Health Care Utilization and Costs of Veterans Evaluated for Traumatic Brain Injury Through Telehealth

Abstract

Background:

Mild traumatic brain injury (TBI) is prevalent among Afghanistan (Operation Enduring Freedom [OEF]) and Iraq (Operation Iraqi Freedom [OIF]) Veterans. With clinical video telehealth (CVT), Veterans screening positive for potential deployment-related TBI can receive comprehensive TBI evaluations by providers at specialized centers through interactive video communication.

Introduction:

We examined health care utilization and costs for Veterans during the 12 months before and after being evaluated through CVT versus in-person.

Materials and Methods:

We examined OEF/OIF Veterans receiving comprehensive evaluations at specialized Veterans Affairs facilities from October 2012 to September 2014. Veterans evaluated through CVT and in-person at the same facilities were included. We used a difference-in-difference analysis with propensity score weighted regression models to examine health care utilization and costs between TBI evaluation groups.

Results:

There were 554 Veterans with comprehensive evaluations through CVT (380 with and 174 without confirmed TBI) and 7,159 with in-person evaluations (4,899 with and 2,260 without confirmed TBI). Veterans in the in-person group with confirmed TBI had similar increases in outpatient, inpatient, and total health care costs as Veterans who had TBI confirmed through CVT. However, Veterans with a confirmed TBI evaluated in-person had greater increases in rehabilitation and other specialty costs.

Discussion:

When visits are in-person, Veterans may have opportunities to discuss more issues and concerns, whether TBI-related or not. Thus, providers might make more referrals to rehabilitation and specialty care after in-person visits.

Conclusion:

Veterans receiving in-person evaluations who were diagnosed with TBI had similar increases in health care costs as Veterans with TBI confirmed through evaluations through CVT.

Introduction

Traumatic brain injury (TBI) is an alteration of brain function or other evidence of brain pathology, that is caused by external force.¹ TBI can range from mild to severe resulting from injuries caused by incidents including falls, vehicle crashes, and being hit by objects.^2,3 For Veterans of the conflicts in Afghanistan and Iraq (Operation Enduring Freedom [OEF], Operation Iraqi Freedom [OIF], and Operation New Dawn [OND]), between 15% and 20% may have experienced a mild TBI while being deployed.^2,4

–8 Mild TBI is a condition that can manifest as affective, somatic, and cognitive symptoms (e.g., headaches, problems with sleep, balance, and/or memory, irritability, and sensitivity to light).⁹ Although these symptoms usually resolve in a matter of hours to a few months, some of these symptoms may persistent for prolonged periods and become chronic.¹⁰

Veterans with TBI generally use more services with higher health care costs than Veterans not diagnosed with TBI.¹¹ To identify Veterans who may benefit from treatment and services, the United States Department of Veterans Affairs (VA) implemented a national clinical reminder in April 2007 to screen for TBI.¹² Because the presence of other symptoms or conditions (e.g., depression, post-traumatic stress disorder [PTSD], pain, and insomnia) may result in a positive screening on the TBI clinical reminder,^6,13
–15 a positive screen is not a definitive diagnosis of TBI. Consequently, there is a need to ensure access to comprehensive evaluations for thorough assessments and development of appropriate treatment plans. Veterans with positive results on the TBI clinical reminder are referred to a multidisciplinary team for a comprehensive TBI evaluation.

The comprehensive evaluation is conducted by TBI specialists (physiatrist, neurologist, and neuropsychiatrist) who examine etiological variables associated with TBI severity, neurological sequelae, and symptoms. These practitioners also conduct physical examinations and pharmacological reviews.¹⁴ However, Veterans also face numerous challenges that may make it difficult for those with possible TBI to be evaluated in-person at a VA facility. Previous studies suggest that PTSD-related problems with anger, trust, and social situations, as well as mild TBI-related problems with memory and attention, negatively impact appointment attendance, as can other factors including distance to facilities, frequent appointments, competing daily demands, and financial constraints.¹⁵

To improve access to comprehensive evaluations, the VA Physical Medicine and Rehabilitation Service initiated a pilot program to implement a standardized protocol for providing comprehensive evaluations through clinical video telehealth (CVT). Telehealth has been used across a variety of medical domains (speech pathology, dermatology, atrial fibrillation, etc.) with comparable success to face-to-face appointments across private sector and VA patients.^16

–19 Telehealth has been particularly beneficial for improving care in rural areas with patients reporting easier access to specialty care and satisfaction with the care they received.¹⁷

With CVT, Veterans can travel to their nearest VA medical center or VA community-based outpatient clinic to complete comprehensive TBI evaluations with providers through interactive video communication. Although use of CVT may increase access to comprehensive evaluations, it is unclear whether subsequent health care use differs between those who received the comprehensive evaluation through CVT and those who received in-person evaluations. The aim of this study was to assess the association of CVT use for the comprehensive TBI evaluation with subsequent health care utilization and costs.

