Psychiatric correlates of past incarceration in the national co-morbidity study replication.

BACKGROUND: Mental illness and substance abuse have been increasingly linked to criminal justice system involvement, but this relationship has mostly been by survey of prison populations and inferences of excess rates of disorder made by noting how these rates compare with national population-based surveys of mental disorders. AIMS: The aim of this study is to examine associations between history of mental disorders, including substance misuse, with incarceration history within a single population-based data set. METHODS: Data were from the National Comorbidity Survey Replication, a nationally representative household survey of respondents 18 years and older conducted between 5 February 2001 and 7 April 2003. RESULTS: Multivariate regression analysis showed the strongest independent risk factors for a history of incarceration were being male [odds ratio (OR) = 6.3; p < 0.001], past receipt of welfare payments (OR = 2.1; p < 0.001), longer than 1 week of past homelessness (OR = 2.1; p < 0.001), not being from the northeast of the USA (OR = 0.31; p < 0.001) and a lifetime substance abuse or dependence diagnosis (OR = 4.9; p < 0.001). With the exception of welfare payments, these measures were also independently associated with longer (27 + days) incarceration. CONCLUSIONS: The socioeconomic associates of incarceration history were unexpected, and in line with other, differently conducted studies. The fact that only substance misuse disorders of all those assessed were independently associated with incarceration history was a surprise, given the multiplicity of prison surveys, which have shown higher rates of other serious mental disorders. Although we were unable to include measures of schizophrenia or similar psychosis and used impulse control disorders as surrogates for personality disorder, absence of a relationship between depression and incarceration when measured in the same way and over the same time among those previously incarcerated and those not, raises questions about the weight that should be put on the existing epidemiological perspective of mental disorder among prisoners.

Incarceration among male veterans: relative risk of imprisonment and differences between veteran and nonveteran inmates.

Using data from national surveys of jail and prison inmates conducted in 2002 and 2004, the authors found that male veterans in the age group that entered military service in the early years of the All Volunteer Force (AVF) were at greater risk of incarceration than nonveterans of similar age and ethnicity, whereas veterans who enlisted in later years of the AVF had less risk of incarceration than nonveterans. Although White veterans tend to have greater risk of incarceration than nonveteran Whites, Black and Hispanic veterans were at less risk than their nonveteran peers, although they are at greater risk than White veterans. These patterns are best explained by changes over time and in differential effects across racial/ethnic groups of recruiting practices, accession standards, and in civilian employment opportunities rather than combat trauma or other adverse experiences in the military. For example, reductions in the relative risk for incarceration of veterans during the AVF appear to generally result from increases in recruit qualifications and socioeconomic status due to greater military pay, improved skill in recruiting, and higher accession standards.

Mental health correlates of past homelessness in the National Comorbidity Study Replication.

This study uses data from a nationally representative epidemiologic survey, the National Comorbidity Survey Replication, to investigate the association of mental health and substance use disorders, along with other factors, with past homelessness. Approximately 5% of the 5,251 adults reported having been homelessness. Multivariate analysis showed the strongest independent risk factors for past homelessness were past receipt of welfare payments (odds ratio [OR]=5.7), incarceration for 27 or more days (OR=3.9), exposure to personal violence (OR=2.7), lifetime substance use disorder (OR=2.4), and Black race (OR=2.1). Several non-substance use psychiatric disorders were also significantly, if less strongly (OR 1.4 to 1.6), associated with past homelessness. Past homelessness is associated with a broad array of sociodemographic, economic, and mental health problems. While the association of both substance use and psychiatric disorders with past homelessness was quite strong, non-substance use psychiatric disorders was not as strong an independent risk factor as substance abuse disorders.

Correlates of past homelessness in the National Epidemiological Survey on Alcohol and Related Conditions.

This study sought to investigate correlates of past homelessness, especially mental illness and substance abuse, using data from the National epidemiological survey on alcohol and related conditions (NESARC), a large nationally representative survey conducted in 2001-2002. Multivariate analyses showed that the factors most strongly related to past homelessness were diagnoses of behavioral health conditions which showed consistently stronger association than sociodemographic characteristics, measures of economic well being, or general health indicators. The results presented here confirm in a nationally representative sample a number of single site studies that have also demonstrated the over representation of both mental illness and substance abuse disorders in homeless populations.

