Disparities in Patient-Centered Communication via Telemedicine

Abstract

Purpose: This study investigated disparities in the uptake of telemedicine and the degree of patient-centeredness of telemedicine consultations among vulnerable patient populations. The focus includes rural adults and adults living with psychological distress and a high risk for chronic obstructive pulmonary disease (COPD).

Materials and Methods: In August 2020, a random sample of 932 U.S. adults ≥35 years old with a history of smoking tobacco completed an online survey. Chi-squared analyses were conducted to compare the sociodemographics of participants who did and did not use telemedicine. A series of analysis of variance tests were conducted to examine whether satisfaction with patient-centeredness of telemedicine consultations (i.e., open-endedness, expressed empathy, provider's ability, 5-point Likert scale) differs by rural/urban residence, psychological distress, and COPD risk.

Results: About 25% of the sample (n = 240) reported having used telemedicine. Telemedicine use was associated with younger age, Hispanic ethnicity, and moderate-to-high psychological distress, but not rurality. Participants reported high general satisfaction with the patient-centeredness of telemedicine consultations (M = 4.42 ± 0.73). However, high psychological distress and identifying as a current smoker were associated with less satisfaction across all domains. High COPD risk was uniquely associated with less satisfaction in how providers express empathy remotely.

Conclusion: Individuals with moderate-to-high psychological distress and a high risk for COPD experience challenges accessing high-quality, patient-centered care via telemedicine. As telemedicine becomes ubiquitous in health care, innovative solutions are needed to overcome barriers that prevent providers from delivering patient-centered care and patients from feeling satisfied with their remote consultations.

Introduction

The COVID-19 pandemic has accelerated widespread adoption of telemedicine, a technology that offers synchronous and evidence-based health care remotely via video consultations.¹ For example, privately billed insurance claims for telemedicine increased over 8,300% from April 2019 to April 2020,² the month that social distancing mandates and risk mitigation strategies were initially adopted. There are a number of benefits associated with telemedicine use, including its features to overcome physical and sociocultural barriers that have historically impeded access to care among vulnerable patient populations, including those residing in rural geographic regions.^1,3–5 However, despite increased access, patients express concerns about how the mediated experience may impact the patient/provider dynamic and the quality of care they receive.⁴ As such, simply increasing telemedicine access without monitoring the impact of remote consultations on patient/provider relationships conflates its potential for patient-centeredness.⁶

Patient-centered care is one of the six domains of health care quality, aiming to ensure that health care is responsive and inclusive of the needs, preferences, and values of patients.⁷ This form of care makes patients feel seen, engaged, and respected in their care.⁶ Levinson states that patient-centered care requires providers to “elicit patients' true wishes and to recognize and respond to both their needs and emotional concerns” (pg. 1).⁸ This includes welcoming questions, acknowledging patients' comments, and expressing empathy, which is considerably easier said than done well.⁸ The value of patient-centered communication has been well documented, including a variety of proximal and distal outcomes that are associated with positive patient/provider relationships, informed health decision-making, and desirable health outcomes.^6,8,9 There is evidence about patients being satisfied with their interactions via telemedicine,¹⁰ but questions remain about whether the technology is used to address the needs and values of patients.⁴ As such, research is needed to explore the patient-centered nature of telemedicine among vulnerable patient populations during the COVID-19 pandemic.

Vulnerable patient populations have unique health care needs, requiring more frequent, immediate, and specialized medical attention spanning over a period of time.¹¹ Chronic obstructive pulmonary disease (COPD), for example, is a progressive and incurable condition that requires continual self-management and treatment from a specialized provider. To combat concerns regarding access to care, telemedicine has been effective in disseminating health education about self-management behaviors, monitoring and detecting respiratory exacerbations early on for preventative purposes, and conducting smoking cessation and psychosocial counseling support.¹² While patients with COPD generally report a high rate of psychological distress,¹³ the COVID-19 pandemic resulted in a rapid spike in depression and anxiety among the general population, especially those with an underlying mental health concern.¹⁴ Mental health has been a leading health specialty in the use of telemedicine,¹⁵ and there is evidence supporting its effectiveness in diverse patient populations.¹⁶ Part of the appeal of telehealth solutions is its mediated context, where technological separation empowers patients to seek help for socially stigmatized conditions (e.g., mental health concerns, substance use).¹⁷ However, this same technological separation can also compromise the connections between patients and providers that facilitate meaningful patient-centered care.¹⁸

