This review was conducted following the guidelines of the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) [25] (Multimedia Appendix 1). The protocol was registered in PROSPERO (CRD42024570999, accessed on August 2, 2024). Studies were identified through searches of the following electronic databases: CINAHL, Embase, PsycINFO, PubMed, Scopus, The Cochrane Library, and Web of Science. The search was conducted without time or language restrictions. The research question posed was “Is the use of smartphone apps effective for weight loss?” The studies were searched using specific terms related to the use of smartphone apps (“smartphone application,” “mobile application,” “app,” and “m-health”) and body weight management (“obesity,” “overweight,” “body weight,” “weight loss,” and “weight management”; Multimedia Appendix 2). To ensure the selection of relevant and high-quality studies, the search terms also included keywords related to randomized clinical trials (“randomized,” “randomised,” and “control group”; Multimedia Appendix 2). The last search was conducted from the inception of the electronic databases up until July 23, 2024.

This review was designed to evaluate the effectiveness of smartphone-only interventions (apps) for weight loss. In particular, for the purposes of this study, we considered all weight loss apps, regardless of their specific features, such as goal planning, self-monitoring, motivational support, health education, and artificial intelligence (AI) integration. These key features of the apps included in the meta-analysis are detailed in Multimedia Appendix 3 [26-36].

The review included only studies that (1) involved adult participants (aged ≥18 years), (2) included individuals with overweight or obesity (BMI≥25 kg/m²), (3) evaluated the effectiveness of a smartphone app, and (4) reported weight loss as one of the outcomes. Only randomized controlled trials (RCTs) were included, along with studies that used either a nutritional intervention or no intervention (free diet) as a control.

Studies were excluded if they involved participants who did not have overweight or obesity or who had other comorbidities. Additionally, studies that did not follow an RCT design (eg, pilot studies or secondary analyses of RCTs, in which case the original RCT was retrieved), studies using smartphone apps providing only text message reminders via SMS, and studies where smartphone apps were used in combination with wearable devices (eg, step counters, smartwatches to manage or promote physical activity, or electronic weighing scales connected to the mobile app) were also excluded.

Furthermore, any studies not published in English were excluded. Systematic reviews and meta-analyses that assessed the use of smartphone apps for weight loss were examined to identify additional papers not captured in the initial database search. All extracted studies were reviewed, and the titles and abstracts were screened by 2 independent researchers (AP and EM) using the Rayyan application for systematic reviews to remove duplicates and assess eligibility based on the inclusion criteria. Any rejected papers were reviewed by a third researcher (SM) to confirm or refute exclusion. Disagreements between the 2 researchers were evaluated by the third researcher. Finally, the full texts of all papers were read independently by the same authors and further discussed.

Data extraction was independently performed by 2 researchers (CP and YF) using a Microsoft Excel spreadsheet template that collected information on the following criteria: authors; year of publication; country; sample size; sample characteristics; mean baseline BMI; mean population age; type of intervention and control; intervention duration; follow-up points; whether the study was funded; conflicts of interest; and study outcomes reported as either mean (SD), mean (SE), or mean difference (MD; 95% CI). The quality of the studies included in the meta-analysis was independently assessed using the Revised Cochrane Risk-of-Bias Tool for Randomized Trials. Based on the study methodology, the quality of the studies was rated as low, unclear, or high risk of bias. The following domains were assessed to determine study quality: randomization process, allocation concealment, blinding of participants and personnel to the interventions, blinding of outcome assessment, handling of missing data, and reporting of end-point outcomes in the study [37]. Two independent researchers (AP and EM) evaluated the risk of bias in the included studies using standardized data extraction tables. Any discrepancies were resolved through discussion with a third researcher (SM) until a consensus was reached.

The meta-analysis was conducted using all studies that were homogeneous in terms of interventions, reported outcomes, and included information on the end points analyzed in this study. Data reported as SE and CI were converted to SD. Additionally, if body weight was reported in pounds, it was converted to kilograms. The effect size on weight measured before 3 months was estimated using the weighted mean difference, while the standardized mean difference (SMD) was used for subsequent evaluations. For secondary outcomes such as BMI, WC, and body fat percentage, the effect size was estimated using weighted mean difference. All analyses were conducted using random-effects models. Heterogeneity between studies was estimated using the chi-square test and quantified using the I² inconsistency measure, where 25%, 50%, and 75% indicate low, moderate, and high heterogeneity, respectively [38]. Values above 50% indicate substantial heterogeneity [39]. Finally, funnel plots were used as graphical tools to assess study precision and systematic heterogeneity. A symmetric funnel plot suggested that publication bias was less likely, while an asymmetric plot indicated publication bias. All statistical analyses were performed using Review Manager software (version 5.4; Cochrane Training).