An evaluation of an initiative to improve coordination and service delivery of homeless services networks.

This study examines system changes associated with the implementation of the Collaborative Initiative to Help End Chronic Homelessness, an 11-site multi-agency intervention for chronically homeless adults. Data obtained from key informants on community-level interventions and interorganizational relationships were gathered from an average of 6.6 agencies at each site in four yearly waves. Hierarchical linear modeling was used to examine time trends and bivariate relationships between measures. There were significant increases over the full study period in the use of practices designed to encourage system integration, as well as in interorganizational measures of joint planning and coordination, and of trust and respect, although in later waves of the study these measures leveled off. There were also highly significant and positive cross-sectional associations between the use of practices designed to encourage system integration and direct measures of service system integration as well as between measures of change in these system characteristics.

Mental health and other risk factors for jail incarceration among male veterans.

Data derived from the 2002 Survey of Inmates in Local Jails and the 2000 National Survey of Veterans show that having mental health problems in addition to such sociodemographic characteristics as being a member of a minority group, not being married, having less education, and being younger are risk factors for incarceration among veterans, as they are for the general population. As in previous studies veterans who served during the Vietnam Era and to an even greater extent, those who served in the early years of the All Volunteer Force were at greater risk of incarceration than veterans from the most recent period of the AVF, after controlling for age and other factors.

An evaluation of an initiative to improve Veterans Health Administration mental health services: broad impacts of the VHA's Mental Health Strategic Plan.

In federal fiscal year (FY) 2005 the Department of Veterans Affairs (VA) implemented the comprehensive Mental Health Strategic Plan (MHSP). This study used performance measures from six broad domains to examine changes in the overall delivery of mental health services in the VA since the implementation of the MHSP. Performance measures from fiscal year 2004, the year before implementation of the MHSP, were compared with measures from fiscal years 2005, 2006, and 2007, the first 3 years of MHSP implementation. We combined heterogeneous performance measures within domains through the use of standardized scores or "z-scores." An overall improvement of 0.32 standardized units was observed from FY 2004 to FY 2007, representing moderate to large changes by conventional standards. The domains with the greatest improvement (>1.0 standard deviation units) from FY 2004 to FY 2007 were population coverage/access, outpatient care quality, economic performance (primarily efficiency), and global functioning. There was a 0.3 standard deviation decline in inpatient satisfaction and a slight increase in reliance on inpatient care. Overall improvement in VA mental health care was thus substantial and continuing.

Client-level measures of services integration among chronically homeless adults.

INTRODUCTION: While several major studies have examined services integration at the system or interagency level, there has been far less effort to measure the integration of services at the client-level and its correlates. METHODS: This study presents three client-level measures of services integration, two objective measures, representing the proportion of needed services received and the number of outpatient services received by each client, and one subjective measure, a five-item scale measuring perceived coordination of care among clients' service providers. Data from the evaluation of the collaborative initiative to help end chronic homelessness (CICH) are used to examine bivariate and multivariate relationships of these three client-level measures to two system-level measures of services integration, one addressing interagency services coordination/planning and the other interagency trust/respect as well as to baseline client characteristics among 734 chronically homeless adults in 11 cities. RESULTS: Client-level measures of service integration were not strongly associated to each other or to the system-level measures, except for weak associations (r = 0.10) between one objective client measure and the system-level measure of service coordination and planning, and another between client-level use of outpatient mental health services and system-level trust and respect. Multivariate analysis showed that clients who received a greater array of needed services received more service overall and were more likely to have a diagnosis of PTSD and more medical problems, but less serious alcohol problems. Clients who reported more outpatient mental health and substance abuse visits were significantly more likely to be married, to be veterans, to have more serious drug problems, and to be dually diagnosed. Clients with more serious drug problems reported poorer coordination among their service providers on the subjective measure of client-level service integration. CONCLUSION: Three client-level measures of services integration were, at best, weakly associated with measures of system-level integration. Positive associations between client-level measures of integration and health status, outpatient service use and negative relationships with indicators of substance abuse suggest they may usefully represent the experiences of chronically homeless clients, even though they are not strongly related to system-level measures.

Homelessness in the state and federal prison population.

