Sunday, November 24, 2024
Virtual Heart Care
Telehealth 

Patient and Visit Characteristics of Families Accessing Pediatric Urgent Care Telemedicine During the COVID-19 Pandemic

Editor


Introduction

In 2005, Michael Marmot first introduced the concept of social determinants as the root cause of inequalities in health.1 Since then, health experts have been researching how social determinants of health impact long-term health outcomes. In January 2020, the United States confirmed its first case of coronavirus disease 2019 (COVID-19), the disease caused by the novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus.2 With the emergence of the COVID-19 pandemic and the call for social isolation, telemedicine encounters sharply increased,3 highlighting a new social determinant of health: internet and broadband access.4,5 Studies indicate that access to digital health and telemedicine is dependent on social class and poverty levels.6,7 Indeed, although telemedicine expansion during the coronavirus pandemic improved health care access for some, research shows disparate uptake among those in marginalized communities.8–13 Health disparities during the pandemic, presently unearthed by disproportionate infection rates found among African American and Hispanic communities,14–17 are not limited to adults. A recent study found that of 1,000 children tested for the SARS-CoV-2 virus, infection rates were highest among minority and socioeconomically disadvantaged youth.18 Healthy People 2030, a decades-long wellness initiative supporting disease prevention and health promotion, strives to achieve health equity and eliminate disparities to improve health and well-being throughout the nation.19 With this goal in mind, it is imperative to understand the current landscape of telemedicine access and use as it relates to social determinants of health and the characteristics of families accessing virtual care to support more equitable outcomes, during and beyond the pandemic.

Nemours Children’s Health (NCH) is a pediatric health system of hospitals and clinics in Florida and the Delaware Valley (DV). The Nemours CareConnect (NCC) urgent care telemedicine platform offers synchronous telemedicine 24 h per day, 7 days per week, to families with children from birth to 18 years of age. NCC’s board-certified pediatricians are licensed to provide health care to children residing in multiple states across the United States. Further details of NCC operations have been shared in prior publications,20 including its use in disaster response.21

The purpose of this retrospective, cross-sectional analysis was to evaluate demographic changes among pediatric, urgent care telemedicine users during the coronavirus pandemic. Previously, we described chief complaints of patients and clinician redeployment during this timeframe.22 In this study, we examine data across two unique health care markets, comparing patient–family demographics from January to May 2020 to the same timeframe in 2019.

Materials and Methods

This study was approved by the Nemours’ institutional review board.

PEDIATRICIANS

Six board-certified pediatricians were staffing the NCC platform when the U.S. exposure to the crisis began. Amidst the call for social isolation, primary care and urgent care offices within NCH began to consolidate services. A pool of 28 additional providers from hospital medicine, emergency care, urgent care, and primary care assisted the urgent care telemedicine team. In-person visits were rescheduled to virtual on-demand visits where appropriate.22

PLATFORM

NCC complies with current organizational and technological requirements. Specifically, NCC equipment is H.323 compliant with live video resolution of 4 X Common Intermediate Format (CIF), and has bidirectional connection at a minimum 384 kb per second running 4CIF at 30 frames per second.23 To ensure secure data transmission and to protect patient information, video, audio, and electronic medical record components are fully encrypted.

PROCEDURE

Access to the service was consistent with standard NCC operation.20 A parent initiated a telemedicine visit by downloading the NCC application on their smart device, selecting a provider, and entering a virtual waiting room. Once notified, the provider connected with the family and assessed appropriateness for telemedicine. Patients needing in-person medical evaluation were referred to the appropriate level of care. Complaints that met telemedicine standards were completed with patient guidance and parental assistance through history and virtual physical examination, without use of peripheral devices. Final assessment and treatment plan were discussed with the family to conclude the visit.

The provider documented the encounter in the electronic medical database (Epic Systems Corporation, Verona, WI), and completed any necessary electronic orders after disconnecting from the platform. Parents accessed the AVS from the NCC application at their convenience. A copy of the encounter was sent to the patient’s primary care provider on file.

