IntroductionBackground

Digital technologies have become ubiquitous in all facets of life and work []. Within health care, digital solutions offer a promising opportunity to make processes and operational methods more efficient and effective [,]. 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 []. Similarly, digital technologies can facilitate the cost-efficient implementation of health promotion, prevention, monitoring, and surveillance measures, making them easy to access [-]. These technologies also foster novel care possibilities, enabling sustainable and equitable access to health services across entire populations [-]. 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 [], specific target groups such as adolescents [,] or patients who are chronically ill with specific medical conditions [], 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 []. Conversely, the lack of a standardized lexicon creates communication challenges between the subdisciplines of DiPH [,], while concurrently hindering comparability between national health systems.

Definition of DiPH

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 [] with a population health impact []. According to Winslow [], 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 [].

Hence, we define DiPHI as interventions that address “at least one essential public health function through digital means.” []. To enhance acceptance among the population, DiPHI should consistently follow a needs- and evidence-based design with a participatory and human-centric approach [,,,]. 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 [,,].

Unlike the domain of digital health, which involves the application of health ICT for personalized medicine and telemedicine [] and its subfield mobile health, which primarily focuses on patient monitoring and the use of personal digital assistants in the clinical setting [], 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 [].

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 [].

Study Aims and Objective

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] []) and the addressed essential public health functions as defined by the WHO []. 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 [,].

Methods

Following the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guidelines, we registered our review protocol with the PROSPERO database (CRD42021265562) on August 2, 2021, and details and rationale of the protocol were published in 2022 [].

Eligibility Criteria

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 . 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.

Textbox 1. Eligibility criteria for the scoping review.

Inclusion criteria

PopulationThe study focuses on the community level or above (ie, regional or national) of the general population.InterventionThe paper describes a concrete digital public health intervention (DiPHI) that addresses at least 1 essential public health function through digital means as defined by the World Health Organization (WHO).The DiPHI matches our definition of digital public health.The DiPHI addresses at least 1 essential public health function as defined by the WHO.The DiPHI is paid or reimbursed by the government or health insurance.The DiPHI uses the internet or Bluetooth to provide its core function or service.The DiPHI is the central research object of the paper.Study designAll original peer-reviewed studies, reports, books, book chapters, or peer-reviewed conference papers that have a description of a DiPHI as their primary intervention component.Full text

Exclusion criteria

PopulationThe study population consists of veterans, armed forces, prisoners, inmates, refugees, or asylum seekers.InterventionThe intervention needs to be privately bought by the user without reimbursement by the government or health insurance.The intervention focuses on background management processes.Study designStudy protocols, editorials, commentaries, conference proceedings, or reviews (ie, narrative reviews, scoping reviews, systematic reviews, or meta-analyses).Full textFull text was not available.Literature Search

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 for the general overview and 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.

‎Figure 1. The fundamental concept of the search strategy. Screening Process

We applied a 3-stage screening strategy (ie, titles, abstracts, and full texts) to identify suitable publications (). To reduce the list of reasons for full-text exclusion, the first 3 exclusion criteria mentioned in 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) []. 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 []). 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 [-].

‎Figure 2. Flowchart of the search and screening process. Data Extraction and Qualitative Content Analysis

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 (). 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 [] 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 []. 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 []. The public health functions were based on the 11 WHO functions []. Building on our previous research, ESF and ESPHF are merged to map DiPH for this study []. 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 [] 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 [] where we did not define different categories within target groups (the study by Lantzsch et al [] 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 ) were based on the WHO definitions ().

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.