AIMS: This study sought to investigate the rates and correlates of homelessness (i.e. living on the street or in a homeless shelter), including mental illness, among US adult state and federal prison inmates (ASFPIs). METHOD: Data from a national survey of ASFPIs based on a random sampling survey (N = 17,565) were used to compare the homelessness rate among AFSPIs with that in the general population. Logistic regression was then used to examine the association of homelessness among ASFPIs with factors including symptoms, treatment of mental illness, previous criminal justice involvement, specific crimes, and demographic characteristics. RESULTS: Nine percent of ASFPIs reported an episode of homelessness in the year prior to arrest, 4-6 times the estimated rate in the general US adult population after allowing for age, race/ethnicity, and gender. In comparison to other inmates, these homeless inmates were more likely to be currently incarcerated for a property crime, but also to have had previous criminal justice system involvement for both property and violent crimes, to suffer from mental health and/or substance abuse problems, and to be more likely to have been unemployed and with a low income. CONCLUSIONS: Recent homelessness is far more common among ASFPIs than the general population. Prior incarceration, mental illness, substance abuse and disadvantageous socio-demographic characteristics were all found to be associated with homelessness among prison inmates, suggesting that there are several important factors in addition to efforts to survive with limited resources through criminal acts that influence the rates of homelessness among incarcerated individuals.

Jail incarceration, homelessness, and mental health: a national study.

OBJECTIVE: This study sought to investigate the rates and correlates of homelessness, especially mental illness, among adult jail inmates. METHODS: Data from a national survey of jail inmates (N=6,953) were used to compare the proportion of jail inmates who had been homeless in the previous year with the proportion of persons in the general population who had been homeless in the previous year, after standardization to the age, race and ethnicity, and gender distribution of the jail sample. Logistic regression was then used to examine the extent to which homelessness among jail inmates was associated with factors such as symptoms or treatment of mental illness, previous criminal justice involvement, specific recent crimes, and demographic characteristics. RESULTS: Inmates who had been homeless (that is, those who reported an episode of homelessness anytime in the year before incarceration) made up 15.3% of the U.S. jail population, or 7.5 to 11.3 times the standardized estimate of 1.36% to 2.03% in the general U.S. adult population. In comparison with other inmates, those who had been homeless were more likely to be currently incarcerated for a property crime, but they were also more likely to have past criminal justice system involvement for both nonviolent and violent offenses, to have mental health and substance abuse problems, to be less educated, and to be unemployed. CONCLUSIONS: Recent homelessness was 7.5 to 11.3 times more common among jail inmates than in the general population. Homelessness and incarceration appear to increase the risk of each other, and these factors seem to be mediated by mental illness and substance abuse, as well as by disadvantageous sociodemographic characteristics.

Compensation of veterans with psychiatric or substance abuse disorders and employment and earnings.

Using a national sample of veterans, we examined the relationship between disability income and employment, adjusting for health status and other factors. Veterans Affairs disability income payments had no globally detrimental effect on labor force participation, in that the likelihood of employment was reduced only at payment levels of more than $800 per month. Although unearned income from other sources also did not have a substantial negative effect on labor force participation, veterans who received benefits from the Social Security Administration or welfare payments were less likely to be employed, mostly likely because employment earnings above a certain level in some programs may result in the loss of monetary benefits and health insurance.

Factors associated with receipt of pension and compensation benefits for homeless veterans in the VBA/VHA Homeless Outreach Initiative.

Public support payments may facilitate exit from homelessness for persons with mental illness. We examined data from 10,641 homeless veterans contacted from October 1, 1995 to September 30, 2002 in a collaborative outreach program designed to facilitate access to Department of Veterans Affairs (VA) disability benefits. Those who were awarded benefits (22% of contacted veterans) were more likely to report disability, poor to fair self-rated health, and were more likely to have used VA services in the past. Thus, this program achieved only modest success and was most successful with veterans who were already receiving VA services and who might have received benefits even without the outreach effort.

Trends in the professional mix and cost of outpatient mental health care.

