Abstract
Background
The COVID-19 pandemic expanded telehealth use across healthcare systems, including the Veterans Health Administration (VA). Little is known about how large-scale telehealth rollout affected access to primary care for patients experiencing homelessness.
Objective
To examine the extent to which homeless-experienced veterans used telehealth services in primary care and to characterize users before and after the onset of the COVID-19 pandemic.
Design
Retrospective cohort study, 3/16/2019–3/15/2022.
Participants
394,731 veterans with homelessness diagnoses nationally using 4,068,109 primary care visits.
Main Measures
The outcomes were use of 1 + telehealth visits (video, phone, secure messaging) for primary care during each year. Through multivariable regression models, we examined associations between telehealth use, patient characteristics (e.g., age, sex, race-ethnicity, comorbidity), and VA homeless services use (e.g., homeless-tailored primary care (HPACT), permanent supportive housing).
Key Results
Compared to pre-pandemic, telehealth in primary care among homeless-experienced veterans increased substantially 2 years post-pandemic (video: 1.37% versus 20.56%, phone: 60.74% versus 76.58%). Secure messaging was low over time (1.57–2.63%). In adjusted models, video users were more likely to be young (65 + years: OR = 0.43, CI: 0.42–0.44), women (OR = 1.74, CI: 1.70–1.78), Black (OR = 1.14, CI: 1.12–1.16), Hispanic (OR = 1.34, CI: 1.30–1.38), and with more comorbidities (2 + on the Charlson Comorbidity Index; OR = 1.16, CI: 1.14–1.19), compared to video non-users. HPACT patients were less likely to use video (OR = 0.68, CI: 0.66–0.71) than other primary care patients. This was not observed among users of other VA homeless services.
Conclusions
Despite decreased access to health information technology and low pre-pandemic telehealth use, veterans experiencing homelessness still sustained high use of telehealth in primary care post-pandemic. Women and racial-ethnic minorities had higher video uptake proportionately, suggesting that telehealth may address access disparities among these homeless-experienced patient groups. Identifying and targeting organizational characteristics (e.g., HPACT users) that predict telehealth use for improvement may be key to increasing adoption among VA primary care patients experiencing homelessness.
Supplementary Information
The online version contains supplementary material available at 10.1007/s11606-023-08462-3.
KEY WORDS: telehealth, primary care, homelessness, veterans
INTRODUCTION
The COVID-19 pandemic expanded the availability of telehealth care options, but many patients face challenges using virtual modalities to access essential primary care services.1 Patients are now routinely allowed to communicate with primary care teams using several telehealth modalities, including phone, video, and secure messaging.2 Yet, there exists a “Digital Divide,” that is unequal digital health literacy and access to technology (e.g., smartphones, tablets, laptops) or internet required for patients to participate in telehealth care.3 As virtual care increased out of necessity, researchers documented telehealth disparities by age, income, race-ethnicity, etc., as well as differing patient preferences for its use.3–8 Many have studied how to optimally incorporate telehealth modalities into existing primary care models for at-risk demographic groups.9, 10 Homeless-experienced persons often fall into several of these at-risk categories and may thus be disproportionately affected by the Digital Divide.
Historically, persons experiencing homelessness have faced technological barriers (e.g., lack of mobile telephones) in accessing services and require supportive outreach to engage in primary care.11–13 Small-scale studies have documented attempts to engage homeless-experienced persons using telehealth care, as well as their perspectives regarding this care modality.14, 15 At the same time, telehealth has shown promise and is a tool that may reduce racial-ethnic and other disparities in chronic disease care over time.16 Common examples of telehealth modalities used by primary care teams include conducting outreach phone calls or responding to patient secure messages. Video telehealth adoption, however, has and continues to be low in primary care, as compared to specialty care.6 Observations during the COVID-19 pandemic may offer insights about large-scale telehealth adoption, as well as predictive characteristics of telehealth users, among individuals experiencing homelessness.
In response to the COVID-19 pandemic, the Veterans Health Administration (VA) made telehealth broadly available for all VA primary care patients, including veterans experiencing homelessness.17 The VA has invested heavily in housing and ensuring accessible health care services for more than 33,000 veterans who experience homelessness on any given night across the nation.18 VA funds a breadth of services for veterans with homeless experiences, including outreach for unsheltered veterans (e.g., Health Care for Homeless Veterans, HCHV), transitional housing (e.g., Grant Per Diem programs, GPD), and permanent supportive housing (Housing and Urban Development – VA Supportive Housing, HUD-VASH) programs.19 There are national initiatives that tailor primary care services for a subsegment of high-risk patients experiencing homelessness (Homeless Patient Aligned Care Team, HPACT)20 and that have provided over 100,000 tablets for video visits (Digital Divide consult).21, 22 Early in the pandemic, homeless-experienced veterans were found to use video visits less than housed veterans,6 but a subsequent study showed a higher overall rate of video care driven by homeless-experienced veterans’ higher usage of video for mental health care.23
While some research on tele-mental health services for veterans experiencing homelessness exists,24, 25 there remains a knowledge gap on the extent to which primary care services are delivered via telehealth. It is unknown whether these VA initiatives successfully engaged persons experiencing homelessness in primary care and characteristics of these telehealth users. It is possible that vulnerable populations, especially homeless-experienced persons, may have been left behind as telehealth expanded broadly. Alternatively, robust supportive services and available telehealth options may ensure equitable access to primary care. This national VA study aimed to examine the extent to which homeless-experienced veterans used telehealth services in primary care before and after the onset of the COVID-19 pandemic and characterized such users.
