Patterns of Telehealth Use for Mental Health Treatment Among Hispanic Veterans

Abstract

Background:

Video-to-home telehealth (VTH) is promising for increasing access to mental health (MH) services. VA Video Connect (VVC) facilitates video-based teleconferencing between patients and providers and can reduce barriers while maintaining clinical effectiveness. Little is known about the preferences of Hispanic veterans for VTH.

Methods:

A retrospective cohort investigation of VTH for MH care utilization among veterans having at least one MH visit from October 2019 to September 2020. The veterans consisted of 155,492 Hispanic/Latino and 1,544,958 non-Hispanic/Latino. VVC involved face-to-face synchronous video-based teleconferencing between patients and providers, enabling care at home or another private location. The main measures included the percentage of MH encounters delivered through VVC.

Results:

Compared with non-Hispanic veterans, Hispanic veterans had 3.28% greater percentage of VVC MH encounters. Furthermore, there was a 2.65% increase per month in percentage of VVC MH encounters.

Conclusions:

Contrary to preconceived notions, Hispanic veterans access VTH at higher rates than their non-Hispanic counterparts.

Background

COVID-19 accelerated the implementation of telehealth across the country. Telemental health, in particular, has increased access to mental health (MH) services; and the Department of Veterans Affairs (VA) supported its use even before the COVID-19 pandemic.^1,2 Telemental health, especially video-to-home telehealth (VTH), can overcome barriers to care, such as time and costs associated with travel, stigma related to being seen at an MH clinic, and home- and work-related responsibilities that impede patients from going to clinic.³

The benefits of VTH for veterans have been well documented.^1,3 Veterans who access MH services and receive access to video-enabled tablets demonstrate increased use of MH care through VTH, increased psychotherapy visits across all modalities, and reduced suicidality and emergency department visits.¹ VTH also has the potential to improve adherence to MH treatment, as the majority of veterans who engage in VTH find it is as beneficial as in-person care and conducive to development of strong therapeutic alliances and feelings of being supported and understood.⁴ MH care delivered through VTH is as effective as office-based in-person psychotherapy for veterans with post-traumatic stress disorder (PTSD) and depression.^5
–7

Yet, a disparity of access to telehealth has emerged, a concept known as the digital divide.^2,8 Patients may lack a video-enabled device or a reliable high-speed internet connection, both of which are necessary for VTH. Notably, patients from traditionally disadvantaged racial and ethnic groups, including Black and Hispanic people, are often victims of the digital divide and access telehealth services at lower rates than other racial and ethnic groups.^9
–11 Our group has conducted similar prior studies investigating VTH use among Black, female, and rural veterans.^12,13 However, issues related to the digital divide among veterans remain insufficiently studied, particularly for veterans who identify as Hispanic.

Hispanic veterans represent a distinct subgroup of the Hispanic population in the United States. Although a language barrier to medical or MH care may exist for many Hispanic patients in this country, in theory, no such language barrier exists among Hispanic veterans because English proficiency is a requirement for military service. Hispanic veterans are also unique in that they experience higher rates of PTSD than Hispanic civilians, and they may under-report stress and have higher rates of avoidant behaviors than non-Latino White veterans.^14,15 Resources that increase veterans' access to MH, such as VTH, are, therefore, even more crucial to Hispanic veterans than to nonveteran Hispanic patients.

Our goal was to determine whether the established pattern of VTH use among Hispanic civilian patients can also be found in the veteran population, with attention to patterns evident before and after the onset of the pandemic. Specifically, we wanted to explore the impact of the digital divide on Hispanic veterans to determine whether they are less frequent users of VTH for MH care. We also wanted to know whether there are demographic factors (such as age, gender, race, or rurality) that predispose Hispanic veterans to more (or less) usage of VA Video Connect (VVC, VA's preferred telehealth platform).

Methods

This study was approved by the institutional review boards of the Michael E. DeBakey VA Medical Center (#17E04.H) and Baylor College of Medicine (H-40706).

DATA COLLECTION

We retrospectively examined use of VTH for MH care utilization, using data from the Corporate Data Warehouse. We selected a national cohort of veterans who had completed at least one MH encounter at the Veterans Health Administration (VHA) during Fiscal Year 2020 (FY20, October 2019–September 2020). MH clinic stop codes (500 series) were used to identify MH services accessed by patients, with a 179 secondary stop code to capture VVC visits to a non-VHA location. Demographic data (i.e., age, rurality, gender, race, and ethnicity) that were collected during VHA enrollment through veteran self-report was also examined. The final cohort included 12 months of data for all veterans (155,492 Hispanic/Latino and 1,544,958 non-Hispanic/Latino).

STATISTICAL ANALYSES

We first used descriptive statistics (means and standard deviations or counts and percentages) to describe the national cohort. We then examined differences between Hispanic and non-Hispanic veterans in demographic characteristics, using an independent samples t-test for age and chi-square tests for rurality, race, and gender.