BACKGROUND: In recent years, lower cost non-physician providers have been increasingly used to deliver outpatient mental health services. METHODS: We compared 1997 and 2004 data on outpatient services provided by the Veterans Health Administration to examine trends in outpatient provider mix and to evaluate the association of outpatient provider mix with average per capita outpatient mental health costs. RESULTS: We found a small decline in the proportion of outpatient mental health services provided by psychiatrists and an increase in the proportion of such services provided by nurses. However, greater use of non-physician providers, at the facility level, was associated with greater, rather than lower per capita outpatient costs and there was no significant change over time. CONCLUSION: While an increasing proportion of outpatient mental health visits were with non-psychiatrists, there was not an associated decline in per capita mental health costs. Increasing VA workloads over the past decade have forced increases in psychiatrist panel size, substantially increasing physician efficiency. Additional non-physician staff have supported the development of psychosocial rehabilitation and crisis intervention capacities needed to cope with a 63% decline in inpatient beds and thus have most likely been complements of, rather than substitutes for, psychiatrists.

Use of nationwide outcomes monitoring data to compare clinical outcomes in specialized mental health programs and general psychiatric clinics in the Veterans Health Administration.

There has been a growing interest in the implementation of evidence-based specialized mental health programs. However, there has been little study of the effectiveness of these programs in comparison with standard mental health care in real world mental health systems. This study used a national sample of patients from the Veterans Health Administration to compare changes in mental health status in various specialized mental health outpatient programs and in general psychiatric clinics. Hierarchical linear models were used to compare the association of both regularity and intensity of care in six specialized mental health programs with GAF change scores in patients treated in general psychiatric clinics. While improvements were observed in all programs, two specialized programs performed better overall than general psychiatric care, one was not significantly different, and three had poorer outcomes than general psychiatric clinics. Programmatic differences in target populations accompanied by imperfect risk adjustment for population differences most likely explain why these results differ from those observed in clinical trials. While the analytic strategies demonstrated here may have wider applicability to comparative performance assessment, this study provides a cautionary tale concerning the limits of conclusions that can be drawn from large scale outcomes monitoring efforts.

Access of behavioral health patients to prescribing professionals.

OBJECTIVE: Advanced practice nurses (APNs), including clinical nurse specialists and nurse practitioners, now have prescribing privileges in all states. This study examined the proportion of Department of Veteran Affairs (VA) mental health patients who were seen by any prescribing professional and specifically the proportion and characteristics of patients who were treated exclusively by APNs. METHOD: Logistic regression models were used to examine data on all patients who received care in VA specialty mental health clinics in 2002 (n=767,920). We first identified patient characteristics independently associated with prescriber contact and, secondly, among those with prescriber contact, exclusive contact with an APN. We also compared characteristics of patients seen exclusively by an APN with those who saw both a physician and an APN. RESULTS: The strongest predictors of both whether a veteran saw a prescriber (66.6%) and whether that prescriber was exclusively an APN (6.7%) were indicators of severity. Specifically, more severe diagnoses, such as schizophrenia or bipolar disorder, receipt of VA disability payments and greater service use increased the odds that a veteran would have had a prescriber contact, decreased the odds that they would see an APN exclusively and, among those who saw an APN, increased the odds that they would also see a physician. CONCLUSIONS: Patients who see physician prescribers have more severe mental health problems than those who see APNs. Our results indicate that APNs and physicians treat distinguishable patient populations, suggesting that APNs may not be substitutes for physicians.

Housing outcomes for hospitalized homeless veterans.

This study examines housing status at the time of hospital discharge in a national sample of 3,502 veterans who were homeless at admission to a Veterans Health Administration (VHA) medical center. A supplemental survey on homelessness was added to the annual VHA inpatient census over a 4-year period (1995 through 1998). Data from this survey were used to identify homeless veterans hospitalized on the final day of the fiscal year and to document their housing status at discharge. A signal detection technique, Receiver Operating Characteristic analysis, was used to identify predictors of housing status at discharge. At discharge, only 13% of the veterans in our analytic sample were literally homeless, 40% were doubled up, and 33% were transferred to another an institution; only 13% were living independently. Housing status at admission (specifically, staying with friends or family temporarily rather than literally homeless), treatment in a psychiatric or substance abuse program rather than a medical program, and a greater income level or access to VHA benefits were found to be associated with better housing outcomes. As one might expect, homelessness was infrequently resolved during an acute inpatient hospitalization, but few veterans returned to literal homelessness at the time of discharge.

Does system reform reduce geographic variation in mental health system performance.