Materials and Methods

Study Design and Sample

The study examined the health care utilization and costs of a sample of OEF/OIF/OND Veterans who received comprehensive TBI evaluations in the United States through CVT (between the Veterans' local VA facilities and a distant VA site) or in-person between October 1, 2012 and September 30, 2014. We examined associations between the method of delivery (CVT versus in-person) and changes in health care utilization and costs over 12-month periods before and after the comprehensive evaluation. This study was approved by the Edward Hines, Jr. VA Hospital Institutional Review Board.

We identified Veterans in the VA Managerial Cost Accounting (MCA) National Data Extracts (NDEs) who visited a VA polytrauma clinic (VA clinic stop code 197) and who also had a secondary clinic code indicating CVT (VA secondary clinic codes 690, 692, and 693). We determined which of these Veterans received a comprehensive evaluation at the polytrauma clinic visit using information from the national comprehensive TBI evaluation database managed by Veterans Health Administration's Office of Patient Care Services. We also sampled Veterans who had their comprehensive evaluation in-person from the same VA locations as those in our CVT group, resulting in a sample of 7,713 Veterans.

Data Sources and Measures

Results of each Veteran's comprehensive evaluation were abstracted from the comprehensive evaluation database, derived from VA's electronic health record. Additional clinical and demographic data were from the VA Medical SAS Inpatient and Outpatient datasets,^20,21 including date of birth, most frequently occurring ZIP code of residence, gender, race, marital status, and comorbidities. Educational history, TBI history, etiology of TBI, and history of loss of consciousness after TBI were from the comprehensive TBI evaluation database. Comorbid conditions were determined from diagnoses in VA Medical SAS datasets during the 12-month period before the comprehensive evaluation. Identification of history of mental health diagnoses was based on ICD-9 (International Classification of Diseases, Ninth Revision) codes for depression (300.4, 296.90, 296.20–296.36, 296.50–296.55), PTSD (309.81), anxiety other than PTSD (300.0, 300.2, 300.3), adjustment disorder (308, 309), substance-related disorder (291, 292, 303, 304, 305), bipolar disorder (296.4, 296.6, 296.8, 296.56, 296.00–296.16), and other psychoses (295, 297, 298). Travel time in minutes and hours to nearest VA facility were calculated using ZIP code and geographic information system software (ArcGIS 9.3).

Measures of Health Care Utilization and Cost

Health care utilization data for 12 months before and after the comprehensive evaluation were obtained from the VA MCA NDEs.²² Outpatient utilization was categorized (primary care, rehabilitation care, etc.) based on clinic codes in these databases. For each of these categories, we assessed Veterans' outpatient encounters and costs. Because Veterans may visit multiple clinics while at VA facilities, Veterans may have had more than one encounter per facility visit.

We examined direct health care costs from the VA's (i.e., the payer/provider's) perspective where cost estimates reflect VA's expenditures for each Veteran. Costs for outpatient care and inpatient care provided by a VA facility were obtained from VA MCA NDEs.²² The MCA system extracts information from the VA's accounting and payroll system and combines it with workload information from patient care and administrative departments to produce cost estimates.²² These databases contain estimates of personnel costs, including physicians, nurses, technicians, and other staff, and also costs of supplies and other administrative/overhead expenses of inpatient stays and outpatient encounters.²² Care provided outside VA but paid for by VA were from the VA Fee Basis datasets.²³ We examined total costs per patient, which consisted of total outpatient (primary care, rehabilitation care, polytrauma care, mental health care, other specialty care, and other VA outpatient costs) and total inpatient costs during the 12 months before and after the date of the comprehensive evaluation. All costs were adjusted to 2016 dollars using the Consumer Price Index.

Analysis

Analyses were conducted using Stata/MP version 14.2. To balance characteristics between Veterans who received the comprehensive evaluation through CVT or in-person, we used propensity score weighting.²⁴ We calculated the propensity score for each Veteran as the predicted probability of receiving the comprehensive evaluation through CVT, using multiple logistic regression analysis of the patient characteristics given in Table 1. Each Veteran was assigned a propensity score weight based on this propensity score. For the CVT group, the weight was the inverse of the propensity score; for the in-person group, the weight was equal to the inverse of 1 − propensity score.²⁴ We calculated the standardized differences in characteristics between TBI evaluation groups.

Table 1.

