Skip to main content

Differences in access to virtual and in-person primary care by race/ethnicity and community social vulnerability among adults diagnosed with COVID-19 in a large, multi-state health system

Abstract

Background

Research exploring telehealth expansion during the COVID-19 pandemic has demonstrated that groups disproportionately impacted by COVID-19 also experience worse access to telehealth. However, this research has been cross-sectional or short in duration; geographically limited; has not accounted for pre-existing access disparities; and has not examined COVID-19 patients. We examined virtual primary care use by race/ethnicity and community social vulnerability among adults diagnosed with COVID-19 in a large, multi-state health system. We also assessed use of in-person primary care to understand whether disparities in virtual access may have been offset by improved in-person access.

Methods

Using a cohort design, electronic health records, and Centers for Disease Control and Prevention Social Vulnerability Index, we compared changes in virtual and in-person primary care use by race/ethnicity and community social vulnerability in the year before and after COVID-19 diagnosis. Our study population included 11,326 adult patients diagnosed with COVID-19 between March and July 2020. We estimated logistic regression models to examine likelihood of primary care use. In all regression models we computed robust standard errors; in adjusted models we controlled for demographic and health characteristics of patients.

Results

In a patient population of primarily Hispanic/Latino and non-Hispanic White individuals, and in which over half lived in socially vulnerable areas, likelihood of virtual primary care use increased from the year before to the year after COVID-19 diagnosis (3.6 to 10.3%); while in-person use remained stable (21.0 to 20.7%). In unadjusted and adjusted regression models, compared with White patients, Hispanic/Latino and other race/ethnicity patients were significantly less likely to use virtual care before and after COVID-19 diagnosis; Hispanic/Latino, Native Hawaiian/Pacific Islander, and other race/ethnicity patients, and patients living in socially vulnerable areas were also significantly less likely to use in-person care during these time periods.

Conclusions

Newly expanded virtual primary care has not equitably benefited individuals from racialized groups diagnosed with COVID-19, and virtual access disparities have not been offset by improved in-person access. Health systems should employ evidence-based strategies to equitably provide care, including representative provider networks; targeted, empowering outreach; co-developed culturally and linguistically appropriate tools and technologies; and provision of enabling resources and services.

Peer Review reports

Background

In the United States (U.S.) to date, there have been over 870,000 deaths related to coronavirus disease 2019 (COVID-19) and more than 73 million cases since January 2020 [1]. Unsurprisingly, given the well-documented connection between systemic inequities and structural racism, and the disparate distribution of health burden in the U.S. [2,3,4,5,6,7,8,9], COVID-19 has had a disproportionate impact on racialized groups and socially vulnerable communities. Greater rates of COVID-19 cases, hospitalizations, and deaths have been reported among Hispanic/Latino, American Indian/Alaska Native, Asian, Black/African American, and Native Hawaiian/Pacific Islander populations than among White, non-Hispanic populations. Socially vulnerable communities characterized by factors such as increased poverty and crowded housing have also experienced greater rates of COVID-19 infections and mortality [10,11,12,13,14,15,16].

Symptoms of COVID-19 are highly variable in both extent and type, and can affect multiple organ systems [17,18,19]. As many as one in three patients with COVID-19 report symptoms persisting beyond 4 weeks [17, 20], and following even mild cases some patients experience symptoms lasting months after COVID-19 onset, so called COVID-19 long-haulers [21]. Given this, appropriate post-COVID-19 infection care includes monitoring for persistent post-COVID-19 conditions and sequelae and treating those that arise [22]. Unfortunately, healthcare providers and researchers are becoming increasingly concerned that long-term COVID-19 symptoms and sequelae may disproportionately impact racialized groups for some of the same reasons that they experience greater rates of infection, illness severity, and death—lack of access to high-quality care and difficulty persuading providers that their experiences and conditions are real [23].

During the COVID-19 pandemic, efforts to slow the spread of COVID-19 and mitigate its adverse consequences have led to a remarkable transformation in U.S. healthcare delivery [10, 24]. As early as March 2020, the Centers for Disease Control and Prevention (CDC) called for a prioritization of telehealth care, and at both the state and federal levels reimbursement for telehealth services was markedly expanded [25, 26]. Prior to the COVID-19 pandemic, health systems reported relatively low rates of telehealth use [27], and even those with high adoption performed fewer than 100 virtual visits per day [28]. Now, many health systems conduct hundreds of virtual visits per day [29] and telehealth is considered a key access point for diagnosis, triage, and treatment of health conditions [10, 29, 30]. With this unprecedented increase in telehealth adoption, pre-pandemic care disparities that existed during a time when this care modality was largely unavailable may be reduced or mitigated.

Most early pandemic research has found disparities in access to telehealth for some groups, including groups disproportionately impacted by COVID-19 disease. One study by Chunara, Zhao, Lawrence, et al. [31] showed that from March 19 to April 20, 2020, Black patients in a large healthcare system in New York City had significantly lower odds of any telehealth use (virtual urgent and ambulatory encounters, together) compared with White patients. Pierce and Stevermer [32] found that during a similar time period among Medicare, Medicaid, self-pay, and privately insured patients, compared with White patients and urban patients, Black patients and rural patients in a Missouri academic medical center were less likely to use family medicine telehealth care. For those that did, Black patients were less likely than White patients to have access to audio-video (versus audio-only) family medicine telehealth. In addition, in the first year of the pandemic, individuals across the U.S. with employer-sponsored insurance living in high poverty and rural areas had the smallest increases in telehealth use of any type according to a study by Cantor, McBain, Pera et al. [33]. Conversely, research on a healthcare organization in Southern California serving racially/ethnically and socio-economically diverse patients found that compared with pre-pandemic levels, increases in synchronous telephone or live-video-audio telehealth use of any kind during the pandemic’s first year were largest among Hispanic and low-income patients [34].

Motivation for this study

With adequate access to high-quality primary care, morbidity, adverse clinical outcomes, and negative social consequences (e.g., loss of income and economic instability) such as those of long-term COVID-19 disease can be reduced [35,36,37,38,39]. Primary care providers who know their patients and are aware of their life circumstances are in ideal positions to act as care hubs, coordinating and personalizing COVID-19 recovery, providing referrals to specialty care as appropriate, and addressing barriers to needed services and supports [35]. During the COVID-19 pandemic, health systems have relied more heavily on telehealth to ensure access to these crucial primary care services [40]. Yet, access to telehealth care has not been distributed equitably across racialized groups, income levels, or geographic areas. That said, existing research on telehealth disparities during the pandemic has covered relatively small geographic regions and has been either cross-sectional in nature or short in duration. Moreover, it has largely focused on general patient populations rather than those with COVID-19—patients who may experience outsized benefit from enhanced access to primary care given the very real potential for long-term COVID-19 conditions and sequelae. In addition, existing research has neglected to examine virtual primary care specifically, an important access, triage, and treatment point for COVID-19 patients. In this study, we seek to fill some of these gaps by exploring differences in longer-term access to virtual and in-person primary care by race/ethnicity and community social vulnerability among patients diagnosed with COVID-19 in a large, multi-state health system during the first year and a half of the pandemic, with the goal of understanding in/equity of health system response to COVID-19 disease among a racially/ethnically and socio-demographically diverse group of COVID-19 patients.

