Virtual Savings: Patient-Reported Time and Money Savings from a VA National Telehealth Tablet Initiative

Abstract

Virtual care holds promise for offering services to Veterans Affairs (VA) patients who have barriers to accessing care. In 2016, the VA began issuing video-enabled tablets to Veterans with geographic, clinical, and/or social barriers to in-person care. To complement a national evaluation of system-level implementation and effectiveness of these tablets, we sought to understand whether the VA-issued tablets generated money and/or time savings for patients. We distributed a survey to 2,120 Veterans who received tablets and administered a follow-up survey 3–6 months later. The final analysis included 594 and 399 patients who responded to questions about money and time savings, respectively. We used poststratification survey weighting methods to address potential selection and nonresponse bias. In multinomial logistic regressions and logistic regressions, we examined patient characteristics associated with reported money and time savings. A majority of survey respondents reported that the tablets saved them money (89%) and time (71%). Respondents were more likely to report monetary savings if they lived at a greater distance from the VA, if they experienced travel barriers, and if they did not have a mental health condition. Respondents were more likely to report time savings if they were <45 or ≥65 years of age, employed, and reported more overall technology experience. Findings may inform policy decisions regarding patient targeting and training as VA aims to expand its use of video telehealth technology.

Introduction

Growing evidence supports the effectiveness of telehealth in increasing patient access to care.^1
–3 Evidence regarding cost savings has been mixed, however, and is often dependent on whether the analyses are conducted from the perspective of the system (i.e., payer) or from the societal perspective, which includes the patient perspective.⁴ Many of the benefits attributed to telehealth interventions relate to improving access and decreasing costs for patients and their families.^3,5,6 One systematic review of real-time video interventions found that whenever the patient perspective was included, telehealth was found to be cost saving, whereas when the payer perspective was considered, the proportion of studies reporting cost savings was halved.⁷ Although there is increasing evidence highlighting the impact of including the patient perspective in telehealth cost analyses, there has been less emphasis on the patient characteristics that are associated with monetary and time savings.

In 2016, the Department of Veterans Affairs (VA) launched a pilot program that provided video-enabled tablets to Veterans with geographic, clinical, and/or social barriers to accessing VA health care. The devices were distributed to Veterans who lacked their own device or broadband connectivity and supported the use of peripheral devices (e.g., blood pressure cuffs) and video visits. They also provided access to VA applications and the VA patient portal.⁸ As part of a national evaluation of this initiative, a patient experience survey was conducted with a subset of tablet recipients. In this article, we describe patient-reported monetary and time savings and characteristics associated with those savings.

Methods

Study Population

VA tablet recipients received a baseline survey (N = 2,120) with their tablet shipment between April 1 and September 30, 2017, and 1,294 responded to this survey. A follow-up survey was mailed 3–6 months later (matched n = 764). These analyses include data from 594 and 399 individuals who responded to questions about monetary costs and time savings, respectively ( Appendix Fig. A1 provides a survey response flowchart).

Dependent Variables

Patient-reported savings were determined by asking patients if they agreed that tablets saved them money and time. Responses were dichotomized as “strongly agree or agree” versus “neutral, disagree, or strongly disagree.” Patients were asked to consider transportation, gas, lodging, and food costs and to quantify how much money they saved per appointment; responses were categorized as <$25, $25–50, and >$50. For time savings, patients were asked to report whether they avoided missing work due to the tablets and whether the time saved was “paid” or “unpaid.”

Independent Variables

Variables from the VA electronic health record included age, gender, distance to the VA facility where most care was received, and chronic physical and mental health conditions. Patients self-reported employment status, income, whether they lived alone, previous technology experience, and barriers to accessing VA care (i.e., transportation, outside commitments, and feeling uncomfortable at VA).

Statistical Analysis

We examined the proportion of respondents reporting monetary and time savings and used multinomial logistic regression analysis to determine characteristics associated with higher monetary savings per appointment ($25–50 and >$50) compared with a base category of lower savings (<$25). A sensitivity analysis using an ordered logistic regression is presented in Appendix Table A1. We conducted logistic regression analyses to examine characteristics associated with any time savings and the type of time saved. To address concerns about nonresponse bias, we conducted poststratification weighting that adjusted for the following significant differences between respondents and nonrespondents at baseline: age, rural residence, and mental health conditions.