Patient Characteristics

	TBI CONFIRMED THROUGH COMPREHENSIVE EVALUATION						TBI NOT CONFIRMED THROUGH COMPREHENSIVE EVALUATION
	UNWEIGHTED			WEIGHTED			UNWEIGHTED			WEIGHTED
	IN-PERSON (N = 4,899)	CVT (N = 380)	STD. DIFF.	IN-PERSON (N = 4,899)	CVT (N = 380)	STD. DIFF.	IN-PERSON (N = 2,260)	CVT (N = 174)	STD. DIFF.	IN-PERSON (N = 2,260)	CVT (N = 174)	STD. DIFF.
Gender, %
Female	5.9	6.1	0.7	5.9	5.9	0.1	9.2	12.1	9.2	9.4	14.0	14.2
Male	94.1	93.9	−0.7	94.1	94.1	−0.1	90.8	87.9	−9.2	90.6	86.0	−14.2
Race, %
Black	14.6	11.6	−9.1	14.4	13.4	−3.0	24.0	20.1	−9.3	23.7	26.1	5.6
White	68.5	65.0	−7.5	68.3	72.3	8.7	59.2	58.0	−2.3	59.1	60.5	2.8
Other	6.9	13.2	20.9	7.3	6.5	−3.3	8.9	10.3	4.8	9.0	7.5	−5.4
Unknown	9.9	10.3	1.2	9.9	7.8	−7.4	8.9	10.3	4.8	9.0	7.5	−5.4
Hispanic ethnicity, %
No	84.0	86.1	6.3	84.0	82.2	−4.8	84.6	84.5	−0.3	84.6	83.9	−1.8
Yes	10.3	6.1	−15.5	9.8	11.4	4.5	10.4	7.5	−10.4	10.2	11.3	3.5
Unknown	5.9	7.9	7.8	6.1	6.5	1.7	5.0	8.0	12.6	5.2	4.7	−2.0
Married, %
No	46.6	43.9	−5.3	46.4	46.6	0.5	47.5	39.1	−17.1	46.9	46.6	−0.7
Yes	53.4	56.1	5.3	53.6	53.4	−0.5	52.5	60.9	17.1	53.1	53.4	0.7
Age (years), %
<25	12.3	11.1	−3.9	12.2	10.8	−4.4	9.1	6.3	−10.3	8.9	6.8	−7.8
25–29	32.7	31.3	−3.0	32.6	31.7	−1.9	27.9	27.0	−1.9	27.8	33.5	12.4
30–34	24.5	23.4	−2.4	24.4	23.3	−2.6	20.8	17.2	−9.2	20.6	18.1	−6.2
≥35	30.5	34.2	7.9	30.8	34.2	7.3	42.2	49.4	14.5	42.7	41.6	−2.3
Education, %
High school/GED or less	60.3	64.5	8.7	60.6	61.5	1.8	60.8	63.8	6.3	60.9	59.3	−3.5
Some college	33.6	31.6	−4.2	33.3	30.6	−5.8	31.0	26.4	−10.1	30.7	33.8	6.6
Beyond college	6.2	3.9	−10.2	6.0	7.9	7.2	8.2	9.8	5.4	8.4	7.0	−5.2
Service connected disability, %
0	50.4	44.7	−11.3	49.9	50.9	1.9	50.4	47.7	−5.5	50.2	51.2	2.0
1–20	6.1	8.2	8.1	6.2	5.8	−1.8	6.7	5.7	−3.9	6.6	5.2	−6.2
21–40	7.6	7.1	−1.8	7.5	6.7	-3.2	8.1	7.5	-2.2	8.0	11.9	12.8
41–60	9.2	7.4	−6.8	9.1	6.5	−9.7	10.6	11.5	2.8	10.7	10.4	−1.0
61–80	14.3	17.9	9.7	14.6	13.4	−3.4	14.2	16.7	6.9	14.3	13.4	−2.6
81–100	12.4	14.7	6.9	12.6	16.6	11.5	10.0	10.9	2.9	10.1	7.9	−7.6
Mental health comorbidities, %
PTSD	60.3	64.5	8.6	60.6	64.1	7.1	48.8	48.9	0.1	48.8	44.6	−8.4
Depression	38.3	36.6	−3.5	38.2	40.3	4.4	38.3	34.5	−8.0	38.0	35.0	−6.2
Anxiety disorder	30.4	23.2	−16.3	29.8	36.0	13.1	32.9	32.8	−0.3	32.8	28.2	−10.0
Adjustment disorder	16.4	20.0	9.4	16.6	20.4	9.6	18.3	13.8	−12.2	18.0	20.3	6.0
Substance use disorder	16.9	15.0	−5.1	16.7	18.1	3.5	15.9	16.1	0.4	15.9	16.5	1.7
Travel time, %
<30 min	34.5	6.6	−73.7	32.5	35.9	7.1	37.5	11.5	−63.4	35.6	36.1	1.0
30–59 min	33.3	34.2	1.9	33.4	31.7	−3.6	36.1	34.5	−3.3	36.0	35.1	−1.8
≥60 min	28.1	42.6	30.7	29.2	28.4	−1.7	22.5	41.4	41.3	23.9	24.5	1.4
Unknown	4.0	16.6	42.1	4.9	4.0	−4.3	3.9	12.6	32.0	4.5	4.3	−1.1
Lost consciousness, %
Did lose consciousness	48.8	52.6	7.7	49.0	46.1	−5.9	0	0	0	0	0	0
Did not lose consciousness	51.2	47.4	−7.7	51.0	53.9	5.9	100	100	0	100	100	0
Etiology of TBI, %
Bullet	2.5	1.3	−8.5	2.4	2.8	2.8	0.8	1.1	4.1	0.8	0.3	−6.3
Vehicular accident	21.9	20.8	−2.7	21.8	25.3	8.2	10.1	6.9	−11.6	9.9	7.5	−8.5
Fall	25.1	22.6	−5.9	24.9	22.6	−5.5	14.2	13.2	−2.9	14.1	14.5	1.2
Blast	74.5	72.6	−4.2	74.3	70.9	−7.7	40.5	32.8	−16.2	39.9	35.8	−8.6
NSI category, %
1	17.4	21.6	10.7	17.6	14.5	−8.4	25.6	27.0	3.2	25.8	26.9	2.5
2	23.1	23.9	2.0	23.2	20.2	−7.3	26.0	32.8	14.8	26.4	26.1	−0.6
3	29.0	31.3	5.1	29.1	27.8	−3.0	26.2	26.4	0.5	26.2	24.0	−5.2
4	30.6	23.2	−16.8	30.1	37.5	15.8	22.2	13.8	−21.9	21.6	23.0	3.4
Days enrolled in VA in 12 months before comprehensive evaluation, mean	240.8	240.2	−0.4	240.7	255.8	10.3	246.8	248.2	1.0	246.8	236.8	−7.0