Methods

Study aims

Using a retrospective, observational cohort design, we examined the association between race/ethnicity, community social vulnerability, and use of virtual primary care among adult patients diagnosed with COVID-19 between March and July 2020 at a Providence health system site, comparing virtual primary care use in the year prior to and the year after COVID-19 diagnosis. To explore whether disparities in access to virtual primary care may have been offset by improved access to in-person primary care, we compared concurrent use of in-person primary care in the year before and after COVID-19 diagnosis by race/ethnicity and community social vulnerability among the same patients.

Study data and population

Data for this study came from the Providence health system electronic health record (EHR), which contains information on patient demographic and health characteristics, and healthcare utilization. Providence is a non-profit healthcare system operating across seven states: Alaska, California, Montana, New Mexico, Oregon, Texas, and Washington [41]. Individuals were included in the study if they were at least 18 years of age at the time of their COVID-19 diagnosis, if they did not die in the year following their COVID-19 diagnosis, and if they resided in a state served by Providence. Individuals were excluded from the study if they resided in Texas (n = 46); at the time of this study, the Providence Texas EHR was not fully integrated with the larger Providence EHR and therefore complete data were not available for Texas’ patients. Individuals were also excluded from the study if they had missing community social vulnerability (n = 31) or race/ethnicity (n = 1072) information, resulting in a final study sample comprised of 11,326 individuals.

Variables

Dependent variables

Our dependent variables included two binary variables for whether an individual used virtual primary care and in-person primary care, defined as having at least one virtual or in-person primary care visit, respectively. Primary care visits were identified in the EHR as visits that occurred with any provider delivering care in an outpatient department group designated as primary care. Visits were classified as virtual if they took place in a virtual office setting, or in-person if they took place in an in-person outpatient setting. A virtual office visit could include visits in which patients had audio-video or audio-only capability. Asynchronous messaging was not included in our definition of virtual primary care visits. In order to examine longer-range access to primary care (versus COVID-19 onset-related access), we censored visits that took place during the first 30 days after COVID-19 diagnosis when constructing each of our utilization outcomes, as these visits may have been associated with acute COVID-19 disease-related care.

Independent variables

Data on study participants’ race and ethnicity was used to create a mutually exclusive race/ethnicity variable comprised of seven categories: non-Hispanic White, non-Hispanic Black, non-Hispanic Asian, non-Hispanic American Indian/Alaskan Native (AI/AN), non-Hispanic Native Hawaiian/Pacific Islander (NH/PI), non-Hispanic other, and Hispanic/Latino. If a study participant was recorded as being Hispanic/Latino and any other race, they were categorized as Hispanic/Latino, otherwise they were categorized based on the most up-to-date race/ethnicity information available in the EHR.

Community social vulnerability was defined using the Social Vulnerability Index (SVI) from the CDC, which categorizes an area as socially vulnerable via factors such as poverty, lack of access to transportation, and crowded housing [42]. The SVI determines the social vulnerability of each census tract using a percentile ranking of the proportion of tracts that are equal to or lower in rank than the tract of interest along the SVI theme. For example, a tract ranking of 0.90 indicates that the tract of interest is more vulnerable than 90% of tracts in its state along the SVI theme in question [43]. The SVI ranks each tract on 15 social factors, which are then grouped into four themes: (1) socioeconomic status, (2) household composition and disability, (3) minority status and language, and (4) housing type and transportation [44]. Following the approach outlined by Flanagan, Gregory, Hallisey et al. [45], we created binary variables for each of the four SVI themes which were equal to 1 if an individual resided in a tract that was more vulnerable along the SVI theme in question than 90% or more of tracts in its state, and 0 if otherwise.

The post-COVID-19 diagnosis period was identified based on a study participant’s first positive COVID-19 diagnosis. From this we created a binary variable that identified whether an observation occurred before or after the COVID-19 diagnosis date.

Demographic variables included age group, calculated as the study participant’s age at the date of their COVID-19 diagnosis (categories included < 20, 20–34, 35–44, 45–54, 55–64, 65–74, and ≥ 75); and sex (categories included male and female). Health status variables included a proxy variable for COVID-19 severity, which was a variable indicating whether a study participant was diagnosed with COVID-19 in an inpatient healthcare setting (categories included yes, no); and five binary variables indicating whether a study participant had chronic conditions including diabetes, hypertension, coronary artery disease, chronic kidney disease, and/or congestive heart failure (categories included yes, no). These were the five most common chronic conditions among the study sample that were flagged by Providence as potentially leading to worse COVID-19 disease and outcomes, and are relevant to pre-COVID-19 disease primary care needs and utilization.

Statistical analysis

Descriptive analysis

We examined demographic and health characteristics of our study sample overall and by virtual primary care use status by computing cell sizes and percentages for categorical variables. We descriptively examined primary care use outcomes by computing cell sizes and proportions of the study sample who used each type of primary care, and by computing means and standard deviations for each type of primary care during the study period overall, and separately for the year before and after COVID-19 diagnosis.

Regression analysis

We employed logistic regression and calculated average marginal effects to explore changes in the probabilities of each primary care outcome from the year before to the year after COVID-19 diagnosis by race/ethnicity and community social vulnerability. To do this, we created interaction terms for the post-COVID-19 period and each category of race/ethnicity, and interaction terms for the post-COVID-19 period and each binary SVI theme variable. Following the approach employed by other researchers studying COVID-19 and/or telehealth disparities [46,47,48,49], and because associations between race/ethnicity, community social vulnerability, and access to care reflect persistent structural racism and inequitable distribution of resources in the U.S., we examined both unadjusted differences in use of care, which we consider our main models, and adjusted differences that account for the demographic and health characteristics described above. To address potential heteroskedasticity, we estimated robust standard errors in all regression models [50]; statistical significance was determined at the traditional 5% alpha level. All analyses were performed using Stata version 14.2 [51]. This study was reviewed and approved by the Providence Institutional Review Board.

Results

Population characteristics

Table 1 provides demographic and health characteristics of the study sample overall and stratified by virtual primary care use status. Overall, the study population was largely comprised of Hispanic/Latino (41.7%) and non-Hispanic White (39.5%) individuals. Individuals who used virtual primary care during the study period were more likely to be non-Hispanic White, Black, and Asian. Over half the study sample (56.7%) lived in a census tract categorized as socially vulnerable along at least one community social vulnerability theme, with the housing type and transportation theme being the most common (41.6%). Individuals who used virtual primary care during the study period were less likely to live in tracts categorized as socially vulnerable along each of the community social vulnerability themes.

