Attitudes Toward Telehealth Visits Among Spanish- and English-Speaking Patients in a Safety-Net Gynecological Oncology Clinic

Introduction

Telehealth can be defined as the provision of real-time, interactive health care with the assistance of technology providing audio capability, video capability, or both. ¹ It has been endorsed by the American College of Obstetricians and Gynecologists as a mode of providing care with the potential to improve patient satisfaction and engagement without decreasing the quality of care. ² With the onset of the COVID-19 pandemic, the Centers for Medicare and Medicaid Services eased certain Health Information Portability and Accountability Act (HIPAA) requirements for telehealth, ³ and began reimbursing providers for telehealth visits at the same rates as in-person visits, thus accelerating its uptake. ⁴ Providers of oncologic care have noted the high efficacy of telehealth visits. ⁵ Studies have highlighted similar uptake of telehealth visits among cancer patients with different social vulnerability indices, suggesting both the wide acceptability of telehealth and the possibility of reducing barriers to care. ⁶

Over 37 million people in the United States speak Spanish; of these, over 16 million speak English less than “very well.” ⁷ A previous study among our clinic patients noted that, of those with social needs, approximately 70% reported that Spanish was their preferred language. ⁸ This study sought to determine if patients completing our survey in Spanish have a different experience with telehealth visits compared to patients who completed our survey in English among a Medicaid or uninsured patient population in our gynecologic oncology clinic. We explored whether these differences could be related to challenges posed by language or access to the necessary connectivity and devices. We also sought to determine if telehealth care was an acceptable mode of care for a Spanish-speaking population compared to an English-speaking population. We hypothesized that there would be a difference between these two language groups. For instance, Rodriguez et al. found that, among a large sample of limited English proficiency patients in California, only 4.8% of individuals with limited English proficiency, including Spanish-speakers, had a telehealth visit within a 12-month period, and only 41.1% of those patients had internet access. ⁹ Another study conducted among both primary and specialty care visits showed that non-English language and public insurance were associated with lower uptake of telehealth visits. ¹⁰

Methods

Patients presenting to the gynecologic oncology practice at a large urban public safety net hospital serving a largely non-White, non-English-speaking patient population were offered an opportunity to complete a telehealth survey described below.

Anonymous patient surveys were developed in both English and Spanish using the previously published California Consumer Assessment Survey Instrument (CCASI) by the California Health Care Foundation (CHCF). ¹¹ The CCASI was designed to be adapted by health care organizations to suit their specific populations. Surveys were built in SurveyMonkey and customized to address patients seen for both face-to-face and telehealth visits. Patients chose the language in which they completed the survey. At times, the language in which the patient completed the survey did not match up with their preferred language for medical care as listed in the electronic medical record as many of our patients are bilingual in English and Spanish. For the sake of brevity, we will use “Spanish respondents” to refer to those who completed our survey in Spanish and “English respondents” to refer to those who completed our survey in English. A 4-point Likert scale was used for each question. The adapted CCASI questions were supplemented by categorical questions addressing telehealth access variables and overall patient experience. The only mandatory questions on each survey involved patient demographics and patients could skip certain items as desired. Thus, the total number of responses for each completed survey differed slightly. Patients were scheduled for either in-person or telehealth visits based on a combination of clinician judgment and patient preference. Clinic schedules during the above-mentioned timeframes consisted roughly of two-thirds face-to-face and one-third telehealth visits. During the study period, the electronic medical record allowed for telehealth visits via video; however, the uptake was poor among providers and patients, and thus all of the telehealth visits in this study were conducted via phone.

QR codes linking to self-administered Spanish or English surveys were available in all clinics. Alternatively, surveys were administered in-person or via telephone in the patient’s preferred language with in-person or virtual assistance of clinic staff as needed. Patients were self-selected to complete the questionnaire. All patients scheduled for a clinic visit during the study period were given an opportunity to complete the survey. The introductory paragraph of the survey instrument emphasized that the survey is anonymous, that participation is voluntary, and that a patient’s care will not be impacted by their participation. Questionnaire responses were collected anonymously through SurveyMonkey and were not linked to a medical record. Results were tabulated and exported for quantitative analysis.

