key: cord-1011833-hnd7z97b
authors: Harsono, Dini; Deng, Yanhong; Chung, Sangyun; Barakat, Lydia A.; Friedland, Gerald; Meyer, Jaimie P.; Porter, Elizabeth; Villanueva, Merceditas; Wolf, Michael S.; Yager, Jessica E.; Edelman, E. Jennifer
title: Experiences with Telemedicine for HIV Care During the COVID-19 Pandemic: A Mixed-Methods Study
date: 2022-01-22
journal: AIDS Behav
DOI: 10.1007/s10461-021-03556-7
sha: fc95bfe192bd503911d774e299097e50c6479243
doc_id: 1011833
cord_uid: hnd7z97b

To characterize perspectives and experiences with telemedicine during the COVID-19 pandemic, we conducted a mixed-methods study in two HIV clinics in the US Northeast. Among surveyed patients with HIV (PWH) who had a telemedicine appointment (n = 205), 42.4% perceived telemedicine visits as useful during the pandemic. PWH and clinical staff identified benefits of telemedicine: (1) ability to engage and re-engage patients in care; (2) perceived patient-centeredness and flexibility; (3) opportunity to engage family and multidisciplinary care team members; and (4) opportunity to enhance telemedicine use proficiency through practice and support. Identified barriers included: (1) technical challenges; (2) privacy concerns; (3) loss of routine clinical experiences and interactions; (4) limited objective patient remote monitoring; and (5) reimbursement concerns. Efforts to optimize telemedicine for HIV care should consider strategies to improve technology support for PWH, flexible options to access care, additional platforms to allow patient remote monitoring, and appropriate billing and reimbursement methods. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s10461-021-03556-7.

SARS-CoV-2 virus and the COVID-19 disease it causes has created an unprecedented global health emergency [1] , resulting in a major transformation of the healthcare system. To ensure HIV care continuity during the pandemic, many clinics have rapidly deployed or extensively expanded telehealth. Defined as "the use of electronic information and telecommunication technologies to support and promote long-distance clinical health care, patient and professional health-related education, public health, and health administration," [2] telehealth offers a variety of modalities (e.g., video or telephone-based communications, patient portal) and functions (e.g., medication adherence monitoring) [3, 4] .

Telemedicine, a subset of telehealth focused on providing clinical services [2] , has been applied in the context of HIV care to increase access to quality, timely healthcare; reduce patient travel burden; alleviate stigma; and improve cost-effectiveness [3, [5] [6] [7] , with high patient satisfaction rates [8, 9] , acceptability [10] , improved medication adherence [11] , and HIV viral suppression [12, 13] . Literature on the usage of telemedicine in HIV care during the COVID-19 pandemic is still evolving. Data captured early in the pandemic indicated positive reception among patients with HIV (PWH) toward telemedicine [14] [15] [16] , with several challenges including patients' lack of access to technology and private space, the need for technical training among staff, and limited physical examination [15, 17, 18] . These studies are limited by their small numbers of patient participants and a single-site approach [10, [14] [15] [16] 18] . A recent policy statement by the Infectious Diseases Society of America and the HIV Medicine Association underscores the importance of developing interventions that promote digital health equity [19] ; however, further data are lacking to support providers and programs' understanding on how, exactly, to do so.

Thus, the goal of this exploratory study was to generate timely insights regarding experiences with telemedicine for HIV care from perspectives of patients and clinical staff at two urban settings where telemedicine was rapidly scaled up due to COVID-19. These findings may serve to directly inform efforts to optimize telemedicine in the context of HIV care.

Using a concurrent triangulation mixed-methods design [20] , we conducted a survey and focus groups involving patients and clinical staff at two large urban HIV clinics in the US Northeast. We collected data from May 15, 2020 through August 11, 2020, shortly after the initial peak of the COVID-19 pandemic in these regions. Informed by the RE-AIM implementation science framework [21, 22] , we broadly sought to understand how telemedicine was able to reach patients and its perceived effectiveness for delivering routine care to PWH. Patients were recruited to complete a one-time telephone-based survey to assess experiences with the pandemic and its impact on their health and access to care. In parallel, we invited clinical staff to participate in a virtual focus group and in a brief web-based survey to collect data on demographics and clinical roles. The study protocol was approved by institutional review boards at Yale University and SUNY Downstate Health Sciences University and was HIPAA compliant.