With the rapid shift to telemedicine during this pandemic, we must be careful to not alienate our most vulnerable patients once they gain access to care. As such, there is a need to understand how the human touch of patient-centered care is upheld during the transition to telemedicine from face-to-face formats among these populations.¹⁹ This study sought to investigate disparities in telemedicine uptake and satisfaction with the remote, patient-centered experience according to rural/urban residence, psychological distress, and COPD risk.

Materials and Methods

In August 2020, 932 adults completed a 15-min online survey in one sitting. We worked with Qualtrics Panels to identify a stratified, nationally representative random sample of adults according to their smoking tobacco behaviors (i.e., 50% of smokers who currently smoke every day or some days and 50% of people who have smoked 100 cigarettes in their life but no longer smoke) and their geographic region (50% self-identified as living in a rural or nonurban community and 50% self-identified as living in an urban or nonrural community). Qualtrics Panels draws from a large database of respondents and applies audience segmentation techniques to recruit random, nationally representative samples comprising the most specific, hard-to-reach subgroups.^20,21 Eligibility criteria for this particular study included being ≥35 years old, having smoked ≥100 cigarettes in their lifetime, and living without a COPD diagnosis. Anyone who answered opposite any of these criteria was excluded. The lead researcher secured Institutional Review Board (IRB) approval before data collection.

The cross-sectional survey included items to measure sociodemographic variables (e.g., age, gender, race, ethnicity, education, health insurance status), including geographic residence (i.e., zip code categorized as rural, small town, micropolitan, or metropolitan region based on USDA RUCA values) and status as a current or former smoker. Participants also completed the COPD-population screener (COPD-PS) Questionnaire²² and Serious Mental Illness (SMI) scale (K6).²³ If participants reported having used telemedicine in the past, they were invited to complete the Interview Satisfaction Questionnaire (ISQ)²⁴ to assess their satisfaction with the patient-centeredness of their telemedicine experiences, with strong attention to patient/provider communication.

The COPD-PS is a 5-item clinically validated instrument that measures an individual's risk for COPD according to age and various behavioral and symptom-related indicators.²² Sample items and weighted response options include the following: “During the past 4 weeks, how much of the time did you feel short of breath?” (none of the time = 0; a little of the time = 0; some of the time = 1; most of the time = 2; all of the time = 2) and “Do you ever cough up any ‘stuff,’ such a mucus or phlegm?” (no, never = 0; occasionally = 0; a few days a month = 1; most days a week = 1; yes, every day = 2). Following scoring procedures, an individual receives a score ranging from 0 to 10. For clinical consideration, the score is interpreted dichotomously as “low” (i.e., a score ranging from 0 to 4) and “high” (i.e., a score ranging from 5 to 10). The COPD-PS has demonstrated a high degree of concurrent validity and predictive power.^22,25,26

The SMI scale (K6)²³ is a 6-item psychological screening instrument that assesses the frequency of nonspecific distress related to anxiety and depression. The instrument was developed for use in population-based surveillance surveys, including the U.S. National Health Interview Survey. We asked participants to think about how they have been feeling during the past 30 days and indicate how often they have felt “nervous,” “hopeless,” “restless or fidgety,” “so depressed that nothing can cheer you up,” “that everything was an effort,” and “worthless.” Responses were anchored on a 5-point Likert scale (0 = none of the time; 4 = all of the time), where average scores were categorized as “low” (score = 0–4), “moderate” (score = 5–12), and “high” (score = 13–24). The moderate and high/severe thresholds have been recommended in prior research.^23,27 The instrument resulted in data with excellent internal consistency (Cronbach's alpha = 0.93).

