Digital technologies have become ubiquitous in all facets of life and work [1]. Within health care, digital solutions offer a promising opportunity to make processes and operational methods more efficient and effective [2,3]. This advancement in health care and public health will empower individuals to engage in self-management and actively manage their overall well-being and health [4]. Similarly, digital technologies can facilitate the cost-efficient implementation of health promotion, prevention, monitoring, and surveillance measures, making them easy to access [5-11]. These technologies also foster novel care possibilities, enabling sustainable and equitable access to health services across entire populations [12-16]. Therefore, adopting digital health solutions presents a valuable opportunity to strengthen national health care systems. Eventually, these solutions will transform traditional clinical care and public health structures, fundamentally reshaping health systems on a global scale.

However, to our knowledge, no study has yet attempted to comprehensively map out the heterogeneous landscape of digital public health (DiPH) and digital public health interventions (DiPHIs). Despite existing reviews focusing on selected intervention categories [17], specific target groups such as adolescents [18,19] or patients who are chronically ill with specific medical conditions [20], a comprehensive mapping of interventions in digital health is still missing. In addition, in all existing reviews, the studies analyzed digital health interventions but not DiPHIs, thereby limiting these interventions to their medical and clinical usability. Therefore, this scoping review aims to fill the research gap and comprehensively map out the heterogeneous landscape of DiPHIs. The development, implementation, integration, and evaluation of evidence- and needs-based DiPH interventions require a clear and shared understanding of the specific characteristics of digital health technologies for public health purposes [21]. Conversely, the lack of a standardized lexicon creates communication challenges between the subdisciplines of DiPH [22,23], while concurrently hindering comparability between national health systems.

For this review, DiPH is defined as using information and communication technologies (ICTs) such as digital technologies, services, or tools to achieve traditional public health goals [24] with a population health impact [25]. According to Winslow [26], public health aims at “preventing disease, prolonging life, and promoting physical and mental health and efficiency through organized community efforts...” This definition is still used nowadays by the World Health Organization (WHO) to define essential public health functions. This places public health as the leading discipline in health governance, financing, health information systems, communication, coverage, care, education, and health regulations to protect and promote the health of populations [27].

Hence, we define DiPHI as interventions that address “at least one essential public health function through digital means.” [26]. To enhance acceptance among the population, DiPHI should consistently follow a needs- and evidence-based design with a participatory and human-centric approach [21,25,28,29]. The impetus behind their development should not stem solely from technological capabilities but from DiPHIs potential to mitigate disparities and improve access to health services among different groups of a population [22,25,27].

Unlike the domain of digital health, which involves the application of health ICT for personalized medicine and telemedicine [30] and its subfield mobile health, which primarily focuses on patient monitoring and the use of personal digital assistants in the clinical setting [31], DiPHIs encompass a broader spectrum of services and interventions. These include those in health promotion, such as wearables to promote physical activity; in prevention, such as web-based vaccination registries; in health care, such as telemedicine or electronic health records (EHRs); and in surveillance systems, such as dashboards or tracing systems for infectious disease monitoring on a national level. The goal of DiPHIs is to improve the health and well-being of groups considered vulnerable rather than individuals [25].

By combining the foundational goals of public health with the capabilities of ICT applications, DiPHIs have the potential to substantially strengthen population health, promote well-being, and advance public health objectives. This impact can be realized at the individual, community, and national levels [32].

This scoping review primarily aims to outline a comprehensive range of proposed to implemented DiPHI, including different levels of prevention, health care, and public health research initiatives. Following the aforementioned definition of DiPHI, the second objective is to map the landscape of the existing DiPHI, shedding light on their self-reported digital health functions (based on the evidence standards framework [ESF] for digital health technologies from the National Institute for Health and Care Excellence [NICE] [33]) and the addressed essential public health functions as defined by the WHO [27]. Our scoping review will guide future research to address underexplored DiPHI types by identifying research gaps (eg, underrepresented intervention types). The multinational and interdisciplinary comparative analysis of interventions will generate a data set on various interventions, their characteristics, functions, and use cases, which can support researchers or policy makers to make informed decisions for future DiPHIs planning and upscaling.

It is vital to acknowledge that this scoping review, while intended to illustrate the distinctive characteristics of DiPHI, does not seek to provide exhaustive analyses on cost-effectiveness, efficacy, or barriers and facilitators of implementation. The inherent heterogeneity of DiPHI renders it impractical to address these multifaceted aspects within a single literature review [13,14].

