This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR mHealth and uHealth, is properly cited. The complete bibliographic information, a link to the original publication on https://mhealth.jmir.org/, as well as this copyright and license information must be included.
Given the magnitude and speed of SARS-CoV-2 transmission, achieving timely and effective manual contact tracing has been a challenging task. Early in the pandemic, contact tracing apps generated substantial enthusiasm due to their potential for automating tracing and reducing transmission rates while enabling targeted confinement strategies. However, although surveys demonstrate public interest in using such apps, their actual uptake remains limited. Their social acceptability is challenged by issues around privacy, fairness, and effectiveness, among other concerns.
This study aims to examine the extent to which design and implementation considerations for contact tracing apps are detailed in the available literature, focusing on aspects related to participatory and responsible eHealth innovation, and synthesize recommendations that support the development of successful COVID-19 contact tracing apps and related eHealth technologies.
Searches were performed on five databases, and articles were selected based on eligibility criteria. Papers pertaining to the design, implementation, or acceptability of contact tracing apps were included. Articles published since 2019, written in English or French, and for which the full articles were available were considered eligible for analysis. To assess the scope of the knowledge found in the current literature, we used three complementary frameworks: (1) the Holistic Framework to Improve the Uptake and Impact of eHealth Technologies, (2) the Montreal model, and (3) the Responsible Innovation in Health Assessment Tool.
A total of 63 articles qualified for the final analysis. Less than half of the selected articles cited the need for a participatory process (n=25, 40%), which nonetheless was the most frequently referenced item of the Framework to Improve the Uptake and Impact of eHealth Technologies. Regarding the Montreal model, stakeholder consultation was the most frequently described level of engagement in the development of contact tracing apps (n=24, 38%), while collaboration and partnership were cited the least (n=2, 3%). As for the Responsible Innovation in Health framework, all the articles (n=63, 100%) addressed population health, whereas only 2% (n=1) covered environmental considerations.
Most studies lacked fundamental aspects of eHealth development and implementation. Our results demonstrate that stakeholders of COVID-19 contact tracing apps lack important information to be able to critically appraise this eHealth innovation. This may have contributed to the modest uptake of contact tracing apps worldwide. We make evidence-informed recommendations regarding data management, communication, stakeholder engagement, user experience, and implementation strategies for the successful and responsible development of contact tracing apps.
As the global battle against the COVID-19 pandemic continues, the SARS-CoV-2 virus has infected over 160 million people and claimed over 3.3 million lives by May 2021 [
Contact tracing is a fundamental containment strategy in response to emerging outbreaks. Public health agencies aim to rapidly identify individuals who may have been exposed to a person who is infected to recommend the most appropriate course of action (eg, self-isolation, symptom recording, and testing). The incubation period of this virus can last up to 14 days, during which infected individuals can unsuspectingly contaminate others [
A variety of eHealth solutions, which leverage information and communication technologies for the betterment of health and health care services [
Contact tracing apps can differ according to eight fundamental characteristics. First, their installation can be voluntary or compulsory. Second, the extent of informed consent varies between apps. Third, some apps use a decentralized data management strategy, while others enable linkages with governmental agencies. Fourth, their ability to detect contact between users can rely on technologies such as GPS, Bluetooth, or Quick Response codes. Fifth, the specific algorithms deployed in the back end of these apps will determine their output (eg, the calculation of a risk of infection or the tracing of potential contacts). Sixth, they require varying levels of human oversight, if at all present. Seventh, the degree of interaction with users regarding recommended actions (eg, testing and isolation recommendations) and the extent of interaction with public health agencies can differ. Last, safety protocols for data privacy may also vary [
Given their fundamental mechanism, the effectiveness of contact tracing apps depends in part on the level of uptake and ongoing use [
As contact tracing apps encompass many underlying principles and disciplines, multiple aspects can facilitate or hinder their adoption. One of the main caveats in their implementation is concerns over data security and management. Societies are rightfully preoccupied with the challenges in reconciling civil liberties with public health imperatives in a pandemic context [
Key factors have been suggested as drivers for widespread success of contact tracing apps: integration with local health policy, adaptable workflows in an ever-evolving context, rapid notification systems, the ability to evaluate the effectiveness of the app transparently, and clear communications addressing privacy concerns [
New eHealth initiatives have emerged at an accelerating pace in the last decade; some have seen widespread adoption, whereas others have failed to provide sustained value. These failings can be attributed to design and implementation efforts that were initiated without a good understanding of the interdependencies between technology, societal values, and user experience in a health care setting. Many conceptual frameworks based on implementation science have been developed to evaluate and orient eHealth delivery. These frameworks highlight key factors that predict successful and sustainable eHealth technologies. The urgency of the ongoing public health crisis stimulated the rapid development of contact tracing apps and other eHealth innovations, and this generated a substantial number of related publications. Their coverage of essential design and implementation characteristics for eHealth innovation remains underinvestigated.
