Abstract
This study explored the relationship between trust in physicians and telehealth use during the COVID pandemic in 162 African Americans with diabetes. More than 90% of patients had internet-capable devices and internet service but only 61 patients (39%) had a telehealth visit. Compared to the latter, participants with no telehealth visits had less trust in physicians' ability to diagnose COVID, less trust in physicians' ability to treat via telehealth, and resided in more deprived neighborhoods. There were no differences in age, sex, education, nor literacy. For African Americans with diabetes, health disparities may increase unless fundamental issues such as trust are addressed.
Keywords: African Americans, telehealth, trust
Introduction
The COVID-19 pandemic unmasked substantial health disparities, including the use of telehealth to deliver care.1 Prior to the COVID pandemic, telehealth use was lower in African Americans than Whites (30.1% vs. 38.6%, respectively), and the racial disparity widened during the pandemic (40.1% vs. 60.7%, respectively) and contributed to worse health outcomes in African Americans with diabetes.2–4 Health and computer literacy (eg, lack of internet connectivity and devices) may contribute to the disparity, but the patient-physician construct also may play a role.
This study explored the relationship between trust in physicians, neighborhood deprivation, and telehealth use during the pandemic in African Americans with diabetes.
Methods
From March to April 2020, race-concordant community health workers (CHWs) phone-interviewed a convenience sample of African Americans with diabetes (N = 162) who were participating in diabetes-related research projects (ClinicalTrials.gov NCT03393338 and NCT03466866). The CHWs assessed whether participants had: a telehealth visit with their provider since the pandemic; an internet-compatible device (eg, smartphone); internet availability; knowledge of COVID prevention practices (eg, wear face mask, social distancing) and COVID symptoms (eg, fever, difficulty breathing); trust in physicians' ability to diagnose COVID; and trust in physicians' ability to treat (any condition) via a video visit. Trust items were rated from 1 (not at all) to 10 (extremely). Thomas Jefferson University Institutional Review Board approval was obtained and all participants provided verbal informed consent. Previously collected data included demographic characteristics, health literacy (Literacy Assessment for Diabetes), depressive symptoms (Patient Health Questionairre-9), cognition (short-Montreal Cognitive Assessment), and hemoglobin A1c level. Participants' home addresses were used to generate the Area Deprivation Index score, which represents a neighborhood's relative deprivation or privilege based on income, employment, education, and housing quality.5 Statistical tests included one-way analysis of variance for continuous data, and cross tabulations for categorical variables.
Results
Of 157 participants, 145 (93%) had smartphones; 142 (92%) had internet service; and 136 (87%) could use a smartphone for video calls. Despite this high level of potential access, only 61 participants (39%) had a telehealth visit. Participants with no telehealth visits had lower mean trust scores in physicians' ability to diagnose COVID (7.7 [95% CI 7.2, 8.3] vs. 8.6 [95% CI 8.0, 9.1]), respectively [F = 4.25 (1,156), P < 0.04]; lower trust in physicians' ability to treat any condition via a video visit (6.9 [95% CI 6.4, 7.5] vs. 7.9 [95% CI 7.4, 8.5]), respectively [F = 6.32 (1,156), P < 0.03]; and higher (worse) Area Deprivation Index scores (8.3 [95% CI 7.9, 8.7] vs. 7.5 [95% CI 6.8, 8.2]), respectively [F = 4.86 (1,156), P < 0.03] than participants who had telehealth visits (Table 1). There were no differences in age, sex, education, literacy, hemoglobin A1c, nor knowledge of COVID prevention or COVID symptoms.
Table 1.