The standards for eligibility were constructed in the participants, intervention, and study design format. An overview of the inclusion and exclusion criteria is presented in Textbox 1. Accompanying these criteria, the general requirements of full-text accessibility and the publication language were also considered. The review expanded the publication language limitation to English, German, and Chinese based on the language skills of the reviewer team. Given the scope of our study, which focuses on mapping the characteristics of DiPHI rather than comparing the effectiveness or costs, the review process refrains from specifically incorporating reported outcome variables.

The initial search followed a comprehensive search strategy conducted on February 19, 2021, in CENTRAL, PubMed, and Web of Science. We ran an additional search following the same search strategy on December 1, 2021, for 2 additional databases: ACM Full-Text Collection and IEEE Xplore. The search results for all 5 databases were updated on October 5, 2022. We applied a consistent search string to all databases with adaption according to index and syntax specifications differences. Given the substantial volume of publications aligning with our inclusion criteria, supplementary manual searches were not undertaken. However, we conducted reference list searches for reviews and meta-analyses describing DiPHIs. No additional references were identified through this approach, as all references were already identified through our systematic search.

The search strategy for all databases was structured around 3 themes connected through “AND.” Terms within the themes are combined through “OR” (refer to Figure 1 for the general overview and Multimedia Appendix 1 for the complete search strategy for all databases, including results per term of the initial search). We decided against searching for concrete interventions such as EHRs, specific health or medical apps, or other interventions to reduce the risk of confirmation bias.

We additionally used Medical Subject Headings wherever appropriate during database searches. All other terms were limited to title, abstract, or keywords. No search limits were applied.

We applied a 3-stage screening strategy (ie, titles, abstracts, and full texts) to identify suitable publications (Figure 2). To reduce the list of reasons for full-text exclusion, the first 3 exclusion criteria mentioned in Textbox 1 were summarized under “wrong setting.” Three authors (LM, MF, and KA) were involved in the screening process for eligibility, and each stage had 2 authors partake in the screening independently. After each screening stage, another author (CCP) resolved any disagreements on study selection with independent decisions. The screening was conducted through the free web and mobile app Rayyan (Rayyan Systems Inc) [35]. References published in another language than English, Chinese, or German were excluded during full-text screening. Chinese references were screened only by CCP. Cohen κ for full-text screening was 0.77 (identifying a substantial agreement rate according to the study by Landis and Koch [36]). Full texts were included if the article was published open access, if it was published in a subscription journal licensed by our institution, or if the corresponding author provided a copy. An overview of all excluded full texts, along with the reasons for their exclusion, is included in Multimedia Appendix 2 [37-39].

Two authors (LM and KA) independently extracted data in an Excel 2019 spreadsheet (Microsoft Corp). The extraction sheet was piloted with the first 10 included publications. The extraction for all data except for the technical functions and nontechnical intervention characteristics (which were extracted as free texts) followed a predefined coding table, which was extended if necessary (Table 1). The extraction for all data was based on qualitative description directly provided by the included texts. Discrepancies in data extraction were resolved through discussion between the 2 authors. Missing data for the included publications or data where only a vaguely formulated description was given in the text were requested from authors via email. Where multiple interventions were described within 1 publication, data for each intervention was extracted in a new row. Target groups were defined and extracted based on the digital health application directory by Lantzsch et al [37] due to their precise differentiation between healthy users and patient users. Additional target groups (eg, policy makers, health professionals, or researchers) were added in an iterative process. The digital health functions and the intervention types followed the 2019 edition of the NICE ESF for digital health technologies [33]. The most updated ESF from 2022 reduced its number of individual technical functions and was changed toward a more clinical setting. However, the 2019 framework still included 10 different functions, for example, intervention types covering health promotion, health care, communication, education, and system services. Therefore, we deemed the 2019 version more suitable for our mapping approach. In addition, these functions were supported by the work of Lantzsch et al [37]. The public health functions were based on the 11 WHO functions [27]. Building on our previous research, ESF and ESPHF are merged to map DiPH for this study [29]. Extracting authors selected the option “other” where multiple target groups, functions, or prevention or health care levels appealed.

