Equity in veterans' mental health care: Veterans Affairs medical center clinics versus community-based outpatient clinics

This study examined differences in structures and processes of mental health care at Veterans Administration (VA) primary care clinics, comparing VA medical center (VAMC) clinics to community-based outpatient clinics (CBOCs). A survey was conducted of nurse managers at 46 of 49 primary care clinics (23 VAMC clinics and 23 CBOCs) within a VA health care network in the south central United States. Integration of care and services overall was comparable between VAMC clinics and CBOCs. The service mix differed. Integrated CBOCs more often offered group therapy, medication management, and smoking cessation. Integrated VAMC clinics more frequently used written suicide protocols and depression screening. Distance to offsite specialty care and wait times for referrals were shorter for patients at VAMCs than at CBOCs. The provision of mental health care at CBOCs is comparable to that at VAMC clinics, although differences in patient access to offsite care indicate that full equity was not achieved at the time of the survey. Since 2000, the VA has initiated several programs to address this need.

Blending education, research, and service missions: the Arkansas model

Creating school and community partnerships with academic health centers (AHC) offers one strategy for initiating and sustaining broad-based change in health systems. This article describes the development, initial evaluation, and current iteration of the Arkansas Partners in Behavioral Health Sciences Model, a collaboration between personnel from an AHC and K-12 schools to address behavioral health issues in children. The model's focus on education, research, and service provides an opportunity for AHC faculty and school personnel to collaborate to promote mental health in school-aged youth. Quantitative and qualitative methods have been used to inform development and confirm effectiveness of the program. From 2001 through 2005, more than 2,700 school personnel from 72 of the 75 counties in Arkansas participated in more than 30,000 hours of continuing education. The programs have also targeted students using interactive televideo presentations, supplemental classroom curricula, and an exhibit in a state science museum, resulting in an outreach to more than 2,500 youths. Results of longitudinal and randomized studies also show changes in knowledge, attitudes, and behaviors. In an era of extraordinary need and finite resources for school systems, AHCs are poised to provide the critical link to improve the scientific knowledge and understanding of behavioral health conditions. The current program targets behavioral health, but AHCs also can incorporate other health conditions, scientific topics, and medical interventions to provide a important service for the public and to accomplish an important mission toward health leadership in the community.

Disagreement between parent and adolescent reports of functional impairment

OBJECTIVE

Adolescents' functional impairment has become increasingly important as a criterion for diagnosis and service eligibility as well as a target of therapeutic intervention in mental health settings. This study examines three critical issues in measuring functioning: 1) agreement between parent and adolescent reports of functioning, 2) explanations for disagreement, and 3) clinicians' ratings of functioning compared with parent and adolescent reports.

METHODS

Agreement between parent and adolescent reports of functioning was estimated using the kappa statistic and conditional agreement in a sample of 258 adolescents. Rates of and reasons for expected disagreements between informants were explored in semi-structured interviews (n = 43). ANOVA was calculated for clinician ratings for parent-adolescent pairs categorized on the basis of their agreement or disagreement on impairment. Finally, the independent contribution of parents' or adolescents' reports of impairment on clinician ratings of functioning was examined.

RESULTS

From 12% to 97% of problems reported by one informant were denied by the other. Agreement was particularly poor for questions about relationships with friends, peers' delinquent behaviors, and leisure activities. On average, parents and adolescents were more likely to expect the other would agree with their ratings rather than disagree. Reasons for disagreement included: 1) differences in how parents and adolescents interpreted questions; 2) lack of parental awareness of adolescents' behaviors; and 3) different thresholds for what is considered problematic. Results also demonstrated that clinicians perceive problems reported only by parents as somewhat more serious than problems reported only by adolescents.

CONCLUSIONS

Implications for diagnosis, treatment planning and outcomes measurement are discussed.

Detection and outcomes of substance use disorders in adolescents seeking mental health treatment

OBJECTIVE

To compare detection rates, service use, and outcomes of substance use disorder (SUD) in adolescents seeking mental health treatment.

