Virtual Primary Care Implementation During COVID-19 in High-Income Countries: A Scoping Review

Step 1: Identifying the research questions

The two central research questions were as follows:

Step 2: Identifying the relevant studies

A search strategy was created in collaboration with an academic librarian at the University of Toronto for the two concepts of virtual care and COVID-19. To increase the sensitivity of the search, the concept of primary care was not included as a search concept owing to the broad definition used in this review and the variety of ways primary care could be described in the literature. Instead, the concept of primary care was screened by the review team after the articles were identified using the first two concepts. The search strategy used a combination of subject headings, text words, and other field codes as applicable, and the virtual care concept was adapted from the search strategies used in other published reviews.^13,14

The COVID-19 search strategy was adapted to the Ovid Expert Searcher COVID-19 filter.¹⁵ The subject databases were searched on August 10, 2020 and included the following: Ovid MEDLINE including Epub Ahead of Print, In-Process & Other Non-Indexed Citations, Ovid MEDLINE Daily and Ovid MEDLINE, Ovid EMBASE, and EBSCO’s CINAHL. There were no date limits or language limits placed on the subject databases. An additional Scopus search was run on August 15, 2020 using a date limit of 2019 and 2020 to ensure our search captured the most recent literature related to virtual primary care during COVID-19. Complete search strategies are available in Multimedia Supplementary Appendix SA1.

Step 3: Study selection

Inclusion/exclusion criteria

All types of indexed studies including commentaries, protocol papers, and editorials were included in this review. In addition, articles were considered eligible for this review if it was from high-income countries as defined by the World Bank,¹⁶ that were published in English since December 2019. In this review, we defined virtual care as any technology used to enable bidirectional patient–provider communication remotely. Articles were included if it examined virtual care delivered in a primary care setting during the COVID-19 pandemic.

To further narrow the scope of our search, studies were excluded if they did not address at least one of the two research questions. This ensured that all articles were relevant to the objectives and research questions of this scoping review. Box 1 outlines the inclusion and exclusion criteria used to inform the selection of relevant articles in this review.

Screening abstracts and titles

Each citation was independently screened by two reviewers at the title and abstract level. Any conflicts were resolved by a third reviewer and discussed with the greater research team. A screening guide outlining the eligibility criteria mentioned previously was developed iteratively by the research team and used to guide reviewers during the screening and selection process. A total of seven researchers independently piloted the screening guide on a test sample of 100 abstracts, concluding with an interrater reliability (IRR) of 83%. According to Hallgren,¹⁷ an IRR >80% is considered to be in perfect agreement. Any discrepancies identified during the pilot test were discussed, which were then applied to further refine the screening guide. The screening guide is provided in Multimedia Supplementary Appendix SA2.

Screening full text

Two reviewers independently screened the full-text articles to determine if they met the inclusion criteria. Similar to the title and abstract phase, conflicts were resolved by a third reviewer. Reviewers met regularly to discuss any challenges related to study selection and the eligibility criteria was refined as needed. The reasons for excluding studies were documented and are outlined in the PRISMA diagram in Figure 1.

Step 4: Data items and data collection process

During the data extraction process, a data extraction form was developed, and pilot tested on two articles. The feedback from the pilot test was discussed with the entire research team and further refinements were made to ensure the form’s consistency and comprehensiveness at capturing relevant data. The following article summary information was captured in the finalized extraction form created on Google Forms:

Five reviewers extracted all data from the included articles, with the data from each review being extracted only once by a single reviewer. Single abstraction was used as 100 articles were tested in the pilot stages by five reviewers. The data extraction form demonstrated consistent results among the reviewers. The extracted data were exported from Google Forms into Microsoft Excel for analysis. The data extraction form is provided in Multimedia Supplementary Appendix SA3.

Step 5: Synthesizing and reporting the results

The extracted data were synthesized using a qualitative descriptive approach.⁹ Based on recommendations by Levac et al.,⁹ a qualitative content analysis was conducted to identify major themes related to the main challenges encountered when implementing and delivering virtual primary care services during COVID-19, and the strategies for overcoming those challenges. The themes were developed iteratively in collaboration with the research team over a series of meetings. These discussions also supported the identification of several recommendations to inform decision makers on how best to implement and deliver virtual primary care services during and after the COVID-19 pandemic. The PRISMA-Scoping Reviews checklist was followed to report on our results.¹⁸

Results

Included Studies

The database searches found 10,549 citations that were imported into Covidence for deduplication and study selection. An additional 766 citations were identified from the reference lists of only 43 of the 49 included studies for data extraction. Six articles were not located using Scopus and Web of Science, thus references from those six articles were not included in the supplementary search. After deduplication in Covidence, 6,580 unique citations were screened for eligibility. From the screening phase, a total of 49 articles were included for data extraction. After this, additional 11 articles were excluded from the study because they did not contain relevant information related to at least one of our research questions as outlined in Figure 1. Data from 38 articles were included in the final review.

