A systematic review revealed that behavior interventions and supports improved disordered eating, body dissatisfaction, and extreme weight control behaviors in girls, but not boys [19]. However, the face-to-face interventions are sometimes constrained by accessibility, the costly nature of the format, a global shortage of mental health professionals, or stigma [20].

Digital technologies offered cost-effective alternatives in body dissatisfaction interventions [21]. Although studies identified that social media engagement was associated with higher body dissatisfaction and restricting food [22], digital interventions could offer especially critical support in adolescents [23]. A systematic review assessed the effectiveness of universal body image interventions delivered through a digital platform among young women. Most articles indicated that these interventions were effective in improving at least one body image outcome [24]. However, findings in other studies remained inconsistent. A study involving 127 girls aged 10-13 years, using videos developed by the Dove Self-Esteem Project, revealed that after intervention, there was no significant difference between the intervention group and the control group on body satisfaction [25]. This highlights the need for a comprehensive synthesis of current evidence.

There was no systematic review to explore the effectiveness of digital interventions on body dissatisfaction in children and emerging adults through quantitative synthesis. In this regard, this study aimed to fill this identified gap in the literature through a meta-analysis and to synthesize the effectiveness of various digital interventions on children and emerging adults with body dissatisfaction. By providing meta-analyses of currently available randomized controlled trials in children and emerging adults, this paper aimed to consolidate evidence on the use of digital interventions to treat body dissatisfaction.

This systematic review was reported in line with PRISMA (Preferred Reporting Items for Systematic Review and Meta-analysis) guidelines [26] (the PRISMA checklist is provided in Checklist 1). PubMed, Web of Science, MEDLINE, EBSCO (Elton B Stephens Company), Cochrane Library, the Chinese databases CNKI (China National Knowledge Infrastructure) and WANFANG were searched from inception to April 24, 2024. This process was carried out independently by 2 researchers (LL & JNY). The PICOS (population, intervention, comparison, outcome, and study design) formula was used:

The final search strategy is #1 AND #2 AND #3, with #4 (Type of Study: randomized controlled trial) applied as a filter. Reference lists of publications were also searched for potentially relevant studies (see Multimedia Appendix 1).

This systematic review is based on the synthesis of previously published studies and does not involve the collection of primary data directly from human participants. As such, formal ethical approval, including an institutional review board approval number, informed consent, and compensation, was not applicable. However, we have adhered to the ethical principles of research, including the accurate and transparent reporting of study findings. We have also ensured that all included studies were conducted in accordance with relevant ethical guidelines and regulations.

A total of 808 records were identified from the initial literature search. After removing the duplicates, titles and abstracts were screened, and 42 studies were further reviewed in full text. Finally, 20 studies [25,38-56] met the inclusion criteria, including 9 [41-43,49,50,53-56] studies identified from reference lists. The study selection flow chart is shown in Figure 1.

All the studies included were RCTs, including randomized waitlist-controlled trials (n=7). Only the data in the first phase were extracted in the randomized crossover and waitlist trials. Of the included studies, 20 RCTs with 5251 participants were included, including 2610 participants in the intervention groups and 2641 participants in the control groups. The studies were published between 2000 and 2023 and were conducted in the United States (n=10), Australia (n=3), the United Kingdom (n=3), China (n=1), Indonesia (n=1), Italy (n=1), and Brazil (n=1). Five studies [38-42] had mixed-gender samples, 15 [25,42-56] with only females. Sample sizes of the included studies ranged from 53 to 2000. All studies reported postintervention effects, whereas short-term and long-term follow-up data were reported in 11 [38,42,43,45,49,51-56] and 3 [39-41] studies, respectively. Participants and study characteristics are summarized in Table 1.

The digital interventions mainly included web page (n=6), computer-based sessions (n=1), internet-based sessions (n=2), computer-based videos (n=2), social media–based video (n=1), social media (n=3), such as Facebook and podcasts, mobile apps (n=2), internet games (n=1), animated films (n=1), and chatbot (n=1). Among the included studies, the interventions varied in length from 1 minute to 2 hours per session and in duration, ranging from a single day up to 8 weeks.

Of the 20 included studies, only 4 (20%) were assessed as low risk of bias, while 13 (65%) demonstrated high risk of bias, and 3 (15%) were categorized as moderate risk of bias (see Multimedia Appendix 2).

This systematic review and meta-analysis synthesized evidence from 20 RCTs involving 5251 participants to evaluate the effectiveness of digital interventions (eg, web pages, mobile apps, and social media) on body dissatisfaction and related outcomes. Key findings indicated that digital interventions significantly improved body dissatisfaction, physical appearance comparison, thin-ideal internalization, self-esteem, self-compassion, and depression, with small to medium effect sizes. However, there was substantial heterogeneity across studies, and potential publication bias was detected. Digital interventions did not significantly improve negative affect (including depression and anxiety) and eating behaviors.