Table 1 Characteristics of study sample overall and by virtual primary care use status

With respect to the other demographic and health characteristics of the sample, the majority of individuals (81.3%) were under the age of 65, and there was an almost even split between male (47.8%) and female (52.2%) sex individuals. Older individuals and female sex individuals were more likely to use virtual primary care during the study period. About one quarter of individuals (24.0%) were diagnosed with COVID-19 in an inpatient setting, and about one in five individuals (22.4%) had hypertension; a smaller proportion had diabetes (14.8%), coronary artery disease (10.4%), chronic kidney disease (9.2%), and congestive heart failure (4.8%). Individuals who were diagnosed with COVID-19 in an outpatient setting were more likely to use virtual primary care during the study period, as were individuals without chronic health conditions.

Primary care use

Table 2 provides descriptive data on virtual and in-person primary care use overall and stratified by the year before and after COVID-19 diagnosis. Overall, about one in ten individuals (12.0%) used virtual primary care and about one in three (27.4%) used in-person primary care during the study period. While the proportion of individuals who used in-person primary care remained similar before and after COVID-19 diagnosis (21.0 and 20.7%, respectively), the proportion who used virtual primary care increased substantially after COVID-19 diagnosis (3.6% before to 10.3% after COVID-19 diagnosis). The average number of virtual and in-person primary care visits remained similar from the year before to the year after COVID-19 diagnosis.

Table 2 Primary care use, overall and before and after COVID-19 diagnosis

Differences in primary care use by race/ethnicity

Table 3 gives the average marginal effects (percentage-point differences) for the probability of using each type of primary care for our main and ancillary regression models. In our main model that included race/ethnicity, community social vulnerability, and interactions for the post-COVID-19 period, differences in the likelihood of using virtual primary care were observed by race/ethnicity. For example, in the year prior to COVID-19 diagnosis, non-Hispanic Black individuals had an increased likelihood of using virtual primary care compared with non-Hispanic White individuals [3.30 percentage points (pp), p < 0.05], while individuals identified as non-Hispanic other and Hispanic/Latino race/ethnicity had lower likelihoods of using virtual primary care in the year prior to COVID-19 diagnosis (─2.53 pp., p < 0.05, and ─2.11 pp., p < 0.01, respectively). None of the interaction terms for time period and race/ethnicity were statistically significant, indicating that these differences in virtual primary care use by race/ethnicity persisted in the year following COVID-19 diagnosis. (Differences in virtual primary care use could not be assessed for the non-Hispanic AI/AN group due to small cell size.)

Table 3 Association between race/ethnicity, community social vulnerability, and primary care use among individuals diagnosed with COVID-19

Similar to the differences identified in virtual care utilization, in the year prior to COVID-19 diagnosis, there were significant differences in the likelihood of use of in-person primary care by race/ethnicity, with individuals identifying as non-Hispanic AI/AN (─6.83 pp., p < 0.01), non-Hispanic NH/PI (─9.49 pp., p < 0.01), non-Hispanic other (─6.95 pp., p < 0.001), and Hispanic/Latino (6.01 pp., p < 0.001) race/ethnicity having lower likelihoods of using in-person primary care compared with non-Hispanic White individuals. Again, none of the interaction terms were significant, indicating that these differences persisted in the year after COVID-19 diagnosis.

In ancillary models that adjusted for demographic and health characteristics, nearly all differences in virtual and in-person primary care use by race/ethnicity were similar in magnitude and significance. The only difference that was slightly reduced was the difference in virtual primary care use for non-Hispanic other race/ethnicity individuals (from ─2.53 pp., p < 0.05, to ─2.33 pp., p = 0.058).

Differences in primary care use by community social vulnerability

In our main models that included community social vulnerability, race/ethnicity, and interactions for the post-COVID-19 period, use of virtual primary care was not significantly different by community social vulnerability in either the year before or after COVID-19 diagnosis. However, differences in use of in-person primary care by community social vulnerability were observed. For example, individuals residing in areas categorized as vulnerable based on minority status and language, and housing type and transportation had decreased likelihoods of using in-person primary care in the year before COVID-19 diagnosis (─3.72 pp., p < 0.01, and ─5.61 pp., p < 0.001, respectively). None of the interaction terms for time period and community social vulnerability theme were significant, indicating that these differences persisted in the year after COVID-19 diagnosis. In ancillary models, differences in in-person primary care use by community social vulnerability were similar in magnitude and significance.

Discussion

Our study is one of the first to examine disparities in virtual and in-person primary care use among individuals diagnosed with COVID-19. Findings from our research suggest that, although there was a substantial increase in access to virtual primary care in the wake of the initial wave of COVID-19 infections and telehealth expansions, this did not substantially alter pre-pandemic disparities in access to primary care. Overall, only 12% and 27% of individuals diagnosed with COVID-19 used any virtual and in-person primary care, respectively, during the two-year study period. Disparities in use of virtual primary care were observed by race/ethnicity, and for some racialized groups, may have been compounded by concurrent disparities in use of in-person primary care. Specifically, Hispanic/Latino and non-Hispanic other race/ethnicity individuals were less likely to use virtual and in-person primary care compared with non-Hispanic White individuals in the year prior to COVID-19 diagnosis, and these disparities persisted in the year following COVID-19 diagnosis. Individuals who identified as Non-Hispanic NH/PI were no more likely to use virtual primary care compared with non-Hispanic White individuals, and were less likely to use in-person primary care. On the other hand, we found that during the study period, non-Hispanic Black individuals were more likely to use virtual primary care than non-Hispanic White individuals. These differences were observed when accounting for community social vulnerability and after controlling for demographic and health characteristics.

Previous research on racial and ethnic disparities in access to virtual care during the pandemic provides mixed results. For example, some studies have found that Black individuals had less access to urgent and ambulatory telehealth encounters and family medicine telehealth visits, compared with White individuals [31, 32], while others have found that Black individuals were more likely than White individuals to self-report using telehealth generally because of the pandemic [52]. However, previous studies examined shorter time periods or were narrower in geographic focus. Moreover, previous studies included general patient populations rather than those diagnosed with COVID-19, making direct comparisons to our own findings challenging. In addition, they did not distinguish between Hispanic and non-Hispanic Black patients, and most did not distinguish types of virtual care use (e.g., primary care versus telehealth care more broadly), further compounding comparison issues. That said, our study adds to the growing body of evidence demonstrating that White, non-Hispanic individuals generally enjoyed greater access to virtual care than non-White and Hispanic/Latino individuals during the COVID-19 pandemic—a particularly concerning finding given that racialized groups generally experience greater chronic disease burden than White and non-Hispanic populations [53] and increased COVID-19 infection rates, both of which may necessitate increased access to care. In addition, our study highlights what pre-pandemic research has shown: that racial/ethnic disparities in access to virtual care have existed since the advent of telehealth care, and rather than improving disparities through expansion of telehealth, the pandemic has only exacerbated and/or shed more light on them [48, 49, 54, 55].