The ISQ²⁴ is a 12-item instrument that assesses 4 dimensions of patient-centeredness. Dimensions included opportunities for open-endedness, empathy expressed by the provider, patient confidence in the ability of the provider to facilitate care, and general satisfaction with the provider during the telemedicine experience. The instrument's instructions were adapted to gauge satisfaction with patient-centeredness of telemedicine consultations. Participants were asked to think about when they have talked to a health care provider on a videoconference or video-chat and to indicate their level of agreement with the statements included in the instrument. Sample items from each dimension include the following: “I was able to tell the provider what was bothering me” (i.e., dimension 1: open-endedness); “The provider gave me undivided attention” (dimension 2: empathy); “I had a good deal of confidence in the provider” (dimension 3: confidence in the ability of the provider); and “Overall, I was satisfied with the provider” (dimension 4: general satisfaction). All items were anchored on a 5-point Likert scale (1 = strongly disagree to 5 = strongly agree). Internal consistency in the initial instrument development study ranged from 0.74 to 0.93. In this study, each dimension yielded adequate internal consistency (Cronbach's alpha = 0.86).

Frequency and descriptive statistics were computed with SPSS v27 to summarize the sample. Chi-squared analyses were conducted to examine differences in telemedicine use by sociodemographic variables, psychological distress, and COPD risk. ANOVA (analysis of variance) tests were conducted to examine average differences in communication satisfaction according to rural/urban residence, level of psychological distress, and risk for COPD. When applicable, Bonferroni post hoc tests were conducted to compare satisfaction between categories. Missing data were managed with pairwise deletion procedures.

Results

Participants were largely non-Hispanic (96.0%), white (90.6%), and older in age (M = 62.08 ± 10.13; 35–81). Over 90% of the participants reported zip codes residing in metro- or micropolitan regions. There were 240 (25.8%) telemedicine users, who consulted primary care providers (34.6%), mental health providers (26.3%), cardiologists (9.2%), and pulmonologists (3.0%). All participants had a history of smoking tobacco, with 50% identifying as current smokers (i.e., smoking tobacco every day or some days). Participants generally had a low COPD risk score (M = 4.13 ± 1.14), with only 25.6% classified as having a high COPD risk. Most participants were categorized as having low (69.1%) or moderate (23.3%) psychological distress.

Table 1 shows the sociodemographic characteristics of participants according to telemedicine use, which did not statistically significantly vary by gender, race, income, education, health insurance coverage, or rural/urban geographic region. Telemedicine use did vary by age and ethnicity. Among all the 35- to 44-year-old participants, 14.6% reported using telemedicine compared with the 6.7% who have not used it, χ² (2, N = 931) = 15.03, p < 0.001. Similarly, 5.5% of Hispanic adults reported using telemedicine compared with 2.6% who have not used it, χ² (2, N = 924) = 4.46, p < 0.05.

Table 1.

Sociodemographics of the Sample According to Telemedicine Use, N = 932

VARIABLE	NUMBER (%)
	TELEMEDICINE USER	NONTELEMEDICINE USER
Age, years
35–44	35 (14.6)**	46 (6.7)
45–64	80 (33.3)	277 (40.1)
65–81	125 (52.1)	368 (53.3)
Gender
Male	120 (50.4)	355 (51.4)
Female	118 (49.6)	335 (48.6)
Race
White	213 (88.8)	628 (91.3)
Black or African American	9 (3.8)	20 (2.9)
American Indian/Alaska Native	2 (0.8)	2 (0.3)
Asian/Pacific Islander	8 (3.3)	23 (3.3)
Multiracial	7 (2.9)	14 (2.0)
Other	1 (0.4)	1 (0.1)
Ethnicity
Hispanic	13 (5.5)*	18 (2.6)
Non-Hispanic	224 (94.5)	669 (97.4)
Income
Less than $20K	23 (9.7)	94 (13.7)
$20K–$34,999	28 (11.9)	100 (14.6)
$35K–$49,999	31 (13.1)	103 (15.1)
$50K–$74,999	58 (24.6)	149 (21.8)
$75K or more	96 (40.7)	238 (34.8)
Education
Less than high school	3 (1.3)	16 (2.3)
High school diploma/GED	37 (15.4)	141 (20.4)
Some college	71 (29.6)	225 (32.6)
College degree	129 (53.8)	309 (44.7)
Health insurance coverage
Purchased from employer	69 (28.7)	184 (26.7)
Purchased on own	15 (6.3)	45 (6.5)
Medicare	122 (50.8)	354 (51.3)
Medicaid	18 (7.5)	54 (7.8)
TRICARE	8 (3.3)	24 (3.5)
Other	2 (0.8)	8 (1.2)
Not insured	6 (2.5)	21 (3.0)
Rural/urban continuuma
Metropolitan	200 (84.7)	539 (79.5)
Micropolitan	21 (8.9)	81 (11.9)
Small town	10 (4.2)	29 (4.3)
Rural	5 (2.1)	29 (4.3)

