VA Disability Compensation and Money Spent on Substance Use Among Homeless Veterans: A Controversial Association
Abstract
Objective:
There has long been concern that public support payments are used to support addictive behaviors. This study examined the amount of money homeless veterans spend on alcohol and drugs and the association between public support income, including VA disability compensation, and expenditures on alcohol and drugs.
Methods:
Data were from 1,160 veterans from 19 sites on entry into the Housing and Urban Development–Veterans Affairs Supportive Housing program. Descriptive statistics and nonparametric analyses were conducted.
Results:
About 33% of veterans reported spending money on alcohol and 22% reported spending money on drugs in the past month. No significant association was found between public support income, VA disability compensation, and money spent on alcohol and drugs.
Conclusions:
A substantial proportion of homeless veterans spend some income on alcohol and drugs, but disability income, including VA compensation, does not seem to be related to substance use or money spent on addictive substances.
Substance abuse has been identified as one of the main risk factors for homelessness (1,2). Many homeless adults continue to purchase and use substances while homeless (3,4), even though they lack money for food, housing, and other basic necessities. It is not well understood how much money and what sources of income homeless adults are using to fund their substance use. Moreover, there has been concern about the possibility that public support payments, and disability payments in particular, may be a trigger that facilitates substance use and that disability income is used to purchase alcohol and drugs. One well-cited study of nonhomeless adults with schizophrenia who abused cocaine found that patterns of cocaine use were associated with the timing of disability payments, referred to as the “check effect” (5). There is a long-standing policy debate about whether public support or disability payments made as in-kind provision of goods or services, instead of cash transfers, would reduce this risk. This debate has been concerned with who decides how to spend such transfer payments (the state or the client), because even in-kind provisions are fungible (6,7).
These issues are of special relevance for U.S. veterans who receive disability compensation from the U.S. Department of Veterans Affairs (VA) for service-related injuries and illnesses, some of whom are returning from recent conflicts in the Middle East. Although there has been nearly universal support for providing disability benefits to veterans, examination of the link between benefits and substance use may be important for homeless services and research.
Studies have shown that disability and public support income do not appear to contribute to overall levels of substance abuse among homeless veterans (8–10). These findings largely suggest that homeless veterans do not use their disability income to buy alcohol and drugs. However, these studies did not specifically examine VA disability compensation separately from other disability and public support income and did not analyze high-risk individuals (only those with substance use disorders), and only one previous study, using data from the 1980s, has reported on the amount of money homeless veterans spend on substances (8).
Given the national initiative to prevent and end homelessness among veterans (11), further research is needed to tease out the association between VA disability compensation and substance use among homeless veterans. VA disability compensation has been found to be protective against homelessness (12), but high rates of substance use disorders have been found among homeless veterans (13), suggesting that homeless veterans who receive VA disability compensation may use it to support addictive habits resulting in their homelessness. Many homeless veterans also have problems managing their money, and money mismanagement has been identified as a potential target for intervention in this population (14).
This study assessed the amount of money homeless veterans reported spending on alcohol and drugs and then examined the association between public support dollars received—and VA disability compensation in particular—and dollars spent on alcohol and drugs. It was hypothesized that a substantial proportion of homeless veterans spend money on alcohol and drugs but that there would be no association between income from public support sources, including VA disability compensation, and money spent on alcohol or drugs.
Methods
In 1992, the U.S. Department of Housing and Urban Development (HUD) and the VA collaboratively formed the largest supported housing program for veterans—the Housing and Urban Development–Veterans Affairs Supportive Housing (HUD-VASH) program. Currently, there are HUD-VASH teams in more than 130 VA facilities, and studies have shown that HUD-VASH is effective in housing homeless veterans (15).
Data for this study were from a combined national observational data set and an experimental data set that included information about veterans who entered the HUD-VASH program between 1992 and 2003. Eligibility criteria for the HUD-VASH program during this time included being eligible for VA services, living in a shelter or on the street for at least 30 days, and having a psychiatric or substance use disorder at the time of initial contact. Veterans were referred to the program by clinicians working in specialized VA homeless service programs. After informed consent and approval by the appropriate institutional review boards were obtained, data were collected through interviews with veterans conducted by clinical staff in the HUD-VASH program. Psychiatric diagnoses were made by clinical staff through interviews and review of existing medical records. This study focused on 1,160 homeless veterans from 19 sites on entry into the HUD-VASH program (baseline)—that is, while they were still homeless.
Income was assessed by asking veterans how much money they received in the past 30 days from various sources, which were categorized into three sources for this study: employment, public support (for example, unemployment insurance, welfare, and VA and non-VA disability), and other (for example, friends and family, panhandling, and illegal sources).
Veterans were asked whether or not they had a VA service-connected disability for a psychiatric condition or a general medical (nonpsychiatric) condition. Among those with a VA service-connected disability, veterans were further asked what percentage rating they had received for their psychiatric or general medical condition, which ranges from 0% to 100%, with higher percentages reflecting greater compensation levels.
Veterans were also asked about non-VA disability benefits—specifically, whether they received no non-VA disability benefits, psychiatric disability benefits only, physical disability benefits only, and psychiatric and physical disability benefits.