Study Characteristics

Table 1 describes the characteristics of the studies included in this review. This review identified 38 unique articles published from 11 high-income countries with most publications originating from the United States (n = 21, 55%), United Kingdom (n = 5, 13%), and Australia (n = 4, 11%). More than half of the articles published were commentaries, opinion pieces, or editorials (n = 21, 55%), followed by case studies (n = 8, 21%), and case–control studies (n = 3, 8%). Given the novelty of COVID-19 and the transition to virtual care in the primary care sector, the majority of the articles were published in June (n = 11, 29%), May (n = 9, 23%), and July 2020 (n = 8, 21%), just a few months after the WHO declared the COVID-19 pandemic.

The included articles described clinicians using video communication (n = 32, 84%), telephone communication (n = 27, 71%), and asynchronous text messaging (n = 12, 32%) to continue providing care to patients during the pandemic. Other types of discussed virtual care modalities included patient portals (n = 9, 24%), asynchronous e-mail messaging (n = 7, 18%), and other modalities such as online applications/surveys and remote monitoring devices (n = 6, 16%).

Barriers to Implementation of Virtual Care

Table 2 outlines seven factors associated with virtual primary care implementation and the corresponding barriers and strategies related to each factor that was found in this review. As demonstrated in Table 2 and in Multimedia Supplementary Appendix SA4, more than half of the included articles (n = 20, 53%) focused on equity, specifically on lack of access to internet, smartphones, and internet bandwidth for rural, seniors, and underserved populations. Lack of funding for virtual care (n = 14, 37%) was the second most common topic across the included articles, such as inadequate reimbursement policies for virtual care.

Other common topics included negative patient and clinician perceptions of virtual care (n = 11, 29%), lack of appropriate regulatory policies (n = 10, 26%), inappropriate clinical workflows (n = 9, 24%), lack of virtual care infrastructure (n = 8, 21%), and the need for appropriate virtual care training and education for clinicians (n = 5, 13%). Overall, the articles stated the above seven factors were prominent barriers in implementing virtual care in primary care settings.

Strategies to Implementation of Virtual Care

The articles included in this review outlined a variety of strategies used when implementing virtual care during COVID-19, including those that were based on the authors’ own experiences of shifting to virtual care during the pandemic. As summarized in Table 2, strategies that were discussed to mitigate challenges associated with funding for virtual care services were the most prevalent (n = 15, 39%), followed by strategies to improve clinical workflow and integrate virtual care (n = 13, 34%). Some of these strategies included providing access to interpreter services¹⁹ and expanding the roles of nurses, medical assistants, and other allied health professionals to assist with patient management.²⁰

Other strategies included conducting appropriate virtual care training and education (n = 11, 29%), and creating mitigation strategies to improve virtual care infrastructure and patient equity (each had n = 7, 18%), regulatory policies (n = 5, 13%), and strategies to improve patient and clinician perception of virtual care services (n = 3, 7%). A detailed chart outlining the barriers and strategies proposed or utilized in each article is given in Multimedia Supplementary Appendix SA4.

Key Barriers and Strategies Within the Quadruple Aim Components

The following presents the main barriers and strategies found within the scoping review as categorized within the Quadruple Aim components, with the two components of improving patient experience and improving health outcomes grouped together.

Improving patient experience and health outcomes by integrating equitable access and educational initiatives

Inequitable access was found to be a top barrier in more than half of the articles in this scoping review. Factors included lack of access to the internet, phone, or other devices to connect with physicians, low levels of technical literacy, and poor access to virtual care in rural and underserved populations.

Several of the included articles implemented and proposed a variety of strategies to mitigate disparities in equitable access that can improve the patient’s experience with virtual primary care. For example, Sundar³⁵ proposed an equitable alternative to in-person access of care could be the use of smartphone tools, such as FaceTime, which are easily accessible to populations with limited English proficiency. In another example, Woodall et al.³⁸ and Zhai³⁹ recommended advocating for low-cost internet bandwidth for people with poor access to the internet and increasing funding for expansion of virtual primary care in rural and underserved communities.