Overall, we found no significant differences in use of virtual primary care by community social vulnerability. Yet, we did observe disparities in the use of in-person primary care that were not improved by increased access to virtual care. Specifically, regression analysis revealed that individuals living in areas characterized as vulnerable based on minority status and language, and housing type and transportation were less likely to use in-person primary care and no more likely to use virtual primary care than individuals living in areas not characterized as vulnerable in these ways. These findings held in both models that adjusted only for race/ethnicity and in models that additionally adjusted for age, sex, and chronic conditions.

To the best of our knowledge, previous studies have not used the CDC’s Social Vulnerability Index to explore disparities in access to virtual care during the pandemic. However, existing studies have examined factors related to community social vulnerability, including increased poverty and rural geography, finding pandemic-related disparities in access to virtual care along these dimensions [32,33,34]. In addition, evidence has clearly demonstrated how mechanisms of social stratification at the community level (e.g., segregation, community disinvestment, geographical concentration of poverty) result in differential access to care, and that community characteristics directly and indirectly shape what and where care is available, and the quality of that care [56]. Our findings support and add to existing literature on the association between community-level vulnerability and access to care, indicating that even after accounting for race/ethnicity and controlling for demographic and health characteristics, those living in areas characterized by greater rates of non-White and non-English speaking individuals, and by crowded housing and lack of transportation, experienced disparate access to primary care prior to and during the first year and a half of the pandemic.

Taken together, our findings highlight the urgent need to ensure equitable access to virtual (and in-person) primary care. This is particularly important given additional surges in cases and the emergence of new COVID-19 variants, all of which signal that the pandemic is not likely to end soon and that virtual care will remain an important care modality. Furthermore, there is a growing “care debt” that has the potential to lead to deleterious downstream consequences such as complications from unmanaged health conditions and incapacitation of an already overwhelmed healthcare system [29]. To create health systems that can effectively manage these contingencies, it will be crucial to transition telehealth services from a crisis intervention tool to an equitable and sustainable system for providing proactive patient care.

Evidence-based strategies exist for creating more equitable telehealth and primary care infrastructure [31]. First, conducting targeted patient outreach and actively connecting with individuals and groups who experience barriers to care has been shown to improve access to and utilization of care [57]. For example, in a study by Ospina-Pinillos et al. [58], participatory design methods were used to tailor the website of a virtual mental health clinic to improve outreach to Spanish-speaking individuals, which led to adequate acceptability levels in the website’s homepage, and triage, booking, and video visit systems for Spanish-speakers, and also enabled the clinic to identify the need for tailored assessment tools and greater integration with Spanish-speaking services and communities. In addition, a systematic review of interventions aimed at modifying the healthcare system to better outreach to and serve racialized groups and communities revealed that these interventions were associated with both improved processes of care delivery and reduced access disparities [59]. However, health systems must first be able to identify those experiencing barriers to care and find ways to create meaningful connections with them. This necessitates leveraging our current understanding of the multiple intersecting individual and community factors affecting access to care and addressing them in outreach materials and methods. Furthermore, at a systems level, this means deconstructing current systems which are inherently racist, overtly discriminatory, and implicitly biased, and rebuilding them into more just and healing systems that are acceptable and comfortable for diverse patient populations. For virtual care, this also means conducting additional research on what constitutes effective and trustworthy outreach and communication to diverse populations [31].

Next, developing representative provider networks can improve capacity of and access to care, while at the same time improving quality of care for underserved individuals and communities. Racial/ethnic concordance between patients and providers is associated with improved use of preventive services, satisfaction with care, patient-provider communication quality, and patient participation in care and decision-making [60, 61]. In addition, evidence shows that clinicians from racialized groups are more likely to treat patients from racialized groups, including those who live in medically underserved and vulnerable areas [62]. However, policymakers and health systems must purposefully devote financial and other resources to improving provider representativeness and dismantling racist and discriminatory practices including those that have resulted in a current provider supply that is more White and socioeconomically advantaged than the general U.S. population [63].

Designing culturally appropriate tools and technology that enable and improve access requires adaptations to systems predominantly designed for White, English-speaking individuals. To that end, health systems can collect and incorporate input on telehealth tools and technologies from racialized groups and those with limited English proficiency [64], as evidence indicates that cultural and linguistic tailoring can improve healthcare access and outcomes [58, 65,66,67]. Data collection and user testing should be done in a participatory manner in which cultural adaptations, and knowledge and language translation are co-designed with patients and/or research participants [58]. Health systems can also increase robust adoption of the National Culturally and Linguistically Appropriate Services Standards developed by the U.S. Department of Health and Human Services [68], which are intended to provide health workers and systems with a blueprint for developing equitable, understandable, respectful systems of care.

Policy solutions are also needed to address systemic barriers to care, such as inequitable distribution of healthcare and enabling resources. A recent survey found that limited broadband connectivity and related technology (e.g., computers and smart phones) has created barriers to telehealth during the pandemic [69]. This issue has particularly impacted individuals in rural areas and those over the age of 65. One policy solution is to provide funding for broadband expansion in medically underserved communities. Several initiatives are underway to accomplish this: As part of the American Rescue Plan Act of 2021, the Federal Communications Commission is launching the $3.2 billion Emergency Broadband Benefit program to help Americans with qualifying household incomes obtain high-speed internet [70]. In addition, a $100 million federal pilot program has been implemented to cover eligible costs of broadband connectivity, network equipment, and information services needed to provide connected care services to patients; and the COVID-19 Telehealth Program included $200 million in Congressional appropriations to help healthcare providers provide connected care to patients at their homes or in mobile locations [71, 72]. Time and future research will tell whether these policy solutions have reduced disparities in access to telehealth care.

Other policy and systems-level solutions that have been shown to improve access to primary care among underserved populations and communities include expanding scope of practice laws for and increasing the use of non-physician clinicians; expanding the supply of non-hospital-based clinics such as Federally Qualified Health Centers (FQHCs); increasing the availability of after-hours primary care services; and removing cost-related barriers to primary care such as cost-sharing [73].