ANALYTIC PLAN

For this study, patient data were obtained from the electronic medical database across Nemours’ two primary markets of DV (Delaware, New Jersey, and Pennsylvania) and Florida. Variables included patient draw, age, gender, race, ethnicity, language, median household income, ZIP code, and insurance type, extracted to examine telemedicine use by region during the early months of the coronavirus pandemic from January 2020 to May 2020. Comparison data for these variables was extracted for the same range of months in 2019 to illustrate changes in patient demographics evidenced during the pandemic. Data are reported for unique patients for demographic data (patient age, gender, race, ethnicity, and language), and by visit for all other variables. Emphasis was directed toward patient characteristics that correlate with the social determinants of health (e.g., race, ethnicity, language, income, and geography), in an effort to determine demographics for whom telehealth usage was high (or low) during the increased need for health care during the pandemic.

Results

PATIENT VOLUME BY YEAR: 2019, 2020

During January to May 2019, Nemours pediatricians completed N = 1,812 urgent care telemedicine visits. Within Nemours’ 2 primary regions, Florida patients received 944 visits, and DV patients received 868 visits. In contrast with the same timeframe the year prior, patient visits jumped to N = 4,931 in 2020, for an overall year-over-year increase of 172% in urgent care telemedicine visits. Florida patients received 2,442 of those visits, and DV patients received 2,489 visits.

FLORIDA

Overall patient volume increased by 159% during the study timeframe from 2019 to 2020. Of 832 unique patients in 2019, the mean patient age was 5.33 years (standard deviation [SD] = 4.58, 0–18), which was not significantly different from that of 2020 (n = 2,091; 5.21 years, SD = 4.6, 0–18, p > 0.05). Men represented 52.4% of unique Florida patients during 2019 and 49.9% in 2020 (p > 0.05). A majority of 2019 patients were White/Caucasian (52.9%), which declined to 46.7% in 2020. Similarly, the percentage of Black/African American (6.5%) patients in 2019 declined to 5.8% in 2020. The category “Other” comprised less well-represented racial categories including Native American (n = 2 in 2019; n = 6 in 2020), Asian Indian (n = 5 in 2019; n = 13 in 2020), and patient-families who identified as “Some Other Race.” In 2019, the “Other” category included 24.2% of patients; in 2020, it included 22.4%. Patient-families who selected multiple race categories were classified as “Mixed Race” (1.4% of patients in 2019; 1.0% in 2020). Finally, 23.5% of patient-families in 2020 did not respond to the request to identify their race, an increase of 15% from the prior year. Whereas a change in the racial composition of the Florida patients pre- to postpandemic cannot be determined definitively from our data because of the large number of missing values, utilizing all available race data (i.e., excluding nonresponders), a chi square test indicated no significant change in race across years (p > 0.05). Figure 1 displays the racial composition of the Florida patients across study years.

Fig. 1.

Fig. 1. Patient change in racial makeup by study year, Florida.

Year over year, the percentage of patients who identified as Hispanic increased from 33.5% to 34.1%. Consistent with the request for patient race information, the percentage of patient-families who chose not to respond to the request for ethnicity increased from 13.7% to 22.2% across years. Among the respondents, the proportion of patients identifying as Hispanic or Latinx was significantly higher than expected in 2020, relative to 2019 (χ2 [1, n = 2,344] = 4.961, p < 0.05). Florida patient ethnicity by year is given in Table 1.

Table 1. Florida: Changes in Patient Ethnicity Across Study Years
ETHNICITY 2019   2020  
  n % n %
Non-Hispanic or Latinx 439 52.8 914 43.7
Hispanic or Latinx 279 33.5 712 34.1
Not available 114 13.7 465 22.2
Total 832 100 2,091 100

English and Spanish represented the two most dominant languages spoken by patients in Florida. Over 85% of patients spoke English as their primary language in both years, whereas the percentage of patients who spoke Spanish as a primary language increased by 1.2% from 2019 to 2020. In 2020, linguistic diversity expanded to include patients who spoke Mandarin (n = 2), Nepali (n = 1), Portuguese (n = 2), and Vietnamese (n = 1). Whereas the number of languages spoken by Florida telemedicine patients increased, the relative proportion of languages spoken did not change significantly across years (p > 0.05). Table 2 presents Florida patient primary languages spoken by study year.

