While this scoping review found that rapidly increasing numbers of mobile apps are being created for the evolving COVID-19 outbreak, it is no surprise that evidence-based assessments of these apps do not currently exist. The rapid growth of the commercial market and need for COVID-19 apps has outpaced a standard of improvement for content and functionality of these mobile apps, resulting in an overaccumulation of COVID-19 apps with little evidence of effectiveness despite their availability for public use.
All 14 of the mobile apps appeared to have a singular focus, being either symptom management/assessment, news updates/statistic tracker, or an information sharing/training app; none comprehensively include a full range of features.
Current capabilities for sharing news, statistics, and live updates of COVID-19 through the included apps limit user customization, meaning that the app may share information that is irrelevant to the user’s geographic location or provide notification that are not relevant regardless of whether the user wants them. 71% have live news and statistics communicated through the app, nearly all of which impart global, rather than local or user-selected information. Several of the apps included some type of territorial filtering but remained broad; for example, Canada COVID-19 allowed for province filtering, but does not support specific data for certain cities and towns [22]. BC COVID-19 is specific with features and content for British Columbians but still remains broad in terms of city filtering, remaining as a general tool for the entire province [18]. While many of the apps (77%) had the ability to convey news and statistics, these apps did not frequently provide notifications to users for live updates. Research has shown that mobile app notification can be a useful tool when used at the appropriate frequency with a concise message [30]. It is important for future apps related to COVID-19 to consider personalized notification features to increase app interaction and dissemination of desired content.
Research has shown that apps can be successful in improving symptom management and tracking, specifically apps that allowed clinicians to provide input [31]. One app, HSE COVID-19, had the capability to document users’ in-home isolation and synchronize activity data with Apple’s Health Kit, Fitbit, and a pulse oximeter [17]. These results were then shared with clinicians who remotely monitor patients. Measuring a patient’s oxygen saturation, symptoms and activity allows clinicians to review a patient’s condition in self-isolation, while ultimately avoiding a hospital visit until absolutely necessary, therefore reducing the risk of spreading or contracting the virus by reducing the number of unnecessary visits. This feature may be beneficial for future app developers to consider. Of interest is the fact that only two of the six included apps had feedback surveys, despite research which indicates feedback is a valuable resource for quality improvement of mobile apps [32].
One of the apps, Coronavirus Australia, had the ability to document registered isolation [26]. The Government of Australia used this information to better understand the experience of those in isolation, to create a safeguard for isolated individuals, and to allow public health to conduct appropriate analysis and research. Personal information collected such as name, age, and geographical location were to allow for the government to obtain timely and accurate infection rates. No personal information was disclosed to parties overseas, and was instead stored on servers within Australia. While this app did not allow tracking of the participant in isolation to be shared with other individuals on the app, it allowed public health to stay updated on the patient’s current status through documentation of symptoms and to confirm that isolation was followed with people remaining in the same geographical location as that selected as their location of isolation. Analysts interpreting this data should do so cautiously, as the correct tracking would require for the subject to carry their mobile device with them at all times, and therefore this function may lead to inaccurate mapping.
Only one of the included apps, OSLER Covid Learning Centre, was strictly a resource app for supporting health care providers at William Osler Hospital in Brampton, Ontario [21]. The app hosted a learning centre focused on COVID-19 related topics such as techniques on non-invasive ventilation and co-ventilation of multiple patients. The app shared updated World Health Organization (WHO) situation reports, the Cochrane library collection for critical care management and infection control, and videos and podcasts related to COVID-19. The use of apps for education of hospital staff has the potential to share information in an innovative and engaging format that enables users to learn at their own speed [33] and is a good approach in an evolving situation. Given the current recommendation of self-isolation and social distancing, a mobile app would allow for health care providers in various fields (for example, home nurses, personal support workers, and mobile phlebotomists) to convene on health topics and changes to protocol without potentially perpetuating the spread of COVID-19.