The primary objective of this review is to map and analyze the literature on the design and implementation of COVID-19 contact tracing apps. This was achieved through three distinct questions: (1) to what extent does the available literature discuss features that promote the use of contact tracing apps by interested parties? (2) how have patients and citizens been engaged in the design and implementation of these apps? and (3) does the development of these apps correspond to principles of responsible research and innovation?
Through these questions, we studied how the development of contact tracing apps has taken into account considerations related to the uptake and impact of the innovation, the engagement of end users, and the responsible development of eHealth technologies. We ultimately identified the components required for successful and responsible eHealth development as described in the available literature on contact tracing apps and those that are lacking.
This study is reported according to PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines [
Three complementary frameworks were selected to address each of the research questions. To answer the first research question on factors that promote uptake of contact tracing apps, we selected the Framework to Improve the Uptake and Impact of eHealth Technologies proposed by van Gemert-Pijnen et al [
For the second research question, which addresses the engagement of users in the development of contact tracing apps, we selected the Montreal model established by Pomey et al [
Finally, for the third research question relating to responsible health innovation, we selected the Responsible Innovation in Health Assessment Tool by Pacifico Silva et al [
A systematic literature search was performed by one author (GG) in five databases (PubMed, Scopus, IEEE Xplore, AMC Digital Library, and Europe PMC) using terms specifically related to the research question.
((“contact tracing” [mesh] OR “epidemiological monitoring” [mesh]) AND (“mobile applications” [mesh] OR “algorithms” [mesh] OR “computer security” [mesh] OR “big data” [mesh] OR “computer simulation” [mesh] OR “geographic mapping” [mesh] OR “geographic information systems” [mesh] OR “microcomputers” [mesh] OR “software” [mesh]))
OR (tracing [tw] AND (app [tw] OR apps [tw] OR proximity [tw]))
OR ((contact [tw] OR exposure [tw]) AND notification* [tw] AND (app [tw] OR apps [tw] OR application* [tw]))
OR ((digital* [tw] OR mobile [tw] OR ehealth [tw] OR “eHealth” [tw] OR mhealth [tw] OR “m-health” [tw] OR app [tw] OR apps [tw] OR application* [tw] OR “geolocation*” [tw] OR “location service*” [tw] OR “location system*” [tw] OR “location information” [tw] OR gps [tw] OR big data [tw] OR ((geographic [tw] OR geographical [tw]) AND tracking [tw])) AND (“contact tracing” [tw] OR “contact tracking” [tw] OR “digital epidemiology” [tw]))
The study eligibility criteria (
Each of the 1959 articles was screened by one of two authors (ER and JP), initially excluding those with titles and abstracts unrelated to the topic of study. They then read the remaining articles to determine their relevance to the research question with respect to the inclusion criteria. The senior authors (EO and MPP) screened articles with uncertain relevance for final inclusion or exclusion. Discrepant decisions were resolved through team discussion and consensus. This process resulted in the inclusion of 63 articles, following the search iterations previously described.
