In early 2020, the coronavirus disease 2019 (COVID-19) pandemic emerged, and widespread social-distancing efforts suspended much of the delivery of nonurgent health care. Telehealth proved to be a viable alternative to in-person care, at least on a temporary basis, and utilization skyrocketed. Many Federally Qualified Health Centers (FQHCs) serving low-income patients started delivering telehealth visits in high volume in March 2020 to help maintain access to care.
Before the pandemic, most health centers did not offer telehealth visits for primary care or behavioral health, in large part because of reimbursement policy. In spring 2020, dramatic policy changes removed many of the restrictions on telehealth delivery, and health centers responded by standing up large telehealth programs.
This sudden and dramatic change in health care delivery posed numerous challenges. Health centers had to quickly make changes to technology, workflows, and staffing to accommodate telehealth visits. To support health centers in these efforts, the California Health Care Foundation established the Connected Care Accelerator (CCA) program, a quality improvement initiative that was launched in July 2020. The goal of CCA was to facilitate the transition to telehealth services by providing funding and technical assistance to 45 health centers in California. The study we describe in this article evaluated the progress of health centers that participated in CCA by exploring changes in telehealth utilization and health center staff experiences with implementation.
In this research, we consider recent literature on telehealth implementation in safety net settings. We also present new information on the experiences of the 45 CCA health centers, drawing from data on visit trends, interviews with health center leaders, and surveys of health center providers and staff.
We first present quantitative data on in-person, video, and audio-only visits from February 2019 to August 2021. These data were reported by participating health centers using a standardized reporting tool. We then summarize findings on the experiences and perceptions of health center leaders and staff. These qualitative data were collected through semistructured interviews with more than 100 staff members from 23 health centers. Interviews with staff who were involved in telehealth implementation occurred at three time points: fall 2020, spring 2021, and fall 2021. Lastly, we present findings from a health center provider and staff survey that was fielded in April 2021. The survey was administered to more than 500 providers from 30 health centers and covered knowledge, attitudes, and experiences with telehealth.
Key Findings
Although overall visit volumes remained about the same from the prepandemic to the pandemic study periods, the share of audio-only and video visits dramatically increased during the pandemic, particularly for behavioral health. Monthly visit volumes during the pandemic study period (August 2020–August 2021) were largely similar to those of the prepandemic period (February 2019–February 2020), with the exception of reduced primary care volume during three months in summer and fall (August, October, and November 2020) and reduced behavioral health volume during two months in the fall (October and November 2020). (Data from March 2020 through July 2020 are not included here because of volatility associated with the pandemic at that time.)
Audio-only visits were the highest-volume telehealth modality for primary care and behavioral health throughout the entire pandemic study period; at the end of the study period, however, audio-only visits were eclipsed by in-person visits for primary care but not for behavioral health. Audio-only visits for primary care peaked in April 2020 and remained the dominant telehealth modality throughout the pandemic study period. However, by February 2021, in-person visits regained the position of dominant modality among the three modalities as audio-only visits declined. In the final three months of the pandemic study period (June–August 2021), 66 percent of total visits for primary care were conducted in person, while 30 percent and 4 percent were audio-only visits and video visits, respectively. For behavioral health, audio-only visits exceeded in-person visits and were the dominant modality of care delivery throughout the entire pandemic study period. In the final three months of the pandemic study period, 28 percent of total behavioral health visits were conducted in person, while 52 percent and 20 percent were audio-only visits and video visits, respectively.
The use of video visits varied substantially across health centers, particularly for behavioral health; health centers that delivered numerous video visits and replaced audio-only visits with video visits over time had some common promising practices that could have contributed to their success. From January 2021 to August 2021, video visits represented a median of 3 percent (range: 0–39 percent) of all primary care visits and a median of 14 percent (range: 0–98 percent) of all behavioral health visits across different health centers. Interviews provided insights into the factors that facilitated the use of video visits. Interviewees suggested that telehealth platforms that did not require patients to use patient portals facilitated video telehealth use. The health centers that had telehealth platforms that were embedded in the electronic health record (EHR) (and that required patients to log into patient portals to access) often were preferred by providers compared with stand-alone systems that did not require portal use. However, platforms that did not require portal use and that allowed providers to send a link to the patient so that they could access the video visit were easier for patients to use, particularly those with low digital literacy. Because of challenges with the digital divide, many health centers found it helpful to actively address potential barriers to engagement in both audio and video by adding onboarding processes to screen patients for technology access, offering one-on-one support for information technology (IT) issues, and having providers use their relationships with patients to encourage telehealth use. Health centers found such processes particularly important in facilitating video visits, in addition to scheduling video as the default modality and setting targets for each modality.