Social media is a popular source for news consumption, with users claiming the value of sharing news in real-time and exposure to a wider range of topics [34, 35]. Popular mobile apps such as WhatsApp, TikTok, Facebook, Instagram, and Twitter have been sharing COVID-19 knowledge by creating Information Centres available through their apps. These Information Centres include updates from health organizations, tips on self-isolation, and resources for sharing with loved ones. The information shared on each of these platforms can be specific to the geographical location of the user and reflects information from the recognized Health Organization serving their area. Canadian Facebook users can review COVID-19 information shared by the Government of Canada, found on Canada.ca/coronavirus. WhatsApp has been commissioned by the WHO to begin facilitating information sharing on COVID-19; users can send messages to a number specific to their area to subscribe for COVID-19 updates, reports and statistics in real-time [36]. The strategies used by popular social media support the growing perspective that social media platforms are central to the way people experience news [34, 35].
Many pre-existing apps have also begun including COVID-19 related features, modifying their apps to fit the needs of their users during the pandemic. Headspace, a self-help app used for tackling problems such as anxiety, relationships and sleeplessness, began offering free guided meditations and exercises to all employers and employees to compassionately support teams through the stress and anxiety of COVID-19 [37].
While the development and use of apps could be critical in circulating information and minimizing the spread of disease by encouraging virtual engagement with clinical services, caution must be exercised when selecting which apps to download. Disguising themselves as sources of information on the COVID-19 outbreak, some apps are installing malware and demanding special access once downloaded [15]. To address these fraudulent apps, and to curb the sharing of inaccurate information, the Apple Store and Google Play Store seem to be obscuring searches for “coronavirus”, “covid-19” and other similar keywords [9, 10]. Such searches have yielded “no results found” messages or have returned options from reputable sources such as the Centre for Disease Control and Prevention (CDC) app. Without any clinical validation of apps, users should be wary of sources of any COVID-19 information. Apps affiliated with ministries of health, government organizations, and academic institutions should be prioritized when searching for reliable mobile sources of information for COVID-19.
Further research should also evaluate web-based apps and online dashboards for tracking COVID-19. A number of dashboards are currently available which allow consumers to view the evolving global trajectory of COVID-19 by including confirmed and recovered cases, and deaths [38–43]. Many of these dashboards transform these data into charts and interactive maps to help consumers visualize these statistics and the distribution of cases. With the functionality to group country data and customize filters, these dashboards provide users the opportunity to put these data into a more local and personal context [38–43]. A majority of the dashboards identified through Google Search Engine include those which cite their data from the Coronavirus COVID-19 Global Cases dashboard supported by the Center for Systems Science and Engineering at Johns Hopkins University [39], which itself cites the WHO, the CDC, the European Centre for Disease Prevention and Control, and other federal organizations worldwide. Dashboards have been proven to support the fight against infectious diseases by collating location to events in relationship to disease spread [44]. One hallmark example is the Cholera outbreak in Soho, London. In 1858, John Snow was able to complete hand-spatial analysis to determine that pumps supplying water were the source of Cholera outbreaks [45]. By analyzing the distribution of disease, dashboards have the potential to present the patterns of spread of COVID-19 in ways that may help minimize transmission. While dashboards, and websites were not targeted for inclusion within this paper, it is important to highlight all of the information sources available on mobile devices regarding the advancing COVID-19 crisis to illustrate the current environment of this field.
While this review did not include apps that were not available on an app store or in a language other than English, an app developed in Hangzhou City, Zhejiang, China [46] is worth highlighting. This health status app displays colours of either green, yellow, or red to indicate the current health status of the user [46]. The colours are allocated based on information reported by the individual, including their current health status, travel history, and previous exposure to highly affected areas such as Wuhan, China. Those with a red or yellow code are required to undergo quarantine, while anyone with assigned green can continue going to work. These unified health code systems are promoted throughout China and checked frequently on public transportation and by safety officers to ensure compliance with quarantine regulations to maintain the safety of other civilians [46]. While such a strategy could look promising in maintaining quarantine regulations, it is important to consider that these codes are sensitive to users’ input, and that these answers require no medical verification [46].