Table 1. Coding table for data extraction.Study typeCountryType of interventionImplementation statusPrimary target groupDigital health functionEssential public health functionLevel of prevention, health care, and researchCase reportAfghanistanHealth or medical appProposed interventionHealth insurancesActive monitoringGovernancePrimary preventionCase-control studyAlbaniaDisease surveillance systemPlanned interventionHealth professionalsSimple monitoringFinancingSecondary preventionLongitudinal studyAlgeriaEarly warning softwarePilot studyResearchersCalculateHuman resourcesTertiary preventionCross-sectional studyAndorraElectronic health recordLocally implementedPolicy makersCommunicateHealth information systemPrimary health careEconomical studyAngolaElectronic consultationRegionally implementedHealthy without known risk factorsDiagnoseResearchSecondary health careMixed methods studyAntigua and BarbudaElectronic medication planNationally implementedHealthy with known risk factorsInformSocial participation and health communicationTertiary health careTrial or Experimental studyArgentinaElectronic prescribingInternationally implementedAcute ill not life threateningPreventative behavior changeHealth promotionResearchQualitative studyArmeniaElectronic referralWebsiteChronically ill stableSelf-manageHealth protection——aAustriaElectronic disease registry—Highly vulnerable or unstable health statusSystem serviceDisease prevention——AzerbaijanElectronic vaccination registry—Disaster management professionalsTreatmentPreparedness for public health emergencies——BahamasHealth information system———health care——BahrainImplants——————BangladeshInformation website——————BarbadosPatient portal——————BelarusPatient to provider communication portal——————BelgiumProvider to provider communication portal——————BelizeTelemedicine——————BeninWearable—————

aNot applicable.

Table 2. Applied definitions for prevention and health care.Level of prevention or health careDefinitionSourcePrimary preventionAvoiding new diseases, disabilities, or injuriesWHOa [] and Baumann and Ylinen []Secondary preventionEarly detection of diseases, disabilities, or injuriesWHO [,] and Baumann and Ylinen []Tertiary preventionReducing complication risk of existing diseases, disabilities, or injuriesBaumann and Ylinen []Primary health careCare for acute mild illness, injuries, or medical problems at a primary care provider with a whole-of-society approach (eg, through a general practitioner)WHO [,], Hopayian [], and Bodenheimer and Grumbach []Secondary health careSpecialist or emergency medical care through specialized physiciansHopayian [] and Bodenheimer and Grumbach []Tertiary health careHighly specialized medical health care over an extended period in stationary settings, usually in referral hospitalsHopayian [] and Bodenheimer and Grumbach []

aWHO: World Health Organization.

For the geographic analysis of interventions, we counted every intervention multiple times if it was implemented internationally. For instance, Adler et al [] 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 . 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 [].

ResultsOverview

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 [,,-] 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.

Distribution of Interventions Globally

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 []), 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 [-,,], and 2 (20%) each were applied in Afghanistan and Ethiopia [-,,,]. Furthermore, of the 10 interventions, 1 (10%) each was planned or implemented in Mali, Sierra Leone, and Uganda [,,].

‎Figure 3. World map for the number of included interventions by country, continent, and high-income to low-income countries. Intervention Types and Implementation Status at the Time of Publication

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 [], 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 , 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 [] reported about it, but was already nationally implemented when Herbek et al [] and Schaller et al [] described the platform). This results in a total of 181 displayed interventions. Interventions in >1 country were categorized as internationally implemented.

‎Figure 4. Overview of intervention types and implementation status. Some interventions were described during earlier phases and then again in later periods. Therefore, they are listed once for every reported intervention life cycle in this figure. Target Population and Intervention Setting

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 [,,,,,,,,,,,], 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) [,,]. 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) [,,,,,,,,,, ,,]. Of the 179 interventions, 8 (4%) focused only on secondary prevention, as digital screening and surveillance programs [,,,,,,,], and 3 (2%) improved the health of patients who were chronically ill on a tertiary prevention level [,,]. In contrast, of all included 179 interventions, 10 (6%) were established for primary health care purposes only [,,,,,,, ,,,,], 17 (9%) were designed for secondary health care through specialists or emergency use cases [,,,,,,,,,,-,,,,], and 5 (3%) aimed at tertiary health care in hospital settings [,,,,]. gives an overview of the relative size of each target group and the addressed level of prevention, health care, or research per intervention setting.

‎Figure 5. Addressed target groups and level of prevention, health care, or research in relative distribution per intervention type. The Intersection of Digital Health Technologies and Essential Public Health Functions in DiPHIs

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 , 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 . 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.