Patients can also benefit from receiving educational materials on how to use virtual care services.^32,39,56 This was because patients could lack knowledge on the availability of virtual care services and may perceive the care to be of lower standards. Sundar³⁵ suggested that the educational materials should be provided in different languages as well as communicated using different modalities (e.g., through mail, e-mail, text messaging, or phone call). In addition, Cheng et al.¹⁹ suggested providing access to interpreter services. Many of the articles in this review recognized that by implementing strategies that enable equitable access and increase the community’s knowledge of available primary virtual care services, barriers experienced by low income and marginalized populations could be mitigated.

Reducing per capita health care costs by developing new models of virtual care reimbursement

The lack of appropriate reimbursement policies for health care professionals, including nurses and allied health care professionals was identified as one of the key barriers in our scoping review. Several studies proposed the potential strategy of developing new funding and reimbursement policies to reduce the burden felt by lack of funding,^21,32,36,48 although Fisk et al.⁴⁴ noted the importance of creating funding and administrative frameworks that are consistent among jurisdictions. In fact, inconsistency in funding and administrative frameworks across cities was thought to potentially create a fragmented system resulting in policy and jurisdictional conflicts rather than the creation of an integrated health care system.⁴⁴

Remuneration models that enable workforce substitution were also suggested to help reduce physician burden by utilizing the entire clinical team as well as potentially reduce cost and improve clinical workflow.⁴⁵ Volpp et al.³⁶ suggested the use of capitation models that are based on specific patient characteristics, including home care patients using remote monitoring tools to provide a true reflection of the time utilized to manage complex diseases and enable physicians to be renumerated as such. Finally, fee-for-service models based on time increments, were proposed as a reflection of true effort and infrastructure costs to manage various modalities.²³ However, it could incentivize one modality over others if remuneration is not accurately reflected and this might not be based on patient choice.

Overall, virtual primary care was thought to have the potential to reduce costs in the long run by implementing renumeration models that are patient-centered and improve care coordination. Upfront funding to equitably expand virtual care could reduce per capita cost of health care, as part of the Quadruple Aim^7,36 in the long run both during and after the pandemic.⁷

Improving provider experience by redesigning clinical workflow and integrating virtual care infrastructure

Significant barriers identified in the review were associated with disruptions to workflow when integrating virtual primary care. To improve provider experience and incorporate virtual primary care more seamlessly, Srinivasan et al.²⁰ suggested that organizations need to construct new role responsibilities for providers and provide training for these new responsibilities.

Furthermore, Wherton et al.⁴³ and Verhoeven et al.⁴⁹ proposed that organizations need to adjust practice management by separating COVID and non-COVID patients, reconfigure patient administration systems to enable appointment scheduling templates that distinguish between different appointment systems, and create protocols to account for different aspects of a virtual clinic. For example, considerations could include scheduling to account for patient “no-shows,” recording of decisions and outcomes, and adapting workflows to accommodate for virtual patient check-in/check-out procedure.⁴³

Applicable technology and infrastructure for virtual primary care services was found to be a pertinent barrier in expanding virtual primary care services and impeding provider experience in this review. Olayiwola et al.³⁰ and Peek et al.⁴² suggested that organizations ensure that required software and hardware are available and connected to virtual care services outside primary care to provide a seamless hand-off and care experience to patients. In addition to the equipment, Sherwin et al.³¹ suggested that organizations must also leverage the Fast Healthcare Interoperability Resources Application Programming Interfaces to integrate third-party videoconferencing platforms into the electronic medical record to reduce administrative burden of double documentation.

Above all, Sherwin et al.³¹ believed organizations must also invest in governance models and user experience to help the implementation of virtual care. By incorporating user-centered designs and iterative development of technology, patients and clinicians can be represented in the model of care leading to increased virtual care adoption.^25,43 Overall, many of the articles in this review determined that by redesigning existing clinical workflows to adapt to the integration of virtual care technology and ensuring the appropriate infrastructure is in place to support this technology, provider experience can significantly improve when delivering virtual primary care services.

Discussion

Principal Findings

This scoping review highlights the barriers and possible strategies found in literature for implementing virtual care technologies in a primary care setting within high-income countries. The review includes a broad range of virtual care modalities such as telephone (synchronous) to patient education portal (asynchronous) and corresponding barriers and strategies. To our knowledge, this is the first review of its kind to synthesize literature on virtual care modalities during the COVID-19 pandemic. Our results demonstrate that physicians are using a wide variety of modalities to communicate with patients; however, ancillary factors such as lack of equitable access to care, lack of regulatory policies and funding, poor patient and clinician perception, inappropriate clinical workflows, and lack of virtual care infrastructure can make implementation difficult.

Strategies and facilitators to circumvent these barriers have been recommended to improve care experiences during the pandemic. Thereafter, key barriers and strategies found through our scoping review were mapped to the Quadruple Aim to create a sustainable and recommend system-level improvements in virtual care for patients. Still, some considerations are needed to successfully implement virtual primary care services.