Target groups were defined and extracted based on the digital health application directory by Lantzsch et al [37] due to their precise differentiation between healthy users and patient users. Extracting authors selected the option “other” where multiple functions or target groups appealed. This decision displays the only deviation from the registered study protocol for this review [34] where we did not define different categories within target groups (the study by Lantzsch et al [37] was published after our study protocol, which underlines the rapidly evolving nature of DiPH). The definitions for prevention and health care (ie, last column in Table 2) were based on the WHO definitions (Table 2).

In addition to the previously mentioned topics, we also extracted information on the study design, publication year and type, intervention name, technical features, and nontechnical characteristics. We did not extract data on outcome measures or type of data collection in the included publications because this review aims to conduct an intervention mapping focusing on their characteristics and not on their effectiveness or costs.

For the geographic analysis of interventions, we counted every intervention multiple times if it was implemented internationally. For instance, Adler et al [45] described an intervention that was implemented in 7 countries. Weighting these interventions per country by the number of countries they were implemented in (here 1/7) would have weakened internationally planned or implemented interventions compared to interventions only targeting 1 country. Therefore, we decided to give full weight to each intervention and count the intervention as 1 for every targeted country. While this approach resulted in an artificially higher number of included interventions, it also ensured that international interventions were not discriminated against based on the geographic analysis.

For the qualitative analysis, we grouped the identified references by intervention types based on our study protocol. We then summarized the extracted data for each intervention type, as illustrated in Table 2. Additional intervention types were added in an inductive procedure where needed. In addition, we developed an inductive category system based on 5 randomly selected publications per intervention type to describe the technical functions and nontechnical characteristics of different intervention types. The qualitative content analysis through iterative coding was then conducted for all references of 1 intervention type in MAXQDA 2022.7 (VERBI GmbH) by LM. The procedure followed the inductive content analysis procedure described by Vears and Gillam [46].

In total, we identified 15,701 different publications published until October 5, 2022 (24,767 before deduplication), through our systematic search in 5 scientific databases. Of these, 2987 (19.02%) were assessed for abstract screening and 1609 (10.25%) for full-text screening. The process excluded 1414 records, primarily due to unavailable full-text content and a lack of specificity in describing interventions. Of the screened full texts, the remaining 11.5% (185/1609) were considered eligible for this review. Some of these publications described multiple interventions, whereas, in other cases, the same intervention was presented by >1 publication. Of the 185 publications, 96 (51.9%) were case reports that focused on a detailed intervention design or implementation description, while the aim of the remaining 89 (48.1%) publications lay on the intervention evaluation, most often evaluated through cross-sectional studies (n=37, 20%), trials (n=19, 10.3%), or longitudinal studies (n=18, 9.7%). The characteristics and extracted information of all included 185 references are displayed in Multimedia Appendix 3 [38,45,47-229] instead of the main manuscript. As sometimes multiple references are reported on the same intervention, this scoping review mapped the characteristics of 179 different DiPHI for the included 185 publications.

Figure 3 displays the distribution of the included interventions by country. For this analysis, all interventions were weighted equally regardless of the number of countries they were implemented in. As several interventions were implemented in >1 country, this analysis artificially increased the number of observed interventions to 199. Although most interventions (36/199, 18.1%) were from the United States, continent -wise, Europe took the lead with 38.2% (76/199) reported interventions. Ranked third was Asia with 17.6% (35/199), followed by Africa (26/199, 13.1%), Oceania and Australia (9/199, 4.5%), and South America (6/199, 3%). From a gross domestic product point of view (ie, grouped in high, middle, or low-income countries as classified by the World Bank [230]), most interventions were planned for or implemented in high-income countries (129/199, 64.8%) and were set-up in middle-income countries (58/199, 29.1%). Of the 10 interventions in low-income countries, 3 (30%) targeted Malawi [73-76,82,218], and 2 (20%) each were applied in Afghanistan and Ethiopia [73-76,91,175,222]. Furthermore, of the 10 interventions, 1 (10%) each was planned or implemented in Mali, Sierra Leone, and Uganda [56,111,190].

Under the classification of intervention by NICE ESF, most interventions (49/179, 27.4%) were centered on telemedical services, including telestroke, telecare, and telemonitoring. According to Timpel et al [231], these interventions are characterized using ICT to cover a geographic distance in health care delivery from a health professional to a patient or group of patients. The second most frequently counted interventions were health apps and medical apps (28/179, 15.6%), followed by EHR (23/179, 12.8%). The complete overview of intervention types is displayed in Figure 4, together with their implementation status at the reported time from the included references. As 4 publications reported on the same intervention but at different time phases, all publications were included for this display (eg, Austria’s national EHR ELGA was only planned when Ströher and Honekamp [199] reported about it, but was already nationally implemented when Herbek et al [113] and Schaller et al [177] described the platform). This results in a total of 181 displayed interventions. Interventions in >1 country were categorized as internationally implemented.

