Ovid MEDLINE and Embase were searched for relevant publications published up until July 6, 2020. The search was initially informed by a preliminary unstructured search of QxMD [19] and MEDLINE (via PubMed). The search strategy (Multimedia Appendix 1) was developed using the COVID-19 keywords from the Canadian Agency for Drugs and Technologies in Health search strings [12].

Inclusion criteria for selection of publications were as follows: publications that discussed use of RMTs to deliver health care to patients with COVID-19; publications that referred to specific RMTs as opposed to general recommendations (this criterion was used to ensure barriers could be linked to a specific type of technology); and publications that were available in English, French, or Spanish.

Publications were excluded if the record only described RMTs used in contact tracing or epidemiological surveillance. This criterion was used because the focus of this review was on technologies that provide service or treatment to patients with COVID-19, rather than surveillance of the virus.

An inclusive search strategy for different types of publications was used and included editorials, reviews, and letters to the editor. This inclusive strategy was adopted considering the novelty of COVID-19 (and hence dearth of published research-based studies). Further, the inclusive search strategy reflected the potential benefits of presenting a breadth of information at this early stage of the pandemic.

The primary author (EH) reviewed the title and abstract of publications then extracted key information. Key information extracted included the year, journal, design, aim of the publication, nature of the RMT, whether the RMT was synchronous or asynchronous, information gathered via RMT, equity considerations, benefits of RMT use, and barriers to RMT use.

Publications’ results were synthesized using a “content analysis” approach [20]. A single reviewer (EH) developed a coding framework and mapped the RMTs used and variables monitored. Barriers and benefits of using RMT were coded inductively, and individual codes were grouped into overarching themes. A second reviewer (DOP) independently reviewed the developed coding framework against a random sample (7/48, 15%) of the publications. Further, the second reviewer coded this sample of publications against the finalized coding framework. All discrepancies (7% discrepancy rate) in the coding framework or coding were settled by consensus-based discussion.

To ensure explicit consideration of health equity, relevant considerations were mapped deductively to the PROGRESS-Plus acronym [17]. The PROGRESS-Plus acronym refers to consideration of the following: place of residence, race or ethnicity or culture or language, occupation, gender or sex, religion, education, socioeconomic status, social capital, and “plus”—referring to personal characteristics associated with discrimination, features of relationships, and time-dependent relationships [17].

Of 767 records identified through MEDLINE and Embase searches, and 4 records identified through QxMD, 491 remained after duplicates were removed (Figure 1). After title and abstract screening, 406 records were excluded from the review. Upon full-text review, 37 of 85 records were excluded because they did not include health care for patients with COVID-19 (n=13), did not describe an RMT (n=12), were epidemiological studies and/or only described contact tracing (n=6), were not in English, Spanish, or French (n=5), or were technical reports (n=1). In total, 48 publications were included in the qualitative synthesis.

The most common types of publications identified were descriptive studies or proposals describing the implementation and development of RMTs (n=18) and reviews of multiple RMTs (n=13). Letters to the editor (n=9), editorials (n=5), one retrospective cohort study, one protocol, and one case study were also included (Table 1).

These publications identified 35 distinct types of RMTs (32 asynchronous and 3 synchronous; Table 2). Overall, 35 of the 48 studies identified synchronous technologies, including video (n=25), telephone (n=15), chat (n=8), and mobile or cellular device video (n=3) consultations. In total, 33 studies reported asynchronous technologies, including mobile or cellular apps for patients to manually enter their symptoms or education (n=17), home diagnostic tests (n=6), remote access to electronic health records by provider (n=8), digital pulse oximeters (n=5), biometric clothing (n=5), and remote access to electronic health records by patient (n=5). Data were collected through manual input by patients, caregivers, and health care practitioners (n=45; Multimedia Appendix 2), and automatically through the use of biosensor technology (n=22; Multimedia Appendix 3). The most commonly reported manually inputted outcome measures were “symptoms” (n=18), dyspnea or shortness of breath (n=7), pulse oximetry (n=7), mental health (n=7), temperature (n=6), and presence or absence of fever (n=6). The most frequently reported biosensor measures were temperature (n=5) and cardiac activity (as measured by an electrocardiogram; n=5).