Substantial regional variation in health service use has been reported for both general medical and mental health services. It is unknown however, whether regional variation is reduced during periods of major system reform. Data from the Department of Veterans Affairs' are used to examine performance measures for its 22 regional networks from 1996 to 2001, a period of major system reform. Out of 21 mental health performance measures the coefficient of variation (the standard deviation divided by the mean) increased for 16 measures and declined for only 5. Although regional variation increased only slightly on 12 of these 16 measures it appears system reform clearly does not reduce regional variation across the board and it may in fact increase such variation for some aspects of care.

Using the GAF as a national mental health outcome measure in the Department of Veterans Affairs.

OBJECTIVE: Data from the Veterans Health Administration (VHA) were used to evaluate the strengths and weaknesses of the Global Assessment of Functioning (GAF), a single-item mental health status measure, as an outcome measure for large mental health care systems. METHODS: The sample consisted of VHA mental health patients who had at least two GAF scores 45 days apart in 2002 (N=283,754). First, to evaluate the discriminant validity of the GAF change measures, the authors examined the association of these measures with sociodemographic and clinical characteristics. Facility-level risk-adjusted measures of GAF change were then created in three different clinical samples at more than 130 VHA medical centers, adjusting for patients' sociodemographic characteristics and diagnoses. The internal consistency of the scale created by using these items and their consistency across medical centers over time was evaluated. RESULTS: The analysis supported the discriminant validity of the GAF-derived measures. As expected, veterans who had a diagnosis of schizophrenia or Alzheimer's disease or who had service-connected disability ratings above 50 percent had lower baseline GAF scores and showed less improvement. The overall GAF performance measure had a high level of internal consistency (a standardized alpha of .85) and was highly consistent across facilities over time. CONCLUSIONS: The results of this study provide preliminary empirical support for cautious use of a GAF-derived scale in monitoring changes in average facility-level outcomes over time. However, because of the potential for gaming of the measures and uncontrolled variation in the scale's administration across facilities, the scale should not be used to compare outcomes across facilities.

Continuity of care and clinical outcomes in a national health system.

OBJECTIVE: Continuity of care is widely viewed as a key quality indicator for outpatient mental health care. However, few studies have been conducted of the relationship between continuity of care and client outcomes. This study examined the relationship between measures of three aspects of continuity of care (regularity of care, continuity of treatment across organizational boundaries, and intensity of treatment) and the Global Assessment of Functioning (GAF), a single-item mental health status measure, in a national health care system. METHODS: Three analytic samples were derived from a nationwide Department of Veterans Affairs administrative data set: patients with at least one inpatient GAF rating and a later outpatient GAF rating (N=8,334) and two groups who had at least two outpatient GAF ratings, one group that was beginning a new episode of treatment (N=49,946) and a second group in ongoing treatment (N=123,371). Hierarchical linear modeling was used to control for potential site-level autocorrelation and to adjust for differences in diagnostic status, sociodemographic characteristics, baseline GAF score, and the length of time between GAF ratings. RESULTS: Several positive and significant relationships were found for discharged inpatients and new outpatients. However, only a few of these relationships could be confidently said to be clinically meaningful. Specifically, among discharged inpatients, for every additional month in which an outpatient visit occurred over a six-month period, there was a .69 increase in the GAF change score for a total increase of 4.1 points. Among new outpatients the equivalent values were smaller, at .3 and 1.8. In contrast with the findings for discharged inpatients and new outpatients, high intensity of care was negatively associated with GAF change scores for continuing outpatients. CONCLUSIONS: In contrast with several earlier studies, this study showed positive and statistically significant associations between several continuity-of-care measures and client outcomes. These relationships were observed only in transitional treatment situations, that is, after hospital discharge and at the beginning an episode of outpatient care, situations in which continuity of care may be especially important. However, although positive and statistically significant, the magnitude of these effects may not have been clinically meaningful.

Changes in satisfaction with mental health services among blacks, whites, and Hispanics in the Department of Veterans Affairs.

This paper presents an analysis of changes in satisfaction of minorities with inpatient mental health services provided by the Veterans Health Administration (VHA) during a period of major system change (1995-2001). Post discharge data from 16,223 veterans who received inpatient VHA mental healthservices at 87 medical centers during this period was examined using hierarchical linear models. Blacks were found to have higher satisfaction levels on most measures over the period of study while the satisfaction of whites and Hispanics were not significantly different. There was little change over the study period in the relative satisfaction of minorities and whites. Changes in patterns of VHA mental health care have not adversely affected the satisfaction of minority veterans.