METHOD

Adolescents (n = 237) and their parents or caregivers completed parallel, self-administered versions of the Adolescent Treatment Outcomes Module (ATOM) at intake and 6-month follow-up. SUD was assessed using the Diagnostic Interview Schedule for Children (DISC).

RESULTS

Although 42 (16.6%) adolescents seeking mental health treatment met DISC criteria for any SUD at baseline, clinicians detected only 19 of these (kappa = 0.51). Sensitivity and specificity were high for a screening tool for adolescent SUD (95.2 and 82.0, respectively). Adolescents with undetected SUD were less likely to receive SUD services and to have more legal problems at 6-month follow-up compared to adolescents with detected SUD.

CONCLUSIONS

SUD rates may be high in adolescents seeking mental health treatment, confirming the need for routine screening in this population. Results confirm that a large gap exists between the need for and access to SUD treatments, which may contribute to poorer outcomes for all adolescents with SUD.

How well do automated performance measures assess guideline implementation for new-onset depression in the Veterans Health Administration?

BACKGROUND

Because most guidelines focus on patients with new episodes of depression, algorithms to identify such samples must be accurate. This study examined whether the Veterans Health Administration's (VHA's) electronic medical record database could identify valid cases of new-onset depression.

RESULTS

Of 109 individuals receiving outpatient care at one of three VHA medical centers who were identified with newly diagnosed depressive disorder, 39 (35.8%) actually had documentation of depression diagnosis and antidepressant prescription or other treatment within the previous six months. Good to excellent agreement was found between indicators of guideline-concordant care using automated and manual chart review methods.

DISCUSSION

Electronic medical records can validly identify many cases of new-onset depression, although with a higher-than-anticipated rate of false-positives. Half of depressed veterans received care consistent with clinical guidelines for psychopharmacological intervention, regardless of data source.

SUMMARY

Clinical managers, administrators, and policy advocates must weigh the cost-benefit of administrative versus medical record reviews to assess quality.

Relationship between self-report and clinician-rated impairment in depressed outpatients

This study examined the potential of using the Global Assessment of Functioning (GAF) scale as a valid measure of patient-reported status and predictor of three-month outcomes for 59 patients seeking outpatient treatment for depression. A moderate relationship between the GAF and patient report at baseline and follow-up suggests it may be clinically useful for routine monitoring of depressed patients.

Service utilization and outcomes in medically ill veterans with posttraumatic stress and depressive disorders

This study examined behavioral health service utilization, health-related quality of life, and psychological distress in medically hospitalized male veterans (N = 743) with and without current or lifetime comorbid posttraumatic stress disorder (PTSD) and depressive disorder. Participants completed psychiatric and psychosocial self-report measures at baseline and follow-up. Clinical/functional status and service utilization rates were compared for patients with PTSD only, depressive disorder only, comorbid PTSD/depressive disorder, and neither disorder. Patients with PTSD/depressive disorder were more likely to use mental health/substance abuse services, have longer lengths of stay, and report more psychological distress than others. Results indicate that screening, early detection, and referral are critical in treating these comorbid patients because of increased psychological distress and high service-use rates.

The Depression-Arkansas scale: A validation study of a new brief depression scale in an HMO

Recent trends in mental-health care have increased the need for practical depression instruments. The Depression-Arkansas (D-ARK), a brief, economical, multipurpose instrument, has been validated for assessing major depressive disorder (MDD) and depressive-symptom severity. Psychometric properties of the D-ARK were compared with standard depression scales (Beck Depression Inventory and Geriatric Depression Scale) among 294 adult and 193 senior primary-care patients, respectively, and 163 patients enrolled in cognitive-behavioral depression classes. The severity scale displayed adequate internal reliability (coefficient alpha =.81-.86), high correlation with the BDI-2 (r =.78-.83) and GDS (r =.75), and similar factor structure to the BDI-2. The D-ARK was calibrated against the BDI-2 and GDS, providing familiar severity category cutpoints with the new instrument. This study yields further data supporting the reliability, validity, and practical utility of the D-ARK.