Limitations

This study has some limitations worth noting. First, our study sample is limited to Providence patients in six mostly Mid−/Western U.S. states, which may limit generalizability to Southern and North−/Eastern states. That said, this study provides data on patients across a large, multi-state geographic area that includes both rural and urban areas, enhancing generalizability compared with existing research on smaller geographic areas and largely urban centers. Next, our sample is comprised of patients who tested positive for COVID-19, yet evidence has demonstrated disparities in COVID-19 testing rates among racialized groups and those with limited English proficiency, even as they experience higher COVID-19 infection rates [74,75,76]. Therefore, our sample likely does not include all Providence patients who contracted COVID-19. If patients from racialized groups who contracted COVID-19 were tested at a lesser rate than non-Hispanic white patients, our results likely underestimate disparities in access to care. Despite this, the fact that our COVID-19 positive sample was largely comprised of Hispanic/Latino patients while the larger Providence patient population is primarily comprised of non-Hispanic white patients enhances confidence in our findings. Finally, various issues arise in analyses of electronic health record data and should be taken into consideration when interpreting our findings. For example, if Providence patients received care outside of a Providence setting, it is not recorded in the EHR or included in our analyses. In addition, the EHR data does not contain information on other relevant factors such as socio-economic status or access to enabling resources. However, we did include census tract-level socioeconomic and resource-related variables via the SVI, and thus captured at least some of the variability in these factors and their association with access to care.

Conclusion

The pandemic has further illuminated the persistent inequities that lead to poorer access to care and health outcomes among racialized groups and vulnerable communities. Our study adds to the mounting body of evidence that lays bare these inequities. Using data from a large health system across multiple states, we found disparities in utilization of virtual and in-person primary care by both race/ethnicity and community social vulnerability among individuals diagnosed with COVID-19, some of the same groups of people who have been hit hardest by COVID-19 infections, morbidity, economic consequences, and mortality. The importance of primary care, together with widespread telehealth expansion brought about by the COVID-19 pandemic highlight both an urgent need and unprecedented opportunity to address these disparities, but only if solutions are purposefully designed and implemented to address their root causes [31, 34].

Availability of data and materials

The dataset used in this study was derived from electronic medical records and includes individual-level identifiers and protected health information that are not publicly available due to human subjects protection requirements. We are unable to provide this dataset for public use. However, requests for summary information may be submitted to the corresponding author for consideration.

Abbreviations

AI/AN:

American Indian / Alaska Native

CDC:

Centers for Disease Control and Prevention

COVID-19:

Coronavirus Disease 2019

EHR:

Electronic Health Record

NH/PI:

Native Hawaiian / Pacific Islander

SVI:

Social Vulnerability Index

U.S.:

United States

References

  1. Covid in the U.S. Latest Map and Case Count. New York Times. 2022. Available from: https://www.nytimes.com/interactive/2021/us/covid-cases.html [cited 28 Jan 2022]

    Google Scholar 

  2. Du Bois W. The Philadelphia negro; a social study. Philadelphia: Published for the University; 1899. Available from: https://search.library.wisc.edu/catalog/999852732802121

    Google Scholar 

  3. Hargrove TW. Structural Racism and Inequalities in Health: Footnotes: a publication of the American Sociological Association; 2021. Available from: https://www.asanet.org/news-events/footnotes/apr-may-jun-2021/features/structural-racism-and-inequalities-health [cited 3 Jan 2022]

  4. Beltrán-Sánchez H, Soneji S, Crimmins EM. Past, present, and future of healthy life expectancy. Cold Spring Harb Perspect Med. 2015;5(11) Available from: https://pubmed.ncbi.nlm.nih.gov/26525456/ [cited 3 Jan 2022].

  5. Bailey ZD, Feldman JM, Bassett MT. How structural racism works — racist policies as a root cause of U.S. racial health inequities. N Engl J Med. 2021;384(8):768–73 Available from: https://www.nejm.org/doi/full/10.1056/NEJMms2025396 [cited 3 Jan 2022].

    Article  PubMed  Google Scholar 

  6. Gee GC, Ford CL. Structural racism and health inequalities: old issues, new directions. Du Bois Rev. 2011;8(1):115–32 Available from: https://pubmed.ncbi.nlm.nih.gov/25632292/ [cited 6 Jan 2022].

    Article  PubMed  PubMed Central  Google Scholar 

  7. Williams DR, Lawrence JA, Davis BA. Racism and health: evidence and needed research. Annu Rev Public Health. 2019;40:105–25 Available from: https://pubmed.ncbi.nlm.nih.gov/30601726/ [cited 2 Jan 2022].

    Article  PubMed  PubMed Central  Google Scholar 

  8. National Academies of Sciences E and M, Division H and M, Practice B on PH and PH, States C on C-BS to PHE in the U, Baciu A, Negussie Y, et al. The Root Causes of Health Inequity. In: Communities in Action: Pathways to Health Equity: National Academies Press (US); 2017. p. 1–558. Available from: https://www.ncbi.nlm.nih.gov/books/NBK425845/ [cited 6 Jan 2022]

  9. Churchwell K, Elkind MSV, Benjamin RM, Carson AP, Chang EK, Lawrence W, et al. Call to action: structural racism as a fundamental driver of health disparities: a presidential advisory from the American Heart Association. Circulation. 2020;142:E454–68 Available from: https://www.ahajournals.org/doi/abs/10.1161/CIR.0000000000000936 [cited 6 Jan 2022].

    Article  PubMed  Google Scholar 

  10. Whaley CM, Pera MF, Cantor J, Chang J, Velasco J, Hagg HK, et al. Changes in health services use among commercially insured US populations during the COVID-19 pandemic. JAMA Netw Open. 2020;3(11):e2024984 Available from: https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2772537 [cited 2021 Nov 17].

    Article  PubMed  PubMed Central  Google Scholar 

  11. Dasgupta S, Bowen V, Leidner A, Fletcher K, Musial T, Rose C, et al. Association between social vulnerability and a county’s risk for becoming a COVID-19 hotspot — United States, June 1–July 25, 2020. MMWR Morb Mortal Wkly Rep. 2020;69(42):1535–41 Available from: https://www.cdc.gov/mmwr/volumes/69/wr/mm6942a3.htm [cited 2021 Oct 4].

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  12. COVID-19 Racial and Ethnic Disparities. Centers for Disease Control and Prevention. 2020. Available from: https://www.cdc.gov/coronavirus/2019-ncov/community/health-equity/racial-ethnic-disparities/index.html [cited 3 Jan 2022]

    Google Scholar 

  13. Lopez L, Hart LH, Katz MH. Racial and ethnic health disparities related to COVID-19. JAMA. 2021;325(8):719–20 Available from: https://jamanetwork.com/journals/jama/fullarticle/2775687 [cited 3 Jan 2022].

    Article  CAS  PubMed  Google Scholar 

  14. Mude W, Oguoma VM, Nyanhanda T, Mwanri L, Njue C. Racial disparities in COVID-19 pandemic cases, hospitalisations, and deaths: a systematic review and meta-analysis. J Glob Health. 2021;11:1–15 Available from: https://pubmed.ncbi.nlm.nih.gov/34221360/ [cited 3 Jan 2022].

    Article  Google Scholar 

  15. Hernandez-Romieu AC, Leung S, Mbanya A, Jackson BR, Cope JR, Bushman D, et al. Health care utilization and clinical characteristics of nonhospitalized adults in an integrated health care system 28–180 days after COVID-19 diagnosis — Georgia, May 2020–March 2021. Morb Mortal Wkly Rep. 2021;70(17):644 Available from: /pmc/articles/PMC8084119/. [cited 24 Aug 2021].