Related to design considerations of contact tracing apps for COVID-19 (eg, privacy considerations, citizen inclusion and participatory approach, or incentivization)
Related to user experience or implementation approaches of contact tracing apps for COVID-19 (eg, surveys or focus groups, strategies to mitigate social vulnerabilities, or recommendations for governments communication with the public)
Published before 2019
Published in a language other than French or English
Pertaining exclusively to aspects of computer science or technical developments of contact tracing apps
Pertaining exclusively to epidemiological feasibility or efficacy of contact tracing apps
Authors could not obtain access to the full article
Data from the included studies was charted on an extraction grid (
Considers the problems of a particular situation and offers a corresponding solution (eg, proposal to incentivize a contact tracing app [
Reflects the author’s or journal’s opinion about a subject (eg, contact tracing app effectiveness and data security [
Concentrates on survey or focus group methods for data collection (eg, user acceptability of a contact tracing app [
Studies a particular
Examines what has already been discovered about a subject (eg, systematic evaluation of content and features of a contact tracing app [
Discusses ideas from the literature in a support of arguments about a specific subject (eg, discussion of an intervention to introduce contact tracing technology [
Articles that outlined theoretical recommendations or criticism regarding the overarching concept of contact tracing apps were categorized as being based on theory (eg, ethical considerations of instantaneous contact tracing). In contrast, articles that factually described specific use cases or empirical studies were categorized as being based on practice (eg, description of technical features of a given app and their potential impact on implementation or surveys on user acceptability of a contact tracing app).
Furthermore, the extraction grid included the components of the Gemert-Pijnen et al [
Figures were produced using codes in accordance with the three theoretical frameworks used and other relevant information such as the date of publication, the type of article, and the theory or practice classification. Visualization of these elements shed light on changes in the available information and on gaps in the underreported domains of design and implementation of contact tracing apps. Critical appraisals of the included articles were beyond the scope of this study.
This scoping review generated a total of 1959 records. Following the removal of duplicates and the application of the selection criteria, 63 articles were included in the analysis (
Flowchart of the study selection process.
The included studies (N=63) were published between April 16 and November 6, 2020, in 40 different journals or preprint databases. The majority (n=48, 76%) of included studies were published articles or e-prints, and 24% (n=15) were preprints. The number of published articles peaked in August (n=16, 25%;
Number of monthly publications/preprints, according to theory- and practice-based categorization.
Types of article by month of publication.
Overall, all of the six principles of the Framework to Improve the Uptake and Impact of eHealth Technologies [
Proportion of articles integrating holistic eHealth development principles.
None of the selected articles referred to documentation given to citizens or patients on COVID-19 contact tracing apps, leaving the Montreal model “information” component [
Proportion of articles integrating each level along the patient or citizen engagement continuum.
Among the five domains of the Responsible Innovation in Health Assessment Tool [
Proportion of articles integrating Responsible Innovation in Health domains.
The principal considerations and recommendations regarding contact tracing apps in the COVID-19 pandemic context were drawn from the key findings of analyzed articles and are summarized in
Themes of considerations and recommendations regarding app design and implementation.
Themes | Considerations and recommendations found in the literature (N=162), n (%) |
Data management (eg, the use of blockchain [ |
58 (35.8) |
User experience (eg, rewards to app users [ |
48 (29.6) |
Communication (eg, clarifying false beliefs about the app [ |
23 (14.2) |
Research and implementation methodology (eg, NPTa framework to guide development and evaluation of complex DPTb interventions [ |
18 (11.1) |
Engagement of stakeholders (eg, low-cost wristband in low-socioeconomic areas [ |
15 (9.3) |
aNPT: Normalization Process Theory.
bDPT: digital proximity tracing.