The impact of father absence on adolescent separation-individuation

The authors compared separation-individuation and psychological separation from fathers of 25 adolescent boys who were living with both biological parents with that of 25 boys who were living with their biological mothers in homes in which the fathers did not reside. The results showed that the boys in the 2 groups did not differ on measures of separation-individuation and that the quality of the mother-son relationship mediated several of the assessed separation-individuation manifestations. The authors' initial data analysis with regard to psychological separation showed that boys who lived in homes in which the father was a nonresident were more separated on 2 of the 4 dimensions assessed; however, when controlled for quality of mother-son and father-son relationship, these differences were not significant. The frequency of father contact in homes in which the father did not reside was positively correlated with healthy separation but negatively correlated with functional, attitudinal, and emotional independence from the father. The results of the study support the importance of the quality of a child's relationship with his or her mother and father as a mediator of several dimensions of the separation-individuation process.

Case-mix adjustment of adolescent mental health treatment outcomes

Case-mix adjustment methods are needed to account for differences between providers when the youth they treat have characteristics that adversely affect treatment success. This study explores variables for adjusting mental health treatment outcomes for adolescents and the differential effects of case-mix adjustment on providers' performance. Linear regression modeling was used to identify case-mix variables for five outcomes. Predictive equations for each outcome were developed for models based on intake clinical data alone, clinical data plus administrative data, and clinical data plus data describing youth history and family environment. Variance explained by intake clinical data alone did not increase appreciably with the addition of administrative data or data describing youth history and family environment. Adjusting outcomes changed the relative performance of certain individual providers substantially, but had a more moderate impact on the overall interpretation of providers' performance.

Validity of the Depression-Arkansas (D-ARK) Scale: a tool for measuring major depressive disorder

The results of 2 validation studies for an assessment tool designed specifically for quality improvement and outcomes assessment efforts in mental healthcare are presented in this paper. The studies evaluated a new tool to assess the patient outcomes for major depressive disorder following treatment in routine clinical settings called the Depression-Arkansas Scale (D-ARK). Study 1 included 54 patients recruited from 3 hospital-based clinics (2 mental health clinics and 1 primary care clinic). Study 2 includes 827 patients from 5 clinical settings including a university based outpatient clinic, a VA based mental health clinic, and a managed-care program. These 2 very different studies provide preliminary evidence that the D-ARK may be a useful tool for quality improvement efforts in the mental healthcare setting. Specifically, they indicate that the D-ARK has strong validity when compared to 2 different research assessments, the Structured Clinical Interview for DSM-III-R, Patient Edition (SCID-P) and the Inventory to Diagnose Depression (IDD), and compared to clinical assessments using both the clinical diagnosis and a clinician checklist.

Comparing outcomes of routine care for depression: the dilemma of case-mix adjustment

The purpose of this study was to formulate and test two case-mix models for depression treatment that permit comparisons of patient outcomes across diverse clinical settings. It assessed demographics; eight, diagnostic-specific, case-mix variables; and clinical status at baseline and follow-up for 187 patients. Regressions were performed to test two models for four dependent variables including depression severity and diagnosis. Individual treatment settings were then ranked based on a comparison of actual versus predicted outcomes using regression coefficients and predictor variables. A model inclusive of baseline physical health status and depression severity predicted depression severity, mental health, and physical health functioning at follow-up. A simpler model performed well in predicting depression remission. This study identifies variables to be included in case-mix adjustment models and demonstrates statistical methods to control for differences across settings when comparing depression outcomes.

Toward a level playing field: predictive factors for the outcomes of mental health treatment for adolescents

OBJECTIVE

To understand better the effectiveness of routine treatment for emotional and behavioral problems experienced by adolescents, methods are needed to control for between-provider differences in the distribution of factors that adversely affect treatment success. Such methods are necessary to fairly compare providers' outcomes and to aid clinicians in identifying adolescents for whom routine care may need to be altered. As a preliminary step toward developing a model to adjust treatment outcomes to account for predictive factors, findings from studies of treated samples of adolescents were reviewed to identify the factors that influence the likelihood of treatment success for this population.