For the data analysis, descriptive statistics were conducted to examine the background characteristics, disability benefits, and substance use of the veterans in the sample. Then nonparametric tests were conducted to examine the relation between background characteristics, disability benefits, and the dependent variables—days of substance use and expenditures on substances in the previous month. Spearman correlations were used for continuous or binomial independent variables (for example, age and public support income), and Kruskal-Wallis one-way analysis of variance was used for independent categorical variables (for example, race and service era). Multivariable analyses did not need to be conducted because no background characteristics or baseline data were significantly associated with days of substance use or expenditures (and the results remained the same when multiple regressions were conducted with nights homeless in the past three months). These analyses were repeated for only those with substance use disorders (N=885), which required multiple regression analyses because age, race-ethnicity, service era, and number of nights homeless in the past three months were significantly associated with amount of substance use and expenditures. To adjust for inflated type I error, significance for all analyses was set at the .01 level.
Results
Of the 1,160 homeless veterans, 1,102 (95%) were male, 517(45%) were white, and 552 (48%) were black. The mean±SD age of the sample was 42.97±7.99 years, with a mean of 12.50±1.80 years of education. A total of 657 (60%) served during the Vietnam War era, and only 53 (5%) were married. Nearly half (N=569, 49%) had a drug abuse or dependence diagnosis, and over half (N=733, 63%) had an alcohol abuse or dependence diagnosis. The mean length of homelessness was 3.53±4.06 years, with 31.27±34.08 days homeless in the past three months. Homeless veterans in the sample reported a total income of $406.60±458.54 in the past month, with 55% of that from public support income. In terms of disability benefits, 225 (19%) had a VA service-connected disability, and 225 (19%) received non-VA disability benefits. Among those with a VA service-connected disability, the mean percentage rating was 28.84±22.12 (range 0–100) for a psychiatric condition and 14.94±14.74 (range 0–85) for a general medical condition.
In the total sample, 430 (37%) reported using alcohol in the past month and 486 (33%) reported spending money on alcohol in the past month; 102 (9%) reported spending more than $150. [A table summarizing these findings is available in an online data supplement to this report.] The most common form of alcohol consumed was beer. About 22% of the sample (N=257) reported spending any money on drugs in the past month; 16% (N=183) reported spending more than $150. Among those reporting drug use in the past month, 17% (N=194) reported using “crack,” 11% (N=127) reported using cannabis, and 8% (N= 93) reported using cocaine.
As shown in Table 1 employment income was positively associated with days of alcohol use and money spent on alcohol, as well as with money spent on drugs. Other sources of income (for example, family and friends and panhandling) were also positively associated with alcohol use, money spent on alcohol, drug use, and money spent on drugs. In contrast, public support income was negatively associated with alcohol use and money spent on alcohol. No significant association was found between public support income and drug use or money spent on drugs. In addition, no significant association was found between receipt of VA or of non-VA disability benefits and alcohol and drug use or money spent on alcohol or drugs.
| Alcohol use | Drug use | |||
|---|---|---|---|---|
| Variable | Days | Money spent | Days | Money spent |
| Age | r=–.07 | r=–.07 | r=–.21** | r=–.23** |
| Male | r=–.05 | r=–.04 | r=–.01 | r=–.02 |
| Race-ethnicitya | χ2=11.53 | χ2=7.26 | χ2=48.80** | χ2=74.08** |
| Education (years) | r=–.02 | r=–.04 | r=–.03 | r=.01 |
| Married | r=–.00 | r=–.02 | r=–.02 | r=.00 |
| Service erab | χ2=6.72 | χ2=5.89 | χ2=28.30** | χ2=34.62** |
| Diagnosis | ||||
| Psychotic disorder | r=–.04 | r=–.02 | r=–.07 | r=–.06 |
| Mood disorder | r=–.04 | r=–.07 | r=–.12** | r=–.12** |
| Personality disorder | r=–.04 | r=–.04 | r=–.06 | r=–.09* |
| Posttraumatic stress disorder | r=–.06 | r=–.05 | r=–.05 | r=–.05 |
| Adjustment disorder | r=–.05 | r=–.07 | r=–.03 | r=–.05 |
| Alcohol abuse or dependence | r=.20** | r=.21** | r=.06 | r=.03 |
| Drug abuse or dependence | r=.08 | r=.03 | r=.36** | r=.37** |
| Other disorder | r=–.06 | r=–.05 | r=–.01 | r=–.05 |
| Dual diagnosisc | r=.02 | r=.02 | r=–.01 | r=–.02 |
| Lifetime homelessness (years) | r=.01 | r=.02 | r=.02 | r=–.01 |
| N times homeless in lifetime | r=.05 | r=.06 | r=.06 | r=.04 |
| Nights homeless in past 3 months | r=–.16** | r=–.16** | r=–.21** | r=–.21** |
| Income in past month ($) | ||||
| Employment | r=.10* | r=.11** | r=.07 | r=.11** |
| Public support | r=–.09* | r=–.09* | r=–.07 | r=–.05 |
| Other sources | r=.16** | r=.15** | r=.17** | r=.14** |
| Total income | r=.06 | r=.08 | r=.08 | r=.13** |
| Receives VA service-connected disability benefits | ||||
| Psychiatric condition | r=.02 | r=.03 | r=–.02 | r=–.01 |
| Medical condition | r=–.03 | r=–.01 | r=–.01 | r=–.01 |
| Receives Non-VA disability benefitsb | χ2=6.65 | χ2=8.40 | χ2=2.76 | χ2=2.94 |
TABLE 1 Association between characteristics, income and disability compensation, and substance use among 1,160 homeless veterans
When analyses were repeated with a sample that included only those with substance use disorders, multiple regression analyses showed that VA and non-VA disability benefits were not associated with alcohol and drug use or with money spent on alcohol or drugs. As in the main analyses, employment income was significantly associated with money spent on drugs (beta=.15, p<.001), and income from other sources was significantly associated with number of days of alcohol use (beta=.10, p=.003) and drug use (beta=.22, p<.001) and with money spent on drugs (beta=.35, p<.001).