This study demonstrates that the volume of articles on design and implementation considerations for COVID-19 contact tracing apps grew rapidly in the first months of the pandemic and peaked in August 2020. As the evidence base increased, the proportion of opinion pieces decreased in comparison to surveys and review articles. This is not surprising given the novelty of this technology and the urgency around technological developments triggered by the pandemic. Accordingly, earlier articles mostly discussed principles related to the development and implementation of contact tracing apps mostly in theoretical terms, while the proportion of empirical articles increased starting in July. Despite the growing number of publications, their scope remained limited with respect to design and implementation considerations. Our findings demonstrate that critical design and implementation considerations were lacking in the early academic literature on contact tracing apps.
In fact, less than half of the selected articles cited the need for a participatory process, which nonetheless constituted the most frequently referenced item of the Framework to Improve the Uptake and Impact of eHealth Technologies. Only 40% (25/63) of the articles presented evidence of public or patient engagement in the contact tracing app development process. This stands in stark contrast with the ideal of “co-creation from ideation to operationalization” in eHealth technologies, as described in this framework [
We further analyzed the level of citizen and patient engagement in the development of contact tracing apps by using the Montreal model [
Moreover, when we assessed these studies according to the Responsible Innovation in Health framework [
The incompleteness of the academic literature on the design and implementation characteristics of COVID-19 contact tracing apps stems in part from the novelty of the topic and the need for timely innovations to fight the pandemic. As such, we did not expect the literature to address all the domains of the selected frameworks. However, few academic publications highlighted the need to assess the impact of this intervention or consider organizational challenges related to its deployment, among other key elements to successful and responsible eHealth innovation. Likewise, the limited references to higher forms of upstream end user engagement stand in contrast to this technology’s inherent reliance on widespread adoption. It may be that app developers or researchers did not perceive some of these factors as bearing sufficient relevance for incorporation at the development stage or for later publication. Nonetheless, more than one year since the start of the pandemic, many of these elements are still lacking and were highlighted as essential to the success of COVID-19 contact tracing apps [
Data protection and the looming risk of mass surveillance have understandably dominated the debate on the ethical, legal, and social implications of contact tracing apps. Users must therefore be intelligibly informed of the steps taken to protect their privacy. One consideration is that the risk of the proposed contact tracing app should be compared with that of the frequently used apps that most people have on their cellphones. For example, the New York Times “Privacy Project” recently revealed the tracking of millions of unsuspecting Americans through location-sharing apps (eg, apps used to access directions, weather information, or local fidelity programs) [
Clear and transparent messaging developed with and for citizens can be promoted through simple and familiar means, such as concise information labels inspired by those of the food industry. This approach was used by members of the health care machine learning community to promote the transparent and responsible use of clinical decision support tools [
Key stakeholders of contact tracing apps include potential users, technology developers, policy makers, and funding agencies. As previously noted, the prevailing form of engagement has been through public consultations, which is recommended in the analyzed articles. We would therefore argue that app developers and the various institutions that implement such technologies would benefit from greater upstream collaboration and partnership with individuals from diverse backgrounds, including patients. Engaging key stakeholders early in the process will also help identify the right problem and constraints, eventually narrowing the range of most suitable technological tools. Indeed, even if a contact tracing app perfectly identified high-risk contacts, it would not achieve the desired outcome of reduced viral transmission if it required prohibitively expensive hardware or if it relied on massive viral testing in a strained health care system unable to provide a sufficient number of diagnostic tests.