Almost all publications (46/48, 96%) provided some description of the reported benefits of RMTs. In total, 34 distinct codes were developed inductively from these documents (Multimedia Appendix 4). These were grouped into 10 themes (Table 3): “reduces infection risk” (n=33 articles), “reduces burden of care” (eg, limits hospital beds used; n=28), “supports vulnerable populations” (n=14), “reduces costs” (n=12), “improves patient experience” (n=11), “promotes knowledge development” (n=8), “facilitates navigation through health care system” (n=8), “improves health outcomes” (n=6), “supports public health initiatives” (n=5), and “technology-specific benefits” (n=4).

The most common theme, “reduces infection risk,” included 3 codes: “reduces risk of transmission generally” (n=18), “reduces exposure of health care practitioners” (n=11), and “reduces cross-contamination or clustering” (n=10). A total of 28 codes were mapped to the theme “reduces burden of care,” notably that RMTs “reduce the burden on hospitals” (eg, limit hospital beds used; n=17), “provide continuous accessible data or monitoring” (n=12), and “reduce burden of time on health care workers” (eg, using asynchronous technology or reducing burden overall; n=6). In addition, 5 codes were mapped to the theme “RMTs support vulnerable populations,” including “reduces need for transfer of vulnerable patients” (n=6) and “supports patient mental health” (eg, relieves stress, supports anxious patients; n=5). A total of 3 codes were mapped to the theme “reduces cost,” notably “reduces health care system or public health agency costs” (n=12). A total of 3 codes were mapped to the theme “improves patient experience,” including “improves patient initiative, engagement, autonomy, or self-management” (n=5). In addition, two codes were mapped to “facilitates navigation through the health care system,” notably “facilitated follow-up, continuity of care, or linkage to care” (n=5). The theme “improves health outcomes” contained only one code, “provides rapid identification of infection or clinical deterioration for timely treatment of COVID-19” (n=6). Finally, the theme “supports public health initiatives” contained 2 codes, including “delivers educational messages” (eg, fights disinformation or the “infodemic”; n=5). All other codes were noted by <5 publications.

Many publications (38/48, 80%) reported barriers, challenges, and/or concerns regarding implementation of RMTs (Table 4). In all, 54 distinct codes were developed inductively (Multimedia Appendix 5) and mapped to 15 themes: “equity-related barriers” (n=16), “a lack of RMT implementation guidelines and research” (n=12), “resources required for technology development and implementation” (n=11), “challenging patient experiences of RMTs” (n=10), “confidentiality-related concerns” (n=10), “workforce training” (n=8), “quality of information” (n=8), “communication-related barriers” (n=7), “ethical concerns with RMTs” (n=7), “policy requirements” (n=7), “quality of care concerns” (n=4), “technology-specific barriers” (n=3), “technology integration–related barriers” (n=2), and “financial barriers” (n=2). A total of 8 codes were mapped to the theme “equity-related barriers,” including “lack of access to RMTs in low-resource settings” (eg, patients experiencing homelessness, neighborhoods without access to libraries, households without internet or devices, low-income communities unable to afford RMTs or share RMTs; n=8), “low network quality or internet connectivity or bandwidth,” which can impact quality of care (n=6), and “low patient health literacy” (n=5). A total of 5 codes were mapped to the theme “a lack of RMT implementation guidelines and research,” with 9 publications reporting a “paucity of high-quality data or guidelines to support effective and safe RMT, particularly in acute care.” A total of 7 codes were mapped to the theme “resources required for technology development and implementation,” including “inadequate control of patient flow with some RMTs” (eg, the fluctuating recruitment of patients should be matched with staffing; n=5). A total of 4 codes were mapped to the theme “challenging patient experiences of RMTs,” including “the complexity or intrusiveness of switching to online consultation or remote monitoring and disruption to patient or worker processes and routines” (n=5). A total of 3 codes were mapped to the theme “confidentiality-related barriers,” including “the need to address privacy concern when implementing RMTs” (n=9). A total of 3 codes were mapped to the theme “workforce training,” including “the need for additional workforce education and training in use of RMTs” (n=6). A total of 2 codes were mapped to the theme “quality of information,” including “issues regarding the quality of health information reported or collected” (eg, self-reporting; n=6). Finally, 3 codes were mapped to the theme “ethical concerns with RMTs,” of which the majority (n=5) of the publications reported general ethical concerns. All other codes were reported by <5 publications.