CI, confidence interval; CVT, clinical video telehealth; GED, general equivalency diploma; NSI, neurobehavioral symptom inventory; PTSD, post-traumatic stress disorder; TBI, traumatic brain injury; VA, Veterans Affairs.

Because other factors may have influenced trends over time, we used difference-in-differences methodology to examine whether TBI evaluation modality was associated with differences in health care utilization and costs. Difference-in-differences estimates of health care utilization and costs were obtained from generalized estimating equations (GEE) with inverse probability of treatment weighting to account for pre-/postperiod subject-level correlation. Because the numbers of outpatient visits and inpatient days in the 12-month periods before and after the comprehensive evaluation were non-negative integers, we used a negative binomial distribution in the GEE for our analyses of outpatient and inpatient utilization. For health care costs, we selected the distribution and link functions for the GEE based on results from modified Park and Box–Cox tests.²⁵

The GEE models included indicator variables for time (pre- vs. postperiod), TBI evaluation group, and the interaction between time and evaluation group.²⁶ Because results from the comprehensive evaluation may impact subsequent health care utilization and costs, we examined these outcomes separately for Veterans with and without TBI confirmed through the comprehensive evaluation.

Results

Sample Description

Table 1 presents patient characteristics before and after propensity score weighting for patients with and without a confirmed TBI. Before weighting, the greatest imbalance was for travel time. Among Veterans whose TBI was confirmed through the comprehensive evaluation, 34.5% of the in-person group but only 6.6% of the CVT group had a travel time <30 min. In contrast, 28.1% of the in-person group but 42.6% of the CVT group had a travel time ≥60 min. Propensity score weighting reduced these imbalances.

Health Care Utilization after the Comprehensive Evaluation

Among the 554 Veterans who received the comprehensive evaluation through CVT, 380 (68.6%) were diagnosed with TBI, and among the 7,159 Veterans who received the comprehensive evaluation in-person, 4,855 (68.4%) were diagnosed with TBI. Health care utilization before and after the comprehensive evaluation is given in Table 2.

Table 2.

