Transitioning to Telehealth During the COVID-19 Pandemic: Patient Perspectives and Attendance at an HIV Clinic in San Francisco

Abstract

Approximately 1.2 million people are living with HIV in the United States, with 16,000 in San Francisco. Many HIV-positive individuals have difficulty maintaining follow-up clinic visits under normal circumstances, and this is complicated by the coronavirus disease 2019 (COVID-19) pandemic as many institutions transition to a telehealth-focused model of care to maintain patient and provider safety. However, it was unknown how telehealth would impact patient attendance and perceptions about their care, particularly in populations at high risk of appointment nonattendance. To quantify the impact of telehealth on retention in care for a vulnerable HIV-infected population and to identify patient perspectives of telehealth and its effect on appointment attendance, we studied patients at the University of California, San Francisco (UCSF) 360 Wellness Center, an HIV primary care clinic in urban San Francisco, California. Anonymous surveys were conducted to assess patient perceptions of telehealth, and 21% of patients sent surveys responded. Appointment attendance rates for all visits were analyzed before and after the shelter-in-place order in San Francisco on March 16, 2020. With the transition to telehealth, the overall nonattendance rate decreased by ∼3%. Most common perceived strengths of telehealth included convenience and safety, whereas disadvantages included technical barriers and unfamiliarity. Despite barriers and disadvantages listed by patients, a majority are willing to attend telehealth visits, as 80.5% of respondents reported being equally or more likely to attend telehealth visits. This is a critical finding during the COVID-19 pandemic and beyond; we believe that telehealth can improve appointment attendance for vulnerable populations who often face barriers to receiving health care.

Introduction

The coronavirus disease 2019 (COVID-19) pandemic forced many institutions to rapidly transition to a telehealth-focused model of care.¹ Although this change was necessary to mitigate spread of disease and protect patients and health care workers, it was unknown how patients would perceive and respond to this new mode of health care. Globally, 38 million people are living with HIV, but this population disproportionately experiences health care disparities, social injustices, and poor retention in health care.^2,3 Further, the COVID-19 pandemic may have a very different impact on people living with HIV, especially HIV-positive people of distinct demographics.⁴ HIV care retention is vital, as missed visits are associated with increased mortality, yet up to 35% of HIV primary care patients in the United States do not consistently attend their follow-up appointments.^5–8 It is crucial to maintain consistent care in patients with HIV to control viral load, avoid medication resistance, and prevent new infections.^9–11 In addition, continuity of care is necessary to maintain patient connections with social and other health care services.¹²

Our urban HIV adult primary care clinic at the University of California, San Francisco (UCSF), serves one of the largest US populations of people living with HIV, with ∼16,000 infected across the city.¹³ Our clinic population is majority male (∼80%), and racially diverse (58.8% White, 13% Hispanic, 11.2% African American/Black, 11.2% Asian American/Pacific Islander, and 5.8% Other). Prior research at our clinic showed that the 2-year nonattendance rate for in-office visits was 16.7%. As patient care has shifted significantly to telemedicine due to the COVID-19 pandemic, there has been a great deal of interest in understanding the pandemic's impact on care of HIV-infected populations.^14–16

Studies before the pandemic have shown that using telehealth as a health care delivery method for subpopulations of patients living with HIV may expand access to interventions and increase perceived patient privacy.^17,18 To provide optimal health care to this vulnerable population and decrease the nonattendance rate, we aimed to assess the impact of telehealth on clinic attendance, and understand what patients like and dislike most about using telehealth. Even in our clinic where the majority of patients are local, we hypothesized that telehealth appointments would decrease the nonattendance rate. By addressing perspectives and attendance simultaneously, we believe that we can use patient input to address new telehealth-related barriers that arise while maintaining features of telehealth that patients prefer. In this way, we hope to help our HIV-infected patients maintain consistent clinical care and provide access in a patient-centric way.

Methods

This study was reviewed by the Institutional Review Board (IRB 20-31313) at our institution and received an exemption from further oversight. Statement from IRB: This study qualifies as Exempt under the following Revised Common Rule (January 2018) category: Category 2: Research that only includes interactions involving educational tests (cognitive, diagnostic, aptitude, achievement), survey procedures, interview procedures, or observation of public behavior (including visual or auditory recording).