Our analysis showed that DiPHIs tend to focus on an average of 3 distinct groups within the population. These include health professionals, policy makers, public insurance executives, researchers, disaster management professionals, patients, and healthy individuals. Unsurprisingly, health professionals were the most often targeted group, with 140 (78%) of 179 interventions. They were followed by patients who were chronically ill but stable (101/179, 56.4%). Of all interventions, 51.9% (93/179) were aimed at healthy people with or without risk factors for certain health conditions.

Parallel to the observation of the target population, each intervention affected, on average, 2 different settings, that is, level of prevention, health care, or research as defined previously. Most interventions were health care applications (ie, 100/179, 55.9% for primary, 89/179, 49.7% for secondary, and 55/179, 30.7% for tertiary health care). Out of the 179 interventions, while 16 (8.9%) interventions included a research component [57,84,93,104,128,139,154,163,180,220,221,225], only 3 (1.7%) focused exclusively on research: the Lone Star Stroke Consortium Telestroke Registry, the SAI Databank for electronic health research and evaluation, and the German COVID-19 data donation app (ie, Corona Datenspende) [57,93,128]. For prevention, of the 179 interventions, 13 (7%) exclusively targeted primary prevention (ie, through vaccination registries, behavior change programs, or health education in healthy individuals) [55,62,96,98,101,106,126,140,143,188, 208,223,226]. Of the 179 interventions, 8 (4%) focused only on secondary prevention, as digital screening and surveillance programs [48,61,153,161,171,194,201,219], and 3 (2%) improved the health of patients who were chronically ill on a tertiary prevention level [103,147,202]. In contrast, of all included 179 interventions, 10 (6%) were established for primary health care purposes only [49,67,79,109,124,172,178, 186,210,212,228,229], 17 (9%) were designed for secondary health care through specialists or emergency use cases [52,54,60,65,78,83,88,95,134,135,173-175,179,189,193,216], and 5 (3%) aimed at tertiary health care in hospital settings [70,109,132,157,200]. Figure 5 gives an overview of the relative size of each target group and the addressed level of prevention, health care, or research per intervention setting.

The analysis of the distribution of EPHFs and digital health technology functions matches the aforementioned focus on health care in DiPHIs. The heat map, displayed in Figure 6, highlights that most interventions were found in the cross-section of health care and diagnostic (64/179, 35.8%), treatment (63/179, 35.2%), communication (61/179, 34.1%), and information (57/179, 31.8%) or as system services in health care (42/179, 23.5%). Relatively few interventions, in contrast, were used for calculation (11/179, 6.1%), preventive behavior change (40/179, 22.3%), or public health research (48/179, 26.8%). A subgroup analysis for all intervention types with at least 20 included references is included in Multimedia Appendix 4. This analysis covers telemedicine, EHR, and health or medical apps, offering an overview of the diversity of topics addressed even within the same intervention category.

Given the sample size, we only provide details on the technical functions and nontechnical characteristics of telemedical interventions and EHRs. We chose not to include health apps or medical apps in this analysis due to the considerable heterogeneity in the wide range of purposes and functions of these apps, as discussed in a previous article [232].

Our scoping review has identified a substantial number of publications, underscoring the substantial interest and activity in DiPH. The wealth of publications illustrates the broad spectrum of interventions being explored and implemented worldwide. This diversity reflects the evolving landscape of health care delivery and disease prevention in the digital age.

The distribution of DiPHIs across countries and continents showcases the worldwide interest in leveraging technology to enhance health care but also points out significant differences in implementation adaptation between high-income and low-income countries. A reason for this imbalance negatively impacting the global digital divide might lie in the inequality regarding the digital advance among developed countries compared to developing countries [235,236]. The prevalence of interventions in high-income countries may partly be attributed to more significant resources and infrastructure [237]. This divide is highlighted by a lack of funding, inadequate computer availability, and limited internet skills that limit the expansion and applicability of ICT and digital (public) health [238]. Still, interventions in low-income countries signal the potential for digital technologies to bridge health care gaps in resource-constrained settings and aging societies [237]. Indeed, the Global Digital Health Monitor covers our findings by displaying overall scores for countries regarding their digital health maturity. Developing countries (especially in Africa) rank lower than their developed counterparts in Europe or North America [239]. More rigorous public health research and practice efforts are needed to empower developing countries to apply digital tools for health-related purposes. The WHO’s Digital Health Atlas can serve as such an effort where project leads can voluntarily register their digital health intervention free of charge [240]. Maps such as these will support public health policy, research, and practice in better resource allocation (eg, the map can display already existing subnational initiatives that could be scaled up to the national or international level without creating double-infrastructures), benchmarking between interventions and further developing intervention types. However, the explicit distinction between medically focused digital health interventions and population-centric DiPHIs will be needed for future versions of the atlas.