Results

In total, 92% of respondents (566/617) reported that the tablets saved them money or time; 89% (527/594) reported saving money, and 71% (284/399) reported saving time. Among those who reported monetary savings, 41% reported saving $25–50 and 31% reported saving >$50 per appointment. Among 160 respondents who reported time savings and were employed, 45% reported saving paid time.

Predictors of Monetary Savings

Relative to Veterans who lived within 15 miles of a VA facility, those who lived >40 miles away were more likely to indicate that tablets save them $25–50 (relative risk ratio [RRR]: 4.62, p < 0.0001) and >$50 (RRR: 4.56, p < 0.0001), as opposed to <$25 (Table 1). Veterans with a mental health condition were less likely than those without one to report saving >$50 (RRR: 0.51, p = 0.038). Veterans who reported transportation barriers were more likely to save >$50 as opposed to <$25 (RRR: 3.34, p = 0.011) relative to Veterans without barriers.

Table 1.

Multinomial Logistic Regression Results of Factors Associated with Money Saving

	MONETARY SAVINGS (REFERENCE GROUP: <$25), N = 493
	$25–50		>$50
	RRR (95% CI)	p	RRR (95% CI)	p
Age category, years
18–44	Reference		Reference
45–64	1.00 (0.50–2.00)	1.00	0.89 (0.43–1.86)	0.762
65+	0.88 (0.39–1.96)	0.752	0.64 (0.28–1.49)	0.300
Employment
Not working	Reference		Reference
Working	1.47 (0.81–2.68)	0.202	0.95 (0.51–1.77)	0.871
General health conditions
Good–excellent	Reference		Reference
Poor–fair	0.71 (0.43–1.15)	0.164	1.63 (0.88–3.00)	0.074
Mental health conditions
No	Reference		Reference
Yes	0.88 (0.48–1.62)	0.685	0.51 (0.27–0.96)	0.038
Technology experiences
Not any	Reference		Reference
1–3 types	0.60 (0.28–1.29)	0.190	1.05 (0.45–2.51)	0.910
4–8 types	0.76 (0.34–1.71)	0.508	1.16 (0.46–2.93)	0.751
Distance to VA facilities
≤15 miles	Reference		Reference
15–40 miles	1.64 (0.9–2.99)	0.108	1.01 (0.51–2.00)	0.968
>40 miles	4.62 (2.42–8.83)	<0.0001	4.56 (2.31–8.98)	<0.0001
Travel barriers^a
No	Reference		Reference
Yes	1.72 (0.87–3.4)	0.120	3.34 (1.31–8.49)	0.011
Commitment barriers^b
No	Reference		Reference
Yes	0.71 (0.41–1.23)	0.219	0.99 (0.56–1.74)	0.967
Uncomfortable barriers^c
No	Reference		Reference
Yes	0.95 (0.55–1.63)	0.858	1.38 (0.77–2.46)	0.277

N.B.: Regression models also adjusted for gender, income, and whether the respondent lived alone.

Travel barriers include having problems with travel time to the VA facility where the respondent receives care, difficulty getting transportation to the VA where the respondent receives care, cost of traveling to VA where the respondent receives care, health conditions that make it difficult for the respondent to get the health care they need, and bad weather conditions.

Commitment barriers include having problems with work or school making it difficult for the respondent to get the health care they need, family/caregiving responsibilities making it difficult for the respondent to get the health care they need.

Uncomfortable barriers include having problems related to feeling out of place, uncomfortable, or uneasy at the VA.

CI, confidence interval; RRR, relative risk ratio; VA, Veterans Affairs.

Predictors of Time Savings

Younger (odds ratio [OR]: 2.11, p = 0.029) and older (OR: 3.38, p < 0.0001) age groups were more likely to report time savings relative to middle-aged Veterans (Table 2). Being employed (OR: 5.13, p < 0.0001) and prior use of technology (OR: 3.16, p = 0.019) were associated with a higher likelihood of time savings. Among employed Veterans, no predictors were significantly associated with taking paid versus unpaid leave (Table 2).

Table 2.