METHODS
Study Design and Cohort
This retrospective cohort study examined use of various telehealth modalities (phone, video, secure messaging) among 394,731 VA primary care patients with a diagnosis of homelessness between March 16, 2019, and March 15, 2022. We included three study years (0 = 1 year before COVID-19 onset, 1 = 1 year after COVID-19 onset, 2 = 2 years after COVID-19 onset). For each study year, we identified a cohort of patients who had at least one VA primary care visit and a diagnosis of homelessness within the study year. Homelessness was identified through International Classification of Diseases (ICD-10) codes Z59.0x, Z59.1x, Z59.8x, and Z59.9x.26 Study variables were extracted from VA administrative and electronic health records in the VA Corporate Data Warehouse. This quality improvement effort was deemed exempt from review by the VA Greater Los Angeles Healthcare System’s Institutional Review Board.
Main Measures
The outcome measures were cohort patients having at least one video visit, one phone visit, or one secure message (My HealtheVet) within a study year. We examined the following patient characteristics which have been associated with differential telehealth use in prior research:3–8 age (18–44, 45–64, 65 +), birth sex (male, female), race-ethnicity (non-Hispanic White, non-Hispanic Black, Hispanic, non-Hispanic other, unknown race-ethnicity), Charlson Comorbidity Index (CCI) (0 = no comorbidity, 1 = one comorbidity, 2 + = two or more comorbidities).27 Since VA homeless and mental health services (hereafter, “supportive services”) may facilitate telehealth adoption, we also examined the following: assignment to homeless-tailored primary care (HPACT) (yes/no),20 received supportive services from Health Care for Homeless Veterans (HCHV) (yes/no), Grant Per Diem (GPD) programs (yes/no), Housing and Urban Development – VA Supportive Housing (HUD-VASH) (yes/no), and Mental Health Intensive Case Management (MHICM, VA’s Assertive Community Treatment program) (yes/no).19
Statistical Analysis
We conducted univariate (descriptive) analyses for all study variables by study year. We also conducted bivariate analyses and reported unadjusted percent telehealth use (video, phone, secure messaging) for each study variable by study year. For multivariable analyses, we conducted three logistic regression analyses (one for each outcome measure), adjusting for patient characteristics and VA supportive service variables. Models included fixed effects for each study year to account for secular trends, as well as adjusted for multiple observations of individuals over time. We reported odds ratios and 95% confidence intervals, as well as predicted probabilities (adjusted percent) of telehealth use while holding all covariates at their means.
RESULTS
The study included 1,430,729 VA primary care visits (n = 218,127 patients) nationally 1 year prior to the COVID-19 pandemic; however, the numbers decreased slightly to 1,310,658 visits (n = 189,168 patients) 1 year after and 1,326,722 visits (n = 195,968 patients) 2 years after pandemic onset. Table 1 illustrates the descriptive data (unadjusted percent and sample size) for each study variable by study year. Patient characteristics and VA supportive service utilization remained stable over time. Compared to pre-pandemic, video and phone use among patients increased substantially 1 year after (video: 1.37% vs. 22.29%, phone: 60.74% vs. 88.94%), with a slight decrease 2 years after pandemic onset (video: 20.56%, phone: 76.58%). Secure messaging was low but also increased after pandemic onset (1.57% (year 0) vs. 2.46% (year 1) vs. 2.63% (year 2)) (Table 1).
Table 1.