Patients with limited English proficiency participated in a significantly lower percentage of video visits. To address disparities in access, clinics engaged in a variety of creative solutions to address the digital divide. The population of patients who received health care prior to the pandemic (in August 2019) varied notably from the population of patients who received video visits in August 2020 and 2021, both in primary care and behavioral health. Most notably, patients who preferred a language other than English participated in 45.5 percent of total primary care visits in August 2019 but only 37.7 percent of video visits in August 2020 and 35.8 percent of video visits in August 2021 (p < 0.01). These differences might reflect variations in digital literacy, a lack of instructions in multiple languages, and a cultural preference for in-person care. In interviews, staff described difficulties providing comprehensive interpretive services within and outside the actual telehealth visit (e.g., in communications from the health center about the visit).
Perceptions of whether telehealth provided an acceptable level of care were relatively positive, but perceptions of its sustainability—as well as perceptions of equity and quality of care—were nuanced. Interviewees noted the importance of telehealth in maintaining patient access to care and in overcoming barriers to in-person care, such as transportation. However, some expressed concern about inequities in access, particularly for patients who were most affected by the digital divide. Although many interviewees viewed telehealth as “here to stay,” they noted that patients appeared to be increasingly comfortable with in-person care. Some health centers were seeking more information on what modalities work best for patients, while others were expanding telehealth offerings for patients with access challenges. Staff survey findings also suggest nuanced conclusions about telehealth. Providers and staff who responded to the survey had a positive experience with telehealth overall, with most respondents agreeing with positive statements on the resources and training provided to them, the team-based care and workflows implemented in their clinics, the access to care provided to patients, the encouragement that leadership provided on telehealth implementation, and their satisfaction with the work they have done through telehealth. However, some reported a preference for in-person visits, while others reported a preference for telehealth visits. Furthermore, providers and staff did not agree on the quality of care provided by telehealth overall and specifically for patients with limited English proficiency.
The implementation of telehealth required significant changes to workflow and staff roles. Key facilitators of telehealth implementation were leadership support, patient willingness to use the technology, platforms that were easy to use and access, a sense of urgency within clinics because of the public health emergency, changes in reimbursement policy, and training opportunities for staff. Interviewees and survey participants suggested that support to pursue telehealth among executive-level and clinician leaders facilitated telehealth implementation, built buy-in among staff, and helped coordinate efforts across organizations. Patient and caregiver willingness to try telehealth, as well as access to easy-to-use telehealth technology, also factored into success. Interviewees described a sense of urgency to deliver telehealth to maintain patient access—and to keep health centers afloat financially. This might help explain greater telehealth use at the start of the pandemic study period, when uncertainty about the impact of the pandemic on services and the sense of urgency were highest. Changes in reimbursement policies (particularly for the reimbursement of audio visits at rates equivalent to those of in-person visits) also enabled the shift to telehealth. CCA and other peer-learning and collaborative opportunities also helped health centers share promising ideas for implementation.
Implications for Policy, Practice, and Future Study
As health centers and policymakers consider the role that telehealth should play in safety net settings moving forward and how to implement needed changes, both might want to factor in some of the key findings of this study, particularly with regard to health equity, quality of care, and implementation in low-resource settings. In the following sections, we offer some key considerations for policy, practice, and future study within each of these domains.
More Work Is Needed to Understand How Telehealth Helps and Hinders Health Equity and to Improve Equitable Telehealth Access
Although most studies to date are somewhat inconsistent on differences in telehealth utilization by race and ethnicity, studies (that are not limited to safety net settings or populations) do consistently show that rural populations and patients with limited English proficiency are less likely to engage in video visits. Although this study did not look at differences in utilization by rurality, our findings echo prior data on limited video usage for those with limited English proficiency in particular (Hsueh et al., 2021; Rodriguez et al., 2021; Sachs et al., 2021).
Billing modifiers that differentiate between modalities are needed to further understand disparities. Because visit data collected for this evaluation were reported in the aggregate, not at the individual level, our evaluation team could not explore granular questions on disparities that adjust for multiple demographic characteristics. In future studies, it will be particularly important to find ways to structure utilization data to include information on which modality was used (i.e., video versus audio only). Currently, Medicaid billing modifiers in California do not distinguish between telehealth visit types, which limits the utility of Medicaid data in assessing disparities in use of different modalities and in surfacing quality issues that are specific to audio-only or video visits. Policymakers are considering incorporating and requiring audio-only modifiers to improve data quality, and we argue that this is a key improvement to better track and evaluate telehealth access and quality.