    Article  CAS  Google Scholar 

  16. Karaye IM, Horney JA. The impact of social vulnerability on COVID-19 in the U.S.: an analysis of spatially varying relationships. Am J Prev Med. 2020;59(3):317–25.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Nalbandian A, Sehgal K, Gupta A, Madhavan MV, McGroder C, Stevens JS, et al. Post-acute COVID-19 syndrome. Nat Med. 2021;27:601–15. https://doi.org/10.1038/s41591-021-01283-z Nature research [cited 27 May 2021].

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Struyf T, Deeks JJ, Dinnes J, Takwoingi Y, Davenport C, Leeflang MMG, et al. Signs and symptoms to determine if a patient presenting in primary care or hospital outpatient settings has COVID-19. Cochrane Database Syst Rev. 2021;2021 Available from: https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD013665.pub2/full. John Wiley and Sons Ltd [cited 27 May 2021].

  19. Zhou X, Snoswell CL, Harding LE, Bambling M, Edirippulige S, Bai X, et al. The role of telehealth in reducing the mental health burden from COVID-19. Telemed E-Health. 2020;26:377–9 Available from: www.blackdoginstitute.org.au/getting-help/self-help-tools-apps. Mary Ann Liebert Inc [cited 2021 May 27].

    Article  CAS  Google Scholar 

  20. Logue JK, Franko NM, McCulloch DJ, McConald D, Magedson A, Wolf CR, et al. Sequelae in adults at 6 months after COVID-19 infection. JAMA Netw Open. 2021;4(2):e210830.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Greenhalgh T, Knight M, A’Court C, Buxton M, Husain L. Management of post-acute covid-19 in primary care. BMJ. 2020:370 Available from: https://www.bmj.com/content/370/bmj.m3026 [cited 7 Oct 2021].

  22. Mikkelsen M, Abramoff B. COVID-19: evaluation and management of adults following acute viral illness. Up To Date 2022. Available from: https://www.uptodate.com/contents/covid-19-evaluation-and-management-of-adults-following-acute-viral-illness [cited 10 Mar 2022]

    Google Scholar 

  23. Cooney E. Researchers fear long Covid may disproportionately affect people of color. STAT. 2021. Available from: https://www.statnews.com/2021/05/10/with-long-covid-history-may-be-repeating-itself-among-people-of-color/ [cited 3 Jan 2022]

    Google Scholar 

  24. Iyengar K, Mabrouk A, Jain VK, Venkatesan A, Vaishya R. Learning opportunities from COVID-19 and future effects on health care system. Diabetes Metab Syndr Clin Res Rev. 2020;14(5):943–6.

    Article  Google Scholar 

  25. Hartnett KP, Kite-Powell A, DeVies J, Coletta MA, Boehmer TK, Adjemian J, et al. Impact of the COVID-19 pandemic on emergency department visits - United States, January 1, 2019-May 30, 2020. MMWR Morb Mortal Wkly Rep. 2020;69(23):699–704 Available from: https://pubmed.ncbi.nlm.nih.gov/32525856/ [cited 3 Jan 2022].

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  26. Shachar C, Engel J, Elwyn G. Implications for telehealth in a postpandemic future: regulatory and privacy issues. JAMA. 2020;323(23):2375–6 Available from: https://jamanetwork.com/journals/jama/fullarticle/2766369 [cited 3 Jan 2022].

    Article  CAS  PubMed  Google Scholar 

  27. Harvey JB, Valenta S, Simpson K, Lyles M, McElligott J. Utilization of Outpatient Telehealth Services in Parity and Nonparity States 2010–2015. Telemed J E Health. 2019;25(2):132–6 Available from: https://pubmed.ncbi.nlm.nih.gov/29847224/ [cited 22 Nov 2021].

    Article  PubMed  Google Scholar 

  28. UPMC sees “phenomenal increase” in telemedicine use in COVID-19 crisis - Pittsburgh Business Times. Available from: https://www.bizjournals.com/pittsburgh/news/2020/03/28/upmc-sees-phenomenal-increase-in-telemedicine-use.html. [cited 22 Nov 2021]

  29. Wosik J, Fudim M, Cameron B, Gellad ZF, Cho A, Phinney D, et al. Telehealth transformation: COVID-19 and the rise of virtual care. J Am Med Inform Assoc. 2020;27:957–62 Available from: https://academic.oup.com/jamia/article/27/6/957/5822868. Oxford University Press [cited 27 May 2021].

    Article  PubMed  PubMed Central  Google Scholar 

  30. Alexander GC, Tajanlangit M, Heyward J, Mansour O, Qato DM, Stafford RS. Use and content of primary care office-based vs telemedicine care visits during the COVID-19 pandemic in the US. JAMA Netw Open. 2020;3(10):e2021476 Available from: https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2771191 [cited 4 Jan 2022].

    Article  PubMed  PubMed Central  Google Scholar 

  31. Chunara R, Zhao Y, Chen J, Lawrence K, Testa PA, Nov O, et al. Telemedicine and healthcare disparities: a cohort study in a large healthcare system in New York City during COVID-19. J Am Med Inform Assoc. 2021;28(1):33–41 Available from: https://academic.oup.com/jamia/article/28/1/33/5899729 [cited 22 Sep 2021].

    Article  PubMed  Google Scholar 

  32. Pierce RP, Stevermer JJ. Disparities in use of telehealth at the onset of the COVID-19 public health emergency. J Telemed Telecare. 2020; Available from: https://journals.sagepub.com/doi/full/10.1177/1357633X20963893 [cited 31 Dec 2021].

  33. Cantor JH, McBain RK, Pera MF, Bravata DM, Whaley CM. Who is (and is not) receiving telemedicine care during the COVID-19 pandemic. Am J Prev Med. 2021;61(3):434–8.

    Article  PubMed  PubMed Central  Google Scholar 

  34. Qian L, Sy LS, Hong V, Glenn S, Ryan DS, Morrissette K, et al. Disparities in outpatient and telehealth visits during the COVID-19 pandemic in a large integrated health care organization (preprint). J Med Internet Res. 2021; Available from: http://preprints.jmir.org/preprint/29959/accepted [cited 30 Aug 2021].

  35. Berger Z, Altiery De Jesus V, Assoumou SA, Greenhalgh T. Long COVID and health inequities: the role of primary care. Milbank Q. 2021;99(2):519–41 Available from: https://onlinelibrary.wiley.com/doi/full/10.1111/1468-0009.12505 [cited 4 Jan 2022].