Health Care Utilization for Veterans Whose Traumatic Brain Injury Evaluation Was Conducted In-Person or with Clinical Video Telehealth

	TBI CONFIRMED THROUGH COMPREHENSIVE EVALUATION						TBI NOT CONFIRMED THROUGH COMPREHENSIVE EVALUATION
	PRE		POST		DIFF. IN DIFF. (99% CI)	p	PRE		POST		DIFF. IN DIFF. (99% CI)	p
	CVT (N = 380)	IN-PERSON (N = 4,899)	CVT (N = 380)	IN-PERSON (N = 4,899)	DIFF. IN DIFF. (99% CI)	p	CVT (N = 174)	IN-PERSON (N = 2,279)	CVT (N = 174)	IN-PERSON (N = 2,279)	DIFF. IN DIFF. (99% CI)	p
Outpatient visits, mean
Primary care	8.6	6.5	14.4	15.0	−2.7 (−7.4 to 2.0)	0.14	7.5	6.6	12.7	13.7	−1.8 (−4.5 to 1.0)	0.10
Mental health	4.0	3.0	7.4	7.4	−0.9 (−6.5 to 4.8)	0.70	3.1	2.9	6.5	6.7	−0.3 (−3.6 to 3.0)	0.81
Rehabilitation	0.3	0.4	1.5	1.8	−0.2 (−0.9 to 0.5)	0.42	0.6	0.4	1.2	1.5	−0.5 (−1.2 to 0.3)	0.12
Polytrauma	0.1	0.4	2.9	3.5	−0.3 (−1.3 to 0.7)	0.41	0.1	0.3	1.9	2.2	−0.2 (−0.8 to 0.4)	0.42
Other specialty	0.9	0.9	2.1	3.0	−0.8 (−1.5 to −.1)	0.004	1.2	0.9	3.1	2.9	−0.03 (−1.2 to 1.2)	0.95
Other outpatient	5.5	4.7	7.3	8.2	−1.6 (−3.6 to 0.5)	0.05	5.5	4.4	8.2	6.9	0.2 (−2.3 to 2.6)	0.87
Inpatient days	1.3	0.4	0.5	0.8	−1.1 (−3.2 to 1.0)	0.17	0.5	0.4	0.5	0.8	−0.5 (−1.3 to 0.4)	0.16

CI, confidence interval; CVT, clinical video telehealth; TBI, traumatic brain injury.

Veterans with TBI confirmed through the comprehensive evaluation

Follow-up services, either at VA or non-VA facilities, were recommended for 4,914 (93.1%) Veterans with TBI confirmed through the comprehensive evaluation: 4,558 (93.0%) Veterans who received the evaluation in-person and 356 (93.7%) Veterans who received the evaluation through CVT (p = 0.63) (data not given). According to the care plan that was developed for these Veterans after their diagnosis of TBI, most Veterans were supposed to receive follow-up services within the VA health care system (4,364 [89.1%] Veterans with an in-person evaluation versus 340 [89.5%] Veterans with an evaluation through CVT) (data not given).

Both Veterans who received the comprehensive evaluation in-person or through CVT had increases in the average numbers of outpatient visits from the pre- to the postevaluation periods. The difference-in-differences estimates indicated that the in-person group had significantly greater increases in other specialty care visits (e.g., neurology, cardiology) and other outpatient visits (e.g., ancillary care, home care) from the pre- to postevaluation periods than the CVT group (Table 2). Between the pre- and postevaluation periods, other specialty care visits increased, on average, from 0.9 to 3.0 visits for the in-person group and from 0.9 to 2.1 visits for the CVT group. Consequently, there was a greater increase in specialty care visits for Veterans who received the evaluation in-person (difference-in-differences: 0.8 visits, p = 0.004).

Veterans completing comprehensive evaluation and not receiving a TBI diagnosis

Follow-up services, either at VA or non-VA facilities, were recommended for 1,917 (78.8%) Veterans with TBI not confirmed through the comprehensive evaluation: 1,785 (79.0%) of Veterans who received the evaluation in-person and 132 (75.9%) of Veterans who received the evaluation through CVT (p = 0.33) (data not shown). According to the follow-up care plan, most Veterans were supposed to receive follow-up services within the VA health care system (1,692 [74.9%] of Veterans with an in-person evaluation versus 125 [71.8%] of Veterans with an evaluation through CVT) (data not given).

Among Veterans who did not have a TBI confirmed through the comprehensive evaluation, there were also increases in the numbers of outpatient visits between the pre- and postevaluation periods; however, these increases were not significantly different between Veterans who received the comprehensive evaluation either in-person or through CVT.

Health Care Costs after the Comprehensive Evaluation

Health care costs during the 12-month periods before and after the comprehensive evaluation are given in Table 3.

Table 3.