Data Are Needed to Understand the Impact of Audio-Only Visits on Quality of Care to Inform Reimbursement Policy
The impact of audio-only visits on quality of care needs to be empirically tested. Perhaps one of the most striking findings regarding telehealth utilization since the pandemic started—both in this study and in previous work—is the ongoing use of audio-only visits among health centers. However, even though audio-only visits undoubtedly helped health centers stay afloat when patients and providers were avoiding in-person visits for safety reasons, it is important to remember that audio-only visits, which are delivered either alone or as a component of hybrid care, are relatively untested from a quality perspective. It would be particularly helpful for future research to focus on determining what types of visits are clinically appropriate for the audio-only modality. With this knowledge, health centers could set up workflows in which audio-only visits are offered only under certain conditions so that patients using that modality do not receive inferior care. Until more evidence emerges on effectiveness, policymakers should consider setting up guardrails to prevent overuse of audio-only visits.
Health Centers Should Be Permitted to Serve as Distant Sites on a Permanent Basis
The high and sustained use of telehealth in safety net settings suggests that health centers and their patients derive certain benefits from telehealth. It is clear that some of the policy waivers put in place for the public health emergency, such as permitting health centers to serve as distant sites (i.e., sites where providers delivering telehealth services are located), should be maintained. Prematurely restricting health centers’ ability to deliver telehealth visits would reduce access to care for underserved populations. It would also prevent the research community from documenting the real-world impacts of telehealth on access, quality, and costs outside a public health emergency. This information will be critical to crafting effective long-term policy.
Health Centers Need Resources, Time, and Support to Successfully Implement Telehealth
Health centers can be supported in increasing access for patients with limited English proficiency and other populations with access challenges more generally. One issue of clear concern is telehealth access for populations with limited English proficiency. Policymakers and payers can support health centers with grants, technology solutions, and promising practices to better support patients with limited English proficiency who are engaging in telehealth. Health centers should also engage patients with limited English proficiency in the design and rollout of telehealth platforms.
Health centers should take steps to better support patients facing digital barriers. For those patients facing such barriers as limited digital literacy, lack of technology, lack of broadband, and distrust of telehealth, health centers should consider expanding their capabilities to provide real-time technical support, organize telehealth navigator programs, and provide one-on-one training to patients. Tools and guidance on these and other strategies to facilitate telehealth implementation are widely available. For example, CCA provides several types of online resources (Center for Care Innovations, undated).
Knowledge of what works in video visit implementation already exists, and health centers have many tools to grow their video visit programs. Understanding lessons learned from positive deviants (i.e., health centers that successfully implemented video visits) can inform future implementation efforts. Although audio-only visits were the dominant telehealth modality during the pandemic study period in both primary care and behavioral health, the extent of variation in the use of different visit modalities across participating health centers was noteworthy. Whereas audio-only visits made up the majority of telehealth visits for some health centers, other health centers were able to incorporate many more video visits into their care delivery. Those that grew their programs often had common characteristics, such as leadership support and utilization goals. Some health centers experienced particular success with getting their processes set up for delivering video telehealth—or perhaps self-limiting their use of audio-only visits. It would be helpful for these health centers to share and disseminate their processes.
Appropriate staffing is key to telehealth implementation. Telehealth implementation in a hybrid care model, in which patients receive a mix of telehealth and in-person visits, will require appropriate staffing models. Interviews showed that medical assistants played a critical role in successful implementation by viewing their role as malleable and filling gaps in staffing as needed. Health centers will need to consider ways to further support team-based care and formally integrate telehealth across all roles to spread out the resources needed across the organization. Another way to approach easing the burden of telehealth implementation, where possible, is to do what some health centers have already done: establish a new telehealth coordinator role.
Conclusion
This mixed-methods study followed 45 health centers serving low-income populations for 18 months of the COVID-19 pandemic. Although several studies have been conducted throughout the pandemic to capture telehealth utilization and implementation experiences, this study represents one of the most comprehensive analyses to date on the topic of telehealth implementation in the safety net. A unique contribution of this study is its ability to distinguish between different telehealth modalities. The high and ongoing use of audio-only visits throughout the pandemic study period is noteworthy. Although audio-only visits have clearly played a critical role in maintaining access to care during the public health emergency, their ongoing role in the care delivery of low-income populations requires careful consideration. Telehealth has the potential to increase access to care and deliver care that is more convenient and patient-centered; however, ongoing research is needed to ensure that telehealth is implemented in a way that ensures high-quality care and health equity.