Summed variable n-values may not equal 932 due to missing data.

^a USDA rural/urban continuum code derived from five-digit zip code.

^* p < 0.05; ^**p < 0.01.

GED, high school equivalency diploma.

We also examined telemedicine use according to COPD-PS risk score, SMI K6 psychological distress score, and smoking status (current vs. former). A greater proportion of individuals with moderate and high levels of psychological distress reported using telemedicine compared with their counterparts who do not use it, χ² (2, N = 922) = 35.22, p < 0.001. However, in looking at the proportion of telemedicine users, over half (54.2%) had a low degree of psychological distress, compared with those categorized as meeting criteria for moderate (32.6%) to high (13.1%) distress. Telemedicine use did not vary by smoking status or COPD-PS risk score.

Based on a 5-point Likert-type scale, participants reported considerably high general satisfaction with their telemedicine experience (M = 4.42; SD = 0.73). Consistently, there was high satisfaction with the open-endedness (M = 4.49 ± 0.64), expressed empathy (M = 4.50 ± 0.64), and demonstrated ability of the telemedicine provider (M = 4.49 ± 0.66). Table 2 shows that participants had greater satisfaction with their telemedicine experience if they were former smokers, F(1, 237) = 30.99, p < 0.001, had low psychological distress, F(1, 235) = 5.96, p < 0.01, and low risk for COPD, F(1, 215) = 6.10, p < 0.05. Participants with low psychological distress also had greater satisfaction with their telemedicine providers' open-endedness (p < 0.01), empathy (p < 0.01), and ability to care for them (p < 0.01). Users with low COPD risk also had high satisfaction with their telemedicine consultation. This was in regard to the empathic communication of their provider (p < 0.05).

Table 2.

Patient Satisfaction with Patient-Centeredness of Telemedicine Consultations

VARIABLE	OPEN-ENDEDNESS		EXPRESSED EMPATHY		PROVIDER'S ABILITY		GENERAL SATISFACTION
	M (SD)	95% CI	M (SD)	95% CI	M (SD)	95% CI	M (SD)	95% CI
Rural/urban continuuma
Metropolitan	4.50 (0.64)	4.41–4.59	4.51 (0.65)	4.42–4.60	4.52 (0.64)	4.42–4.60	4.43 (0.72)	4.33–4.53
Micropolitan	4.44 (0.75)	4.10–4.78	4.49 (0.70)	4.17–4.81	4.56 (0.70)	4.24–4.88	4.48 (0.87)	4.08–4.87
Small town	4.46 (0.36)	4.21–4.72	4.40 (0.52)	4.03–4.77	4.43 (0.45)	4.11–4.75	4.37 (0.51)	4.00–4.73
Rural	4.80 (0.30)	4.43–5.17	4.53 (0.65)	3.73–5.34	4.40 (0.83)	3.37–5.43	4.60 (0.55)	3.92–5.28
Smoking status
Current	4.23 (0.73)	4.09–4.36	4.27 (0.73)	4.13–4.40	4.23 (0.73)	4.09–4.36	4.16 (0.83)	4.00–4.32
Former	4.73 (0.43)**	4.65–4.80	4.70 (0.47)**	4.62–4.78	4.72 (0.47)**	4.64–4.81	4.66 (0.51)**	4.57–4.75
COPD riskb
High risk	4.37 (0.69)	4.19–4.55	4.33 (0.61)	4.17–4.49	4.38 (0.66)	4.21–4.55	4.23 (0.73)	4.03–4.41
Low risk	4.52 (0.65)	4.42–4.63	4.55 (0.67)*	4.45–4.66	4.53 (0.67)	4.42–4.64	4.50 (0.72)*	4.39–4.61
Psychological distressc
High	4.20 (0.73)	3.94–4.47	4.19 (0.64)	3.96–4.43	4.14 (0.78)	3.85–4.43	4.13 (0.90)	3.80–4.46
Moderate	4.38 (0.74)	4.22–4.55	4.33 (0.76)	4.16–4.50	4.37 (0.72)	4.20–4.53	4.30 (0.75)	4.13–4.47
Low	4.51 (0.53)**	4.52–4.70	4.66 (0.51)**	4.57–4.75	4.64 (0.53)**	4.55– 4.73	4.56 (0.63)**	4.45–4.67