‎Figure 6. Heat map of the essential public health functions and digital health technologies to map digital public health interventions. Technological Functions and Nontechnical Intervention Characteristics

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 [].

Telemedicine

According to a review by Sood et al [], which included 104 publications, telemedicine “...uses communications networks for delivery of health care services and medical education from one geographical location to another, primarily to address challenges like uneven distribution and shortage of infrastructural and human resources.” The category system includes 5 mayor categories and subcategories for which we assessed all 49 telemedical interventions included in our review [,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,,,,, ,,,,,,,,,-,]. The results are displayed in . Each intervention may exhibit a combination of these features, showcasing how telemedicine is applied in different settings. The most dominant characteristic for included interventions was that the software program was supported by hardware to facilitate clinical care. Besides computers and webcams, this included high-quality cameras, sensors (eg, for blood pressure or blood sugar), ultrasounds, or digital stethoscopes. Nearly half of the interventions (25/49, 51%) described how they used the internet or satellite connections to process and store data. The share of interventions that included a provider-to-provider communication option was slightly lower (17/49, 34%) than those that allowed patient-to-provider communication (23/49, 47%). Nevertheless, 20 (41%) of 49 interventions were characterized by their focus on individual patients, sometimes even through personalized intervention interfaces. However, telemedical interventions may not only focus on clinical care. As stated in the definition, they are also applied to the education context (ie, for patients and health care workers), health promotion, and improving self-efficacy in patients who were chronically ill.

In summary, telemedical interventions exhibit core characteristics that include the use of supporting hardware during consultations, data storage through databases, internet connectivity, bidirectional communication between users, educational modules, user interfaces, and the ability to plan and adjust clinical care plans when needed.

Table 3. Technical functions and nontechnical characteristics of the telemedical interventions (n=49).CategoryInterventions, n (%)ReferencesTechnical infrastructure
Software supported through hardware tools41 (84)[,,,,,,,,,,,,,,,,,,,,, ,,,,,,,,,,,,, ,,-,]a-c
Data acquisition, transmission, and storage in a database33 (67)[,,,,,,,,,,,,,,,,,,,,,,, ,, ,,,,,,]a
Internet network and connectivity25 (51)[,,,,,,,,,,,,,,,,,,,,,, ,,]
Web-based user interfaces22 (45)[,,,,,,,,,,,,,,,,,,,,,]
Data security procedures18 (37)[,,,,,,,,,,,,,,,,,,,]a
Centralized server application16 (33)[,,,,,,,,,,,,,,,]b
Multiaccess, multiprofile web application16 (33)[,,,,,,,,,,,,,,,,,]a
Clinician modules and units14 (29)[,,,,,,,,,,,,,,,]a
Patient modules and units6 (12)[,,,,,]bPatient-provider interaction
Bidirectional communication (patient-provider)23 (47)[,,,,,,,,,,,,,,,,,,,, ,,]c
Clinical care treatment planning and adjustments22 (45)[,,,,,,,,,,,,,,,,,,,,,,]a-c
Bidirectional communication (provider-provider)17 (35)[,,,,,,,,,,,,,,,,]
Remote monitoring of vital signs and clinical status through monitoring kits and sensors17 (35)[,,,,,,,,,,,,,,,,]
Compliance and adherence monitoring11 (22)[,,,,,,,,,,]
Custom and predefined questionnaires6 (12)[,,,,,]User-centered features
Patient-centered interactions possible20 (41)[,,,,,,,,,,,,,,,,,,]b,c
Self-monitoring and tracking10 (20)[,,,,,,,,,]
Personalization (eg, avatars, screen names, communication preferences, reminders, and alerts)8 (16)[,,,,,,,]
Alarm detection and notification6 (12)[,,,,,]Use of patient data
Data interpretation and analysis12 (24)[,,,,,,,,,,,]
Data visualization tool13 (27)[,,,,,,,,,,,-,]a
Data- and guideline-driven decision support11 (22)[,,,,,,,,,,,,]aSupport and engagement
Education and counseling modules22 (45)[,,,,,,,,,,,,,,,,,,, ,,]c
Counseling and motivation module10 (20)[,,,,,,,,,]
Patient empowerment and self-efficacy5 (10)[,,,,]

a3 studies [,,] described the same intervention.

bIntervention 1 by Hartvigsen et al [].

cIntervention 2 by Hartvigsen et al [].