    Article  PubMed  PubMed Central  Google Scholar 

  36. Role of primary care in the COVID-19 response. World Health Organization. 2021. Available from: https://apps.who.int/iris/handle/10665/331921 [cited 4 Jan 2022]

  37. Key Messages: The Role of Primary Health Care in COVID-19 Response & Recovery. The Primary Health Care Performance Initiative. Available from: www.improvingphc.org. [cited 4 Jan 2022]

  38. Strengthening the frontline: how primary health care helps health systems adapt during the COVID-19 pandemic - OECD. Organisation for Economic Co-operation and Development. 2021. Available from: https://read.oecd-ilibrary.org/view/?ref=1060_1060243-snyxeld1ii&title=Strengthening-the-frontline-How-primary-health-care-helps-health-systems-adapt-during-the-COVID-19-pandemic [cited 4 Jan 2022]

  39. Lewis C, Seervai S, Shah T, Abrams M, Zephyrin L. Primary care and the COVID-19 pandemic: The Commonwealth Fund; 2020. Available from: https://www.commonwealthfund.org/blog/2020/primary-care-and-covid-19-pandemic [cited 4 Jan 2022]

  40. Using Telehealth to Expand Access to Essential Health Services during the COVID-19 Pandemic. Centers for Disease Control and Prevention. 2020. Available from: https://www.cdc.gov/coronavirus/2019-ncov/hcp/telehealth.html [cited 4 Jan 2022]

  41. Providence. Available from: https://www.providence.org/. [cited 29 Nov 2021]

  42. CDC/ATSDR’s Social Vulnerability Index (SVI). Centers for Disease Control and Prevention. 2021. Available from: https://www.atsdr.cdc.gov/placeandhealth/svi/index.html [cited 29 Nov 2021]

    Google Scholar 

  43. CDC/ATSDR SVI Frequently Asked Questions (FAQ). Centers for Disease Control and Prevention. 2021. Available from: https://www.atsdr.cdc.gov/placeandhealth/svi/faq_svi.html [cited 29 Nov 2021]

    Google Scholar 

  44. CDC SVI Documentation 2018. Centers for Disease Control and Prevention. 2021. Available from: https://www.atsdr.cdc.gov/placeandhealth/svi/documentation/SVI_documentation_2018.html [cited 29 Nov 2021]

    Google Scholar 

  45. Flanagan BE, Gregory EW, Hallisey EJ, Heitgerd JL, Lewis B, Flanagan BE, et al. A social vulnerability index for disaster management. J Homel Secur Emerg Manag. 2011;8 Available from: http://www.bepress.com/jhsem/vol8/iss1/3 [cited 5 Aug 2021].

  46. Dickinson KL, Roberts JD, Banacos N, Neuberger L, Koebele E, Blanch-Hartigan D, et al. Structural racism and the COVID-19 experience in the United States. Heal Secur. 2021;19(S1):S-14–26. https://doi.org/10.1089/hs.2021.0031.

    Article  Google Scholar 

  47. Cohen-Cline H, Li HF, Gill M, Rodriguez F, Hernandez-Boussard T, Wolberg H, et al. Major disparities in COVID-19 test positivity for patients with non-English preferred language even after accounting for race and social factors in the United States in 2020. BMC Public Health. 2021;21(1):1–9 Available from: https://bmcpublichealth.biomedcentral.com/articles/10.1186/s12889-021-12171-z [cited 4 Jan 2022].

    Article  CAS  Google Scholar 

  48. Goel MS, Brown TL, Williams A, Hasnain-Wynia R, Thompson JA, Baker DW. Disparities in enrollment and use of an electronic patient portal. J Gen Intern Med. 2011;26(10):1112–6 Available from: https://pubmed.ncbi.nlm.nih.gov/21538166/ [cited 4 Jan 2022].

    Article  PubMed  PubMed Central  Google Scholar 

  49. Anthony DL, Campos-Castillo C, Lim PS. Who isn’t using patient portals and why? Evidence and implications from a national sample of US adults. Health Aff. 2018;37(12):1948–54. https://doi.org/10.1377/hlthaff201805117.

    Article  Google Scholar 

  50. Huber PJ. The behavior of maximum likelihood estimates under non-standard conditions. In: Le Cam LM, Neyman J, editors. Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability. Berkeley: University of California Press; 1967. p. 221–33.

    Google Scholar 

  51. Stata Statistical Software. Release 14. College Station: StataCorp LP; 2015.

    Google Scholar 

  52. Campos-Castillo C, Anthony D. Racial and ethnic differences in self-reported telehealth use during the COVID-19 pandemic: a secondary analysis of a US survey of internet users from late march. J Am Med Inform Assoc. 2021;28(1):119 Available from: /pmc/articles/PMC7499625/. [cited 4 Jan 2022].

    Article  PubMed  Google Scholar 

  53. Smedley BD, Stith AY, Nelson AR. Introduction and literature review. In: Unequal treatment: confronting racism and ethic disparities in health care. Washington, D.C.: National Academies Press; 2002. Available from: https://www.nap.edu/read/10260/chapter/1.

    Google Scholar 

  54. Park J, Erikson C, Han X, Iyer P. Are state telehealth policies associated with the use of telehealth services among underserved populations? Health Aff (Millwood). 2018;37(12):2060–8. Available from: https://pubmed.ncbi.nlm.nih.gov/30633679/ [cited 4 Jan 2022].

  55. Senft N, Butler E, Everson J. Growing disparities in patient-provider messaging: trend analysis before and after supportive policy. J Med Internet Res. 2019;21(10) Available from: https://pubmed.ncbi.nlm.nih.gov/31593539/ [cited 4 Jan 2022].

  56. Dimick J, Ruhter J, Sarrazin MV, Birkmeyer JD. Black patients more likely than whites to undergo surgery at low-quality hospitals in segregated regions. 2017;32(6):1046–53. https://doi.org/10.1377/hlthaff20111365.

  57. Musa D, Schulz R, Harris R, Silverman M, Thomas SB. Trust in the health care system and the use of preventive health services by older black and white adults. Am J Public Health. 2009;99(7):1293–9 Available from: https://pubmed.ncbi.nlm.nih.gov/18923129/ [cited 4 Jan 2022].

    Article  PubMed  PubMed Central  Google Scholar 

  58. Ospina-Pinillos L, Davenport T, Diaz AM, Navarro-Mancilla A, Scott EM, Hickie IB. Using participatory design methodologies to co-design and culturally adapt the Spanish version of the mental health eClinic: qualitative study. J Med Internet Res. 2019;21(8) Available from: https://pubmed.ncbi.nlm.nih.gov/31376271/ [cited 4 Jan 2022].

  59. Fisher TL, Burnet DL, Huang ES, Chin MH, Cagney KA. Cultural leverage: interventions using culture to narrow racial disparities in health care. Med Care Res Rev. 2007;64(5 Suppl) Available from: https://pubmed.ncbi.nlm.nih.gov/17881628/ [cited 10 Mar 2022].

  60. Saha S, Komaromy M, Koepsell TD, Bindman AB. Patient-physician racial concordance and the perceived quality and use of health care. Arch Intern Med. 1999;159(9):997–1004 Available from: https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/485025 [cited 4 Jan 2022].

