Abstract
Background:
Telehealth's applicability may be limited for vulnerable populations including rural communities. While broadband access is a known barrier to telehealth use, other factors may influence a person's ability or preference to use telehealth.
Objective/Purpose:
To compare characteristics of telehealth users versus nontelehealth users in a rural health care network.
Methods:
We surveyed a stratified random sample of 500 adult patients in August 2021 about telehealth use. We used descriptive statistics to compare characteristics of telehealth users with nontelehealth users. Telehealth was defined in three different ways as follows: (1) phone or video visit, (2) video visit, and (3) patient portal use.
Results:
Mean age of the 206 respondents was 60 years, 60.7% were female, 60.4% had some college education; 84.9% had home internet, and 73.3% used the internet independently. Video telehealth use was independently associated with younger age (<65), having some college education, being married/partnered, and being enrolled in Medicaid. When telehealth included a phone option, disability was positively associated with telehealth use, and living in a rural town versus metropolitan/micropolitan area was negatively associated with telehealth use. Being younger, married/partnered, and having some college education were significantly associated with patient portal use.
Conclusion
: Videoconferencing and patient portal use pose barriers to those who are older and have less education. However, these barriers disappear when telehealth is available through telephone.
Introduction
Telehealth use has dramatically increased in the United States since the beginning of the COVID-19 pandemic, 1,2 with some research indicating a 63-fold increase in telehealth Medicare visits between 2019 and 2020. 3 Despite this rapid growth, it is clear that there are disparities in both willingness to use telehealth and actual utilization based on broadband access, age, ethnicity, geography, and education. 4 –13 The rapid expansion of telehealth during COVID has also demonstrated that type of telehealth technology matters in regard to capacity and willingness to engage. While telehealth has been touted as a great equalizer in terms of access to health care and was notably essential during the first months of the COVID pandemic, the threat of a widening digital divide is evident. 14
Telehealth can include remote visits with a health care provider, as well as use of a patient portal for scheduling appointments, receiving test results, and communicating with a health care provider. With the exception of provider visits conducted over the phone (audio only), these services require digital access. In their digital health equity framework, Crawford and Serhal describe “digital determinants of health,” which include digital health literacy, beliefs about its potential helps or harms, and values and cultural norms, which are in turn shaped by social determinants of health, such as socioeconomic status. 15
Studying digital health equity in a rural area is of special interest. On the one hand, remote and frontier communities have been using telehealth for years due to the geographic challenges of attending visits in person and the paucity of both primary care providers and medical specialists. 7,16,17 On the other hand, the lack of broadband or other reliable high-speed internet services in rural areas has been well documented. 18 –20
Beyond broadband access, there are other potential barriers to use, including digital literacy 21,22 and established rural norms and values. 23 There are also structural barriers to use, including reimbursement for telehealth services. For example, before the pandemic, the Centers for Medicare and Medicaid Services only covered telehealth visits taking place in a health care provider's office; it did not cover telehealth visits from a patient's home. 3
The purpose of this analysis is to characterize utilization of telehealth by adult patients in a rural health care network.
Methods
We administered a survey to a stratified random sample of 500 adult patients (age ≥18) within our rural health care network in upstate New York in August 2021. To ensure equal representation of patients with different levels of access to digital technology in the surveyed population, we stratified the sample into equal numbers of patients who had and had not signed up for the patient portal. Patients who were not signed up to use the network's patient portal were mailed a paper copy of the survey, while patients who had signed up for the patient portal were emailed a link to an electronic version of the survey. One reminder phone call was made to all nonresponders 2 weeks after survey was disseminated. If an individual was not reached by phone, a paper copy of the survey was mailed regardless of patient portal status.
The survey included questions about access to and use of the internet, smartphones, computers/tablets; experience with telehealth for health care provider visits, use of the patient portal, and general videoconferencing (e.g., Zoom); health status, including self-rated health, disability status, and access to a primary care provider; and sociodemographics, including gender, age, education, race/ethnicity, living situation, marital status, financial distress, employment status, and insurance status. The patient's zip code of residence was used to categorize participants by Rural–Urban Commuting Area (RUCA) codes as a measure of rurality.
TELEHEALTH DEFINITIONS
The analyses were conducted separately using three different definitions of telehealth. First, we categorized study participants as ever telehealth users (phone/video) if they answered “yes, using a regular telephone” or “yes, using a computer, tablet or smartphone” to “Have you ever had a health care appointment that was not done in person (where you were not in the clinic/doctor's office)?” Second, from the above category, we identified a subset of participants as ever telehealth users (video only). Third, we categorized study participants as patient portal users if they indicated that they had access to the health care system's patient portal and had used it occasionally, sometimes or extensively in the past 12 months. For each outcome, participants who answered “not sure” were categorized as nonusers.