This meta-analysis on studies evaluating the effects of weight loss apps stems from the need to explore an area of growing interest for both the scientific community and the general public. In an era where mobile technology is playing a central role in managing health and personal well-being—especially with the widespread use of smartphones and internet connectivity—the interest in these apps is motivated by their potential effectiveness in supporting individuals with weight management. These apps integrate behavioral strategies, monitoring, and nutritional education directly on mobile devices, making them accessible and customizable, particularly given the significant nonadherence of a large part of the population to international healthy lifestyle recommendations. This meta-analysis aims to deepen our understanding not only of the effectiveness of such tools but also to identify areas for improvement and current challenges, with the prospect of developing an innovative app that can address some of the issues found in the existing literature.

Most of the studies included in this meta-analysis reported results on weight loss at 3 and 6 months, but few explored the impact on other parameters such as BMI, WC, and body fat percentage. The results of the meta-analysis indicate that smartphone app–based interventions lasting between 4 and 6 months resulted in a significant reduction in body weight (n=8 studies; n=1466 participants; SMD −0.33, 95% CI –0.48 to –0.17; P<.001; I²=49%). However, no statistically significant reductions in body weight were observed before 3 months or in other analyzed time intervals. The weight loss observed in this meta-analysis is lower compared to 2 previous studies. One study reported significant weight losses of –1.99 and –2.8 kg at 3 and 6 months, respectively, analyzing a population consisting of both normal-weight individuals and those who were individuals with overweight or obesity [42]. Another study reported a weight reduction of –2.18 kg at 3 months and –1.63 kg at 12 months, but it included smartphone apps with additional components such as pedometers and weight scales as well as patients with overweight or obesity and metabolic conditions [43]. This meta-analysis specifically aimed to investigate the effect of exclusive use of smartphone apps on body weight in individuals with uncomplicated overweight and obesity, comparing it with standard nutritional treatments or no treatment at all. Regarding other outcomes, it was found that smartphone app interventions lasting between 4 and 6 months led to a significant reduction in BMI (–0.76 kg/m²). Additionally, there was a significant reduction in body fat percentage at 3 months (–0.79%) and between 4 and 6 months of intervention (–0.46%). This meta-analysis confirms that smartphone app–based interventions appear to be ineffective in promoting clinically significant weight loss, with minimal and unsustainable effects over the long term, consistent with existing literature. Indeed, many studies, including those by Unick et al [44] and Al Naabi et al [45], have highlighted that while weight loss apps are often effective in the short term, the ability to maintain achieved results diminishes over time. This phenomenon is primarily attributed to a lack of continuous personalized support and strategies to address stalls or regressions in the weight loss journey. Supporting this is the finding that approximately 75% of users discontinue use of these apps within a short period [46].

Several reasons for this trend have been analyzed in a review [47], with a key issue being the limited personalization of apps. Many apps rely on standardized approaches that do not adequately consider individual needs, such as pre-existing medical conditions, dietary preferences, or physical limitations. The lack of individualization can reduce the overall effectiveness of the app and increase the risk of dropout. This is supported by the analysis of studies in this meta-analysis, which indicates that most of the tested apps lack personalization, while better results are achieved by apps that incorporate some form of customization, such as the study by Carter et al [28], which provided personalized feedback and resulted in a nearly 5-kg reduction in weight at 6 months. Furthermore, many apps tend to focus on quantitative monitoring of weight and calories [48], neglecting psychological and emotional aspects, such as motivation and stress management, which influence long-term results. Indeed, it has been demonstrated that a lack of motivation can limit the acceptance and effectiveness of even the most well-designed mHealth apps [49].

Another key aspect in weight loss through the use of smartphone apps can be attributed to the level of app engagement. A recent study has indeed evaluated the effectiveness of engaging with mobile apps in achieving weight loss in adults with overweight and obesity in a real-world context [50]. In particular, after 12 weeks, as many as 53% of participants classified as active users (active days during the observation period is 100%) experienced a clinically significant weight loss (≥5%) compared to 20% of inactive users (in-app activity <33%) [50]. The same authors also demonstrated a dose-response effect, whereby an increase in in-app activity was associated with greater weight loss [50].