The study was approved by the institutional review board (IRB) at the Olive View-UCLA Education and Research Institute. The requirement to consent patients for the study was waived by the IRB given that this was an anonymous survey. The surveys were conducted between August through September 2020 and January through February 2021. These time periods were selected because they occurred outside of a major COVID-19 surge, to avoid time periods when hospital resources were significantly diverted, as this could have impacted adequate collection of data.

Data were tabulated and descriptive statistics were calculated. Comparisons of Spanish-speaking versus English-speaking patient cohorts and between those completing the questionnaire at a face-to-face versus telehealth visit were calculated using Fisher’s exact or chi-squared tests, as appropriate, with alpha set at 0.05 using STATA version 17.

Results

Over the study period, 211 out of 333 eligible patients (63%) completed questionnaires. Of the questionnaires completed, 117 (55%) were filled out in Spanish and 94 (45%) in English. Most were seen for face-to-face visits. Face-to-face visits were conducted in the clinic as per usual protocol. Telehealth visits were offered by the gynecologic oncology service via both video and phone format; however, all patients marked as completing telehealth visits had their visits by phone and none of the patients had a video visit.

Patient demographics among Spanish and English respondents were markedly different (Table 1). Patients who completed the survey in English, regardless of visit modality, were younger than those who completed the survey in Spanish (in-person visits: median age 52 for Spanish- versus 45 for English-speaking patients, p = 0.008; telehealth visits: median age 57 for Spanish- versus 33 for English-speaking patients, p = 0.00026). Given that many of our patients are bilingual, the preferred language for a medical visit and the language in which the survey was completed were at times different.

Patients who completed the survey in either language indicated that they believed they could receive the same quality of care via telehealth as at an in-person visit (41.7% versus 41.0%, p = 0.93). However, patients who completed the survey in Spanish and who were seen in person for their visit were more likely to respond that the provider listened carefully to them compared to English respondents who were seen in person (100% versus 94.7%, p = 0.038). Spanish respondents seen in person were also more likely to think that the in-person visit was better than a phone visit compared to their English-speaking counterparts (100% versus 94.5%, p = 0.036) (Table 2). Aside from these differences, nearly all patients, regardless of visit modality or language, reported satisfaction with their visits across several domains.

Both Spanish and English respondents had good access to the necessary equipment and resources needed to complete a telehealth visit (Table 3). One notable difference is that the patients who completed the survey in Spanish reported less frequently that they had the ability to install a telehealth app or to have someone who could assist them, and this difference was most notable in those who saw a provider in person (difference for all visit modalities: Spanish respondents 68.7% versus English respondents 83.9%, p = 0.013; difference for in-person visits: Spanish respondents 69.1% versus English respondents 85.5%, p = 0.012).

English respondents seen for an in-person visit were more likely to schedule a telehealth visit for their follow-up visit versus Spanish respondents (29.2% versus 15.9%, p = 0.044) and were also more likely to have taken time off work to attend their visit (30.3% versus 23.9%, p = 0.015) (Table 4).

Discussion

In a representative sample of Spanish- and English-speaking patients accessing services at a gynecologic oncology clinic in an urban safety-net hospital after the onset of the COVID-19 pandemic, we found that telehealth use was highly acceptable for most patients regardless of language preference. This is consistent with previous findings among gynecologic oncology patients but diverges from some other studies that noted decreased patient satisfaction. ^{12 –16} We also found that both Spanish and English respondents were able to access the tools and services needed to make a telehealth visit successful. Most patients reported interest in using a telehealth app and the ability to access the hardware, internet connection, and assistance needed to complete a video visit or to install an app if needed. This finding is consistent with another similar study conducted during the COVID-19 pandemic. ¹⁷ Our use of phone visits as our modality of providing telehealth care also mirrors the high uptake of audio-only telehealth care in safety-net populations. ¹⁸

Despite largely similar findings among the Spanish and English respondents in our study, we found evidence of possible health disparities. Patients who completed the survey in Spanish were more likely to think that their in-person visit was better than a phone visit, were more likely to feel that their provider listened to them, and were less likely to be able to install a telehealth app or find someone who could help when compared to their English-speaking counterparts. This suggests that the patients who completed our survey in Spanish may have a preference for in-person visits, which is reflected in the lower percentage of Spanish-speaking patients who were scheduled subsequently for a telehealth visit.