Our analysis shows that DiPHIs often target multiple population groups and settings. Health professionals are a primary target, reflecting technology integration into health care practice and education. The emphasis on patients who are chronically ill but stable underscores the potential of digital interventions to improve care for this population. Moreover, interventions in diverse prevention, health care, and research settings illustrate their versatility across the health care continuum. This observation was supported by the heat- map representation of EPHFs and ESFs, highlighting a significant concentration of digital interventions in health care–related functions such as diagnostics, treatment, communication, and information. This reflects the predominant role of digital interventions in enhancing clinical care and health care management. However, there is room for further exploration of interventions addressing prevention and public health research functions.

Health apps, telemedical services, and EHRs emerged as the dominant intervention types in our review. Health apps are well known for their heterogeneity [232], which provides the potential to address all areas of public health, including health promotion, clinical care, and rehabilitation [241]. Telemedical services stood out as a substantial focus of our included references, aligning with their capacity to extend universal health coverage reach across geographic boundaries and increase health care delivery speed while lowering treatment costs and response times to control disease outbreaks [242]. EHRs, in contrast, can be essential tools for public health surveillance in providing data on the population’s health status and the health care system’s performance [39,114,243]. These findings highlight the multifaceted nature of DiPHIs and their adaptability to various health care needs and contexts in public health.

Overall, the analysis underscores the heterogeneity within DiPHIs, even within the same intervention category, as we have displayed through a qualitative analysis of included publications for telemedical interventions and EHRs. This diversity necessitates precise definitions and standardized terminologies to facilitate effective communication, collaboration, and evaluation. Clear definitions can also aid in benchmarking and comparing interventions across different health care systems. The detailed analysis of technological functions and nontechnical characteristics of both intervention types revealed the complexity and variation within each intervention type. Identifying core characteristics and added features can guide intervention development and enhance our understanding of their capabilities. In 2019, WHO published their recommendations on digital interventions for the clinical setting, which provides a definition for each intervention (eg, client-to-provider telemedicine), synonyms for these tools, and recommendations regarding their use cases [244]. While this overview serves as a good starting point, it requires extensions for nonclinical settings and an overview of core functions and technical features to allow benchmarking. The heterogeneity in DiPHI forces us to rethink how national health system assessments should be conducted to allow for comparable results. Instead of asking, Does your country have an EHR system?” it is critical to ask for the availability of an intervention with specific characteristics: “Does your country have an intervention with the following characteristics, targeting these population groups, and being applied for at least one of the following settings?”

As the field continues to evolve, it is essential to prioritize rigorous research and evaluation to assess these interventions’ effectiveness, safety, and impact on population health and health care accessibility. This review serves as a valuable resource for policy makers, researchers, and health care practitioners seeking to navigate the dynamic and diverse field of DiPHIs.

Originally conceived as a systematic review, our research protocol was registered with PROSPERO to conduct a comprehensive synthesis of the existing literature on DiPHIs. The systematic approach is typically characterized by a rigorous process involving reviewing evidence to answer narrowly defined research questions. However, as our study progressed, we recognized that our research question and the evolving DiPH landscape were better suited for a scoping review methodology. While systematic reviews focus on specific research questions and outcomes, the strengths of scoping reviews for effectively mapping and characterizing interdisciplinary and comprehensive topics (such as the diverse spectrum of DiPHIs in our case, including their characteristics, settings, and target populations) [245] were more applicable for our study aim. This change allowed us to adapt to the dynamic nature of the field. It provided a broader perspective essential for understanding the multifaceted nature of digital interventions. This transition underscores the importance of selecting the most suitable methodology to achieve comprehensive insights into the subject matter. The selection of our information sources demonstrated the interdisciplinarity of DiPH by combining literature databases from public health and computer science. This rigorous strategy mitigates the risk of missing essential publications. The broader inclusion of publications from 3 languages (ie, English, German, and Chinese) is another strength of our study, given the rising number of publications on digital health and DiPH from German-speaking and Chinese-speaking countries. Throughout the process, a minimum of 2 researchers independently screened all references, adhering to the 4-eye principle. Any disagreements in decisions regarding inclusion or exclusion were resolved by another author (CCP) to ensure the reliability of the results. This study followed the PRISMA-ScR [245] to provide high-quality, transparent research results. The filled checklist is provided in Multimedia Appendix 5.