Demographic and Study Variables by Study Year for Veterans Experiencing Homelessness
| 1 year before COVID onset 3/16/2019–3/15/2020 n = 218,127 |
1 year after COVID onset 3/16/2020–3/15/2021 n = 189,168 |
2 years after COVID onset 3/16/2021–3/15/2022 n = 195,968 |
|
|---|---|---|---|
| Patient characteristics | |||
| Age (mean, SD) | 55.61 (14.24) | 56.04 (14.14) | 56.44 (14.41) |
| Age category % (n) | |||
| 18–44 | 22.85 (49,836) | 22.38 (42,334) | 22.84 (44,765) |
| 45–64 | 51.48 (112,287) | 50.29 (95,124) | 46.83 (91,767) |
| 65–74 | 19.23 (41,946) | 20.96 (39,655) | 23.27 (45,598) |
| 75 + | 6.44 (14,058) | 6.37 (12,055) | 7.06 (13,838) |
| Birth sex % (n) | |||
| Male | 88.03 (192,007) | 88.00 (166,462) | 87.46 (171,388) |
| Female | 11.97 (26,120) | 12.00 (22,706) | 12.54 (24,580) |
| Race/ethnicity % (n) | |||
| Non-Hispanic White | 48.92 (106,701) | 48.47 (91,688) | 48.17 (94,388) |
| Non-Hispanic Black | 37.14 (81,017) | 37.17 (70,322) | 37.12 (72,734) |
| Hispanic | 6.85 (14,941) | 6.83 (12,912) | 6.72 (13,171) |
| Non-Hispanic other | 2.53 (5511) | 2.61 (4933) | 2.62 (5135) |
| Unknown | 4.56 (9957) | 4.92 (9313) | 5.38 (10,540) |
| Charlson Comorbidity Index % (n) | |||
| 0 | 52.24 (113,956) | 50.84 (96,181) | 53.99 (105,806) |
| 1 | 19.5 (42,539) | 19.57 (37,013) | 19.37 (37,961) |
| 2 + | 28.26 (61,632) | 29.59 (55,974) | 26.64 (52,201) |
| VA supported services | |||
| Assignment to Homeless-Tailored Primary Care Team, or Homeless Patient Aligned Care Team (HPACT) | 6.39 (13,928) | 5.72 (10,820) | 5.08 (9948) |
| Outreach Services from Health Care for Homeless Veterans Program (HCHV) | 46.45 (101,324) | 43.16 (81,641) | 46.38 (90,886) |
| Transitional Housing Program, or Grant Per Diem Program (GPD) | 10.93 (23,840) | 10.05 (19,002) | 9.86 (19,328) |
| Permanent Supportive Housing Program, or Housing and Urban Development—VA Supportive Housing (HUD-VASH) | 45.89 (100,095) | 49.79 (94,195) | 48.37 (94,783) |
| Mental Health Intensive Case Management (MHICM) | 2.05 (4476) | 2.41 (4551) | 2.15 (4213) |
| Study outcome measures | |||
| Used video | 1.37 (2988) | 22.29 (42,170) | 20.56 (40,291) |
| Used phone | 60.74 (132,496) | 88.94 (168,246) | 76.58 (150,063) |
| Used secure messaging | 1.57 (3419) | 2.46 (4645) | 2.63 (5152) |
Table 2 illustrates unadjusted percent telehealth use (video, phone, secure messaging) for each study variable during the most recent (year 2) data. Compared to the youngest age group (18–44 years old), 65 + -year-old patients had lower video use (14.15% vs. 29.35%) and secure messaging (1.47% vs. 3.91%). Conversely, 65 + -year-old patients had higher phone use (79.96%) compared to the youngest patients (71.91%). Compared to men, women had higher video use (32.57% vs. 18.84%), phone use (80.39% vs. 76.03%), and secure messaging (6.03% vs. 2.14%). Compared to non-Hispanic Whites, non-Hispanic Blacks had higher video use (22.22% vs. 18.43%), but lower phone use (75.74% vs. 77.37%) and secure messaging (2.41% vs. 2.79%). Patients experiencing homelessness who were assigned to homeless-tailored primary care (HPACT) had lower telehealth use for all three modalities: video, phone, and secure messaging (Table 2) than those assigned to usual primary care. Similarly, patients who used any other VA supportive services, except those who had intensive mental health case management (MHICM), had lower telehealth use for all three modalities (Table 2).
Table 2.
Unadjusted Telehealth Use (Percent) by Demographic and Study Variables for Veterans Experiencing Homelessness (2 Years After COVID-19 Onset)
| Used video % (n) |
Used phone % (n) |
Used secure messaging % (n) |
|
|---|---|---|---|
| Patient characteristics | |||
| Age category | |||
| 18–44 | 29.35 (13,139) | 71.91 (32,190) | 3.91 (1751) |
| 45–64 | 20.42 (18,743) | 76.66 (70,347) | 2.75 (2525) |
| 65 + | 14.15 (8409) | 79.96 (47,526) | 1.47 (876) |
| Birth sex | |||
| Male | 18.84 (32,285) | 76.03 (130,302) | 2.14 (3670) |
| Female | 32.57 (8006) | 80.39 (18,761) | 6.03 (1482) |
| Race/ethnicity | |||
| Non-Hispanic White | 18.43 (17,398) | 77.37 (73,024) | 2.79 (2638) |
| Non-Hispanic Black | 22.22 (16,165) | 75.74 (55,088) | 2.41 (1752) |
| Hispanic | 26.06 (3433) | 77.28 (10,179) | 3.09 (407) |
| Non-Hispanic other | 20.10 (1032) | 75.81 (3893) | 2.67 (137) |
| Unknown | 21.47 (2263) | 74.75 (7879) | 2.07 (218) |
| Charlson Comorbidity Index (CCI) | |||
| 0 | 21.59 (22,845) | 71.31 (75,455) | 2.59 (2741) |
| 1 | 19.78 (7509) | 79.97 (30,357) | 2.85 (1082) |
| 2 + | 19.04 (9937) | 84.77 (44,251) | 2.55 (1329) |
| VA supported services | |||
| Assignment to Homeless-Tailored Primary Care Team, Homeless Patient Aligned Care Team (HPACT) | |||