Regardless of the high acceptability and interest in telehealth visits among the study population, the majority of follow-up visits scheduled after the index visit studied were booked as in-person visits. This finding underscores that in-person evaluations remain an important aspect of gynecologic oncologic care even during a pandemic. This practice aligns with examples of combined telehealth surveillance algorithms which rely at least partially on face-to-face visits, ¹⁹ which allow for important aspects of in-person care such as a physical exam. Multiple studies have highlighted patient anxiety around an inability to have a physical examination during a telehealth oncology visit, which may be driving the demand for in-person visits. ^15,16,20

Our study has several strengths. We targeted a Spanish-speaking population receiving gynecologic oncology care in an urban safety-net hospital, a group that has not been specifically studied to date. We used a standardized questionnaire written at a sixth-grade reading level to lower barriers to data collection due to education level. Our patients were surveyed promptly after their clinic visit, reducing the possibility of recall bias. We also looked at a series of domains, including patient experience, attitudes, and objective access metrics that may impact a patient’s perception of telehealth services.

Limitations of our study include the fact that it was performed at a single institution, which limits its generalizability. We also found a much lower survey response rate among our telehealth patients, and we did not circle back to those who did not respond to the initial survey request. Due to de-identified data collection, we were unable to compare the two groups to determine whether face-to-face respondents differed from telehealth respondents in any significant way, but from baseline characteristics alone, our English-speaking patients were younger than our Spanish-speaking ones, which could lead to greater acceptability and familiarity with telehealth visits. Though our aim is to determine whether disparities exist between English and Spanish-speaking patients, language preference alone may not be an adequate marker for racial and socioeconomic differences in telehealth uptake. For instance, in a study by Ukoha et al., Black patients were more likely to express a lack of interest in telehealth obstetrical visits due to concerns about privacy and general mistrust of the health care system; this difference would not be elicited by studying language preference alone. ¹⁷ Other telehealth studies in obstetrics have also found decreased access to telehealth care among patients of lower socioeconomic status based on Medicaid insurance status or address. ^21,22

Conclusions

Both Spanish- and English-speaking gynecologic oncology patients in an urban safety-net hospital find telehealth to be adequate for their care, though Spanish-speaking patients demonstrate a preference for in-person care. In combination with face-to-face visits, telehealth visits can meet patient needs using resources already available to them. Both groups largely have the means to complete a telehealth visit, allaying our initial concerns about widening health disparities between Spanish- versus English-speaking patients. Based on our data, telehealth visits conducted by phone should continue to be an important option for telehealth care for this population, as patients already demonstrate a high level of satisfaction with phone visits, and a significant minority of both English and Spanish respondents reported an inability to use a telehealth app. ²³ Further research is needed to understand whether video visits or other telehealth visit types result in better care and higher satisfaction for these patients compared to phone visits. Finally, we see two major threats to the ongoing provision of audio-only telehealth. For one, providers were not penalized for violating HIPAA Privacy, Security, and Breach Notification Rules during the Public Health Emergency; after May 2023, this is no longer the case. Audio-only telehealth can involve the use of standard landline phones, which bypasses the need to comply with the HIPAA Security Rule for electronically protected health information. ²³ However, many providers use phones utilizing voice over internet protocol or Wi-Fi technology, which is not exempt in the same way. Health care providers will need to carefully discern how picking up a phone can trigger different privacy considerations depending on how the phone call is transmitted. Second, the choice of type of telehealth visit is potentially impacted by reimbursement rates. In California, where this study was conducted, Medi-Cal currently offers payment parity for any type of telehealth visit (such as video or audio-only phone visit) even after the expiration of the federal Public Health Emergency for COVID-19 in May 2023. ²⁴ We advocate for continued reimbursement of audio-only telehealth visits as they are vital to our patient population.