As previously described, the potential risks of contact tracing apps may disproportionately affect minority and marginalized groups [
A simple and intuitive user interface may not only enhance the user experience at an individual level, it may also improve uptake at the population level. Apps that focus on interactive design features can help users better understand how to use the app correctly and more effectively, whereas apps that are focused on information may reduce the assimilation of this information by users if they are not visually appealing [
Simple and useful apps will likely incentivize uptake and use. This can be supported by developing apps that require readily available personal technology, personalized updates on pertinent and accurate information aligned with local guidelines, minimal disruption to daily functioning (eg, minimizing battery use [
Financial incentives should also be considered. A study that focused on tracking the use of Germany’s official contact tracing app, Corona-Warn-App, found that app uptake is more prevalent among older populations, individuals with pre-existing conditions, and those with high levels of education and income. Additionally, the study reported that information interventions, in the form of short videos addressing privacy, effectiveness, and app functionality issues, were useful in increasing users’ knowledge about the app but were not effective in driving uptake. On the other hand, interventions that provide a monetary incentive (as low as €1 [US $1.22], €3 [US $3.67], or €5 [US $6.11]) upon installation were found to be useful in increasing uptake [
Empirical evidence is urgently needed to determine whether the benefits of a given contact tracing app significantly outweigh its risks. Although contact tracing technologies were used in prior Ebola and influenza outbreaks, there is limited empirical evidence on their effectiveness. As for their impact in reducing viral transmission during the COVID-19 pandemic, it is mainly based on mathematical simulation models that used varied assumptions and methodologies [
In addition, academic reports on contact tracing apps and related eHealth innovations must adhere to established reporting guidelines such as CONSORT-EHEALTH (Consolidated Standards of Reporting Trials of Electronic and Mobile Health Applications and Online Telehealth) for eHealth and mHealth interventions [
The findings of this scoping review must be interpreted considering certain limitations. First, this study includes articles published up to November 6, 2020. As the volume of publications continues to grow, a systematic review focused on narrower questions related to contact tracing apps may become relevant. We attempted to maximize the reach of our search by including multiple databases and by developing a rigorous study selection process. Relevant articles only available in the gray literature or exclusively in governmental databases may nevertheless have been missed, although official apps developed by governments, such as the National Health Service contact tracing app, were included in several of the articles analyzed. Furthermore, given the purposefully broad question posed by this study, the reproducibility of data extraction and charting presented some challenges. We addressed them by relying on well-established and complementary frameworks that were particularly appropriate to the research question. Moreover, multiple authors reviewed the process and provided supportive statements when a particular component or domain was considered to be present in a given article.
The emerging academic literature on contact tracing apps reveals significant knowledge gaps regarding their design and implementation. Key stakeholders are thus limited in their ability to critically appraise this eHealth innovation. Most of the included studies lacked fundamental aspects of the successful eHealth development and implementation framework. Similarly, few articles described the impact of contact tracing apps on the environmental, organizational, and economic domains, which are essential to evaluate responsible innovation in health. Among the studies that described a form of public participation, nearly all of them relied on consultation as opposed to collaboration or partnership. These overlooked components of eHealth development and implementation may have contributed to the modest uptake of contact tracing tools worldwide. They suggest a critical gap between theory and practice, whereby numerous academic sources promote a holistic and participatory approach to eHealth innovation, but few products incorporate them. Partnerships between app developers, researchers, policy makers, and users early in the development process will narrow this gap. Transparent, systematic, and comprehensive reporting of COVID-19 contact tracing app outcomes will further enable their critical appraisal. The lessons learned about the social acceptability of contact tracing apps as they were deployed at an unprecedented pace and scale must serve in future iterations of this innovation and in the development of other eHealth technologies aimed at sustainably supporting public health. They must attest to the importance of stakeholder engagement, problem identification, minimal system disruptions, longitudinal outcome measurement, and use incentivization.
PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) Checklist.
Additional database search strategies.
Database of included articles and analysis.
Consolidated Standards of Reporting Trials of Electronic and Mobile Health Applications and Online Telehealth
Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews
This research was supported by the Observatoire international sur les impacts sociétaux de l'IA et du numérique and the Fonds de recherche du Québec. MPP also receives funding from the Fonds de recherche du Québec pour le soutien à la bourse de chercheur senior. We would like to thank Eleonora Bogdanova for her invaluable help in the literature search and Olivier Demers-Payette, Nathalie de Marcellis-Warin, and Christophe Mondin for their technical and methodological insight.
None declared.