| Yes | 13.76 (1369) | 73.06 (7268) | 1.41 (140) |
| No | 20.92 (38,922) | 76.76 (142,795) | 2.69 (5012) |
| Outreach Services from Health Care for Homeless Veterans Program (HCHV) | |||
| Yes | 20.57 (18,696) | 75.70 (68,800) | 2.52 (2294) |
| No | 20.55 (21,595) | 77.33 (81,263) | 2.72 (2858) |
| Transitional Housing Program, or Grant Per Diem Program (GPD) | |||
| Yes | 18.85 (3643) | 74.81 (14,460) | 1.83 (353) |
| No | 20.75 (36,648) | 76.77 (135,603) | 2.72 (4799) |
| Permanent Supportive Housing Program, or Housing and Urban Development—VA Supportive Housing (HUD-VASH) | |||
| Yes | 18.81 (17,829) | 73.88 (70,025) | 2.39 (2270) |
| No | 22.20 (22,462) | 79.10 (80,038) | 2.85 (2882) |
| Mental Health Intensive Case Management (MHICM) | |||
| Yes | 19.63 (827) | 79.33 (3342) | 1.99 (84) |
| No | 20.58 (39,464) | 76.51 (146,721) | 2.64 (5068) |
Similar to unadjusted results, Table 3 illustrates the adjusted odds ratios (OR) and 95% confidence intervals (CI) for video use, phone use, and secure messaging use among cohort patients, adjusting for all study covariates. Compared to the youngest age group (18–44 years old), older patients were less likely to use video (45–64 years old: OR = 0.66, CI: 0.65–0.68; 65 + years old: OR = 0.43, CI: 0.42–0.44) and secure messaging (45–64 years old: OR = 0.77, CI: 0.74–0.81; 65 + years old: OR = 0.41, CI: 0.38–0.43), but more likely to use phone (45–64 years old: OR = 1.15, CI: 1.13–1.17; 65 + years old: OR = 1.19, CI: 1.17–1.22). Compared to men, women were more likely to use video (OR = 1.74, CI: 1.70–1.78), phone (OR = 1.46, CI: 1.43–1.49), and secure messaging (OR = 2.59, CI: 2.47–2.71). Compared to non-Hispanic Whites, non-Hispanic Blacks were more likely to use video (OR = 1.14, CI: 1.12–1.16), but less likely to use phone (OR = 0.92, CI: 0.91–0.93) and secure messaging (OR = 0.77, CI: 0.74–0.80). Compared to non-Hispanic Whites, Hispanics were more likely to use video (OR = 1.34, CI: 1.30–1.38) but less likely to use secure messaging (OR = 0.90, CI: 0.84–0.97). Compared to patients with no chronic comorbidity, patients with 1 or 2 + chronic comorbidities were more likely to use all three types of telehealth modalities (Table 3).
Table 3.
Predictors of Telehealth Use (Video, Phone, Secure Messaging) for Veterans Experiencing Homelessness
| Used video Odds ratio (95% CI) |
Used phone Odds ratio (95% CI) |
Used secure messaging Odds ratio (95% CI) |
|
|---|---|---|---|
| Patient characteristics | |||
| Age category (ref = 18–44) | |||
| 45–64 | 0.66 (0.65–0.68)* | 1.15 (1.13–1.17)* | 0.77 (0.74–0.81)* |
| 65 + | 0.43 (0.42–0.44)* | 1.19 (1.17–1.22)* | 0.41 (0.38–0.43)* |
| Birth sex | |||
| Female | 1.74 (1.70–1.78)* | 1.46 (1.43–1.49)* | 2.59 (2.47–2.71)* |
| Race/ethnicity (ref = NH White) | |||
| Non-Hispanic Black | 1.14 (1.12–1.16)* | 0.92 (0.91–0.93)* | 0.77 (0.74–0.80)* |
| Hispanic | 1.34 (1.30–1.38)* | 0.98 (0.96–1.01) | 0.90 (0.84–0.97)* |
| Non-Hispanic other | 1.04 (0.98–1.10) | 0.89 (0.86–0.93)* | 0.77 (0.68–0.88)* |
| Unknown | 1.11 (1.07–1.16)* | 0.85 (0.83–0.88)* | 0.73 (0.66–0.81)* |
| Charlson Comorbidity Index (ref = 0) | |||
| 1 | 1.08 (1.05–1.10)* | 1.58 (1.55–1.60)* | 1.35 (1.29–1.42)* |
| 2 + | 1.16 (1.14–1.19)* | 2.23 (2.19–2.27)* | 1.47 (1.40–1.55)* |
| VA supported services | |||
| Assignment to Homeless-Tailored Primary Care Team, Homeless Patient Aligned Care Team (HPACT) | 0.68 (0.66–0.71)* | 0.82 (0.80–0.84)* | 0.59 (0.52–0.66)* |
| Outreach Services from Health Care for Homeless Veterans Program (HCHV) | 0.90 (0.88–0.91)* | 0.94 (0.93–0.96)* | 0.86 (0.83–0.90)* |
| Transitional Housing Program, or Grant Per Diem Program (GPD) | 0.98 (0.96–1.01) | 0.98 (0.96–1.00) | 0.80 (0.74–0.86)* |
| Permanent Supportive Housing Program, or Housing and Urban Development—VA Supportive Housing (HUD-VASH) | 0.79 (0.78–0.81)* | 0.76 (0.75–0.77)* | 0.85 (0.82–0.88)* |
| Mental Health Intensive Case Management (MHICM) | 0.99 (0.94–1.05) | 1.26 (1.19–1.32)* | 0.60 (0.51–0.70)* |
| COVID year onset (ref = 1 yr before) | |||
| 1 year after COVID onset | 21.66 (20.86–22.49)* | 5.43 (5.33–5.52)* | 1.59 (1.53–1.66)* |
| 2 years after COVID onset | 19.6 (18.90–20.38)* | 2.20 (2.17–2.23)* | 1.74 (1.67–1.81)* |
Logistic regression analysis was conducted for each telehealth modality (video, phone, secure messaging) after adjusting for study variables, study year, and repeated observations (i.e., patient clustering effect)
*p < .001