This study was conducted in New Haven, Connecticut and Brooklyn, New York in the context of existing research collaborations [23, 24] focused on reducing tobacco use among PWH [23] . We therefore, actively recruited PWH with current tobacco use by electronic medical record (EMR) documentation. In addition, we recruited a subset of patients without current tobacco use from the New Haven-based clinic. Additional inclusion criteria were: (1) ≥ 18 years of age, (2) English-speaking, and (3) able to give informed consent. We sought to equally sample younger and older patients given our expectations that the COVID-19 pandemic may have differential impacts by age due to varying levels of social isolation, comorbidities, and technology access [25] [26] [27] [28] . Clinical staff from these clinics were invited to participate to gain insights from a diversity of providers. All participants provided verbal informed consent for participation in research activities and were provided a $30 gift card for study participation.

During the study period, patients at the New Havenbased clinic who were due for a visit either made a telemedicine appointment through the Epic-based MyChart patient portal or were contacted by clinic staff to schedule a telemedicine visit. Clinicians at this site had the option to conduct telemedicine visits via a video conference link within MyChart or by directly calling the patient's phone number for an audio-only visit. At the Brooklyn-based site, telemedicine visits were similarly conducted in this flexible manner. Patients were contacted directly by clinic staff or could contact the clinic themselves to schedule a telemedicine appointment, which occurred either as a video visit through the DoxyMe platform or by phone for an audioonly visit.

We collected data from the two groups of patient and provider stakeholders in parallel; data collection took place between May 15, 2020 and August 11, 2020.

The "COVID-19 and PWH Survey" (see Supplementary Survey) was administered by telephone by research staff. Data were collected and managed using Research Electronic Data Capture (REDCap) system hosted at Yale University [29, 30] . The survey was informed by the work of Wolf and colleagues [31] and was modified based on multidisciplinary input from HIV clinicians, researchers, and a community partner. Validated measures were incorporated as applicable [32] [33] [34] . We collected data on several domains, including sociodemographic characteristics, COVID-19 related experiences and impacts, and experiences with and preferences regarding telemedicine. Participants were asked about recent completion of telemedicine visits based on experiences in the 2 months prior to survey date.

To assess patient perceptions of the utility of telemedicine, we examined responses to the following items: 1. "How useful have your medical appointments with telemedicine/telehealth been during the COVID-19 emergency?" and 2. "How useful do you think it will be to have medical appointments with telemedicine/telehealth after the COVID-19 emergency is over?" Participants were asked to rate responses using a 5-point Likert scale; responses were dichotomized as "useful" ("extremely" or "very") vs. "not useful" ("moderately," "slightly," or "not at all"). Additionally, in post-hoc sensitivity analyses, we expanded the definition of "useful" to include "moderately" ("extremely," "very," or "moderately") vs. "not useful" ("slightly" or "not at all"). Participants were then prompted to provide free-text comments that research staff documented verbatim.

Independent Variables Participant characteristics included demographics (age, race, ethnicity, gender identity); use of and adherence to HIV medications; depression [33] and anxiety [34] symptoms; and telemedicine experiences.

Most recent HIV biomarkers (CD4 cell count, presence of a detectable HIV viral load defined as > 50 copies/mL), and engagement in HIV care based on documentation of at least one HIV clinic visit in the 6 months prior to survey date consistent with current standards [35] were extracted from the EMR.

Using the cloud-based Zoom© videoconferencing platform, members of the investigative team conducted two focus groups with each site (n = 4 total). Participants were invited by local study team members via emails. Grand tour (i.e., broad questions) and follow-up probes were used to assess experiences using telemedicine during the COVID-19 pandemic to provide care to PWH (see Supplementary Focus Group Guide). Focus groups were digitally recorded and transcribed and then reviewed by a member of the research team for accuracy and de-identification. Upon focus group completion, participants were asked to complete a brief REDCap-based survey regarding their demographics and clinic role.

Descriptive statistics were used to characterize participant characteristics and patient telemedicine experiences during the pandemic. Univariable and multivariable logistic regression models were built to examine associations between patient characteristics and patient ratings on how useful telemedicine visits were (among those who had completed a telemedicine visit during the pandemic only) and perceptions of all patient participants on how useful it will be to have telemedicine after the pandemic is over. Variables of interest were selected a priori to include in the multivariable models, namely age, smartphone ownership, and site. Additionally, variables that were significant at p < 0.05 in unadjusted analyses were also included in the multivariable models. Statistical analyses were performed using SAS version 9.4 software (Copyright© 2013, SAS Institute Inc., Cary, NC, USA). Open-ended responses to these items were analyzed using content analysis [36] (described further below).