We studied patients at the UCSF 360 Wellness Center, an HIV primary care clinic that serves a total of 966 patients at the UCSF, during 2019–2020. We compared appointment attendance rates for physician, nursing, social work, and nutrition visits before and after the city-wide shelter-in-place order on March 16, 2020, and analyzed its effect on our subsequent transition to telehealth visits. Binomial tests were used to compare rates of total nonattendance before COVID-19 (before March 16, 2020) and during COVID-19 (after March 16, 2020). In addition, our study collected and analyzed patient perceptions of telehealth from June 2020 to July 2020 through an anonymous survey administered through a secure online patient portal. The survey included multiple-choice questions and free-text responses. Patients who did not have portal access were reached by telephone. Responses were analyzed and coded for themes, which were organized by perceived advantages and disadvantages.

Results

Appointment nonattendance rates

We analyzed attendance records among the clinic's 966 patients from October 2019 to June 2020. Between October 1, 2019, and March 16, 2020, telehealth comprised 5.3% of total clinic appointments (244/4693) and in-person visits comprised 94.7% of appointments (4449/4693). During this time, in-person visits had a 16.5% nonattendance rate, whereas telehealth visits had a 0% nonattendance rate (Fig. 1A). Between March 17, 2020, and June 30, 2020, telehealth comprised 86.9% of all clinic appointments (1418/1632) and in-person visits comprised 13.1% of total appointments (214/1632). During this time, telehealth visits had a nonattendance rate of 13.7%, whereas in-person visits had 6.1% nonattendance rate (Fig. 1B).

FIG. 1.

Telehealth Attendance and Patient Preferences. (A, B) Bar graphs show the number of in-person and telehealth appointments before and after COVID-19 shelter-in-place orders, and appointment outcomes, by type of visit. Binomial tests comparing rates of total nonattendance before COVID-19 versus during COVID-19 are significant (p < 0.01). (C, D) Survey attitude results for two questions. COVID, coronavirus disease.

Patient perspective survey results

In total, 202 patients (21% of those surveyed) responded to the Telehealth Patient Perspective Survey. And 46.9% patients responded that they preferred in-person visits, 11.3% preferred telehealth visits, whereas 41.8% liked both in-person and telehealth visits equally (Fig. 1C). When asked about likelihood of attendance based on appointment type, 20.5% reported they would be more likely to attend in-person visits, 10.5% reported they would be more likely to attend telehealth visits, and 69.0% said they would be equally likely to attend in-person and telehealth visits (Fig. 1D).

Thematic analysis was performed from the Telehealth Patient Perspective Survey (n = 202). Themes were organized by perceived advantages and disadvantages of telehealth visits. Regarding advantages, the most frequent theme was “convenience” (i.e., increased flexibility with time, travel, and scheduling), followed by “safety” (i.e., patient feels more safe attending virtually) and “personal patient discussions” (i.e., privacy benefits for personal health communications such as therapy or patient laboratory results) (Table 1). Regarding disadvantages, the most frequent theme was “technical issues” (i.e., problems with internet, smart device, or within video platforms), followed by “unfamiliarity” (i.e., patient distrust or unfamiliarity with telehealth platforms), “lack of human contact and connection” (i.e., less personal visits, constraints with preferred provider and traditional health care visit components), and “communication” (i.e., difficulties with appointment link, instructions, or clinic difficulties) (Table 1).

Table 1.

Thematic Analysis of Perceived Advantages and Disadvantages of Telehealth

	Themes	Codes	Select quotes
Telehealth advantages	Convenience	Increased flexibility with time Less travel Easier to schedule	“Convenient.” “Flexibility in terms of date and time.” “Sometimes I don't feel well enough physically to come in.” “I'm agoraphobic and this is a huge benefit for me to get medical treatment.”
	Safety	Less potential exposure	“Necessity. I would not want to infect or get infected.”
	Personal discussions	Privacy during communications Good for therapy or discussing laboratory results	“Doctor sharing screen with information to point to for education, for example showing an example of medication bottle and change in condition.”
Telehealth disadvantages	Technical issues	Internet problems Video platform issues	“Wouldn't work no matter what device I used. Ended up being a phone call.” “Check all sound both ways but the app doesn't allow for that.”
	Unfamiliarity and distrust	Lack of experience Distrust of video platform	“I did not like having to use Zoom for HIPAA-type information.” “Not knowing if my doctor is joining me or if he's still with a patient.”
	Lack of human contact and interaction	Visits feel less personal Limited medical utility	“I feel that of all the instances I had my vitals taken at regular appointments… this was an instance where it would've been helpful to have that info.”
	Communication	Link to appointment Clinic-side difficulties	“Provider did not show up for the appointment.” “It's usually difficult to follow all the steps to get to the video meeting.”
Refusal	Refuse telehealth	Refuse telehealth	“No interest in videophone communications.” “Not interested in telehealth.”

