Evaluation approaches used to assess the quality of weight loss apps: a systematic review (Preprint)

BACKGROUND Mobile applications (apps) have emerged as a convenient and accessible solution to support weight management. More than 28,000 apps related to weight loss are available across various platforms. However, there is a lack of understanding of the most effective approach to evaluate the quality of these apps. Existing studies have focused only on popular apps or specific user groups. OBJECTIVE To identify the approaches employed to assess the quality of weight loss apps and to determine which app features are considered important for enhancing their effectiveness. METHODS This systematic review was conducted in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. A comprehensive literature search was carried out across four databases: PubMed, Embase, Medline, and Web of Science. As inclusion criteria, studies were eligible if they specifically assessed weight-loss apps among healthy adult users (aged ≥18 years) and were published between January 1st 2019, and June 30th 2024. Studies were excluded if they focused on non-digital interventions, were not in English, involved clinical, military, or athletic populations, or were review articles, meta-analyses, conference abstracts, or reports. Search terms were derived from the concepts of quality, weight loss, and mobile applications. Data extraction focused on the approaches used to evaluate app quality. RESULTS Eleven studies met the inclusion criteria, evaluating a total of 46 distinct weight loss apps. Seven generic app evaluation approaches and two supporting frameworks were identified, with the most frequently used being the Mobile App Rating Scale (MARS) (n=39 apps), Evidence-based Strategies (EBS) Assessment (n=25 apps), and Six Sigma (n=25 apps). Only two approaches, MARS and the System Usability Score (SUS), have been validated to evaluate mobile apps. A total of 8 feature categories were identified as present in the apps across the studies. The most frequently observed were nutrition education (5), self-monitoring and tracking (5), exercise content & tools (5), behavioural support (4), social features (4), coaching and feedback (3), planning and goal setting (3), and technical functionality (2). Nine features were also recommended by the study authors to enhance app effectiveness through behaviour change. These features are progress reports (4), self-monitoring (2), reminders (3), gamification (2), and expert monitoring (1), comprehensive nutrition databases (3), food entry options (2), barcode scanning of calorie content (2), and affordability (2). Only the initial five are associated with behaviour change elements as per the BCT Taxonomy framework. CONCLUSIONS A range of approaches are currently employed to evaluate the quality of weight loss apps. This review identified seven commonly used evaluation approaches and two supporting frameworks, with MARS being the most frequently applied. Additionally, this study identifies a set of common and key features that should be prioritised in the development of weight loss apps for adults living with obesity to potentially enhance their overall effectiveness.