We used a rapid assessment process [37] followed by an inductive process of iterative coding to identify themes using content analysis [36] . Members of the investigative team independently reviewed each transcript line-by-line to develop and refine the codebook and reach consensus on codes. Themes were then generated based on coded quotations and discussion by two investigators initially and then with input from the broader research team.

Analyses of survey data were done in parallel with qualitative focus group data. Themes developed based on focus groups were then merged with responses from the openended patient survey items.

Among a random sample of 755 patients who met the eligibility criteria, we attempted to contact 719 patients, among whom 40.2% (n = 289) were not successfully contacted and 20.4% (n = 147) declined participation. The most common reasons for declining were refusal (49%) and not enough time/too busy (25.2%). A total of 283 patients consented to participate. Missing responses (n = 10) were observed during data cleaning and therefore excluded from the analysis. The sociodemographic characteristics of the patients who completed all sections in the telephone survey (n = 273) are described in Table 1 .

Among patient participants who owned a smartphone (74.4%, n = 186), the majority reported that they were able to use their smartphone for videoconferencing (96.2%) and accessing the internet (97.3%). Most patient participants had a telemedicine appointment (75.1%, n = 205) in the 2 months prior to the survey ( Table 2) . Among these patients, most had a telemedicine phone visit (85.9%) with fewer reporting telemedicine with video (7.3%) or both telephone and video (6.8%). The majority of telemedicine visits were with the participant's HIV provider (i.e., MD, PA, APRN) (66.7%, n = 182). Other telemedicine visits were with a specialist/ consultant (14.7%), behavioral health provider (i.e., social worker, counselor) (10.3%), nurse (4%), pharmacist (3.7%), and nutritionist (1.5%). With regard to reasons for telemedicine use, most patients (93.7%, n = 192) reported having a telemedicine visit in place of a routine face-to-face appointment. Others indicated their use of telemedicine to address other symptoms/urgent issue (5.4%) and possible COVID-19 symptoms (1%).

Overall, telemedicine visits were viewed favorably: 74.3% (n = 153) felt that their telemedicine visit was better than/ just as good as a traditional (face-to-face in person) visit ( Fig. 1) , and 67.5% (n = 139) reported that they probably/ definitely would recommend telemedicine to someone else (Fig. 2) . Approximately two-thirds of patients felt that it was important to be able to have routine medical visits (63.4%, n = 173) or urgent medical appointments (68.1%, n = 186) via telemedicine during the COVID-19 pandemic (Table 2) .

Among patients who engaged in telemedicine visits (n = 205), 42.4% of patients felt that telemedicine visits were useful during the pandemic (Table 3 ). In bivariate analyses, patients who had not recently been engaged in care were more likely to report that telemedicine was useful during the pandemic than patients who had recently been engaged in care (OR 2.64 [CI 1.05 to 6.60]; p = 0.04). In addition, patients who did not own a smartphone were less likely to report that telemedicine was useful during the pandemic than patients with a smartphone (OR 0.27 [CI 0.13 to 0.59]; p < 0.001). After adjusting for age, site, smartphone ownership, and recent engagement in care, not having a smartphone was associated with lower odds of finding telemedicine useful during the pandemic (OR 0.29 [CI 0.13 to 0.67]; p < 0.01). There was no significant difference in perception of telemedicine usefulness during the pandemic by age, site, or recent engagement in care (all p values > 0.05).