Primary care patients assigned to HPACT were less likely to use video (OR = 0.68, CI: 0.66–0.71), phone (OR = 0.82, CI: 0.80–0.84), and secure messaging (OR = 0.59, CI: 0.52–0.66). Similarly, patients who used HCHV and HUD-VASH services were less likely to use video, respectively (OR = 0.90, CI: 0.88–0.91; OR = 0.79, CI: 0.78–0.81), phone (OR = 0.94, CI: 0.93–0.96; OR = 0.76, CI: 0.75–0.77), and secure messaging (OR = 0.86, CI: 0.83–0.90; OR = 0.85, CI: 0.82–0.88). While there were no differences in video use, GPD (OR = 0.80, CI: 0.74–0.86) and MHICM users (OR = 0.60, CI: 0.51–0.70) were less likely to use secure messaging than their counterparts. However, MHICM users were more likely to use phone visits as part of their intensive case management (OR = 1.26, CI: 1.19–1.32).
Overall, fully adjusted models showed nearly a 22-fold increase in video visits for 1 year after (CI: 20.86–22.49) and 20-fold increase 2 years after pandemic onset (CI: 18.90–20.38), compared to the baseline year (Table 3). Phone visits increased fivefold (CI: 5.33–5.52) and twofold (CI: 2.17–2.23) in years 1 and 2, respectively, while secure messages increased 60% (CI: 1.53–1.66) and 70% (CI: 1.67–1.81), compared to pre-pandemic. The Appendix illustrates adjusted percent telehealth use (video, phone, secure messaging) for each study variable during the most recent study year.
DISCUSSION
Despite a Digital Divide and low pre-pandemic use, telehealth in VA primary care increased significantly among veterans experiencing homelessness during the COVID-19 pandemic. To our knowledge, this is one of the largest studies to examine nearly 400,000 homeless-experienced persons, in over 4 million primary care visits, during a 3-year period that encompassed an unprecedented national disaster that necessitated telehealth adoption. Our study cohort was observed to increase use of video and phone visits at rates comparable to the general VA population.17 Post-pandemic, we found that 1 in 5 veterans experiencing homelessness participated in video visits, and the majority contacted their primary care teams by phone. High telehealth use was maintained beyond the first year of COVID-19 and as in-person primary care services returned. Findings were stable when we adjusted for patient factors that may affect telehealth use and show potential for virtual primary care use among a population severely affected by digital and other access barriers.11, 13 Secure messaging rates also increased to albeit a much lesser extent—only used by 2% of homeless-experienced veterans. Consistent with prior research,28 low secure messaging use may suggest that this primary care modality may not be preferred by persons experiencing homelessness or may have more barriers to adoption. It may be more difficult for veterans experiencing homelessness to use VA secure messaging, because secure messaging requires a formal enrollment process involving identity verification, which is not required for video and phone visits. Contextual (e.g., provider variation in telehealth adoption) and patient-level factors (e.g., patient digital health literacy and preference) that support or impede telehealth use among persons experiencing homelessness will need further exploration to ensure equitable access to primary care.
Our study of veterans experiencing homelessness found that telehealth usage patterns largely mirror that of the larger VA population. While homeless-experienced veterans tended to be younger, women, Black, and with fewer comorbidities29 than housed veterans, we still observed disproportionately higher use among those who were young, women, racial-ethnic minorities, and with multiple comorbidities. In our homeless-experienced cohort, we saw known age-related disparities in telehealth use30 over time, specifically 28% of 18–44-year-olds versus 15% of 65 + -year-olds used video visits (Appendix). We again found that women were open to31 and thus engaged with telehealth at higher rates than men, specifically 29% of women versus 19% of men used video visits and 6% of women versus 2% of men used secure messaging. Research on racial-ethnic disparities in telehealth use has been mixed;4–8 however, we did not see lower rates of telehealth use among racial-ethnic minorities in our cohort, specifically 19%, 22%, and 24% of White, Black, and Hispanic patients, respectively, used video visits. Finally, we observed that telehealth, especially phone visits, is helpful care management tool used by homeless-experienced veterans with multiple chronic comorbidities,32 specifically 71% with no comorbidity versus 85% with 2 + comorbidities used phone visits. Young, women, racial-ethnic minority, and high comorbidity veterans had higher video uptake proportionately, suggesting that telehealth may address access disparities among these homeless-experienced patient groups.