All the included studies have been conducted in the last 2 decades, but only 2 studies have been done in developing countries. This indicates that high-quality RCTs need to be conducted further in this area, especially to explore their effectiveness on body dissatisfaction in different ethnic and cultural backgrounds in developing countries [57]. Ten out of twenty included studies selected different theories or models to motivate participants to reduce the pursuit of thin-ideal internalization, to improve emotional regulation, such as self-compassion, and to develop coping strategies with the involvement of parents, peers, and the media. The cognitive dissonance theory, cognitive behavioral theory, and sociocultural theory were recognized as the most widely mentioned theories to reduce body dissatisfaction in the studies. The average sample size was high 200s, but most studies included participants of less than 100. This would result from the narrow age range and the diagnosis of anorexia nervosa and other eating or psychiatric disorders among children and emerging adults in the inclusion criteria. The participants’ ages ranged from 7 to 25 years old, and 6 studies included those younger than 14 years old in private or public schools as participants. Nearly 80% of participants were females in this review, and 9 studies only included female college students or undergraduates as participants, which accounted for 16.7% of the total participants. This can indicate that females would be more likely to experience body dissatisfaction, like other studies [58].

The meta-analysis revealed that digital interventions showed a small to medium effect across most outcomes in body satisfaction and internalized outcomes, and the result was similar to another systematic review [24]. In the systematic review, although there was no quantitative synthesis, body dissatisfaction, body esteem, body appreciation, and other body image outcomes were also included in the review to explore the effectiveness of digital interventions in emerging adults and emerging adults. The results showed 8 out of 15 studies reported digital interventions were effective in improving at least one body image outcome from pre-post interventions with mostly small to medium effect sizes. This means digital interventions are superior to some traditional interventions, such as brochures, documentaries, newspapers, or magazines. It may result from the advantages of digital interventions, such as real-time feedback and motivation enhancement [59], and anonymity of platforms [60], which could reduce the shame and social anxiety emerging adults face when dealing with body image-related issues. However, given that effect sizes were small to medium, it would be caused by the following reasons. First, the intervention duration of this meta-analysis was 6 weeks on average. Second, digital intervention had its limitations, such as being incapable of genuinely caring about one’s feelings during the intervention process. This indicated that digital interventions with an in-person element would be associated with greater effectiveness in body image [61].

Digital interventions did not significantly reduce the negative affect (including depression and anxiety) and eating behaviors. However, another 2 systematic reviews [61,62] found a significant effect of digital interventions on depression and anxiety among emerging adults and emerging adults. This would result from the contents of the digital interventions. The digital interventions of the 2 systematic reviews specifically targeted mental health disorders, including depression and anxiety. However, this meta-analysis mainly targeted improving body dissatisfaction with depression and anxiety as secondary outcomes in some studies included. Furthermore, it indicated that digital psychotherapy did not significantly reduce both eating disorders and restrained eating in this meta-analysis. This would be caused by the following reasons. First, no more than 4 RCTs were included in this meta-analysis to explore the effects of digital interventions on eating disorders or restraint eating. Second, the age of participants in this meta-analysis ranged from 7 to 25 years. Especially for the emerging adults who were predominant participants, who have already established eating behavior habits, which would be so ingrained and resistant to change through short-term digital psychological therapy. According to the Transtheoretical Model [63], the process of behavior change is complicated with 6 different stages, and it would not be easy to change eating behaviors. In addition, the digital interventions in the 20 included RCTs were aimed at addressing body dissatisfaction rather than specific eating disorder symptoms, which might account for the insignificant outcome in changing eating behaviors.

The findings suggest that digital interventions can serve as scalable and cost-effective supplementary tools to improve body dissatisfaction in children and emerging adults. Clinicians may recommend digital interventions as adjuncts to traditional face-to-face therapies, particularly for individuals with limited access to in-person mental health services due to cost, stigma, or geographic barriers. Body dissatisfaction, thin-ideal internalization, eating disorders, and negative affect should be addressed simultaneously in a single intervention to verify the effectiveness of digital interventions in the future. Future studies should be conducted in Asia and some other low-income countries, as this review found that studies in these regions are lacking. Furthermore, the quality of included studies held room for improvement.

This systematic review has several limitations. First, the high heterogeneity in meta-analysis may result from the use of different scales to assess the same outcomes, even using different versions or varying score systems (eg, summed versus mean scores) of the same scales. However, the SMD was selected to express the size of the intervention effect, instead of the mean difference. Second, confidence in the meta-analysis was limited because waitlist groups were set in 7 studies, and assessment-only control (receive no intervention) groups were set in another 4 studies, rather than active control groups. Third, there was insufficient data on long‐term follow‐up and a potential gap that future RCTs can look to fill. Specific tactics could be analyzed to maintain engagement in body image interventions for children and emerging adults, such as integrating digital interventions into existing curricula. Fourth, the quality of included studies held room for improvement, with most of the RCTs included being rated as “unclear risk” for bias. Fifth, publication bias may have influenced the results, particularly for outcomes with small sample sizes, highlighting the need for larger, registered RCTs.

Digital interventions could help children and emerging adults improve body satisfaction or dissatisfaction, physical appearance comparison, thin-ideal internalization, self-esteem, self-compassion, and depression. Because of the limitations, the results should be generalized with caution. In the future, high-quality RCTs with longer intervention duration and long-term follow-up should be conducted, especially in different races and cultures in transitional countries.