Discussion

We analyzed patient perspectives regarding in-person appointments and telehealth appointments during the COVID-19 pandemic, whereas simultaneously tracking the appointment nonattendance rate. Importantly, our study followed the nonattendance rate of our urban HIV care clinic for 6 months before the shelter-in-place order, and nearly 4 months after the transition to telehealth. When comparing the major form of appointment (in-person vs. telehealth) at the clinic, the overall nonattendance rate decreased by ∼3% when the clinic transitioned to telehealth. Based on results from our patient survey, we believe that this reduction is, in part, due to bypassing obstacles to appointment attendance that were more apparent in person. As described in a recent study, using telehealth during the COVID-19 pandemic has helped reduce certain barriers that exist in rural health care settings for people living with HIV, such as addressing difficulties with transportation.¹⁹ Our results complement these findings, as they show that telehealth may also combat barriers faced in the urban setting including an increase in patient privacy, reduction in stigma, and an overall improvement in health care accessibility. Notably, although the overall nonattendance rate decreased after the shelter-in-place order, the telehealth nonattendance rate increased to 13.7% and the in-person nonattendance rate decreased to 6.1%. We believe it is likely that this finding is due to the fact that the rapid transition to telehealth initially introduced new challenges such as accessing wireless internet and appointment instructions, and that those who kept in-person appointments during the pandemic likely had an important reason not to switch to telehealth, health related or otherwise, which motivated their in-person attendance. However, it will be important to continue to assess nonattendance rates in our clinic throughout the near future to better understand whether this trend continues.

Further, our survey results show that although more patients report a preference for in-person appointments than for telehealth, the nonattendance rate for the clinic decreased with the transition to telehealth appointments. As patients and providers continue to learn how to engage with a predominantly virtual health care delivery platform, we suspect that we will improve access to care for patients who face barriers when engaging with HIV health care services, and thereby increase patient follow-up. Our clinic has already begun implementing changes based on patient survey responses to further improve telehealth communication, decrease unfamiliarity, and continue to encourage virtual patient visit attendance. We plan to again examine patient perspectives and attendance to assess efficacy and ultimately maintain telehealth as a long-term option for patient care within our clinic.

Limitations of our study include a small sample size, and the fact that our patient population is in San Francisco, a large urban center that may not reflect the geographic diversity of other areas of the country. In addition, patient attendance rates were only followed up for 3 months at the time of data analysis; with more time, we may see different trends in nonattendance rates. Finally, we are unable to assess whether there was a difference between the 21% of patients who responded to the survey and the 79% who did not, as we did not collect any information from nonresponders. We acknowledge that telehealth comes with its own limitations and some patients do not have the necessary technology to attend these appointments; however, we anticipate that a hybrid in-clinic and telehealth model will help reach most patients who require care.

Our study is novel as it is one of the few to show both patient perspectives with simultaneous tracking of nonattendance rates within the same HIV care clinic. We found that patients prefer in-person visits to telehealth, but the majority of patients would still be equally likely to attend either, a critical finding during a pandemic wherein patient health and safety are of increased concern. These findings offer valuable insight into patient perspectives and open possibilities for improving long-term appointment attendance for more vulnerable patient populations who often face significant barriers to receiving health care. As health care providers across the country have learned to use telehealth throughout the past year, it is likely that many clinics will elect to permanently offer this option for their patients beyond the present pandemic. Further, we speculate that telehealth may become a long-term mainstay for HIV care, as it appears to offer benefits over in-person visits including more privacy, less stigma, and added convenience. These qualities may have an especially positive impact on improving patient utilization of specialized appointments, such as mental health services, which may be more attractive through telehealth. Importantly, this may help to improve linkages to mental health services, as this has been shown to facilitate HIV medication adherence.²⁰ Therefore, we believe that our results can be used to inform future work at other institutions, and that lessons learned here can be applicable beyond the HIV population to include management of other chronic conditions. Further research into both barriers for patient attendance and utilization of telehealth beyond the COVID-19 pandemic will help to advance continuity in HIV care.

Authors' Contributions

All authors had full access to all of the data in the study: I.C.A contributed to conceptualization, methodology, data collection, data curation, and writing and editing original draft. K.J. contributed to methodology, data collection, data curation, and writing and editing original draft. A.T. contributed to statistical analysis, data collection, data curation, and writing and editing original draft. J.M. contributed to project administration, data collection, review, and editing. B.B. contributed to conceptualization, supervision, review, and editing.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This project was not funded. A.T. is supported by Medical Scientist Training Program T32GM007618.