We will follow the 2020 JBI methodological guidance for scoping reviews [26,27], developed by the JBI Scoping Review Methodology Group. Accordingly, our scoping review is structured to follow seven stages, which consist of (1) title and research question, (2) identifying inclusion criteria, (3) identifying a search strategy, (4) evidence screening and selection, (5) data extraction, (6) data analysis, and (7) presentation of results. To guarantee adherence to reporting, we will also follow the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) checklist [28].

The title was developed based on the PCC (population, concept, and context) mnemonic [26], with a focus on the concept and context, namely wearables and smartphones for modifiable risk factors. The title does not explicitly mention the study population due to our broad inclusion criteria (eg, any adults). Motivated by gaps in previous work [20,29] and a lack of overview in the field, this scoping review aims to systematically describe (1) which modifiable risk factors are most often studied in wearable and smartphone-based metabolic health research, and (2) to what extent are measures of modifiable risk factors consistent across studies in terms of measurement methods.

Eligible studies will be peer-reviewed, written in English, published in 2019 and after (to reflect the most current literature and build on previous work [30]); and will include a full-text version. Given our research aims, eligible studies will: (1) measure lifestyle and physiological factors relevant to metabolic health, use (2) smartphones or wearables worn on the wrist, finger, arm, and chest, and (3) analyze the generated wearables or smartphone data for outcome assessment related to metabolic health. Qualitative studies will be excluded. Building on previous work [30], we will include wearable devices that cost <€500 (US $570) per device hardware, consistent with price cut-offs in previous work [30]. We will include membership-based devices. The study population will include adults aged 18 years and older. We will use institutional journal subscriptions and interlibrary loan services to access studies, and in cases where no full text is available, authors will be contacted with a waiting period of 7 days before study exclusion. Textbox 1 presents the study’s inclusion and exclusion criteria.

Building on the most recent reviews in the wearable, digital health, and metabolic health domains [20,29], we identified key search terms capturing wearables or smartphones and prominent modifiable risk factors associated with metabolic diseases [6]. A full list of key search terms can be found in Table S2 in Multimedia Appendix 1. We will search the following 5 major databases, that are Scopus, Web of Science, PubMed, Cochrane Centralized Register of Controlled Trials, and SportDiscus. Identified studies will be imported into the systematic review tool Rayyan developed by the Qatar Computing Research Institute [31].

Study evidence screening and selection will be performed based on the inclusion and exclusion criteria (see Textbox 1). We will first screen the title and abstract based on the predefined criteria, followed by screening the full text. This process will be conducted by 2 reviewers in parallel, and any discrepancies in study eligibility will be resolved by a third reviewer.

For each eligible study, we will extract information using a standardized data charting form. In total, 2 reviewers will manually chart the extracted information, and a third reviewer will resolve any discrepancies. The data charting form is presented in Table 1. The data charting form is adapted from prior work [25].

Following the data charting process outlined in Table 1, the gathered data will be synthesized and summarized in descriptive tables and figures. According to our first research question, we will categorize the modifiable risk factors per each study. Next, we will aggregate the overall prevalence of each modifiable risk factor and the proportion of studies that measure single or multiple modifiable risk factors. Based on our second question, we will describe the measures used to operationalize the most prevalent factors, and we will overview the overlap (vs discrepancies) in measurement, that is, how consistently risk factors are measured across studies. The findings will be summarized and communicated through tables and figures. Relevant findings will be communicated to diverse stakeholders, such as digital metabolic health researchers, healthcare professionals, digital health interest groups, metabolic disease patient organizations, and health insurance providers, through presentations and workshops.

No ethics approval is required for this study as it does not involve conducting trials or collecting primary data.

Prior research [8,29,32] has demonstrated that multiple modifiable risk factors jointly contribute to metabolic disease outcomes. In parallel, a growing number of studies are leveraging wearables and smartphones to track metabolic diseases in daily life (for example, [33,34]). However, it remains unclear whether existing studies examine multiple modifiable risk factors simultaneously or focus on a single factor in isolation. Our results are expected to reveal whether, and the degree to which, studies examine a range of modifiable risk factors in parallel, thereby offering a comprehensive approach, or if they predominantly isolate a single factor, which would suggest a gap in integrating broader determinants of metabolic diseases. Anticipated findings may indicate that certain risk factors, such as physical activity or diet, are more commonly tracked, while others, like stress or sleep, might be underrepresented, thereby revealing potential gaps in the current research focus. Our results could also highlight a strong consistency in the way risk factors are measured (eg, uniform use of validated metrics across studies) or substantial variation in measurement approaches. In the latter case, this inconsistency could point to challenges in comparing or integrating findings across different studies, thereby underscoring the need for standardization in risk factor measurement using wearables and smartphones.

The scoping review is expected to extend the existing literature in several ways. First, Lee et al [20], have begun to outline the characteristics of wearable devices to track cardiometabolic outcomes, but have not examined the degree to which studies examine different risk factors collectively, nor the consistency in measurement across studies. Our review aims to fill these gaps by systematically assessing the extent to which different risk factors are measured together (rather than in isolation), and by evaluating the consistency in measurement across studies. Previous reviews have primarily focused on the application of wearables or digital health technologies within the context of a single disease such as obesity or diabetes [23,35-37], or a single risk factor (eg, nutrition, physical activity, or stress) [14,38-41]. In contrast, our scoping review aims to provide a more comprehensive risk-factor perspective in a broader metabolic health framework.

To the best of our knowledge, this protocol is among the first systematic attempts to outline the current landscape of modifiable risk factors in digital metabolic health research, with a particular focus on the consistency of risk factor measurement across studies. We expect this research to contribute to existing knowledge by systematically scoping the variability in wearable and smartphone-based measurements, thus paving the way for improved measurement standards and comparability across studies. This scoping review will have several limitations, such as only including literature restricted in English, and quantitative empirical studies, which may omit more recent qualitative or industry developments. Our results will be specific to body-worn wearables costing up to €500 (US $570), a price point that may still be considered expensive. As such, some wearables included in our analysis may still be financially out of reach for many, reflecting a potential barrier to accessibility. We will also exclude other potentially relevant devices that are not worn on the body such as breath analyzers. Furthermore, there is often no consensus on the definition of metabolic health in the literature [42], which may lead to varied interpretations of metabolic risk factors. However, drawing from previous studies, we have compiled an extensive list of modifiable risk factors, encompassing both lifestyle behaviors and physiological indicators [43-50]. Future research may compare how different risk factors, and measurement approaches, predict metabolic disease outcomes.

Overall, this scoping review aims to synthesize existing research on how wearable and smartphone-based studies track modifiable risk factors in digital metabolic health. This work may potentially motivate the development of measurement reporting standards, thereby improving the consistency and applicability of measurements in digital metabolic health studies.