Each scale is based on 5-point Likert-type (1 = strongly disagree; 5 = strongly agree); α = 0.86 for each scale.

^a U.S. Department of Agriculture Rural/Urban Continuum Code categories.

^b COPD-PS, where scores for “high risk” = 5–10 and “low risk” = 0–4.

^c Serious Mental Illness (K6), where scores for “low” = 0–4, “moderate” = 5–12, and “high” = 13–24.

^* p < 0.05; ^**p < 0.01.

CI, confidence interval; COPD-PS, chronic obstructive pulmonary disease population screener.

Discussion

Efforts to ensure equitable access to telemedicine across vulnerable patient populations remain a significant priority; however, attention must be paid to ensuring that all patients are obtaining patient-centered care once they access the remote services. We investigated the use of telemedicine among adults with a history of smoking tobacco and surveyed their satisfaction with the patient-centeredness of telemedicine consultations. Findings of this study provide insight to the quality-of-care disparities that persist via telemedicine despite increasing access among disadvantaged groups, including those with high psychological distress and a high risk for COPD.

Disparities in telemedicine use according to rural/urban residence and socioeconomic status did not exist. As expected,²⁸ 35- to 44-year-old participants were more likely to use telemedicine than not. However, a relatively similar proportion of 45- to 64-year-old and 65- to 81-year-old participants reported using telemedicine (33.3% and 52.1%, respectively), which was nearly double and triple the proportion of 35- to 44-year olds (14.6%). There was also an association between telemedicine use and identifying with the Hispanic ethnicity. This finding is supported by evidence that Hispanic adults are accepting of telemedicine and find it advantageous for overcoming barriers to health care access and to ensure greater precision of care.²⁹ Also consistent with prior research,¹⁰ participants who used telemedicine were generally satisfied with the patient-centered nature of their remote, video-based consultations. Unique to prior research, we found that satisfaction with the patient-centeredness of the remote consultation varied according to the degree of psychological distress and risk for COPD reported by participants.

Mental and behavioral health providers find telemedicine beneficial to their practice because they can see patients wherever they are and reach patients who otherwise would not receive care.³⁰ In this study, we found that participants with moderate and high psychological distress were more likely to use telemedicine than those with low psychological distress. However, participants with low-to-moderate psychological distress reported greater satisfaction with the patient-centeredness of telemedicine experiences. This included satisfaction with how their providers facilitated open-ended communication that was empathic, and how they demonstrated their ability to serve the patient. Telemedicine is an effective tool to facilitate effective mental and behavioral health treatment¹⁶; however, patients who may benefit the most from the expansion of telemedicine during the COVID-19 pandemic may be left behind when it comes to accessing high-quality, patient-centered care. This poses a significant concern for those who are most likely to report moderate-to-high psychological distress, including younger women, racial/ethnic minorities, and individuals with a low socioeconomic status.²⁷