Nevertheless, while this scoping review exhibits numerous strengths, it also presents certain limitations. We did not create publication alerts for the selected databases, limiting our results to references published before October 6, 2022. However, we conducted manual reference list checks for reviews and meta-analyses to identify additional publications that were potentially missed in our search. As no further relevant findings were added from the manual search, we are confident that our scoping review draws a holistic picture of the landscape of DiPHIs globally. Furthermore, we are convinced that the included references are sufficient for an initial intervention mapping of DiPHIs. Nonetheless, it is worth noting that certain intervention types might be underrepresented, such as early warning software, electronic medication plans, electronic prescribing, or emerging intervention types that were not yet reported during our search (eg, artificial intelligence tools). Here, we invite other researchers to build upon our first results and continue the mapping process. Following the PRISMA-ScR [245], assessing the quality of the included publications was not applicable, which could lead to questions regarding the reporting accuracy of the included publications. Nevertheless, as we focused solely on the functions and use cases of the interventions and not on their outcomes, this limitation remains minor.

Furthermore, the challenge of intervention specificity emerged as a notable consideration. The varying levels of detail in the descriptions of digital interventions in the literature posed difficulties in discerning their precise nature. This lack of specificity hindered the ability to categorize and analyze interventions effectively, as it often left critical questions unanswered. For instance, understanding the core technological functions, nontechnical attributes, determined use cases, and defined user groups of a given digital intervention is essential for its accurate classification and evaluation. Although in our case, data extraction was conducted by 2 authors (LM and KA) independently and authors were contacted in case of uncertainty, the absence of such information generally can impede efforts to compare interventions, assess their impact, and make informed decisions regarding their implementation or scalability. This is why we advocate for clearer reporting standards for digital health and DiPH interventions. Finally, we must point out that although the volume of included publications was substantial, a distinct portion could not be included in our review due to factors such as unavailable full-text content and a lack of specificity in intervention descriptions. This highlights challenges in accessing relevant literature and emphasizes the need for more detailed reporting in published studies, particularly regarding the specifics of digital interventions. In an era where timely access to information is critical, the inaccessibility of full-text articles poses an obstacle to researchers and stakeholders seeking to explore and build upon existing knowledge in DiPHIs.

Due to the heterogeneity of DiPHIs, it was not feasible for this scoping review to report detailed information for every intervention type. Therefore, we encourage other researchers to uptake our methodology, conduct reviews on specific intervention types, and refine our initial intervention mapping results. More critically, a shared and multidisciplinary understanding of key terminology in digital health and DiPH is needed. Consequently, consensus finding methods (eg, Delphi approaches) need to be applied to internationally create a shared understanding of DiPHI.

In addition, assessments of the sustainability of DiPHIs are crucial to map the implementation status. As Iwelunmor et al [246] and Kaboré et al [247] already highlighted, the sustainability of DiPHIs is an essential component of the whole life cycle of implemented interventions. Therefore, it is crucial to consider whether an intervention has been sustainably implemented or stopped after the initial pilot funding phase has ended. Further research should be conducted to analyze, through a longitudinal design or pre-post comparison, how sustainably DiPHIs are implemented and whether differences between intervention types exist.

This scoping review has shown that DiPHIs are distinctly diverse in their use cases between individual intervention groups and within such groups. Therefore, using precise terminology when planning, implementing, or evaluating DiPHIs is crucial. For instance, instead of phrasing an intervention as an EHR, one should ask for specific technological functions and nontechnical characteristics, determine use cases (from a digital health and public health perspective), and define user groups. This approach facilitates cross-disciplinary collaboration in intervention development and research and fosters international benchmarks and comparability across global DiPH systems. These insights offer valuable guidance for policy makers, researchers, and health care practitioners interested in using the potential of digital interventions to improve population health and health care accessibility.