Lastly, the SARS-2003 outbreak had public health recommendations similar to those now being practiced for COVID-19, including wearing face masks and social distancing [47]. Two unique citations obtained through a grey literature search provided insight on apps modelled on the SARS-2003 outbreak (Table 6, Appendix B). One of the strategies implemented in Singapore during the SARS-2003 outbreak involved electronic tracking of individuals entering public hospitals. Staff and visitors were required to wear credit-card sized radio frequency identification (RFID) cards, and these records were saved by the hospital for 20 days [48]. If an individual became infected, health officials would examine who had been in contact with them based on RFID tracking records [48]. Today, mobile apps in South Korea, such as Corona 100 m, have used this strategy through Bluetooth technology [49]. Participants no longer require credit-card sized recognition cards but can anonymously be tracked through their mobile devices. The information is anonymous to users of the apps; that is, users are able to see where de-identified individuals with COVID-19 have been while government health officials and mobile developers are able to identify who each user is. Further, when a patient is recognized as infected with COVID-19, the Ministry of Health and Welfare in South Korea examines the apps to ensure that the patient stays in quarantine and follows rules of contact [50]. In addition, TraceCovid, one of the included apps in this study, implemented similar strategies utilizing Bluetooth frequencies to identify individuals who has crossed paths. All users who have downloaded the app have their paths and contact with other individuals traced and are able to share this information with Government authority if identified as COVID-19 positive. However, apps for the COVID-19 outbreak largely fail to consider the eHealth technologies that were implemented during the SARS-2003 outbreak. SARS-2003 apps wirelessly transmitted symptoms to health or infectious disease centres and allowed for remote monitoring of patients, communicating through Short Message Service (SMS) notifications on nearby SARS exposure for users who choose to share their phone location [47]. Given the 17-year gap, apps created for COVID-19 should look to past successes and failures of SARS-2003, creating apps that are multifaceted and able to combine all features of tracking, notification, statistics, updates, and symptom management.
mHealth apps remain an affordable, portable, and cost-effective strategy in delivering healthcare [51]. They also constitute an excellent platform for data collection. Some apps, such as Corona-care, allows healthcare workers to access and track patient symptoms remotely, while also collecting longitudinal data for research institutions. This app is accessible to patients who may have COVID-19 and have been seen by a clinician regarding their condition. Another application called Castor COVID-19 was developed for research purposes to collect population level data on COVID-19 by asking participants to report symptoms daily. As these apps can only be downloaded in a research context, they were not included for data extraction, but they were still described within the discussion to highlight the mobile app landscape during COVID-19. While helpful in supporting research efforts and in collecting data for healthcare purposes, it is important to consider privacy, security and additional expenses to the users of these apps, as well as the effects of these factors on adoption rates. In continuous location tracking apps, similar to TraceCovid and Corona100m, users agree to share information by consenting to a location tracking request [15]. Research has shown that 1/5th of cell phone users have turned off location tracking features on their phone, citing worries of privacy invasion [40, 50]. Overtime, users have been found to become comfortable sharing private information with larger corporations when a long-term business relationship between users and services providers has been established [52]. Given that many apps created for COVID-19 are created by government organizations, users may be more comfortable sharing private information. However, privacy issues can remain constrained regardless of the business relationship, with users choosing not to sharing location when in a sensitive area they prefer not to be disclosed, such as place of work or home address. Security concerns pose another adoption constraint, worrying users of potential attacks through malware infection such as viruses [53]. It has been suggested that all mobile apps available on application stores be created in accordance with standard guidelines to ensure that security procedures are put into place [54]. While the evolving pandemic of COVID-19 moved quickly, developers should still be held to a high standard when creating apps to ensure privacy and security measures are undertaken. The overall quality, security and privacy concerns, and additional costs related to the applications should also be taken into account. For example, apps with location tracking function in the background present a challenge in terms of data usage and internet access. When developing apps, data consumption and the reliance upon internet access should be considered in order to reduce user monetary burden and maximize the level of access to pertinent information. The application development process could address such concerns though involvement of end-users in order to incorporate the consumer perspective adequately.