In addition, several studies have demonstrated that gamification makes the app more enjoyable and motivating to use [51], provides emotional support that sustains user motivation [51], and most importantly, helps users promote health improvements and achieve their goals such as weight loss and physical activity [52].

Furthermore, the significance of the app’s design should not be underestimated. It should appeal to users, meet their needs, and feature a straightforward interface that integrates seamlessly with other devices. Ideally, it should be free or without hidden costs [49].

Another important aspect for weight management apps is that they provide food recommendations that consider not only the nutritional needs of the users but also the food culture [53]. This involves ensuring that the foods suggested are not only healthy and appropriate for weight management but also accessible and compatible with local food preferences. For example, an app aimed at the Italian market could emphasize the use of foods typical of the Mediterranean diet, while in Asia, it could focus on low-fat diet options rich in rice and vegetables. In fact, many smartphone apps have too limited choice of pre-entered meals in their databases, which often do not include ethnic foods [54]. Furthermore, they should integrate algorithms that consider variations in body composition, such as the ratio of fat to lean mass, which can differ significantly between ethnicities [8]. These aspects are essential to ensure that the apps are used and effective for a wide range of users even in different contexts. Therefore, all strategies aimed at encouraging greater use of the app, and consequently success in weight loss, should be considered in order to develop more effective mobile apps for obesity management.

Finally, several studies have highlighted the lack of scientific validation of many of the apps currently on the market for weight loss [42]. This could potentially increase stress and anxiety, as well as promote poor body image, and maladaptive eating and exercise behaviors, especially in younger individuals [49,55]. A study on nutrition apps found that some nutritional values varied by up to 50% from those reported by the German Food Database [56]. Therefore, apps without scientific validation as well as those designed by nonmedical institutions often lack expert knowledge and unreliable and nontransparent content, raising concerns about the accuracy of health information and its use in health management [56,57]. This represents a problem both for users and health professionals who could recommend their use.

The identified challenges in the literature also represent opportunities for developing a new app that can address these limitations and incorporate features that users have identified as critical for acceptability, engagement, and use. In addition, new smartphone apps could be integrated with AI systems, which may enhance personalization by analyzing large datasets, including genetics, biomarkers, diet, and lifestyle, to create tailored nutritional and physical activity plans based on the user’s unique characteristics [56,58]. Leveraging AI algorithms, these apps may automatically adjust goals based on user progress and provide suggestions to enhance body weight management [59].

Furthermore, apps that incorporate the combined use of AI and behavioral science techniques (eg, rewards, challenges, positive reinforcement, automatic monitoring, and personalized reminders) can enhance adherence and motivation, thereby reducing the risk of abandonment of weight management apps [60-62]. AI algorithms can detect inconsistencies and errors in user-recorded data, thereby enhancing the accuracy of analyses and rendering the recommendations more effective [63].

Ultimately, AI can equip health care providers with predictive models to identify individuals who may have difficulty losing weight and identify patients at risk of obesity-related complications. This enables proactive interventions through personalized strategies, fostering precision medicine [63]. Thus, these new mobile apps for weight management could be integrated into primary care systems by, for example, providing patients with self-monitoring tools and personalized feedback [64], facilitating communication between doctor and patient, and reducing the workload of health care providers.

In this context, several studies have demonstrated that smartphone apps can be effective tools for both weight management and the management of chronic conditions such as diabetes and cardiovascular diseases [65,66]. Additionally, these approaches could be integrated into public and community health programs, enhancing obesity prevention through digital education strategies and continuous monitoring of dietary habits and physical activity. Indeed, several studies have demonstrated that mobile apps can aid in preventing obesity by promoting adherence to healthier eating patterns and increasing physical activity among children and adolescents [67].

In fact, it is widely recognized that younger populations are heavy users of technology, especially smartphones and apps [18,68]. For this reason, weight management apps are designed also with them in mind, with easy-to-use interfaces and interactive features that fit well with their daily technological habits and preferences.

New weight management apps that incorporate AI, among other features, must overcome several challenges, including data privacy protection, the need for human oversight, and the risk of excessive dependence on technology. Therefore, integrating AI with the support of health care professionals may be the key to maximizing benefits without compromising the human aspect of interventions.