Interventions may be needed to increase telehealth uptake among patients assigned to homeless-tailored primary care (HPACT) clinics or who receive certain VA supportive services. Our study found that 15% of cohort patients assigned to HPACT versus 21% of those who received VA primary care elsewhere used video visits (Appendix). Even though HPACT selectively serves the most vulnerable patients experiencing homelessness,20 which likely contributes to this disparity, there may be room for implementation approaches that help increase adoption of virtual care among HPACT patients. Permanent supportive housing (HUD-VASH) is another setting that we could continue to target for telehealth interventions,22 since 18% of HUD-VASH users versus 22% of non-users used video visits. While HPACT or HUD-VASH users may be accustomed to in-person contact as required by certain supportive services (e.g., clothing distribution, medication delivery), it is worth exploring how best to optimize hybrid in-person and telehealth care to ensure equitable primary care access for the highly vulnerable veterans who use these services. On the other hand, video use was the same for veterans experiencing homelessness who participate in Mental Health Intensive Case Management (MHICM) and Transitional Housing (GPD) services, with minor differences in use of phone and secure messaging. Taking a positive deviance approach33 to study how MHICM and GPD programs successfully encourage telehealth use will likely inform best practices on engaging persons experiencing homelessness in virtual care.
While our national scope and large sample size is a clear strength in this study, limitations still apply. First, we used ICD-10 diagnostic codes to define our cohort of veterans experiencing homelessness. Our cohort excluded patients with homeless experiences who did not receive diagnostic coding, due to lack of standardized screening and coding of housing status, or in cases where informal supportive services were offered.34 Nonetheless, this cohort represents a patient population that willingly engages with VA services and can exercise preference for in-person versus virtual care. Second, administrative studies can be affected by variable telehealth service coding,35 where telehealth visits may be miscoded early on but hopefully remedied later during the pandemic. Third, our analyses did not include information on where patients lived or whether they had access to high-speed/broadband internet. Most (84%) of our homeless-experienced cohort resided in urban areas, which generally have adequate internet coverage areas,36 so this would not have likely changed our conclusions. Additionally, our administrative data sources did not contain patient-reported information on digital health literacy and preference, nor did we have information on provider variation in telehealth delivery for homeless-experienced veterans. Finally, since we did not analyze VA-funded services in the community, findings may or may not generalize to care provided by non-VA healthcare systems for veterans experiencing homelessness. However, prior studies show that the large majority of primary care services are delivered to veterans by the VA.37
The VA and other healthcare systems should continue to monitor use of and consider how best to incorporate telehealth services into primary care for patients experiencing homelessness. Ensuring equitable access to telehealth will help healthcare systems to engage its most vulnerable patients using all available modalities. Greater choice in how to access primary care services may in turn reduce utilization of acute care services (e.g., emergency visits, hospitalizations),38 as well as mitigate other disparities (e.g., racial-ethnic) in health and health care. Since patient characteristics remain immutable, identifying and targeting organizational characteristics (e.g., supportive housing users) that predict telehealth use for improvement may be key to increasing adoption among VA primary care patients experiencing homelessness.
Supplementary information
(DOCX 18.3 kb)
Acknowledgements:
The authors would like to acknowledge Jessica Beroes-Severin for her assistance in manuscript preparation.
Funding
Dr. Leung is supported by Career Development Award IK2 HX002867 from the United States Department of Veterans Affairs Health Services Research & Development Service. This work was supported by the VA Office of Connected Care (XVA 65-127, MPI Der-Martirosian & Leung), VA Office of Primary Care (Primary Care Analytics Team), and the Veterans Emergency Management Evaluation Center (VEMEC) at the VA.
Data Availability
Under the Health Insurance Portability and Accountability Act (HIPAA), the dataset used in this study cannot be shared publicly because it contains patient-level Protected Health Information/Personally Identifiable Information (PHI/PII) from the Veterans Health Administration. To gain access to this data, interested researchers must complete credentialing to conduct VA research, as well as data use agreements with the Primary Care Analytics Team (pcat@va.gov) and other relevant VHA data owners.
Declarations:
Conflict of Interest:
The authors declare that they do not have a conflict of interest.
Disclaimer:
The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the US Department of Veterans Affairs or the US government.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Webber EC, McMillen BD, Willis DR. Health care disparities and access to video visits before and after the COVID-19 pandemic: findings from a patient survey in primary care. Telemed J E Health. 2022;28(5):712–719. doi: 10.1089/tmj.2021.0126. [DOI] [PubMed] [Google Scholar]
- 2.Mehrotra A, Ray K, Brockmeyer DM, Barnett ML, Bender JA. Rapidly converting to “virtual practices”: outpatient care in the era of Covid-19. NEJM Catalyst. 2020 Apr 1;1(2).