Health Care Costs for Veterans Whose Traumatic Brain Injury Evaluation Was Conducted In-Person or with Clinical Video Telehealth

	TBI CONFIRMED THROUGH COMPREHENSIVE EVALUATION						TBI NOT CONFIRMED THROUGH COMPREHENSIVE EVALUATION
	PRE		POST		DIFF. IN DIFF.	p	PRE		POST		DIFF. IN DIFF.	p
	CVT (N = 380)	IN-PERSON (N = 4,899)	CVT (N = 380)	IN-PERSON (N = 4,899)	DIFF. IN DIFF.	p	CVT (N = 174)	IN-PERSON (N = 2,279)	CVT (N = 174)	IN-PERSON (N = 2,279)	DIFF. IN DIFF.	p
Health care costs, $
Outpatient costs, mean	4,172	4,015	8,896	9,582	−843 (−3,007 to 1,322)	0.32	3,654	3,812	7,754	8,057	−145 (−2,183 to 1,891)	0.85
Primary care	1,545	1,387	2,592	2,311	123 (−705 to 952)	0.70	1,247	1,352	1,997	2,014	88 (−489 to 666)	0.69
Mental health care	1,115	967	2,031	2,147	−263 (−1,324 to 797)	0.52	797	932	1,731	1,923	−56 (−1,191 to 1,078)	0.90
Other specialty care	331	400	863	1,366	−434 (−736 to −131)	<0.001	451	390	1,391	1,282	49 (−689 to 787)	0.86
Polytrauma care	31	206	1,638	1,707	106 (−495 to 707)	0.65	37	163	953	1,192	−113 (−405 to 179)	0.32
Rehabilitation care	47	83	248	388	−104 (−196 to −11)	0.004	108	87	213	311	−119 (−240 to 2)	0.01
Other outpatient	1,105	972	1,524	1,662	−271 (−828 to 286)	0.21	1,015	888	1,468	1,336	5 (−636 to 646)	0.99
Inpatient costs, $	2,023	628	1,573	1,008	−829 (−3,056 to 1,397)	0.34	1,038	530	1,147	1,093	−454 (−2,324 to 1,416)	0.53
Total health care costs, $	6,195	4,644	10,537	10,824	−1,839 (−4,730 to 1,052)	0.10	4,692	4,342	9,103	9,246	−494 (−4,355 to 3,348)	0.74

Values in parenthesis denote 99% confidence interval.

CVT, clinical video telehealth; TBI, traumatic brain injury.

Veterans with TBI confirmed through the comprehensive evaluation

Health care costs were generally higher in the postevaluation period. The increase was similar between Veterans who received the comprehensive evaluation in-person and through CVT. However, the difference-in-differences estimates indicated that Veterans who were evaluated in-person had significantly greater increases in rehabilitation and other specialty care costs between the pre- and postevaluation periods. Between the pre- and postevaluation periods, rehabilitation costs increased from $83 to $388, on average, for the in-person group and from $47 to $248 visits, on average, for the CVT group. Consequently, there was a greater increase in rehabilitation costs for Veterans who received the evaluation in-person (difference-in-differences: $104, p = 0.004). Moreover, there was no significant difference in the change in inpatient or total costs. Between the pre- and postevaluation periods, total costs increased from $4,644 to $10,824 for the in-person group and from $6,195 to $10,537 for the CVT group (difference-in-differences: $1,839, p = 0.10).

Veterans completing comprehensive evaluation and not receiving a TBI diagnosis

There was no significant difference in the increase in total outpatient, inpatient, or total health care costs from the pre- to postevaluation periods between Veterans in the in-person or CVT groups.

Discussion

Although overall outpatient and total costs were generally higher in the 12 months after than before the comprehensive evaluation, Veterans who received the comprehensive evaluation in-person and were diagnosed with TBI had similar increases in outpatient, inpatient, and total health care costs as Veterans who had TBI confirmed through CVT. However, Veterans with a confirmed TBI evaluated in-person had greater increases in rehabilitation and other specialty care costs.

Veterans who received the comprehensive evaluation through CVT tended to live farther away from VA facilities, which may make them less likely to schedule or return for rehabilitation and specialty visits. Consequently, it might be useful to explore the extent to which these visits could also be provided through CVT. Although the primary appointment reason may be for a comprehensive evaluation, Veterans tend to use any appointment as a time to discuss additional concerns. When an appointment for a comprehensive evaluation is conducted through CVT, it is possible that it is more focused as two parties (i.e., providers at two different facilities) are involved. When the visit is in-person, a Veteran may have more opportunities to discuss additional issues and concerns, whether TBI-related or not. Thus, providers might make more referrals to rehabilitation and specialty care after in-person visits. The importance of recognizing how telehealth may alter the nature of visits has been discussed in the literature and conceptual models have been posited in an effort to systematically delineate the effects of substituting video for face-to-face consultation.²⁷ Evidence from previous exploratory studies suggests that telehealth and in-person visits do in fact differ in terms of physician–patient communication styles, with physicians exerting more control over the dialog and patients assuming more passive roles. Further research into methods to empower Veterans with better communication skills to articulate their needs during telehealth visits may be important as care is increasingly delivered virtually.