We then examined differences between Hispanic and non-Hispanic veterans in the percentage of MH encounters delivered through VVC (i.e., number of VVC encounters for MH divided by all MH encounters) monthly across FY20, using linear growth curve models. The dependent variable was the percentage of MH encounters that were VVC (from 0% to 100%). The first model included ethnicity (Hispanic = 1 and non-Hispanic = 0) and time (FY20, where October 2019 was coded as 0, November 2019 as 1, and so forth, all the way to September 2020, coded as 11) as fixed independent variables and enabled examination of differences in the percentage of MH encounters that were VVC by ethnicity, as well as change in the percentage of MH encounters that were VVC over time.

A second model also included the interaction between ethnicity and time as a fixed independent variable to examine whether change over time differed between Hispanic and non-Hispanic veterans. Intercept and time were included as random effects in all models.

In addition, to gain a better understanding of Hispanic veteran VVC use 6 months into the COVID-19 pandemic, we investigated age, rurality, gender, and racial differences in percentage of VVC MH encounters in the first 6 months of the COVID-19 pandemic. We conducted three analysis of covariance models in which the average percentage of MH encounters that were VVC in August and September 2020 was the dependent variable; and age, rurality, gender, and race were independent variables. Each model controlled for the average percentage of VVC MH encounters in the 6 months before the pandemic (defined as April 2020). Since not every veteran had an MH encounter in August and/or September of 2020, analyses were repeated using PROC MI and MIANALYZE intent-to-treat procedures in SAS. All statistical analyses were conducted using SAS Version 9.4 (SAS Institute, Cary, NC, USA).

The Institutional Review Board for Baylor College of Medicine and Affiliated Hospitals approved this study. As data were collected from the Corporate Data Warehouse, participants were not directly consented by the investigators.

Results

Demographic characteristics of veterans overall and by ethnicity are reported in Table 1. Compared with non-Hispanic veterans, a greater percentage of Hispanic veterans were <55 years, urban, male, and White; but effect sizes for all ethnic differences were small (Ф between 0.02 and 0.13).

Table 1.

Demographic Characteristics Overall and Comparisons by Ethnicity

	OVERALL ( n = 1,700,450)	HISPANIC/LATINO ( n = 155,492)	NON-HISPANIC/LATINO ( n = 1,544,958)	COMPARISON BETWEEN HISPANIC AND NON-HISPANIC
	OVERALL ( n = 1,700,450)	HISPANIC/LATINO ( n = 155,492)	NON-HISPANIC/LATINO ( n = 1,544,958)	χ ² p -VALUE	EFFECT SIZE ( Φ )
Age (n = 1,483,770)
<55 years of age	700,612 (47.22%)	82,317 (59.87%)	618,295 (45.93%)	<0.0001	0.08
≥55 years of age	783,158 (52.78%)	55,172 (40.13%)	727,986 (54.07%)	<0.0001	0.08
Rurality (n = 1,479,703)
Rural/highly rural	421,767 (28.50%)	18,462 (13.46%)	403,305 (30.04%)	<0.0001	0.11
Urban	1,057,936 (71.50%)	118,748 (86.54%)	939,188 (69.96%)	<0.0001	0.11
Gender (n = 1,519,124)
Male	1,289,343 (84.47%)	120,248 (85.18%)	1,169,095 (84.84%)	0.001	0.02
Female	229,781 (15.13%)	20,927 (14.82%)	208,854 (15.16%)	0.001	0.02
Race (n = 1,581,453)
White	1,128,081 (71.33%)	112,777 (87.47%)	1,015,304 (69.90%)	<0.0001	0.13
Black	392,866 (24.84%)	8,810 (6.83%)	384,056 (26.44%)
Asian	22,259 (1.41%)	952 (0.74%)	21,307 (1.47%)
Native Hawaiian/Pacific Islander	19,344 (1.22%)	3,377 (2.62%)	15,967 (1.10%)
American Indian/Alaskan Native	18,904 (1.20%)	3,009 (2.33%)	15,894 (1.09%)

Across the 12 months, the mean number of veterans with at least one MH encounter was 53936.17 (standard deviation [SD] = 2472.75) for Hispanics and 529403.75 (SD = 22954.50) for non-Hispanics. Across the 12 months, the mean number of monthly encounters was 2.07 (SD = 0.09) for Hispanic veterans and 2.17 (SD = 0.06) for non-Hispanic veterans.