STATISTICAL ANALYSIS
To compare characteristics of telehealth users versus nontelehealth users, the chi-squared test was used for categorical variables, and where appropriate Fisher's exact test was used. We then used multivariable logistic regression to identify characteristics that impact the odds of telehealth use. Statistical significance was defined as p < 0.05. Analyses were conducted using IBM SPSS Statistics version 27.
This study was approved by the Mary Imogene Bassett Hospital Institutional Review Board. The survey included a comprehensive study information sheet, and participants provided tacit consent by returning the completed survey.
Results
A total of 206 respondents completed surveys for a response rate of 41.2% (25.2% completed online, 68.4% completed through mail, 6.3% completed through phone). The mean age of respondents was 60 years; 60.7% were female and 60.4% had some college education (Table 1).
Characteristics of Study Participants by Digital Technology Availability and Experience
RUCA, Rural–Urban Commuting Area.
DIGITAL TECHNOLOGY ACCESS AND EXPERIENCE
The vast majority of participants (79.1%) reported having a computer or tablet at home, 84.9% had home internet, 74.1% reported having a smartphone, and 73.3% reported having ever used the internet independently. Age category, education, marital status, living situation, employment status, and household income were significantly different among all aspects of technology availability and use (Table 1). Respondents who were enrolled in Medicare were less likely to report independent internet use (p < 0.001) and smartphone ownership (p < 0.001).
Respondents who were enrolled in Medicaid were less likely to report having a computer (p = 0.024), having home internet (p = 0.009), and having a smartphone (p = 0.036). Self-reported good/very good/excellent health was associated with having a computer (p = 0.043). Significantly more females reported having a smartphone than males (p = 0.039). There were no differences in technology access/use by rurality, time at address, financial hardship, and disability.
DIGITAL TECHNOLOGY AVAILABILITY AND EXPERIENCE BY TELEHEALTH STATUS
Video telehealth use and patient portal use were associated with having a home computer, having home internet, having a smartphone, using internet, and experience with videoconference technology (Table 2). There was no association between phone/video telehealth use and having home internet. Nearly three-quarters of respondents had a smartphone; 11% of those without a smartphone were patient portal users.
Digital Technology Availability and Experience by Telehealth Status
TELEHEALTH USE
Forty-one percent of respondents reported ever having used telehealth by phone or video; 22.9% reported ever having used telehealth by video (Table 3). Telehealth phone/video use was associated with rurality (p = 0.009), with the highest proportion of users (51.2%) located in isolated rural areas. A greater proportion of those who reported having a disability had ever used telehealth phone/video, although this association was not statistically significant (p = 0.056). In the multivariable analysis, disability was positively associated with phone/video telehealth use (p = 0.021), while living in a rural town versus a metropolitan/micropolitan area was negatively associated with phone/video telehealth use (p = 0.025) (Table 4).
Characteristics of Study Participants by Telehealth Status
SD, standard deviation.
Multivariable Regression of Study Participant Factors Associated with Telehealth Use
<0.05.
<0.01.
CI, confidence interval; OR, odds ratio.
Video telehealth use was associated with younger age (<65), more education (at least some college), living with others, living in an isolated rural Census tract compared with metro/micropolitan area or rural town, not having Medicare, and having a higher income. After adjustment for all other factors, independent predictors of video telehealth use were younger age (p = 0.041), more education (p = 0.003), being married/partnered (p = 0.025), and being enrolled in Medicaid (p = 0.011).
Patient portal users were more likely to be younger, female, more educated, married/partnered, living with others, employed, and living in an isolated rural Census tract. Those with Medicare were less likely to be patient portal users. In the multivariable model, younger age (p = 0.006), more education (0.035), and being married/partnered (p = 0.015) were independently associated with patient portal use (Table 4).
Discussion
The purpose of this analysis was to characterize utilization of telehealth within a rural health care network. The survey results illuminate the complexity of our rural population's experience with telehealth technologies. Consistent with other national and international data, older adults and those with less education were less likely to use telehealth modalities requiring use of digital technology; that is, video telehealth and the patient portal. 24,25 In addition to concerns about digital literacy, older adults may be more likely to have cognitive or sensory impairments that make using video- or web-based telehealth technologies impractical. 26 –28
However, when telehealth visits included a traditional telephone option, the only group that was more likely than others to engage were those who indicated that they had a disability. The most prominently reported disabilities were mobility issues, which can make travel for medical appointments difficult.