- 3.Rodriguez JA, Clark CR, Bates DW. Digital health equity as a necessity in the 21st century cures act era. JAMA. 2020;323(23):2381–2382. doi: 10.1001/jama.2020.7858. [DOI] [PubMed] [Google Scholar]
- 4.Reed ME, Huang J, Graetz I, et al. Patient characteristics associated with choosing a telemedicine visit vs office visit with the same primary care clinicians. JAMA Netw Open. 2020;3(6):e205873. doi: 10.1001/jamanetworkopen.2020.5873. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Rodriguez JA, Betancourt JR, Sequist TD, Ganguli I. Differences in the use of telephone and video telemedicine visits during the COVID-19 pandemic. Am J Manag Care. 2021;27(1):21–26. doi: 10.37765/ajmc.2021.88573. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ferguson JM, Jacobs J, Yefimova M, Greene L, Heyworth L, Zulman DM. Virtual care expansion in the Veterans Health Administration during the COVID-19 pandemic: clinical services and patient characteristics associated with utilization. J Am Med Inform Assoc. 2021;28(3):453–462. doi: 10.1093/jamia/ocaa284. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Weber E, Miller SJ, Astha V, Janevic T, Benn E. Characteristics of telehealth users in NYC for COVID-related care during the coronavirus pandemic. J Am Med Inform Assoc. 2020;27(12):1949–1954. doi: 10.1093/jamia/ocaa216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Eberly LA, Kallan MJ, Julien HM, et al. Patient characteristics associated with telemedicine access for primary and specialty ambulatory care during the COVID-19 pandemic. JAMA Netw Open. 2020;3(12):e2031640. doi: 10.1001/jamanetworkopen.2020.31640. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Uscher-Pines L, Sousa J, Jones M, et al. Telehealth use among safety-net organizations in california during the covid-19 pandemic. JAMA. 2021;325(11):1106–1107. doi: 10.1001/jama.2021.0282. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Cheng MKW, Allison TA, McSteen BW, Cattle CJ, Lo DT. The Adoption of video visits during the COVID-19 pandemic by VA home based primary care. J Am Geriatr Soc. 2021;69(2):318–320. doi: 10.1111/jgs.16982. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kopanitsa V, McWilliams S, Leung R, et al. A systematic scoping review of primary health care service outreach for homeless populations. Fam Pract. 2023;40(1):138–151. doi: 10.1093/fampra/cmac075. [DOI] [PubMed] [Google Scholar]
- 12.Health Quality O. Interventions to improve access to primary care for people who are homeless: a systematic review. Ont Health Technol Assess Ser. 2016;16(9):1–50. [PMC free article] [PubMed] [Google Scholar]
- 13.Howells K, Burrows M, Amp M, et al. Exploring the experiences of changes to support access to primary health care services and the impact on the quality and safety of care for homeless people during the COVID-19 pandemic: a study protocol for a qualitative mixed methods approach. Int J Equity Health. 2021;20(1):29. doi: 10.1186/s12939-020-01364-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Adams CS, Player MS, Berini CR, et al. A telehealth initiative to overcome health care barriers for people experiencing homelessness. Telemed J E Health. 2021;27(8):851–858. doi: 10.1089/tmj.2021.0127. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Garvin LA, Greenan MA, Edelman EJ, Slightam C, McInnes DK, Zulman DM. Increasing use of video telehealth among Veterans experiencing homelessness with substance use disorder: design of a peer-led intervention. J Technol Behav Sci. 2023 Sep;8(3):234-45. [DOI] [PMC free article] [PubMed]
- 16.Nouri S, Khoong E, Lyles C, Karliner L. Addressing equity in telemedicine for chronic disease management during the Covid-19 pandemic. NEJM Catalyst. 2020 May 4;1(3).
- 17.Heyworth L, Kirsh S, Zulman D, Ferguson JM, Kizer KW. Expanding access through virtual care: the VA’s early experience with Covid-19. NEJM Catalyst. 2020 Jul 1;1(4).