Equity groups were mapped deductively to the PROGRESS-Plus acronym (Table 5). Several publications (n=23) noted how remote monitoring provided information on characteristics of populations that could stratify health opportunities and outcomes (Multimedia Appendix 6). The most frequently reported characteristic was the “Plus” code (n=17). This “Plus” code refers to specific patient populations (ie, patients who distrust the health care system, those with chronic conditions or comorbidities, immune-suppressed patients, pregnant women, acutely unwell patients, patients with cognitive impairment, nonadherent patients, seniors, and youth). The next most common PROGRESS-Plus code was “place of residence.” A total of 11 publications commented on patient place of residence as impacting the implementation of RMTs, including rural and remote residences, nursing homes, or homelessness. A total of 7 publications reported socioeconomic factors, such as a lack of community funding in low- and middle-income countries, lack of access to technology, or low-income patients. A total of 3 publications reported on race or ethnicity or culture or language, occupation, and gender or sex, respectively. Occupations considered were frontline workers and veterans. Finally, in terms of PROGRESS-Plus equity considerations, 2 publications commented on education, specifically the health literacy level of patients, which could impact their ability to understand infection control information or assess the quality of unregulated health information. Only one publication commented on social capital and none on religion.

A rapid review was conducted to present the different types of RMTs deployed in response to the COVID-19 pandemic, the perceived benefits of these technologies, and the perceived barriers or challenges to their use in clinical work. There was special emphasis placed on equity considerations.

Results indicate that many different RMTs are used when providing health care to patients with COVID-19. Most RMTs mirrored the real-time interpersonal communication processes involved in prepandemic care, except that the care was being delivered with physical distance—for instance, the use of videoconferencing (25/48, 52%) and teleconferencing (15/48, 31%) rather than in-person appointments, and accessing medical records remotely (8/48, 17%) rather in the office. Video was likely favored because of the substantial advantage of being able to use nonverbal cues and assessment. These innovations were pertinent to monitoring, triage, and diagnosis (eg, symptom tracking using mobile apps).

The most common reported benefit of RMTs was that it reduced the risk of transmission of COVID-19 (n=18), including to health care practitioners (n=10). This is unsurprising, considering that concern about managing infection rates is one of the forces driving rapid implementation of RMTs across many specialties [67]. Further, with increasing pressure on the health care system, RMTs could aid in reducing the burden on hospitals (n=17), health care system costs (n=10), and the burden of time of health care practitioners (n=16).

Use of these RMTs began or substantially increased in response to COVID-19. If perceived advantages are borne out, this may lead to wider adoption beyond the pandemic.

There may be several implications when considering the breadth of RMTs already in use. First, both health care providers and patients may become accustomed to the remote modality and its benefits, including increased convenience, meaning that RMT use could continue beyond the duration of the pandemic. Health care during the pandemic has increased patient involvement in their own care (eg, symptom monitoring). This may lead to a permanent shift in health care culture, in which patients’ active participation is incorporated into the health provision of other conditions. The development of health-related software for signs and symptom monitoring and triage could also continue and become refined and expand the tools through which quality health care is delivered. When considering whether or which changes might endure, it might be helpful to distinguish between advantages of RMTs that are only relevant to COVID-19 infection risk and those where the benefits might persist beyond COVID-19 (eg, widespread use of health-related apps to better monitor symptoms for other conditions). It is possible and likely that the processes surrounding their use will be refined to address and reduce concerns and barriers.