An unexpected finding was that individuals living in rural towns were less likely to have ever used telehealth through phone or video than their more urban counterparts, while there was no difference in telehealth (audio or video) among those living in isolated rural areas. This could be due to the limitations of the rural coding system used (RUCA) as an indicator of rurality relevant to broadband access. It could also be that adults living in isolated rural areas may have more experience with telehealth out of necessity (e.g., long travel distances to health care provider) before the COVID-19 pandemic.
Our results are similar to those from the 2021 National Health Interview Survey, showing that 37.0% of adults used telemedicine (defined as an appointment with a health care provider by video or phone) in the past 12 months (vs. 41.0% of our patients who reported having ever used telehealth by phone or video). 29 There was also higher use of telemedicine among females and those with more education.
Interestingly, when the data are limited to respondents living in noncore regions (based on the NCHS Urban-Rural Classification Scheme for counties), the national estimate of telemedicine use dropped to 27.5%, whereas in our population the proportion of users was highest in the remote rural group (see the Limitations section).
As expected, video telehealth use and patient portal use were strongly associated with having a computer, having/using home internet, having a smartphone, and having experience with videoconference technology. However, there was no association between phone/video telehealth use and having home internet, pointing to the likelihood that those without home internet are using phone-based telehealth versus using video telehealth from another location. Thus, the availability of a telephone option is key to broader telehealth access in this rural population. This has policy implications as CMS is considering eliminating reimbursement for most audio-only telehealth visits after the public health emergency ends. 30
Despite the prevalence and acceptance of patient portal use, individuals who were older and less educated were not as likely to utilize the portal, a finding that is consistent with the literature, 31,32 and is concerning in terms of health equity. Portals were developed with the intention of increasing not only patient centeredness and quality of care but also equity by purportedly improving access. 33 While our study design limits our ability to draw conclusions about what may drive utilization, it is clear that patient portals are not being used equally by all groups.
Along with age and education, being partnered or married was associated with patient portal use. It is unclear what is driving the association. However, it is known that marriage enhances health through partners supporting one another in illness as well as regulating each other's health-related behaviors. 34,35 Such behavior regulation could include encouraging a partner to utilize the patient portal.
LIMITATIONS
There are several limitations to note. First, our analysis relies on self-reported data, and thus social acceptability bias may be present. Second, our survey question asking about telehealth use through computer, tablet, or smartphone assumed that a person using these devices would be using video.
In addition, other factors besides patient technology access and preferences related to telehealth could be contributing to the use of phone versus video telehealth, such as health care provider preference and provider access to video telehealth technology. We have learned from interviews with staff overseeing telehealth across the institution's network that not all health care providers were set up to provide care through video telehealth and not all health care providers were willing to provide care through telehealth.
Finally, measurement of rurality using RUCA codes may not have been indicative of broadband access in our region. We found that when using RUCA codes to categorize rurality, the ZIP codes categorized into the highest level of rurality (isolated rural) were more likely to report video telehealth use than the next lower level of rurality (rural town).
CONCLUSIONS AND IMPLICATIONS FOR PRACTICE
When considering implementing telehealth technologies, health care organizations need to be aware of potential barriers to use beyond broadband access. Patients who are older and have less education are less likely to use video telehealth. However, disparities disappear with the availability of telehealth by phone. Thus, it is important to address policy barriers to conducting audio-only telehealth visits beyond the public health emergency.
Furthermore, to prevent a widening of the digital divide and exacerbating health disparities in rural populations, we recommend further exploration of the implications regarding access to care and health outcomes among those who are unable or unwilling to use the patient portal or video-based telehealth. While this analysis contributes to the evidence base for describing who is using which types of telehealth technologies in the peripandemic era, evidence describing how telehealth changes a person's ease of receiving health care and ultimately health outcomes is still lacking. 36,37
Footnotes
Acknowledgments
The authors thank our study participants for their willingness to engage with this project.
Authors' Contributions
W.B. contributed to conceptualization, methodology (equal), formal analysis, writing—original draft (lead), and writing—review and editing (equal). K.P. assisted with conceptualization, investigation (lead), methodology (equal), formal analysis (lead), writing—original draft, and writing—review and editing (equal). M.S. performed methodology (equal), visualization, and writing—review and editing. N.K. developed methodology (equal), formal analysis, visualization, and writing—review and editing. A.F. conducted investigation, writing—review and editing. M.K. contributed to writing—review and editing.
Disclosure Statement
The authors have no disclosures.
Funding Information
This study has been funded by a research grant from the National Institute of Health Care Management (NIHCM) Foundation.