Among all surveyed patients (n = 273), 27.5% indicated that telemedicine appointments would be useful after the COVID-19 pandemic was over (Table 4 ). In bivariate analyses, a 10-year increase in patient age was associated with higher odds of finding telemedicine useful after the pandemic (OR 1.31 [CI 1.02 to 1.69]; p = 0.03). Additionally, patients who were in Brooklyn, NY were more likely to endorse that telemedicine would be useful after the pandemic than patients who were in New Haven, CT (OR 1.96 [CI 1.14 to 3.38]; p = 0.02). In multivariable analyses, a 10-year increase in age remained to be significantly associated with higher odds of finding telemedicine useful after the pandemic (OR 1.38 [CI 1.05 to 1.81]; p = 0.03). Owning a smartphone and site were not associated with the odds of finding telemedicine useful after the pandemic (all p values > 0.05). In the sensitivity analyses (see Supplementary Table VI and Table VII Clinical staff discussed how they perceived telemedicine as helpful to connect with patients and mitigate patients' feelings of isolation and fears during the pandemic, and reduce no-show rates. Importantly, telemedicine during the pandemic offered an opportunity to re-engage individuals who were out of care as well as discuss new opportunities to maintain health that may have been previously de-prioritized. Telemedicine visits also allowed clinical staff to teach patients new skills, particularly self-administration of injectable medications, to avoid treatment interruptions. Patients similarly described the benefits of telemedicine including to keep their appointments with providers and address pandemic-related concerns. *Numbers do not add up to 100% due to small amounts of missing data † Two participants who indicated "other" provided additional information that they were students ‡ ≤ 50 copies/L § Two participants indicated "sometimes" and 269 indicated "yes"

The ease of scheduling and flexibility were some of the main advantages of telemedicine expressed by clinical staff and patients. Clinical staff expressed the importance of meeting patients where they are through telemedicine, and patients underscored that telemedicine reduces burden of travel with minimal disruption of their daily routines. Clinical staff and patients described concerns about decreased Participants were asked to check all that apply † Ten participants who indicated "other" had telehealth visits with primary care providers (n = 3), care navigators (n = 2), care coordinator (n = 1), pain medicine (n = 1), home attendant (n = 1), physical therapist (n = 1), and dentist (n = 1) 

Clinical staff and patients reported technology-related challenges as common barriers to telemedicine participation, including lack of access to the required technologies such as smartphones, and skills and confidence in using the telemedicine platforms. Among clinical staff, not having a dedicated work phone for telemedicine and inconsistent use of the videoconferencing platforms were described as challenges during the initial phase of telemedicine implementation. Providers also acknowledged that patients had varying levels of technology literacy and access. Connection issues during telemedicine visits caused discomfort and was a hindrance to telemedicine participation for some patients. To overcome the challenges, providers spoke about regular practice to develop proficiency and patient outreach to troubleshoot. These solutions were perceived to improve the quality of the telemedicine experience.

Clinical staff reported challenges with ensuring privacy and confidentiality related to telemedicine describing their experience of conducting telemedicine visits from home and navigating disruptions in their living spaces. Some patients reflected on the difficulty of speaking about personal information during telemedicine visits. On the other hand, providers noted that they were able to more easily engage patients' family members and other members of the care team to deliver patient-focused care.

Overall, clinical staff and patients expressed support for telemedicine use for HIV care after the COVID-19 pandemic is over. Providers believed that telemedicine has made physical and behavioral health care more accessible to patients. Provision of these services should continue with appropriate reimbursement and staffing support to assist patients in accessing the technology effectively. Patients and providers emphasized the importance of assessing patient preferences and needs for telemedicine or in-person visits to improve acceptability and care continuity. Patients expressed their appreciation to telemedicine use for routine visits with their providers, particularly for individuals who are older, with mobility issues, or unable to travel. However, patients would also like to have the option of in-person appointments, for example when physical exams were needed. Additionally, providers recommended incorporating innovative mobile technology that can enhance the quality of remote physical exam and monitoring.

To our knowledge, this is the first study to explore multiple stakeholders' experiences and perspectives of telemedicine use in HIV care during the COVID-19 pandemic.