Logistic Regression Results of Factors Associated with Time Savings

	ANY TIME SAVINGS, N = 373		TYPE OF TIME SAVED (PAID AS REFERENCE), N = 154
	OR (95% CI)	p	OR (95% CI)	p
Age category, years
18–44	2.11 (1.08–4.10)	0.029	Reference
45–64	Reference		0.42 (0.17–1.03)	0.058
65+	3.38 (1.79–6.38)	<0.0001	0.44 (0.15–1.33)	0.146
Employment
Not working	Reference		Reference
Working	5.13 (2.62–10.02)	<0.0001	1.04 (0.43–2.51)	0.931
General health conditions
Good–excellent	Reference		Reference
Poor–fair	1.17 (0.66–2.07)	0.582	1.34 (0.58–3.11)	0.493
Mental health conditions
No	Reference		Reference
Yes	0.67 (0.37–1.22)	0.192	1.60 (0.65–3.95)	0.306
Technology experiences
Not any	Reference		Reference
1–3 types	1.76 (0.75–4.16)	0.196	0.39 (0.12–1.21)	0.102
4–8 types	3.16 (1.21–8.28)	0.019	0.69 (0.21–2.24)	0.532
Distance to VA facilities
≤15 miles	Reference		Reference
15–40 miles	1.42 (0.70–2.88)	0.327	1.93 (0.67–5.56)	0.219
>40 miles	1.11 (0.57–2.20)	0.749	1.47 (0.59–3.69)	0.404
Travel barriers^a
No	Reference		Reference
Yes	0.56 (0.24–1.33)	0.187	1.67 (0.56–4.97)	0.352
Commitment barriers^b
No	Reference		Reference
Yes	0.70 (0.39–1.25)	0.229	0.49 (0.20–1.20)	0.117
Uncomfortable barriers^c
No	Reference		Reference
Yes	1.23 (0.70–2.17)	0.475	0.62 (0.25–1.52)	0.297

N.B.: We also adjusted for gender, income, and whether the respondent lived alone.

Uncomfortable barriers include having problems related to feeling out of place, uncomfortable, or uneasy at the VA.

OR, odds ratio.

Discussion

Most survey respondents reported money and time savings associated with virtual care. Monetary savings were most pronounced among Veterans living a greater distance from VA or experiencing travel barriers and those without mental health conditions. Time savings were most pronounced among both younger and older Veterans, employed Veterans, and those with greater technological experience.

Prior studies have reported that telemedicine can be cost-effective^2,9

–12 and can increase access to care³ for a wide range of conditions. Other studies have found that telemedicine may increase costs when the evaluation is conducted from the perspective of the health care system.^3,6 This study adds to existing literature by focusing on personal resources—tablets issued by a health care system can save specific groups of patients both money and time. It also demonstrates that relatively simple survey strategies can be used to collect patient cost information when more complex valuation approaches are not feasible. Our approach enabled us to estimate patient time and money savings using broad categories; however, our methods do not measure the value of the tablet initiative, which could be elicited using a willingness to pay approach.¹³

When patients experience distance and travel barriers, adopting only a payer perspective in telehealth cost analyses may overlook important savings from the societal perspective. With almost 5 million (24.1%) U.S. Veterans living in rural areas,¹⁴ it is likely that many of these individuals experience barriers to accessing VA care; measuring their potential benefits from telehealth interventions is crucial for policy makers. Future studies should also consider how rural residence may impact telecommunications connectivity. Prior work among this population has found that broadband connectivity issues in rural areas are still a barrier to receiving telehealth for many Veterans.⁸ Our finding that those with mental health conditions were less likely to report monetary savings might be due to the fact that Veterans with mental health conditions in our sample were 1.2 times more likely to receive VA travel reimbursement. Our finding that technological experience predicts time savings suggests that health systems can optimize patient time savings by emphasizing baseline technology training.

Interpretation of these findings may be limited by nonresponse bias. We addressed differences in demographic characteristics between respondents and nonrespondents using poststratification survey weighting; however, the groups still differed with respect to tablet usage. Respondents were 6.3 percentage points more likely to be tablet users, which could upwardly bias the observed monetary and time savings. We also note a possible downward bias of the monetary savings estimates, since we did not include savings associated with visit co-pays. This may affect up to 19.6% of patients whose VA Priority Group status indicated the possibility of co-payments for traditional clinic visits.

Overall, our study provides valuable information about the monetary and time savings experienced by Veterans who receive VA-issued video-enabled tablets. Findings may inform policy decisions regarding patient targeting and training as VA aims to expand its use of video telehealth technology.

Footnotes

Acknowledgments

We acknowledge the contributions of Liberty Greene, MS, and Tolessa Gurmessa, MD, data analysts at the VA Palo Alto Center for Innovation to Implementation, and Rachel Kimerling, PhD, and Daniel Blonigen, PhD, who contributed to survey instrument development. Views expressed are those of the authors and do not necessarily represent views of the Department of Veterans Affairs.