Table 2. Florida: Primary Languages Spoken by Telemedicine Patients Across Study Year
  2019   2020  
LANGUAGE n % n %
English 720 86.5 1784 85.3
Mandarin —   2 0.1
Nepali —   1 0.0
Portuguese —   2 0.1
Spanish 29 3.5 98 4.7
Sign Spanish 1 0.1 —  
Vietnamese —   1 0.0
Not available 82 9.8 203 9.7
Total 832 100 2,091 100

Patient-family income ranged from $24,458 to $116,697, and the median household income (MHHI) of $56,502 did not change across years (p > 0.05). However, telemedicine visits in which patients used government-supported health insurance (including Medicaid, Florida Healthy Kids, or Children’s Health Insurance Program [CHIP]) increased significantly from 33.2% in 2019 to 41.2% in 2020 (χ2 [1, n = 3371] = 18.24, p < 0.01). To examine geographic differences in patient location across years, ZIP codes were defined as rural based on their classification as such in the Federal Office of Rural Health Policy (FORHP); all others were classified as urban. Based on this criterion, 2.0% of Florida patients lived in rural ZIP codes in 2019 and in 2020 (p > 0.05).

DELAWARE VALLEY

During the study months, DV patient volume increased by 187% from 2019 to 2020. Several changes in patient demographic characteristics were noted. Mean patient age in 2019 (n = 836 unique patients) was 4.19 years (SD = 4.36, 0–18). In 2020 (n = 2,378), the mean patient age was 4.79 years (SD = 4.66, 0–20), which was significantly older than patient age across the same timeframe in 2019 (F [1, 3,212] = 10.587, p < 0.01). Men represented 53.5% of DV patients in 2019, similar to 2020 (51.2%; p > 0.05). Patients represented diverse racial backgrounds. The percentage of African American patients who utilized urgent care telemedicine increased from 25% in 2019 to 29% in 2020, whereas a decrease in White/Caucasian patients (59% to 56%) was observed during the same period. The category of “Other” race demonstrated a small decline from 2019 to 2020 (7.9% vs. 6.7%). This category includes patients of Native American descent (n = 1 in 2019; n = 3 in 2020), in addition to Pacific Islander (n = 0 in 2019; n = 4 in 2020). Changes in the racial composition of patients utilizing telemedicine in DV across years approached statistical significance (χ2 [4, n = 3191] = 8.039, p = 0.09). DV patient racial demographics across years are given in Figure 2.

Fig. 2.

Fig. 2. Patient change in racial makeup by study year, Delaware Valley. The category “Asian” is aggregated to include Asian Indian patients, as well as patients with racial identities from the Far East.

The percentage of patients identifying as Hispanic remained consistent across years (p > 0.05). DV patient ethnicity across study years is given in Table 3. Although nearly all (≥98%) patients spoke English as their primary language, linguistic diversity increased in 2020 relative to 2019. In 2019, 4 primary languages were spoken by DV patients. In 2020, 10 unique languages were identified, including those spoken in regions of the Middle East, East Asia, India, Europe, and the Caribbean. Whereas six additional languages were spoken among patients in 2020, this difference across years was not significant (p > 0.05). Table 4 provides DV patient primary language by year.

Table 3. Delaware Valley: Changes in Patient Ethnicity Across Study Years
ETHNICITY 2019   2020  
  n % n %
Non-Hispanic or Latinx 748 89.5 2,116 89.0
Hispanic or Latinx 84 10.0 247 10.4
Not available 4 0.5 15 0.6
Total 836 100 2,378 100

Table 4. Delaware Valley: Primary Languages Spoken by Telemedicine Patients Across Study Year
  2019   2020  
LANGUAGE n % n %
Arabic 1 0.1 1 0.0
Chinese 1 0.1 —  
English 825 98.7 2,335 98.2
Haitian —   1 0.0
Hindi —   1 0.0
Portuguese —   2 0.1
Sign language —   1 0.0
Spanish 4 0.5 21 0.9
Tamil —   3 0.1
Turkish —   2 0.1
Vietnamese —   1 0.0
Not available 5 0.6 10 0.4
Total 836 100 2,378 100