The most commonly reported barrier themes were a lack of guidance (n=12) and increased resources needed (n=11) for implementation, development, and use of RMTs to treat COVID-19. These barriers are likely due to the novel nature of COVID-19 and thus could become less relevant over time. Another main concern cited by several publications was that rigorous privacy and security settings would be necessary to protect patient information (n=9). However, despite emphasizing the importance of privacy and security, only 5 publications of 18 describing implementation of a specific RMT reported on security and privacy features or policies of the software used [24-26,32,35]. Additionally, two publications noted that use of RMTs could break down the humanitarian core of care as well as patient-provider communication [43,50]. One way to promote effective patient communication is to design user-friendly technology with two-way communication [68]. Although patient involvement in technology development can be used to effectively tailor the technology to the specific needs of the patient [69], no publications reported this in practice. This is unfortunate considering the impact it can have on the success of an RMT [70]. Future technologies should involve rigorous user evaluation based on feedback from patients. The extent of patient involvement in RMT implementation should be thoroughly described to support use of the technology. Lastly, it will take time and resources to bring RMTs to scale; information regarding clinical utility and cost will help ascertain which should be prioritized for investment of resources to aid in this development.

Equity factors also proved important to consider when implementing RMTs. Health interventions should be tailored according to population needs or they risk increasing health inequities [17]. The most common overall barrier theme was concerns regarding equitable use of RMTs (n=15). This is concerning considering marginalized groups are already disproportionately impacted by COVID-19. There is a higher incidence of infection, as well as poorer outcomes, among racialized communities and ethnic minorities [71-73]. More than half of the included publications (n=28) noted challenges that might be faced by minorities or emphasized the importance of training health practitioners in equitable digital health implementation and supporting low-income communities. Alternatively, possible benefits were reported in some papers (n=7), including the ability to support vulnerable populations—for example, medically vulnerable individuals, or those living in rural or remote communities. Overall, there appeared to be conflicting perspectives on whether RMTs would decrease or increase health inequities. There is evidence that intervention-generated inequities—those caused by use of interventions that provide limited benefit to vulnerable populations—can decrease health outcomes among marginalized groups [13]. It is important to track characteristics that may stratify health outcomes in order to assess which RMTs will benefit vulnerable populations. Overall, 23 of 48 total papers (48%) and 7 of 18 (39%) implementation and development papers reported on characteristics of populations that stratify health outcomes. It is critical that an increasing number of publications start to analyze data based on characteristics that influence equitable outcomes to better support vulnerable groups and reduce health inequities.

This work has several time-related limitations due to the rapid nature of this review. Only one screener (EH) reviewed the titles, abstracts, and full texts of the records and coded during the selection process. Further, given the delay between the use of an intervention and publication of its evaluation, the technologies presented here only reflect what was used during a specific window of time and as new technologies are rapidly developed and implemented, there will be a need to reassess new and emerging benefits of and barriers to patient care. Further periodic reviews should be conducted to assess how the use of RMTs evolves. In addition, the quality of the analysis was limited. Furthermore, the frequency of reporting does not necessarily equal the magnitude of importance or the impact on outcomes for patients. Lastly, certain papers reported on multiple technologies but did not provide benefits and barriers specific to each RMT individually, which limits the generalizability of our findings. As such, most benefits and barriers reported in this paper were categorized generally as applicable to both synchronous and asynchronous RMTs. Undoubtedly, many of these benefits and barriers will be more or less relevant depending on the technology (eg, reduced isolation is more applicable for treatment via video consultation than for use of a digital pulse oximeter). Future reviews should focus on separating benefits and barriers according to technology type if possible.

This rapid review summarizes RMTs that were used in the early stages of the COVID-19 pandemic and provides insights on the benefits these technologies could provide for future use. It also highlights perceived barriers to implementation of RMTs that should be addressed in ongoing development projects. Guidelines and policies developed for implementing RMTs ought to be mindful of the identified barriers. RMTs have an important role to play in supporting patients and communities through these unprecedented times. One key recommendation is to establish best practices in the development of RMTs so they are both equitable and effective going forward.