Disclosure Statement

No competing financial interests exist.

Funding Information

This study was supported by VA's Office of Rural Health Enterprise Wide Initiative and the eHealth Partnered Evaluation Initiative (QUERI, PI Timothy Hogan, Bedford Mass).

Appendix

Appendix Table A1.

Ordered Logistic Regression Results of Factors Associated with Money Saving

	OR (95% CI)	p
Age category, years
18–44	Reference
45–64	0.92 (0.56–1.52)	0.752
65+	0.77 (0.44–1.34)	0.356
Employment
Not working	Reference
Working	1.1 (0.74–1.66)	0.632
General health conditions
Good–excellent	Reference
Poor–fair	1.43 (1–2.03)	0.047
Mental health conditions
No	Reference
Yes	0.61 (0.4–0.94)	0.026
Technology experiences
Not any	Reference
1–3 types	1.12 (0.69–1.83)	0.646
4–8 types	1.22 (0.71–2.11)	0.473
Distance to VA facilities
≤15 miles	Reference
15–40 miles	1.11 (0.64–1.91)	0.716
>40 miles	2.71 (1.65–4.44)	<0.0001
Travel barriers^a
No	Reference
Yes	2.17 (1.24–3.81)	0.007
Commitment barriers^b
No	Reference
Yes	1.02 (0.69–1.51)	0.913
Uncomfortable barriers^c
No	Reference
Yes	1.27 (0.84–1.92)	0.251

N.B.: Regression models also adjusted for gender, income, and whether the respondent lived alone.

Uncomfortable barriers include having problems related to feeling out of place, uncomfortable, or uneasy at the VA.

CI, confidence interval; RRR, relative risk ratio; VA, Veterans Affairs.

References

Doolittle

, Spaulding

, Williams

. The decreasing cost of telemedicine and telehealth. Telemed J E Health, 2011; 17:671–675.

Fletcher

, Hogan

, Keegan

, et al. Recent advances in delivering mental health treatment via video to home. Curr Psychiatry Rep, 2018; 20:56.

Ashwood

, Mehrotra

, Cowling

, Uscher-Pines

. Direct-to-consumer telehealth may increase access to care but does not decrease spending. Health Aff (Millwood), 2017; 36:485–491.

Jacobs

, Barnett

. Emergent challenges in determining costs for economic evaluations. Pharmacoeconomics, 2017; 35:129–139.

Jones

, Etherage

, Harmon

, Okiishi

. Acceptability and cost-effectiveness of military telehealth mental health screening. Psychol Serv, 2012; 9:132–143.

Dixon

, Hollinghurst

, Edwards

, et al. Cost-effectiveness of telehealth for patients with depression: Evidence from the Healthlines randomised controlled trial. BJPsych Open, 2016; 2:262–269.

Wade

, Karnon

, Elshaug

, Hiller

. A systematic review of economic analyses of telehealth services using real time video communication. BMC Health Serv Res, 2010; 10:233.

Zulman

, Wong

, Sligtham

, et al. Making connections: Nationwide implementation of video telehealth tablets to address access barriers in veterans. JAMIA Open, 2019; 2:323–329.

Lindgren

, Wacker

, Suess

, et al. Telehealth and autism: Treating challenging behavior at lower cost. Pediatrics, 2016; 137(Suppl. 2):S167–S175.

10.

Forbes

, Broman

, Johnson

, et al. Implementation of telehealth is associated with improved timeliness to kidney transplant waitlist evaluation. J Telemed Telecare, 2018; 24:485–491.

11.

Grustam

, Severens

, van Nijnatten

, et al. Cost-effectiveness of telehealth interventions for chronic heart failure patients: A literature review. Int J Technol Assess Health Care, 2014; 30:59–68.

12.

Darkins

, Kendall

, Edmonson

, et al. Reduced cost and mortality using home telehealth to promote self-management of complex chronic conditions: A retrospective matched cohort study of 4,999 veteran patients. Telemed J E Health, 2015; 21:70–76.

13.

Snoswell

, Smith

, Scuffham

, Whitty

. Economic evaluation strategies in telehealth: Obtaining a more holistic valuation of telehealth interventions. J Telemed Telecare, 2017; 23:792–796.

14.

Holder

. Veterans in rural America: 2011–2015. In: US Census Bureau American Community Survey Reports. Washington DC: US Census Bureau, 2017.