Year-over-year, patient income levels increased by $139 (MHHI: $65,873 for 2019; $66,012 for 2020) but did not differ significantly (p > 0.05). The percentage of telemedicine visits having patients who used government-supported health insurance (including Medicaid or CHIP insurance) increased marginally from 36.9% of visits in 2019 to 39.1% in 2020 (p > 0.05). Geographically, the percentage of patients seeking telehealth visits across urban and rural neighborhoods changed slightly across years, with the percentage of patients in rural areas seeking telemedicine visits declining marginally (6.7% to 5%). This change approached, but did not achieve, statistical significance (p = 0.07).

Discussion

During the early days of the COVID-19 pandemic, pediatric, urgent care telemedicine encounters in our sample increased by 172% from 2019 baseline. This surge in patient volume highlighted differences in patient demographic characteristics among telemedicine users during the pandemic era. Unique changes were observed across two U.S. regions during the study’s timeframe. In Florida, significant increases in use of subsidized health insurance for telemedicine and changes in patient ethnicity were observed. In DV, a slight decline in rural patients utilizing telemedicine was observed, approaching statistical significance. In Florida and DV, increased linguistic diversity was observed, underscoring a pragmatic need for language services that optimize patient access, care, and quality.

Telemedicine utilization increased by roughly 8% points among Medicaid users in Florida during the pandemic. The Centers for Disease Control and Prevention noted that increased telemedicine visits during COVID-19 may be related to policy changes and regulatory waivers from the Centers for Medicare and Medicaid Services.3 Increased utilization of telehealth by the Medicaid population in Florida coincided with the 2020 enactment of the Families First Coronavirus Act (FFCRA).24 Under FFCRA, states received a temporary increase in the federal matching rate for the emergency period if they provided continuous eligibility for enrollees in Medicaid/CHIP through the end of the month of the emergency period.25 Performance indicators demonstrated increased Medicaid enrollment after the FFRCA enactment.26 This rising enrollment may have contributed to increased Medicaid utilization on the NCC platform in Florida, highlighting the influence of governmental support and policy change to enhance telemedicine access during a crisis.

Unique patterns in ethnic/racial patient demographics were observed across regions, including a significant year-over-year increase in the proportion of Hispanic patients accessing virtual care in Florida, and a small, although nonsignificant increase in the proportion of African American patients accessing telemedicine in DV. These differences in patient demographics across states and time align with a recent AAP report27 in which the incidence of COVID-19 was evaluated in pediatric patients residing in California and Illinois over an 11-month period. Highlighting the increased incidence of COVID-19 among Hispanic and African American pediatric patients, the authors commented how the surge affected each demographic differently as it shifted with time and by state, underscoring the importance of examining state-level differences over time when evaluating disparities. Of note, the percentage of caregivers who chose not to disclose the patient’s ethnicity/race (California: 19%; Illinois: 26%)27 also was similar to that of our Florida sample (22% ethnicity; 24% race).

Families in rural communities accessed our platform at low rates, 2% in Florida and 5% in DV. These data are not surprising in light of the growing literature suggesting that marginalized communities are disproportionately impacted by COVID-19 and that rural families remain isolated. It has long been known that the lack of broadband access challenges adoption of virtual care for those residing in rural populations.28 Rural families report less access to pediatric care, higher rates of chronic conditions, and overall poorer health status.29 It is unclear whether more patients in rural communities sought telemedicine visits during our study timeframe. We examined completed visits only, necessitating patients having broadband access. This limitation may have obscured any potential changes in virtual care among patients in rural communities owing to technology constraints, thus reflecting a selection bias in the sample. In addition, more research is necessary to evaluate pandemic era, regional declines in telemedicine use among rural patients—as evidenced by the trend among rural DV patients in this study—and determine whether they reflect unintended impacts of imposed lockdowns for those who rely on broadband access outside the home.