During interviews with providers, conducted in conjunction with this study, one of the primary concerns using CVT was the perceived limitation of evaluating physical symptoms over a virtual modality.²⁸ Perhaps providers are discovering more issues with Veterans in-person because they can more directly assess physical complaints, resulting in more referrals for rehabilitation and specialty visits. Moreover, the comprehensive evaluation includes a gait assessment to check balance and coordination; providers have reported challenges assessing this over CVT if there is no appropriate space in the examination room or if the camera cannot be wheeled into the hallway to follow the patient.²⁹

Telehealth, including CVT, is increasingly being used to enhance access to care. The number of Veterans receiving VA care through telehealth is growing at ∼22% annually. In fiscal year 2013 (October 1, 2012 through September 30, 2013), 608,900 VA patients received care through any type of telehealth, representing 1,793,496 episodes of care through telehealth. Among these, 202,823 Veteran received care through CVT.²² For Veterans with TBI, telehealth has been used to conduct comprehensive TBI evaluations, provide follow-up care, and facilitate rehabilitation.

Despite referrals for a comprehensive TBI evaluation, travel distance to specialized facilities may be burdensome for patients and create a barrier to receipt of comprehensive evaluations. In one study, Veterans living >30 min from their VA facility were less likely to be screened for TBI,⁹ demonstrating that travel distance can be a barrier at multiple points in the TBI diagnosis process. Realization of this burden was a key driver for the development of VA's pilot program to offer comprehensive evaluations through CVT. Overall, recent estimates indicate that 45% of patients who receive care in VA through telehealth live in rural areas and may have limited access to VA health care.²⁹

Limitations

Patients were not randomly assigned to receive the comprehensive evaluation through CVT or in-person. Although propensity scoring was used to balance patient characteristics between groups, there may have been unmeasured differences between the groups impacting results. Veterans who received the comprehensive evaluation through CVT had lower outpatient costs for rehabilitation and other specialty care; however, we are unable to determine whether these lower costs were related to availability of these services at VA community-based outpatient clinics. In addition, our analysis was from the VA system's perspective. We did not include non-VA costs or examine costs from the patient's perspective to address issues such as how much they saved in travel expenses and in time that they did not have to take from work to travel to the appointment. This could be important future work. Moreover, we did not examine whether the differences in costs were associated with differences in patient outcomes. However, evaluation of patient satisfaction and provider experiences are ongoing at select pilot sites using CVT for the comprehensive evaluation.

Conclusions

Veterans who received the evaluation in-person and were diagnosed with TBI had similar increases in health care costs as Veterans who had TBI confirmed through CVT. However, there were significantly greater increases in rehabilitation and other specialty care costs for the in-person group. Additional research would be useful to elucidate sources of these patterns and any resulting impacts on health outcomes.

Footnotes

Acknowledgments

This study was supported by the Department of Veterans Affairs, Office of Research and Development, Health Services Research and Development Service Quality Enhancement Research Initiative as award RRP 11–418. The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the U.S. government.

Disclosure Statement

No competing financial interests exist.

References

Menon

, Schwab

, Wright

, Maas

. Demographics and clinical assessment working group of the international and interagency initiative toward common data elements for research on traumatic brain injury and psychological health. Position statement: Definition of traumatic brain injury. Arch Phys Med Rehabil , 2010; 91:1637–1640.

Hoge

, McGurk

, Thomas

, Cox

, Engel

, Castro

. Mild traumatic brain injury in U.S. Soldiers returning from Iraq. N Engl J Med, 2008; 358:453–463.

Okie

Traumatic brain injury in the war zone. N Engl J Med, 2005; 352:2043.

Tanielian

, Jaycox

, eds. Invisible wounds of war: psychological and cognitive injuries, their consequences, and services to assist recovery. Santa Monica, CA: RAND Corporation, 2008.

Wilk

, Bliese

, Kim

, et al. Relationship of combat experiences to alcohol misuse among U.S. soldiers returning from the Iraq war. Drug Alcohol Depend, 2010; 108:115–121.

Evans

, St. Andre JR, Pape TLB, Steiner ML, Stroupe KT, Hogan T, Weaver FM, Smith BM. An evaluation of the VA TBI screening process among OEF/OIF Veterans. PM R , 2013; 5:210–220.

Sayer

NA.

Traumatic brain injury and its neuropsychiatric sequelae in war veterans. Ann RevMed, 2012; 63:405–419.

Defense and Veterans Brain Injury Center. DoD numbers for traumatic brain injury worldwide since. 2000. Available at http://dvbic.dcoe.mil/sites/default/files/DoD-TBI-Worldwide-Totals- 2000-2014-Q1-Q3-Dec1–Dec2014.pdf (last accessed January 27, 2015 ).

Vanderploeg

, Curtiss

, Luis

, et al. Long-term morbidities following self-reported mild traumatic brain injury. J Clin Exp Neuropsychol, 2007; 29:585–598.