    Article  CAS  PubMed  Google Scholar 

  61. Shen MJ, Peterson EB, Costas-Muñiz R, Hernandez MH, Jewell ST, Matsoukas K, et al. The effects of race and racial concordance on patient-physician communication: a systematic review of the literature. J Racial Ethn Health Disparities. 2018;5(1):117 Available from: /pmc/articles/PMC5591056/. [cited 4 2022].

    Article  PubMed  Google Scholar 

  62. Moy E, Bartman BA. Physician race and care of minority and medically indigent patients. JAMA. 1995;273(19):1515–20 Available from: https://jamanetwork.com/journals/jama/fullarticle/388474 [cited 4 Jan 2022].

    Article  CAS  PubMed  Google Scholar 

  63. Talamantes E, Henderson MC, Fancher TL, Mullan F. Closing the gap — making medical school admissions more equitable. N Engl J Med. 2019;380(9):803–5 Available from: https://www.nejm.org/doi/full/10.1056/NEJMp1808582 [cited 6 Jan 2022].

    Article  PubMed  Google Scholar 

  64. López L, Green AR, Tan-McGrory A, King R, Betancourt JR. Bridging the digital divide in health care: the role of health information technology in addressing racial and ethnic disparities. Jt Comm J Qual Patient Saf. 2011;37(10):437–45 Available from: https://pubmed.ncbi.nlm.nih.gov/22013816/ [cited 4 Jan 2022].

    PubMed  Google Scholar 

  65. Agate S. Unlocking the power of telehealth: increasing access and services in underserved, urban areas. Harvard J Hisp Policy. 2017;29:85–97 Available from: https://go.gale.com/ps/i.do?p=AONE&sw=w&issn=10741917&v=2.1&it=r&id=GALE%7CA634432016&sid=googleScholar&linkaccess=fulltext [cited 4 Jan 2022].

    Google Scholar 

  66. Chandler J, Sox L, Kellam K, Feder L, Nemeth L, Treiber F. Impact of a culturally tailored mHealth medication regimen self-management program upon blood pressure among hypertensive Hispanic adults. Int J Environ Res Public Health. 2019;16(7) Available from: https://pubmed.ncbi.nlm.nih.gov/30959858/ [cited 4 Jan 2022].

  67. McCall T, Bolton CS, McCall R, Khairat S. The use of culturally-tailored telehealth interventions in managing anxiety and depression in African American adults: a systematic review. Stud Health Technol Inform. 2019;264:1728–9 Available from: https://pubmed.ncbi.nlm.nih.gov/31438314/ [cited 6 Jan 2022].

    PubMed  Google Scholar 

  68. National Culturally and Linguistically Appropriate Services Standards. US Department of Health and Human Services. Available from: https://thinkculturalhealth.hhs.gov/clas/standards. [cited 4 Jan 2022]

  69. Bailey V. Limited Broadband Poses a Significant Barrier to Telehealth Access. mHealth Intelligence. 2021. Available from: https://mhealthintelligence.com/news/limited-broadband-poses-a-significant-barrier-to-telehealth-access [cited 4 Jan 2022]

    Google Scholar 

  70. H.R. 1319 - American Rescue Plan Act of 2021. 2021. Available from: https://www.congress.gov/bill/117th-congress/house-bill/1319/text

  71. Connected Care Pilot Program. Federal Communications Commission 2021. Available from: https://www.fcc.gov/wireline-competition/telecommunications-access-policy-division/connected-care-pilot-program [cited 4 Jan 2022]

    Google Scholar 

  72. COVID-19 Telehealth Program (Invoices & Reimbursements). Federal Communications Commission. 2022. Available from: https://www.fcc.gov/covid-19-telehealth-program-invoices-reimbursements [cited 4 Jan 2022]

    Google Scholar 

  73. Kona M, Houston M, Gooding N. The effectiveness of policies to improve primary care access for underserved populations: an assessment of the literature. 2021. Available from: https://www.milbank.org/publications/the-effectiveness-of-policies-to-improve-primary-care-access-for-underserved-populations/ [cited 31 Jan 2022]

    Google Scholar 

  74. Grigsby-Toussaint DS, Shin JC, Jones A. Disparities in the distribution of COVID-19 testing sites in black and Latino areas in New York City. Prev Med (Baltim). 2021:147 Available from: https://pubmed.ncbi.nlm.nih.gov/33647352/ [cited 4 Jan 2022].

  75. Duber HC, Kim HN, Lan KF, Nkyekyer E, Neme S, Pierre-Louis M, et al. Assessment of disparities in COVID-19 testing and infection across language groups in Seattle, Washington. JAMA Netw Open. 2020;3(9):e2021213 Available from: https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2770951 [cited 4 Jan 2022].

    Article  PubMed  PubMed Central  Google Scholar 

  76. Lieberman-Cribbin W, Alpert N, Flores R, Taioli E. Analyzing disparities in COVID-19 testing trends according to risk for COVID-19 severity across New York City. BMC Public Health. 2021;21(1):1–8 Available from: https://bmcpublichealth.biomedcentral.com/articles/10.1186/s12889-021-11762-0 [cited 4 Jan 2022].

    Google Scholar 

Download references

Acknowledgements

We would like to acknowledge analytic and project management support from Megan Holtorf and Aisha Gilmore, and manuscript review from Lisa Angus at the Center for Outcomes Research and Education.

Funding

This work was funded by William E. and Thelma F. Housman Foundation. The funder played no role in the design of the study; collection, analysis, and interpretation of data; or in writing this manuscript.

Author information

Authors and Affiliations

Authors

Contributions

Conceptualization, DJG, SER, and HCC; Data curation, DJG; Formal analysis, DJG; Funding acquisition, HCC; Supervision, SER and HCC; Visualization, DJG; Writing – original draft, DJG, SER, KM, and HCC; Writing – review & editing, DJG, SER, KM, and HCC. All authors have read and approved the final manuscript.

Corresponding author

Correspondence to Katherine Marsi.

Ethics declarations

Ethics approval and consent to participate

The protocol for this study was approved by the Providence Saint Joseph’s Health Institutional Review Board (IRB #2021000329). As this was a retrospective study involving well over 10,000 people, we obtained a HIPAA waiver of consent.

Consent for publication

Not applicable.

Competing interests

The authors have no competing interests to disclose.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Govier, D.J., Cohen-Cline, H., Marsi, K. et al. Differences in access to virtual and in-person primary care by race/ethnicity and community social vulnerability among adults diagnosed with COVID-19 in a large, multi-state health system. BMC Health Serv Res 22, 511 (2022). https://doi.org/10.1186/s12913-022-07858-x

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12913-022-07858-x

Keywords

  • Telehealth
  • Primary care
  • Disparities
  • Race
  • Ethnicity
  • Social vulnerability
  • Outpatient care