Figure 1 shows the mean percentage of MH encounters that were VVC over time by ethnicity. In FY20, there was a main effect of ethnicity [b = 3.28, standard error (SE) = 0.06, t(17E5) = 53.65, p < 0.0001] and a main effect of time [b = 2.65, SE = 0.01, t(53E5) = 958.40, p < 0.0001]. Compared with non-Hispanic veterans, Hispanic veterans had nearly a 3.28% greater percentage of MH encounters that were VVC. Furthermore, there was a 2.65% increase per month in percentage of VVC MH encounters. Importantly, there was a significant interaction between ethnicity and time [b = 0.80, SE = 0.01, t(53E5) = 83.98, p < 0.0001], such that the effect of time for Hispanic veterans [b = 3.38, SE = 0.01, t(49E4) = 340.16, p < 0.0001] was greater than the effect of time for non-Hispanic veterans [b = 2.57, SE = 0.01, t(48E5) = 896.93, p < 0.0001].

Fig. 1.

Percentage of mental health encounters that are VA Video Connect, by ethnicity.

Among Hispanic veterans, age category, rurality, gender, and race each significantly predicted the percentage of VVC MH encounters 6 months after the pandemic started (see Table 2). Specifically, the percentage of MH encounters that were done over VVC 6 months after the pandemic began were significantly higher for Hispanic veterans who were <55 years of age [F(1, 58,210) = 1611.06, p < 0.0001], Hispanic veterans who were urban [F(1, 57,271) = 11.44, p = 0.0007], and Hispanic veterans who were female [F(1, 60,402) = 338.01, p < 0.0001]. Furthermore, there were differences in VVC encounters by race, [F(4, 50,915) = 3.69, p = 0.005]. The percentage of VVC MH encounters was significantly higher among Hispanic veterans who were Hawaiian/Pacific Islander (p = 0.004) and Black (p = 0.04) compared with White. Intent-to-treat analyses revealed parallel findings.

Table 2.

Mental Health Encounters for Hispanic Veterans by Age, Rurality, Gender, and Race

VETERAN CHARACTERISTIC	ADJUSTED MEAN PERCENTAGE OF MH ENCOUNTERS CONDUCTED THROUGH VVC 6 MONTHS AFTER THE COVID-19 PANDEMIC BEGAN
Age
<55 years of age (n = 61,196)	35.91_a
≥55 years of age (n = 43,786)	22.45_b
Rurality
Rural/highly rural (n = 13,969)	30.33_a
Urban (n = 91,066)	32.03_b
Gender
Male (n = 92,981)	30.45_a
Female (n = 15,687)	38.72_b
Race
White (n = 90,260)	30.98_a
Black (n = 6,942)	32.39_b
Asian (n = 738)	33.85_ab
Native Hawaiian/Pacific Islander (n = 2,673)	34.17_b
American Indian/Alaskan Native (n = 2,317)	32.65_ab

Within a given characteristic, groups with the “a” subscript differ from groups with the “b” subscript at p < 0.05.

MH, mental health; VVC, VA video connect.

Discussion

We collected national data from MH visits occurring in FY20, grouped by Hispanic status, and found that when we controlled for time, Hispanic veterans demonstrated increased VVC usage compared with non-Hispanic veterans: averaging across all time points, Hispanic veterans had 3.28% more MH care delivered through VVC than non-Hispanic veterans. This difference in VVC usage continued to widen over the first 6 months of the pandemic. Although both Hispanic and non-Hispanic veterans showed increased VVC use once the pandemic began, the slope rose significantly more steeply in the Hispanic group: average increase per month for Hispanic veterans was 2.65%. This finding challenges the conventional narrative that Hispanic patients access telehealth services rates at lower rates than non-Hispanic patients.

Prior studies evaluating VTH usage among Hispanic patients suggest several possible reasons for why this trend is seen in the general Hispanic population, including provider workflows that do not adequately integrate interpreter services and the lack of language-concordant guides for training patients in the use of video visits. Other possible factors include the facts that Black and Hispanic patients may be less likely to have a source of usual care and, therefore, more likely to access the emergency department; that they are more likely to have severe COVID-19 infections and, therefore, more likely to access the emergency department, and that that they may have higher rates of mistrust in digital appointments, specifically, and the medical establishment more generally (as a consequence of systemic racism).^10,11

One reason why this trend is not also seen in the Hispanic veteran population is that English proficiency is a requirement for military service. Furthermore, military training and veteran status may provide Hispanic veterans a special skill set that enables higher rates of VTH usage. However, English proficiency exists on a spectrum, and some people who speak English as a second language may feel more comfortable conducting certain types of encounters (e.g., MH visits) in their native language.¹⁶ This warrants further investigation.

In subgroup analyses of Hispanic veterans, specifically, we found higher rates of VVC use among urban veterans than rural veterans, among female veterans than male veterans, among Black veterans than White veterans, and among younger veterans (age <55 years) than older veterans. These findings are in line with prior studies by our group^12,13 When all these factors are considered together, our data should empower providers to offer VTH services to Hispanic veterans.