Comparison of Participants With and Without Telehealth Visits
| No telehealth visit (n = 96) |
Telehealth visit (n = 61) |
P value | |
|---|---|---|---|
| Age, yearsa | 60.4 (58.2, 62.6) | 59.9 (58.6, 62.1) | .75 |
| Sex, (n, %), female | 65 (67.7%) | 43 (70.5%) | .71 |
| Education, yearsa | 13.0 (12.6, 13.3) | 13.3 (12.8, 13.8) | .33 |
| Area Deprivation Indexa,b | 8.3 (7.9, 8.7) | 7.5 (6.8, 8.2) | .03 |
| Literacya,c | 49.1 (47.3, 50.9) | 50.9 (48.9, 52.9) | .19 |
| PHQ-9a,d | 7.7 (6.6, 8.7) | 7.9 (6.5, 9.2) | .84 |
| s-MoCAa,e | 9.8 (9,3, 10.4) | 10.1 (9.5, 10.7) | .51 |
| Hemoglobin A1ca | 8.6 (8.1, 9.0) | 8.5 (7.9, 9.0) | .77 |
| Trust in Physician Ability to Diagnose COVIDa,f | 7.7 (7.2, 8.3) | 8.6 (8.0, 9.1) | .04 |
| Trust in Physician Ability to Treat via Video Visitsa,f | 6.9 (6.4, 7.5) | 7.9 (7.4, 8.5) | .01 |
| Knowledge of COVID Symptomsa,g | 5.8 (5.5, 6.0) | 6.1 (5.9, 6.3) | .09 |
| Knowledge of COVID Preventiona,h | 6.9 (6.7, 7.1) | 7.0 (6.6, 7.3) | .75 |
Mean (95% Confidence Interval).
Pennsylvania State Decile Score. Scores range from 1 to 10, with higher scores indicating greater disadvantage.
Literacy Assessment for Diabetes. Scores range from 0 to 60, with higher scores indicating better literacy.
PHQ-9. Scores range from 0 to 27, with higher scores indicating more depressive symptoms.
s-MoCA. Scores range from 0 to 16, with higher scores indicating better cognitive function.
Rated from 1 (not at all) to 10 (extremely).
Rated from 0 to 7, with higher scores indicating greater knowledge.
Rated from 0 to 8, with higher scores indicating greater knowledge.
PHQ-9, Patient Health Questionnaire-9; s-MoCA, Short-Montreal Cognitive Assessment.
Discussion
In this sample of African Americans with diabetes, physician mistrust and neighborhood deprivation were associated with low telehealth use, and not associated with demographic characteristics, health literacy, glycemic control, depression, nor cognition. Access to telehealth-enabled devices was not an obstacle, as more than 90% of participants had internet-capable devices and internet service and 87% knew how to use smartphones for video calls. Together, these data suggest that telehealth use by African Americans is more nuanced than having access to the necessary technology, and implicate aspects of the physician-patient relationship.
This exploratory study is limited by the small sample, uncertain generalizability, and lack of comparable data in Whites. Nevertheless, the findings indicate that physician mistrust may be a determinant of telehealth use in underprivileged African Americans and that it needs to be better understood. Telehealth has the potential to increase or reduce health care inequities. For African Americans with diabetes, health disparities may increase unless fundamental issues such as trust are addressed.
Conclusion
Physician mistrust and neighborhood deprivation are associated with low telehealth use in African Americans with diabetes. Health disparities may increase unless fundamental issues such as trust are addressed in this high-risk population.
Authors' Contributions
Drs. Rovner, Casten, Chang, Hollander, and Rising all participated in study concept and design; acquisition, analysis, or interpretation of data; drafting and critically revising the manuscript; obtaining funding: Statistical analysis: Dr. Casten.
Author Disclosure Statement
The authors have no conflicts of interest to disclose.
Funding Information
This research was supported by a grant from the Pennsylvania Department of Health (SAP #400077081) and the National Institute of Diabetes and Digestive and Kidney Diseases (R01DK114033).
References
- 1. Rivera V, Aldridge MD, Ornstein K, Moody KA, Chun A. Racial and socioeconomic disparities in access to telehealth. J Am Geriatr Soc 2021;69:44–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Anthony DL, Campos-Castillo C, Lim PS. Who isn't using patient portals and why? Evidence and implications from a national sample of US adults. Health Aff (Millwood) 2018;37:1948–1954. [DOI] [PubMed] [Google Scholar]
- 3. Lam K, Lu AD, Shi Y, Covinsky KE. Assessing telemedicine unreadiness among older adults in the United States during the COVID-19 pandemic. JAMA Intern Med 2020;180:1389–1391. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Centers for Disease Control and Prevention. COVID-19 cases, data, and surveillance: Hospitalization and death by race/ethnicity. 2020. https://www.cdc.gov/coronavirus/2019-ncov/covid-data/investigationsdiscovery/hospitalization-death-by-race-ethnicity.html Accessed October 10, 2020.
- 5. Kind A, Buckingham W. Making neighborhood-disadvantage metrics accessible—the neighborhood atlas. N Engl J Med 2018;378:2456–2458. [DOI] [PMC free article] [PubMed] [Google Scholar]