In DV, and to a lesser extent in Florida, our data showed increased language diversity among telemedicine patients during the pandemic. Although not statistically significant, we argue its practical significance from an “access to care” standpoint as it highlights a potential need for health care systems to embrace enhanced language translation services and culturally competent patient care instructions, as appropriate, to meet postpandemic patient demographics. Indeed, a recent University of California, Los Angeles (UCLA) report on the Hispanic perspective during COVID-19 recommended that culture and population demographics should be considered when designing online platforms.30 The UCLA taskforce also recommended expediting training and certification for health care interpreters during public health emergencies.30 Future research is needed on the implementation and adoption of seamless translational options for families when connecting for a virtual visit.

Whereas our pediatric data are not limited to COVID-19 cases, they illustrate variations in patient use of telemedicine among vulnerable populations during the COVID-19 pandemic. Although it is presently unclear to what extent the observed diversity will be sustained postpandemic, given its high patient satisfaction,20,22 it is reasonable to assume that some percentage of the patients who utilized telemedicine during this time will utilize it again in the future. Consequently, as the populace of telemedicine patients broadens, it is incumbent upon health care systems to facilitate operational avenues that support equity of access and quality of care in the midst of growing patient diversity. Likewise, policy makers must support continued reimbursement of telemedicine for patients holding subsidized health insurance following the pandemic. Finally, continued efforts are greatly needed to bring broadband technology to isolated regions.

The current analysis is limited to one pediatric platform during the pandemic. Additional research is needed to further understand the inequities in access through telemedicine on a larger scale. Our correlational study results should be interpreted in the context of several considerations. The pandemic affected each state differently. Cases did not surge at the same time in each state, and local authorities uniquely responded to the threat. Social distancing guidelines and timing of business closures varied. Consequently, telemedicine volume increases may have differed systematically based on state and locality. Patient draw may be influenced by the location of our clinics and hospitals. Patients may use our platform because they either live near one of our locations or have been to one of our hospitals.

Conclusion

The ability of health care systems to accommodate and welcome a more diverse patient populace—one that is open to telemedicine as a vehicle for patient care—is at an inflection point. Pandemic influences on changes in patient demographics and insurance coverage illuminate the need for ongoing, targeted mitigation strategies to ensure that telemedicine will meet the needs of a more diverse population of users in the months and years following the pandemic. Such strategies include ongoing federal support for Medicaid reimbursement for telehealth services, broadband expansion, and language and translation-based services, among others. Future research should examine the extent to which the pattern of increased patient diversity observed during the pandemic remains in the years to come. Such knowledge will best position health care systems and providers to employ appropriate technologies and enable operations to ensure that a diverse range of patient needs for virtual health care access can be met, equitably.

Acknowledgments

The authors acknowledge the contributions of telemedicine providers Andrea Ali-Panzarella, MD, Monica Barajas, MD, Rene Chalom, MD, Mattilie Gednas, ARNP, Wayne Ho, MD, Robert Karch, MD, and Rachel Schare, MD, as well as the staff from the Nemours Children’s Health System technology team and the operational support staff who provided assistance and guidance during the coronavirus pandemic. The authors especially thank those providers whose unwavering dedication to patient care led them to become new adopters of telemedicine during this extraordinary time of need. The authors also acknowledge Nemours analysts Sean McElroy and Stacey Milunsky for their expertise and analytic insights into Nemours’ patient geolocation data. Their valuable contributions in service of this work are much appreciated.

Authors’ Contributions

P.S.-J. conceptualized the study, collected and interpreted the data, and drafted the article. J.M.-B. conceptualized the study, collected and interpreted the data, and critically revised the article. C.M.Z.-G. conceptualized the study, analyzed and interpreted the data, and critically revised the article. All authors agreed to the final version of the article and are accountable for this work.

Disclosure Statement

All authors are employed by Nemours Children’s Health System and they report no other competing interests to disclose.

Funding Information

This study was funded by The Nemours Foundation.

REFERENCES





Source link