In our sample, three-quarters of PWH in our study had participated in telemedicine visits during the pandemic, mostly conducted by telephone. Although the majority of those who participated in telemedicine visits felt that their telemedicine visit was better than/just as good as an in-person visit, less than half perceived telemedicine visits as useful during the pandemic and a smaller minority perceived that telemedicine would be useful after the pandemic. Factors associated with patient preference for telemedicine during the pandemic were lack of recent engagement in HIV care, smartphone availability, and site. Increase in patient age and location in Brooklyn, NY appears to be associated with more favorable perspectives regarding telemedicine usefulness after the pandemic. Our study identified four main benefits of telemedicine, namely (1) ability to engage and re-engage patients in care and prioritize new and existing health concerns; (2) perceived patient-centeredness and flexibility; (3) opportunity to engage family and multidisciplinary patient care team members; and (4) opportunity to enhance telemedicine use proficiency through practice and support. Barriers to telemedicine use include (1) technical challenges; (2) privacy concerns; (3) loss of routine clinical experience and in-person interactions; (4) limited objective patient remote monitoring (e.g., blood pressure, medication adherence); and (5) concerns regarding reimbursement. We found that the majority of providers strongly endorsed maintaining telemedicine services beyond the pandemic. PWH also expressed support for telemedicine use after the pandemic, and underscored the importance of respecting patient preferences for telemedicine or in-person visits based on specific consultation purposes, for example scheduling in-person visits when physical exams are needed. In particular, we found that patients who had not been engaged in recent HIV care were more likely to perceive telemedicine as useful during the COVID-19 pandemic, and that patients in Brooklyn, NY were more likely to support that telemedicine was useful during and after the pandemic. While these findings may be driven by challenges with in-persons visits due to job inflexibility, distance to clinic, transportation barriers including reliance on public transportation, and stigma, they merit further investigation and suggest that telemedicine may offer an important alternative approach to in-person HIV care.

Our study additionally highlights that despite the relatively high number of smartphone ownership among PWH, video-enabled telemedicine visits occurred much less frequently than telephone-based visits. Video visits allows the exchange of visual information and non-verbal cues that are important in facilitating better provider-patient rapport and some aspects of physical examination [38] . However, such visits require high speed internet services and more complex setup that may present barriers to our sample of PWH who were older individuals and of lower socioeconomic status.

Our findings mirror other studies that examine telemedicine use for HIV care during the pandemic. One study in an HIV ambulatory clinic in Baltimore reported 70% of telemedicine visits were conducted by telephone [39] , while another study in a Midwestern clinic opted to implement telephone-only telemedicine visits to alleviate technology access barriers among PWH receiving care during the pandemic [40] . In the rapidly evolving literature, researchers also examined sociodemographic factors affecting PWH participation in telemedicine visits during the pandemic, with varying results. Some studies reported older age (50-65 or > 65 years old), non-White, Medicaid insurance PWH were less likely to engage in a video visit [41] , while others found no differences in age, gender, race, or ethnicity between PWH who completed in-person visits compared to those received telephone visits [40] . Differences in study designs, patient access to technology and digital health readiness, and telemedicine implementation policies, practices, protocols make it difficult to compare and generalize from these studies [42] .

Future research to improve PWH engagement in telemedicine should consider equity-focused strategies to address the digital divide by assessing patients' technical readiness and providing programs to improve technology literacy and access. Some examples include considering the use of online patient portal as a determinant of patients' readiness and capacity to use telehealth [41] , conducting consistent assessment of patient characteristics (e.g., age, race and ethnicity, language, income, and educational level) to inform tailored interventions to promote equitable digital health access [19] , and adapting a multilevel social-ecological framework in research and practice on digital health intervention planning, implementation, and evaluation [43] . Our study reveals several practice implications for facilitating telemedicine uptake that include increasing training for clinical and support staff to assist in connecting visits, developing policies and procedures that support remote patient monitoring (e.g., at-home sexually transmitted infection testing), providing financial support including reimbursement for telemedicine, and developing scheduling protocols to incorporate different visit types based on both patient's technical readiness and provider's assessment of patient's medical needs.

Our study has some limitations. First, both sites were located in the US Northeast, potentially limiting generalizability of our findings. Second, due to sample size, we were unable to parse apart the telemedicine experience related to telephone vs. video visits, which could have provided further insights into future telemedicine efforts. In addition, given our modest sample size, we may not have been adequately powered to detect statistically significant associations between measured factors and outcomes of interest. Third, our sample was limited to English-speaking individuals and thus may not generalize to non-English-speaking patients who may be at greater risk for health care interruptions and challenges associated with the pandemic. Fourth, although we used validated measures where possible, we incorporated measurement tools that have not been validated to specifically assess the COVID-19 pandemic related experiences. Fifth, given the rapidly evolving context with the pandemic, our findings may not reflect current perspectives and experiences. Sixth, our sample was limited to PWH and may not generalize to individuals without HIV.

Our study provided timely insights into the telemedicine experiences for HIV care including barriers and benefits by engaging diverse stakeholders. Efforts to optimize telemedicine implementation for HIV care should consider strategies to improve technology support for PWH, patient-centered care options to maintain care, innovative platforms to allow remote monitoring, and establishing and implementing appropriate funding, billing and reimbursement methods to enable universal availability.