The results of this study highlight the limited effectiveness of smartphone apps in their current form and provide numerous insights for future research. In particular, future smartphone apps could provide long-term support through personalized assistance during challenging times, integrating digital coaching, web-based communities, notifications, and gamification to ensure dynamic and digital communication with users. Additionally, they could provide a highly personalized approach by adapting to the user’s dietary preferences, goals, and physical and psychological characteristics, using AI to deliver dynamic recommendations. They could integrate psychological and motivational support, such as mindfulness practices and emotional coaching, to address the emotional aspects of weight management. Furthermore, it is imperative that new apps for body weight management incorporate considerations for data management and privacy, the requirement for human supervision, and the risk of developing an overreliance on technology. Support from health care professionals may thus be crucial in optimizing the benefits of these apps while preserving the human element of the interventions. Finally, it is essential to ensure the scientific validation of these nutritional interventions by developing apps in collaboration with experts and conducting clinical testing to confirm their effectiveness and reliability.

This study has several strengths. First, each step in the process of conducting this systematic review (ie, study selection, data extraction, and methodological assessment) was performed independently by different reviewers. Furthermore, the results obtained were reported as MD, with the exception of weight loss, which was expressed as an SMD, in order to provide useful and clinically relevant insights into the potential effects of smartphone apps on weight loss and changes in body composition. To warrant the rigor of this study, it was decided to evaluate and include only peer-reviewed papers, thereby excluding those found in the gray literature. This approach ensured greater robustness of the results, although it may have excluded some potentially significant papers. Finally, only studies using smartphone apps were included, while those using additional devices that could have influenced the results were excluded.

Furthermore, the analysis focused on the effects of the apps on individuals with overweight or obesity, excluding individuals with comorbidities to minimize bias in the outcomes of the meta-analysis. However, it is important to note that this study has several limitations. The number of studies included in the meta-analysis is limited, and the majority of them were assessed as having unclear or high risk of bias, primarily due to the challenges in blinding participants to the intervention. Furthermore, we excluded studies with additional tools; although this approach improves the methodological rigor and specificity of findings, it may limit generalizability to real-world and practice settings, where smartphone apps are often used with wearable devices. Similarly, the choice to exclude individuals with comorbidities may also reduce the generalizability of the findings, as obesity is often associated with metabolic and cardiovascular diseases. Moreover, this meta-analysis did not take into account other variables that could affect responses to weight loss, such as genetic, psychological, economic, and social factors, as well as gender differences [69]. Finally, the results of this meta-analysis may not be directly applicable to all global populations. The included RCTs were conducted primarily in Europe, the United States, and Asia, contexts characterized by differences in dietary patterns, lifestyle habits, and health care systems. Furthermore, the effectiveness of apps may vary based on access to technology and levels of digital literacy.

For these reasons, the results obtained in this study should be interpreted with caution; however, they provide numerous insights for future clinical research aimed at developing more effective smartphone apps for body weight management, including in more complex clinical settings, also considering their applicability to different populations and health systems.

The current landscape of weight loss apps offers interesting insights but also presents significant challenges. Through this meta-analysis, the authors aim to provide a solid and critical knowledge base that can be used for developing new, more effective, and sustainable solutions over time. The lack of personalization and continuous support represents a significant shortcoming, indicating the need to develop more advanced apps tailored to individual needs. To enhance the effectiveness of these tools, future research should explore the integration of advanced features such as adaptive feedback driven by AI, which could analyze personal user data and provide dynamic recommendations for diet, physical activity, and motivational support. Additionally, gamification systems and digital notifications should be developed to increase engagement and persistence in app use. Beyond weight loss, future studies should assess additional parameters to determine app effectiveness, such as long-term adherence, changes in body composition, quality of life, relationship with food, and stress management. Furthermore, ensuring the scientific validation of these apps will be crucial, involving experts in nutrition and health technology to develop reliable, evidence-based tools. These improvements could not only enhance the impact of apps on weight management but also transform them into effective tools for health prevention and promotion, integrating into primary care pathways and public health strategies. In conclusion, by leveraging existing strengths and addressing the identified challenges, it is possible to design an app that more comprehensively meets user needs, providing personalized, scientifically validated support that can enhance long-term well-being.