- 18.de Sousa T, Andrichik A, Cuellar M, Marson J, Prestera E, Rush K, Abt Associates. (2022) The 2022 annual homelessness assessment report (AHAR) to congress. The U.S. Department of Housing and Urban Development. https://www.huduser.gov/portal/sites/default/files/pdf/2022-AHAR-Part-1.pdf
- 19.Tsai J, Kasprow W, Rosenheck R. Latent homeless risk profiles of a national sample of homeless veterans and their relation to program referral and admission patterns. Am J Public Health. 2013;103(S2):S239–S247. doi: 10.2105/AJPH.2013.301322. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Tsai J, Havlik J, Howell BA, Johnson E, Rosenthal D. Primary care for veterans experiencing homelessness: a narrative review of the Homeless Patient Aligned Care Team (HPACT) model. J Gen Intern Med. 2023;38(3):765–783. doi: 10.1007/s11606-022-07970-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Garvin LA, Hu J, Slightam C, McInnes DK, Zulman DM. Use of video telehealth tablets to increase access for veterans experiencing homelessness. J Gen Intern Med. 2021;36(8):2274–2282. doi: 10.1007/s11606-021-06900-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Wray CM, Van Campen J, Hu J, Slightam C, Heyworth L, Zulman DM. Crossing the digital divide: a veteran affairs program to distribute video-enabled devices to patients in a supportive housing program. JAMIA Open. 2022;5(2):ooac027. doi: 10.1093/jamiaopen/ooac027. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Ferguson JM, Wray CM, Jacobs J, et al. Variation in initial and continued use of primary, mental health, and specialty video care among Veterans. Health Serv Res. 2023;58(2):402–414. doi: 10.1111/1475-6773.14098. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Raja PV, Gabrielian S, Doran N. Access to care for veterans with serious mental illness during the COVID-19 pandemic. Psychiatr Serv. 2021;72(11):1324–1327. doi: 10.1176/appi.ps.202000898. [DOI] [PubMed] [Google Scholar]
- 25.Connolly SL, Stolzmann KL, Heyworth L, 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. 2022;77(2):249–261. doi: 10.1037/amp0000895. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Peterson R, Gundlapalli AV, Metraux S, et al. Identifying homelessness among veterans using VA administrative data: opportunities to expand detection criteria. PLoS One. 2015;10(7):e0132664. doi: 10.1371/journal.pone.0132664. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Quan H, Sundararajan V, Halfon P, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005;43(11):1130–1139. doi: 10.1097/01.mlr.0000182534.19832.83. [DOI] [PubMed] [Google Scholar]
- 28.Jones AL, Gelberg L, deRussy AJ, et al. Low uptake of secure messaging among veterans with experiences of homelessness and substance use disorders. J Addict Med. 2021;15(6):508–511. doi: 10.1097/ADM.0000000000000785. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Tsai J, Mehta K, Mongtomery AE, Elbogen E, Hooshyar D. Changing demography of homeless adult populations. Perspect Public Health. 2021;141(3):177–184. doi: 10.1177/1757913920919796. [DOI] [PubMed] [Google Scholar]
- 30.Kruse C, Fohn J, Wilson N, Nunez Patlan E, Zipp S, Mileski M. Utilization barriers and medical outcomes commensurate with the use of telehealth among older adults: systematic review. JMIR Med Inform. 2020;8(8):e20359. doi: 10.2196/20359. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Goldstein KM, Bosworth HB, Gierisch JM. Can right-sizing the use of virtual care improve access to equitable, patient-centered care for women veterans? J Gen Intern Med. 2023 Jul;38(Suppl 3):868-870. [DOI] [PMC free article] [PubMed]
- 32.Mohamed S, Neale M, Rosenheck RA. VA intensive mental health case management in urban and rural areas: veteran characteristics and service delivery. Psychiatr Serv. 2009;60(7):914–921. doi: 10.1176/ps.2009.60.7.914. [DOI] [PubMed] [Google Scholar]
- 33.Rose AJ, McCullough MB. A practical guide to using the positive deviance method in health services research. Health Serv Res. 2017;52(3):1207–1222. doi: 10.1111/1475-6773.12524. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Gabrielian S, Yuan A, Rubenstein L, Andersen RM, Gelberg L. Serving homeless veterans in the VA Desert Pacific Healthcare Network: a needs assessment to inform quality improvement endeavors. J Health Care Poor Underserved. 2013;24(3):1344–1352. doi: 10.1353/hpu.2013.0124. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Yeramosu D, Kwok F, Kahn JM, Ray KN. Validation of use of billing codes for identifying telemedicine encounters in administrative data. BMC Health Serv Res. 2019;19(1):928. doi: 10.1186/s12913-019-4753-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Lai J, Widmar NO. Revisiting the digital divide in the COVID-19 era. Appl Econ Perspect Policy. 2021 Mar;43(1):458-464. [DOI] [PMC free article] [PubMed]
- 37.Liu CF, Chapko M, Bryson CL, et al. Use of outpatient care in Veterans Health Administration and Medicare among veterans receiving primary care in community-based and hospital outpatient clinics. Health Serv Res. 2010;45(5 Pt 1):1268–1286. doi: 10.1111/j.1475-6773.2010.01123.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Wranik WD, Price S, Haydt SM, et al. Implications of interprofessional primary care team characteristics for health services and patient health outcomes: a systematic review with narrative synthesis. Health Policy. 2019;123(6):550–563. doi: 10.1016/j.healthpol.2019.03.015. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
(DOCX 18.3 kb)
Data Availability Statement
Under the Health Insurance Portability and Accountability Act (HIPAA), the dataset used in this study cannot be shared publicly because it contains patient-level Protected Health Information/Personally Identifiable Information (PHI/PII) from the Veterans Health Administration. To gain access to this data, interested researchers must complete credentialing to conduct VA research, as well as data use agreements with the Primary Care Analytics Team (pcat@va.gov) and other relevant VHA data owners.