The online version contains supplementary material available at https:// doi. org/ 10. 1007/ s10461-021-03556-7.

First case of 2019 novel coronavirus in the United States

Expanding HIV care through telehealth. CAR-EAction Newsletter

Use of telemedicine technologies in the management of infectious diseases: a review

A review of telehealth innovations for HIV Pre-Exposure Prophylaxis (PrEP)

Integration of telehealth services in the healthcare system: with emphasis on the experience of patients living with HIV

Telehealth programs. Health Resources and Services Administration

Infectious diseases society of America position statement on telehealth and telemedicine as applied to the practice of infectious diseases

HIV patient and provider feedback on a telehealth collaborative care for depression intervention

Mixed-methods evaluation of a telehealth collaborative care program for persons with HIV infection in a rural setting

It is time to include telehealth in our measure of patient retention in HIV care

Telephone nurse counseling improves HIV medication adherence: an effectiveness study

Impact of availability of telehealth programs on documented HIV viral suppression: a cluster-randomized program evaluation in the veterans health administration

Improved virologic suppression with HIV subspecialty care in a large prison system using telemedicine: an observational study with historical controls

Symptoms, stress, and HIV-related care among older people living with HIV during the COVID-19 pandemic

Black and latino persons living with HIV evidence risk and resilience in the context of COVID-19: a mixedmethods study of the early phase of the pandemic

Assessing HIV-related stigma in healthcare settings in the era of the COVID-19 pandemic

Results and response to a 10-state COVID-19 practice impact survey in the Mountain West region

Development of telemedicine infrastructure at an LGBTQ+ clinic to support HIV prevention and care in response to COVID-19

Advancing Digital health equity: a policy Paper of the Infectious Diseases Society of America and the HIV medicine association

Best practices for mixed methods research in the health sciences. Bethesda: National Institutes of Health

Evaluating the public health impact of health promotion interventions: the RE-AIM framework

An extension of RE-AIM to enhance sustainability: addressing dynamic context and promoting health equity over time

A SMART approach to treating tobacco use disorder in persons with HIV (SMARTTT): Rationale and design for a hybrid type 1 effectiveness-implementation study

Working with HIV clinics to adopt addiction treatment using implementation facilitation (WHAT-IF?): rationale and design for a hybrid type 3 effectiveness-implementation study

Growing older with HIV/AIDS: new public health challenges

An examination of the social networks and social isolation in older and younger adults living with HIV/AIDS

Management of human immunodeficiency virus infection in advanced age

Social determinants of potential ehealth engagement among people living with HIV receiving ryan white case management: health equity implications from project TECH

The REDCap consortium: building an international community of software platform partners

Research electronic data capture (REDCap)-a metadata-driven methodology and workflow process for providing translational research informatics support

Awareness, attitudes, and actions related to COVID-19 among adults with chronic conditions at the onset of the US outbreak: a cross-sectional survey

The AUDIT alcohol consumption questions (AUDIT-C): an effective brief screening test for problem drinking. Ambulatory Care Quality Improvement Project (ACQUIP)

PROMIS Short Form v1.0-Depression-4a-26Jun2016

PROMIS Short Form v1.0-Anxiety-8a 31May2019

Guidelines for the use of antiretroviral agents in adults and adolescents with HIV

Three approaches to qualitative content analysis

Encyclopedia of social measurement

Differences in the use of telephone and video telemedicine visits during the COVID-19 pandemic

Visit completion during the telemedicine transition in early months of the pandemic

A midwestern academic HIV clinic operation during the COVID-19 pandemic: implementation strategy and preliminary outcomes

Visit trends and factors associated With telemedicine uptake among persons with HIV during the COVID-19 pandemic

The impact of COVID-19 on HIV care provided via telemedicine-past, present, and future

Focusing on digital health equity

All authors contributed to the study conception and design. Material preparation, data analysis, and data interpretation were performed by DH, EJE, YD, and SC. The first draft of the manuscript was written by DH and EJE. All authors critically reviewed and edited the manuscript. All authors read and approved the final manuscript. 

The authors declare that they have no conflicts of interest.

The study protocol was approved by the institutional review boards at Yale University (#2000028033) and SUNY Downstate Health Sciences University (#1593449-1).

Verbal consent to participate was obtained from all study participants prior to the start of data collection.