10.

Eden

, Stevens

, United States, Department of Health and Human Services, Institute of Medicine (U.S.), Committee on Traumatic Brain Injury. Evaluating the HRSA Traumatic Brain Injury Program [Internet]. Washington, DC: National Academies Press; 2006 [cited 2018 October 26].

11.

Kehle-Forbes

, Campbell

, Taylor

, Scholten

, Sayer

. Does co-occurring traumatic brain injury affect VHA outpatient health service utilization and associated costs among veterans with posttraumatic stress disorder? An examination based on VHA administrative data. J Head Trauma Rehabil, 2017; 32:E16–E23.

12.

VHA directive. 1184. Screening and evaluation of traumatic brain injury (TBI) in Operation Enduring Freedom (OEF), Operation Iraqi Freedom (OIF), and Operation New Dawn (OND) veterans. Washington, DC: Veterans Health Administration, 2017. Available at https://www.va.gov/vhapublications/ViewPublication.asp?pub_ID=5376 (last accessed March 5, 2019 ).

13.

Carlson

, Nelson

, Orazem

, Nugent

, Cifu

, Sayer

. Psychiatric diagnoses among Iraq and Afghanistan War Veterans screened for deployment-related traumatic brain injury. J Trauma Stress, 2010; 23:17–24.

14.

Scholten

, Sayer

, Vanderploeg

, Bidelspach

, Cifu

. Analysis of US Veterans Health Administration comprehensive evaluations for traumatic brain injury in Operation Enduring Freedom and Operation Iraqi Freedom Veterans. Brain Inj, 2012; 26:1177–1184.

15.

Sayer

, Rettmann

, Carlson

, et al. Veterans with history of mild traumatic brain injury and posttraumatic stress disorder: Challenges from provider perspective. J Rehabil Res Dev, 2009; 46:703–713.

16.

Yurkiewicz IR1, Lappan

, Neely

, Hesselbrock

, Girard

, Alphonso

, Tsao

. Outcomes from a US military neurology and traumatic brain injury telemedicine program. Neurology, 2012; 79:1237–1243.

17.

Brown-Connolly

NE.

Patient satisfaction with telemedical access to specialty services in rural California. J Telemed Telecare, 2002; 8:7–10.

18.

Reynolds

, Vick

, Haak

. Telehealth applications in speech-language pathology: A modified narrative review J Telemed Telecare, 2009; 15:310–316.

19.

Rush

, Hatt

, Gorman

, Janicki

, Polasek

, Shay

. Planning telehealth for older adults with atrial fibrillation in rural communities: Understanding stakeholder perspectives. Clin Nurs Res, 2019; 28:130–149.

20.

VIReC. VIReC research user guide: fiscal year 2014 VHA medical SAS outpatient datasets and inpatient encounters dataset. Hines, IL: US Department of Veterans Affairs Health Services Research and Development Service, VA Information Resource Center, 2015.

21.

VIReC. VIReC research user guide: fiscal year 2014 VHA medical SAS inpatient datasets. 2nd ed. Hines, IL: US Department of Veterans Affairs Health Services Research & Development Service, VA Information Resource Center, 2015.

22.

Phibbs

, Barnett

, Harden

, King

. Research guide to decision support system national cost extracts. Department of Veterans Affairs, Menlo Park, CA: Health Economics Resource Center of Health Service R&D Services, 2008.

23.

Gidwani

, Hong

, Murrell

. Fee basis data: a guide for researchers. Menlo Park, CA. VA Palo Alto, Health Economics Resource Center, 2015.

24.

Zeng

, An

, Scully

, Barrington

, Patel

, Nichol

. The impact of value-based benefit design on adherence to diabetes medications: A propensity score-weighted difference in difference evaluation. Value Health, 2010; 13:846–852.

25.

McCullagh

, Nelder

. Generalized linear models. 2nd ed. London: Chapman and Hall, 1989.

26.

Karaca-Mandic

, Norton

, Dowd

. Interaction terms in nonlinear models. Health Serv Res, 2012; 47:225–274.

27.

Miller

EA.

Telemedicine and doctor–patient communication: A theoretical framework for evaluation. J Telemed Telecare, 2002; 8:311–318.

28.

Martinez

, Hogan

, Lones

, Balbale

, Scholten

, Bidelspach

, Musson

, Smith

. Evaluation and treatment of mild traumatic brain injury through the implementation of clinical video telehealth: Provider perspectives from the Veterans Health Administration. PM R, 2017; 9:231–240.

29.

Mazmanian

. How VA is driving telemedicine. The Business of Federal Technology, 2014. Available at https://fcw.com/Articles/2014/02/13/How-VA-is-driving-telemedicine.aspx (last accessed March 5, 2019 ).