EHRs

Following the broadly accepted definition by the International Organization for Standardization (2019), an EHR is “a data repository regarding the health and health care of a subject of care where all information is stored on electronic media” []. According to this terminology, we developed a category system with 4 major topics and subcategories, which was applied to the 26 publications that described 23 different EHR systems [,,,,,-,,,,,,,,,,, ,,,,,,,,]. The results are displayed in . Nearly all interventions (18/23, 78%) allowed health care professionals access to the EHR after the patient consented or an active relationship (eg, due to treatment of chronic diseases) existed between the 2 parties. Although not as often described, most EHR systems (14/23, 61%) stated that patients were able to access their data. Data sharing was possible in 61% (14/23) of the interventions, and another 61% (14/23) described data protective procedures for accessing and sharing data. Although fewer times, interoperability with other health care systems was described (11/23, 48%). We observed differences regarding the stored and accessible data. Unsurprisingly, nearly all interventions (21/23, 91%) reported on storing health and clinical data. However, only 30% (7/23) of the interventions included personal data on the record holder or their immunization records. Only 17% (4/23) of the interventions allowed clinicians to upload their notes to the system.

Table 4. Technical functions and nontechnical characteristics of the electronic health records (EHRs; n=23).CategoryInterventions, n (%)ReferencesTechnical infrastructure
Integration and interoperability with external health care systems11 (48)[,,,,,,,,,,,,,,]a,b
Mobile platform7 (30)[,,,,,,,,]b,c
Centralized server6 (26)[,,,,,,,,,]a,b
Communication protocols6 (26)[,,,,,,,]b,c
Cloud-based system4 (17)[,,,,]c
Data synchronization3 (13)[,,,]Data transfer and privacy
Data sharing possible14 (61)[,,,,,,,,,,,,,,,,]a-c
Data security protocols embedded14 (61)[,,,,,,,,,,,,,,,,,]a-c
Opt-out approach4 (17)[,,,,,,,]a-c
Opt-in approach4 (17)[,,,,]cStored data
Medical history (eg, diagnoses, medical procedures, laboratory results, and treatments)21 (91)[,,,,,,,,,,,,,,,,,,,,, ,,,]a-c
Medication plan14 (61)[,,,,,,,,,,,,,,,]b,c
Patient demographic data7 (30)[,,,,,,,,]b
Immunization records7 (30)[,,,,,,,]c
Financial and administrative data related to health care services5 (22)[,,,,]c
Clinician notes4 (17)[,,]EHR system features and modules
Health professional access to patient data18 (78)[,,,,,,,,,,,,,,,,,,,,,]a-c
Patient access to data14 (61)[,,-,,,,,,,,,,,,,]a-c
User interfaces11 (48)[,,,,,,,,,,,,,]b,c
Clinical case–finding features8 (35)[,,,,,,,,,,,]a,b
Patient management modules3 (13)[,,]
Data used for real-time surveillance and monitoring3 (13)[,,]

a3 studies [,,] described the same intervention.

b3 studies [,,] described the same intervention.

c2 studies [,] described the same intervention.

DiscussionPrincipal Findings

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 [,]. The prevalence of interventions in high-income countries may partly be attributed to more significant resources and infrastructure []. 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 []. Still, interventions in low-income countries signal the potential for digital technologies to bridge health care gaps in resource-constrained settings and aging societies []. 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 []. 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 []. 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 [], which provides the potential to address all areas of public health, including health promotion, clinical care, and rehabilitation []. 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 []. 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 [,,]. 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 []. 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.

Strengths and Limitations

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) [] 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 r