Considerations For Virtual Primary Care Implementation

Providing equitable access to care is an important factor in improving patient experience and health outcomes as the current virtual primary care implementation has exacerbated the existing disparities in rural and underserved populations.⁵⁷ Moreover, COVID-19 has disproportionately affected communities, specifically low-income and minority populations, older adults (>65 years), and those with limited English proficiency.⁵⁸ Therefore, the equitable delivery of virtual primary care is imperative to address the high risk of COVID-19 transmission in these communities.⁵⁷

To develop strategies to mitigate disparities in equitable access, data from virtual visits of underserved populations should first be collected and analyzed to understand the impact of sociodemographic factors on access to virtual primary care.⁵⁹ Data collection could be stratified based on relevant factors such as race/ethnicity, socioeconomic status, and age. Measuring these relevant factors could be imperative in any long-term policy development to improve population health.⁵⁹

It is also important to consider the lack of regulatory policies within the privacy regulations that could also hinder the expansion of virtual primary care.⁶⁰ Considering primary care is the first level of contact between the patient and the overall health system, trust in technology used by the clinics is important to ensure patients remain safe and do not have to default to in-person consultation, specifically during the pandemic.⁶¹ Lack of regulatory policies could also negatively impact patient outcome and greatly influence clinical experience, thereby threatening the two tenants of the Quadruple Aim.⁶⁰

The results from this review indicate that implementing appropriate reimbursement models for virtual care can reduce health care costs. Consultation charges of virtual primary care have been found to be lower than in-person, face-to-face consultations in several studies.^59–61 In addition, the marginal cost per patient could reduce if the cost of service remains constant and patient volume increases.⁶²

Costs associated with implementing virtual primary care services (e.g., recurring costs in videoconferencing equipment) can hinder scaling; however, to counteract the costs of implementation, an increase in the volume of patient consultations is needed.⁶² Furthermore, funding models play a pivotal role in determining productivity and reducing the cost per patient, which may hold the potential to improve patient care such as by encouraging adoption of new innovations aimed to reduce health care visits through a capitation mode.⁶²

Finally, provider burnout levels were high before the pandemic, and COVID-19 is exacerbating burnout rates owing to the added levels of stress associated with insufficient resources, long hours, occupational hazards, and rapidly shifting hospital policies.⁶³ Workflow redesign to adapt virtual care platforms in practice are critical in improving providers’ experience and is associated with six times lower odds of burnout.⁶⁴ High provider burnout is also associated with reduced adherence to treatment plans, resulting in poor clinical outcomes leading to poor patient experience and negatively impacting patient care.⁶³

Overall, this review demonstrated that integrating equitable access, developing appropriate renumeration models, redesigning clinical workflows and incorporating virtual care infrastructure into our existing health care system can help improve all components of the Quadruple Aim. However, a system-level perspective is needed, and important regulatory policies are required to support the coordination of the many stakeholders involved. This review highlighted some key recommendations within the Quadruple Aim that could be considered during the development and implementation of virtual care in primary care settings:

Limitations

Although this scoping review used rigorous and transparent methods to identify articles relevant to the research questions, this review may not have identified all articles related to virtual care in a primary care setting during COVID-19. Although we defined primary care broadly, the subjective nature of what constitutes a primary care setting and/or primary care service may have led to some articles not having been captured in this review. However, this led to the inclusion of a wide range of primary care settings and primary care services including some studies (n = 8) that did not explicitly focus on a primary care setting. Specifically, these articles were included as they contained valuable information related to our research questions, thus increasing the external validity of our recommendations. This review may also have been limited by articles published in the English language.

Conclusion

The COVID-19 pandemic has exacerbated several health system inefficiencies. However, it has also provided an opportunity to transform the health care system and bring about long-lasting changes using virtual care. The purpose of this scoping review was to examine challenges, strategies, and knowledge acquired from high-income countries that have experiences in implementation and delivery of virtual primary care services during the COVID-19 pandemic.

This review identified several barriers and the corresponding strategies to mitigate those barriers. These strategies must account for the factors related to virtual primary care implementation such as equity, regulatory policies, technology and infrastructure, education, clinician and patient experience, clinical workflows, and funding for virtual care. In utilizing strategies that take these factors into account, an integrated virtual primary care system that meets the requirements of the Quadruple Aim can be created.

Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this article.

Supplementary Material

Multimedia Supplementary Appendix SA1

Multimedia Supplementary Appendix SA2

Multimedia Supplementary Appendix SA3

Multimedia Supplementary Appendix SA4