Telemedicine use was not associated with COPD risk score or status as a tobacco smoker. However, nearly three-quarters (72.5%) of telemedicine users were assigned a low COPD risk score and 52.5% were former smokers. Similarly, adults with low COPD risk (e.g., younger, fewer respiratory symptoms, former smoker) reported a higher degree of general satisfaction with the patient-centeredness of their telemedicine consultations compared with those with high risk. Smoking status is an indicator of COPD risk; however, we examined its unique effect on patient-centeredness according to whether the participant is an ever-smoker (i.e., smoked 100 cigarettes in their life as per the COPD-PS Questionnaire²²) and a current (i.e., every day, some day) or former smoker. We found that participants who were ever-smokers but did not meet the respiratory conditions indicative of high COPD risk felt high satisfaction with how their provider expressed empathy during the remote consultation. Extracting the unique effect of smoking status on patient-centeredness, we found that former smokers reported having a higher degree of satisfaction with the open-endedness and ability of their provider compared with current smokers. Current smokers experience considerable stigma in health care contexts, which may stifle open and empathic communication about respiratory symptoms and risk for COPD with their providers, both of which are essential for timely diagnosis and treatment shared decision-making.³¹ Greater attention to patient-centered interactions during telemedicine consultations with current tobacco smokers will be important for promoting the timely detection and continual self-management of COPD among those at the highest risk.

There are limitations that should be acknowledged. Despite the rapid shift toward telemedicine during the COVID-19 pandemic, only 25% of participants in this study reported having received health care services via telemedicine (i.e., video-based consultations). Although this statistic appears considerably low, research has demonstrated that patients are using multiple venues of telehealth to access and engage in health care with their providers, including electronic health record secure messaging and mobile health applications.¹⁸ Also, this is a cross-sectional study and results should not be generalized for all adults with a history of smoking tobacco or the duration of the pandemic. Longitudinal research examining trends in psychological distress and comparing levels of satisfaction with the patient-centeredness of in-person versus remote care would be useful. Other variables of interest to pursue during this research would include experience using telemedicine and the prior relationship with the health care provider seen via telemedicine.

As telemedicine becomes ubiquitous in health care, inequities that compromise access to high-quality care will prevail among vulnerable patient populations. Older age and low health literacy are among the most common patient-centric factors that contribute to poor telemedicine acceptance.¹⁹ Providers, however, also report hesitancy using telemedicine because of health insurance reimbursement delays and because they feel the quality of their care and connections with patients are deteriorating given the physical separation.^4,18,19 As a result, telemedicine consultations have been generally provider-centered, rather than patient-centered.^32,33 To optimize the potential of telemedicine to assist all patients equitably, innovative solutions are needed to overcome barriers that prevent providers from delivering patient-centered care and patients from feeling satisfied with their remote consultations. One potential solution that warrants future investigation is to explore technologies that supplement or simulate in-person interactions during telemedicine consultations in attempts of avoiding violations to the ritualistic expectations of clinical visits (e.g., nurse intake, vital signs).¹⁸ Another is to develop and evaluate evidence-based health care communication protocols for telemedicine practice to help providers create a patient-centered experience.

Authors' Contributions

Conceptualization: S.R.P., B.E.B., and C.L.B. methodology: S.R.P., B.E.B., and C.L.B.; formal analysis: S.R.P.; investigation: S.R.P.; resources: S.R.P.; data curation: S.R.P.; writing—original draft preparation: S.R.P., B.E.B., and C.L.B.; writing—review and editing: S.R.P., B.E.B., and C.L.B. All authors have read and agreed to the published version of the article.

Disclosure Statement

Drs. Paige and Bunnell are employed by Doxy.me, LLC. Dr. Bylund does not have a conflict of interest.

Funding Information

This research was supported by the National Heart, Lung, and Blood Institute (Grant No. F32HL143938; PI Samantha R. Paige) and the National Institute of Mental Health (Grant No. MH118482; PI Brian E. Bunnell).