Our study had several strengths. We used a large nationally representative sample of veterans across various time points, allowing robust statistical analyses. One important limitation of our study is the way in which demographic data are collected by the VHA. Veterans self-report their racial and ethnic identity, and racial/ethnic identity among Hispanic people is often complex and difficult to standardize; veterans may not feel that the prescribed categories adequately represent them. This is also a retrospective observational study subject to limitations of its study design.

Conclusion

Contrary to preconceived notions of Hispanic patients, Hispanic veterans access VTH at higher rates than their non-Hispanic counterparts. Future studies could evaluate unique aspects of Hispanic veterans qualitatively, examining specific reasons for their increased uptake and clarifying their enablers and barriers, as well as their thoughts regarding their racial and ethnic identities. More research is also needed to compare Hispanic veterans to Hispanic civilians directly.

Footnotes

Disclaimer

The opinions expressed are those of the authors and do not necessarily reflect those of the Department of Veterans Affairs, the U.S. government, or Baylor College of Medicine.

Disclosure Statement

No competing financial interests exist.

Funding Information

This study is supported by a grant from the VA Office of Rural Health, Veterans Rural Health Resource Center-Salt Lake City and partly the result of the use of facilities and resources of the Houston VA HSR&D Center for Innovations in Quality, Effectiveness and Safety (CIN13-413) and the VA South Central Mental Illness Research, Education and Clinical Center, which played no role in the design and conduct of the study; in the collection, analysis, and interpretation of the data; and in the preparation, editing, or censuring of the article.

References

Gujral

, van Campen

, Jacobs

, et al. Mental health service use, suicide behavior, and emergency department visits among rural US veterans who received video-enabled tablets during the COVID-19 pandemic. JAMA Netw Open, 2022; 5:1–14.

Connolly

, Stolzmann

, Heyworth

, et al. Patient and provider predictors of telemental health use prior to and during the COVID-19 pandemic within the Department of Veterans Affairs. Am Psychol, 2021; 77:249–261.

Lindsay

, Hogan

, Ecker

, et al. The importance of video visits in the time of COVID-19. J Rural Health, 2021; 37:242–245.

Fletcher

, Hogan

, Keegan

, et al. Recent advances in delivering mental health treatment via video to home. Curr Psychiatry Rep, 2018; 20:1–16.

Acierno

, Gros

, Ruggiero

, et al. Behavioral activation and therapeutic exposure for posttraumatic stress disorder: A noninferiority trial of treatment delivered in person versus home-based telehealth. Depress Anxiety, 2016; 33:415–423.

Egede

, Acierno

, Knapp

, et al. Psychotherapy for Depression in older veterans via telemedicine: A randomised, open-label, non-inferiority trial. Lancet Psychiatry, 2015; 2:693–701.

McClellan

, Osbaldiston

, Wu

, et al. The effectiveness of telepsychology with veterans: A meta-analysis of services delivered by videoconference and phone. Psych Serv, 2022; 19:294–304.

Eberly

, Kallan

, Julien

, et al. Patient characteristics associated with telemedicine access for primary and specialty ambulatory care during the COVID-19 pandemic. JAMA Netw Open, 2020; 3:1–12.

Ryskina

, Shultz

, Zhou

, et al. Older adults' access to primary care: Gender, racial, and ethnic disparities in telemedicine. J Am Geriatr Soc, 2021; 69:2732–2740.

10.

Weber

, Miller

, Astha

, et al. Characteristics of telehealth users in NYC for COVID-related care during the Coronavirus pandemic. J Am Med Inform Assoc, 2020; 27:1949–1954.

11.

Rodriguez

, Betancourt

, Sequist

, et al. Differences in the use of telephone and video telemedicine visits during the COVID-19 pandemic. Am J Manag Care, 2021; 27:21–26.

12.

Lindsay

, Caloudas

, Hogan

, et al. Getting connected: A retrospective cohort investigation of video to-home telehealth for mental health care utilization among women veterans. J Gen Intern Med, 2022; 37(Suppl 3):778–785.

13.

Hogan

, Amspoker

, Walder

, et al. Differential impact of COVID-19 on the use of tele–mental health among veterans living in urban or rural areas. Psychiatr Serv, 2022 [Epub ahead of Print]; DOI: 10.1176/appi.ps.202100363.

14.

Pittman

JOE

. Latino veterans with PTSD: A systematic review. Behav Sci (Basel), 2014; 4:320–340.

15.

Gradus

JL.

Epidemiology of PTSD. National Center for PTSD. n.d. Available at https://www.ptsd.va.gov/professional/treat/essentials/epidemiology.asp [last accessed May 26, 2022 ].

16.

Yellowlees

, Parish

, Gonzalez

, et al. Clinical outcomes of asynchronous versus synchronous telepsychiatry in primary care: Randomized controlled trial. JMed Internet Res, 2021; 23:1–15.