Discussion and conclusions
About a third of homeless veterans in this study reported spending money on alcohol in the past month, and about a fifth reported spending money on drugs in the past month, underscoring the importance of substance abuse treatment for this population and the necessity of publicly financed substance use treatment services. However, no positive association was found between public support income and money spent on alcohol or drugs, suggesting that the amount of income homeless veterans receive from disability and other public support sources does not influence their amount of substance use. This finding is consistent with the small body of literature that has similarly found no relation between public support income and substance use (8–10) and extends those findings by specifically examining the amount of money spent on substances and separating VA disability compensation from other public support income. No association was found between VA disability compensation and substance use, which may help allay fears that veterans are using taxpayer money to buy alcohol and drugs and may also contribute to the larger policy debates over disability payments and substance abuse in the general population (6,7). Thus efforts to provide disability compensation to eligible veterans, even those with substance use problems, should be continued.
Several caveats about this study should be acknowledged. This study was limited to observational, cross-sectional data, and causality of associations found cannot be inferred. Data were based on self-reported income and substance use in the past month, which is a limited time frame, and there are potential response biases with self-report. Only veterans referred to the HUD-VASH program were included in the sample, which may not be representative of all homeless veterans. Moreover, the sample was predominantly male, and most participants had served before the Persian Gulf War era. Therefore, generalizability to females and veterans from more recent conflicts is unclear. Strengths of the study included a large sample of homeless veterans, specific information about money spent on alcohol and drugs, appropriate use of nonparametric analysis, and findings that contribute to the literature.
1 : Risk factors for homelessness. Epidemiologic Review 15:546–556, 1993Crossref, Medline, Google Scholar
2 : Latent homeless risk profiles of a national sample of homeless veterans and their relation to program referral and admission patterns. American Journal of Public Health 103(suppl 2):S239–S247, 2013Crossref, Medline, Google Scholar
3 : Individual characteristics of the literally homeless, marginally housed, and impoverished in a US substance abuse treatment-seeking sample. Social Psychiatry and Psychiatric Epidemiology 43:831–842, 2008Crossref, Medline, Google Scholar
4 : Sobering up on the streets: homeless men in alcoholics anonymous. Society 47:333–336, 2010Crossref, Google Scholar
5 : Disability income, cocaine use, and repeated hospitalization among schizophrenic cocaine abusers—a government-sponsored revolving door? New England Journal of Medicine 333:777–783, 1995Crossref, Medline, Google Scholar
6 : Disability payments and chemical dependence: conflicting values and uncertain effects. Psychiatric Services 48:789–791, 1997Link, Google Scholar
7 : Welfare reform and substance abuse. Milbank Quarterly 83:65–99, 2005Crossref, Medline, Google Scholar
8 : Do public support payments encourage substance abuse? Health Affairs 15(3):192–200, 1996Crossref, Google Scholar
9 : The relationship of public support payments to substance abuse among homeless veterans with mental illness. Psychiatric Services 48:792–795, 1997Link, Google Scholar
10 : Outcomes after initial receipt of Social Security benefits among homeless veterans with mental illness. Psychiatric Services 51:1549–1554, 2000Link, Google Scholar
11 Secretary Shinseki details plans to end homelessness for veterans. Press release. Washington, DC, US Department of Veterans Affairs, Office of Public and Intergovernmental Affairs, 2009Google Scholar
12 : Association of substance use and VA service-connected disability benefits with risk of homelessness among veterans. American Journal of Addiction 20:412–419, 2011Crossref, Medline, Google Scholar
13 : Alcohol and drug use disorders among homeless veterans: prevalence and association with supported housing outcomes. Addictive Behaviors 39:455–460, 2014Crossref, Medline, Google Scholar
14 : Homelessness and money mismanagement in Iraq and Afghanistan veterans. American Journal of Public Health 103(suppl 2):S248–S254, 2013Crossref, Medline, Google Scholar
15 : Cost-effectiveness of supported housing for homeless persons with mental illness. Archives of General Psychiatry 60:940–951, 2003Crossref, Medline, Google Scholar
