Association of social isolation and loneliness with telehealth use among older adults during COVID ‐19
Year: 2023
Type: article
Abstract: While rapid expansion of telehealth during the COVID pandemic is considered a major success in novel healthcare delivery, disparities exist in both access and use. Older adults are particularly vulnerable due to lower average technological literacy, as well as higher rates of social isolation and loneliness.1 Social isolation and loneliness have been increasingly linked to a range of poor health outcomes, including dementia, functional decline, cardiovascular morbidity, and death.2, 3 Previous studies have found high social isolation and loneliness among older adults,1, 4 with separate studies finding lower rates of telehealth use.5 Our study is the first to examine these factors together. Telehealth is associated with equivalent or improved outcomes on multiple quality measures.6 Thus, toward the broader goal of identifying strategies to improve care for high-risk older adults, our objective was to examine relationships between telehealth use and social isolation/loneliness among older adults during the first wave of the pandemic. This was a cross-sectional analysis of the 2020–2021 National Health and Aging Trends Study (NHATS) data and associated COVID supplement. The primary independent variables were social isolation and loneliness. Social isolation was measured using a NHATS-validated scoring algorithm based on the Berkman-Syme Social Network Index (SNI).7 Loneliness was assessed using a single-item: "How often do you feel lonely?" The primary outcome was telehealth use, operationalized as self-reported video visit usage with usual healthcare provider. We computed frequencies for our primary variables of interest, and correlations were estimated using Pearson's r. Logistic regression was used to calculate odds ratios for telehealth use associated with social isolation and loneliness, adjusting for covariates (i.e., age, gender, race/ethnicity, Medicaid, health status, anxiety/depression, dementia, mobility difficulty, metropolitan status, census division). Sensitivity analyses were performed to assess for mediation by anxiety/depressive symptoms. All analyses were completed using STATA15 (StataCorp, College Station, TX), accounting for NHATS' complex survey design. Characteristics of the sample (n = 2309) are shown in Table 1. Telehealth usage increased from 6% (before COVID) to 24%. The majority of participants were somewhat isolated (63%), with fewer in the least (18%) or most (19%) isolated categories. Approximately one-third felt lonely on some days (26%) or most/every day (9%). Social isolation and loneliness were uncorrelated (r = 0.02). In the adjusted model, participants who were most isolated were significantly less likely to use telehealth as compared to those least isolated (aOR 0.49, CI 0.29–0.83) (Table 2). Loneliness was not associated with telehealth usage. In sensitivity analyses, results were similar with and without anxiety/depressive symptoms in the model, suggesting no significant mediating effects. Telehealth is both acceptable to older adults and associated with improvements in healthcare quality.6, 8 We found that socially isolated older adults were less likely to use telehealth during the first wave of the pandemic. This is concerning, as limited access to care may compound risk for poor health outcomes in an already at-risk group. A limitation is the lack of visit frequency data in the dataset. Another limitation is that our social isolation measure lacks inclusion of technology-enabled social connections (e.g., video chat, social media). Going forward, measures are needed that carefully balance conceptual clarity with changing societal understandings of social connection/integration. Consistency in measurement is also needed—the field of social isolation research has been hampered by variability in terminology and measures, complicating synthesis of results across studies.1 Interestingly, we found no association between loneliness and telehealth use. It is plausible that our single-item measure of loneliness lacked adequate precision and that a more robust measure is needed (e.g., the UCLA Loneliness Scale1), though a recent pandemic-era study found comparable classification between these 1-item and 3-item scales.9 Loneliness is an important determinant of functional decline, dementia, and death among older adults,3 and efficient and accurate identification of lonely older adults is needed to reduce these outcomes. As telehealth use continues to expand within health systems, efforts are being made to identify those without telehealth access and assist with connection. Some proposals for increasing access, such as institutional technical assistance, employing community health workers to support telehealth connection,10 and engaging caregivers, are promising. Health systems will need to intentionally identify1 older adults who are most isolated or do not have caregiver support, as these individuals may be left out from the benefits of telehealth and may be at high risk for poor health outcomes. While system-wide efforts to identify socially isolated or lonely individuals may be helpful, efforts to overcome these concerns will need to be individualized,4 presenting a remarkable but worthy challenge for improving care of older adults. Rebecca J. Howe: study concept and design, analysis and interpretation of data, preparation of manuscript. Janice F. Bell: study concept and design, analysis and interpretation of data, preparation of manuscript. Julie T. Bidwell: analysis and interpretation of data, preparation of manuscript. Joshua J. Fenton: analysis and interpretation of data, preparation of manuscript. Grace P. Amadi: analysis and interpretation of data, preparation of manuscript. Alicia L. Agnoli: study concept and design, analysis and interpretation of data, preparation of manuscript. The authors declare no conflicts of interest. N/A. Rebecca J. Howe received support from the National Center for Advancing Translational Sciences, National Institutes of Health, through grant number UL1 TR001860. Julie T. Bidwell received support from the National Institutes of Health/National Institute of Nursing Research through grant number K01NR018880. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
DOI: 10.1111/jgs.18270
Institution: University of California, Davis
Language: English
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W4319063430