METHOD

Medline and PSYCInfo databases were searched for studies of treated adolescents that reported the association between expert-nominated predictive factors and outcomes. Thirty-four studies met inclusion criteria.

RESULTS

Significant predictors identified in these studies include diagnosis, baseline severity of symptoms and functional impairment, family dysfunction, and previous treatment. Several expert-nominated factors have not been adequately studied in treated samples.

CONCLUSIONS

Much basic work is needed before a convincing body of empirical evidence can explain predictive factors for adolescent mental health treatment outcomes. Future efforts should determine a reduced set of predictive factors that can be measured with minimal burden to providers.

Psychiatric practice variations in the diagnosis and treatment of major depression

OBJECTIVE

Practice variations in the diagnosis, treatment, and outcomes of patients with major depression were examined within six psychiatric practices participating in a national outcomes-management project.

METHODS

Six of 20 psychiatric clinics met selection criteria for this study and provided a database of 5, 106 patients. Patients completed the BASIS-32, the Short-Form-36 Health Survey, and a Beginning Services Survey. Treatment information was also obtained directly from the clinician or through a medical record review.

RESULTS

Although 73.1 to 77 percent of patients screened positive for a depressive disorder, only 18.5 to 36.8 percent were diagnosed with major depression (p<.001). Between 39 and 72 percent of patients received psychotropic medications, a significant difference across sites (p<.001). In addition, the number of psychotherapy sessions was significantly different across sites (p<.001).

CONCLUSIONS

Patient care varies considerably across psychiatric practices, a finding that is particularly relevant for developers of performance indicators and risk-adjustment strategies for mental health.

Making the grade: update on report card initiatives for 1997

Report cards for healthcare services are increasingly in the news, offering the hope that objective information on the quality of health plan and providers services will eventually enable purchasers and consumers to make selections based on true value. Following is a series of five brief articles that review ongoing report card initiatives in private and public sectors of the behavioral healthcare system. The first four articles review actual report cards designed to hold organizations--particularly managed care--accountable for the quality of their services. The last article reviews research on performance measurement across all segments of the behavioral healthcare industry.

Measuring treatment effectiveness. Part one: Newly emerging outcomes databases for organizations

Provider organizations are under increased pressure from payors to produce outcomes data that demonstrate the effectiveness of their treatment services. More and more, providers are collaborating on joint outcomes projects that allow them to benchmark their results with similar organizations. Here are five brief descriptions of such projects. They span a wide spectrum of organizations and include hospitals and comprehensive healthcare systems, behavioral group practices, child welfare agencies, and psychosocial rehabilitation programs. Part one begins below, part two, on page 45.

The comorbidity of post-traumatic stress disorder and suicidality in Vietnam veterans

Vietnam veterans (N = 232) recruited from three sources were assessed for suicidal thinking and behaviors, and symptoms of posttraumatic stress disorder (PTSD) and depression. Findings support the notion that suicidal thoughts are prevalent in this group, with veterans in psychotherapy reporting a greater likelihood of such symptoms (82.6%) than veterans in the community (35.7%) or those seeking assistance through a veteran's outreach center (66.7%). Thoughts of ending one's life and a previous suicide attempt were significantly correlated with a diagnosis of PTSD (r = .53, p < .001; and r = .33, p < .001, respectively). Veterans with a diagnosis of PTSD and depression or dysthymia were also more likely to report suicidal thinking and behaviors than veterans with only one of the diagnoses.

Children of disaster in the second decade: a 17-year follow-up of Buffalo Creek survivors

OBJECTIVE

To conduct a long-term follow-up of child survivors of a devastating human-caused disaster.

METHOD

Child survivors (2-15) of the Buffalo Creek dam collapse, first evaluated in 1974, 2 years postdisaster, were reevaluated 17 years postdisaster when they were adults. Of the original 207 children, 99 were located and reevaluated using ratings on the Psychiatric Evaluation Form, the Impact of Event Scale, and the SCL-90 and lifetime and current diagnoses from the Structured Clinical Interview for DSM-III-R.

RESULTS

Ratings of psychiatric symptoms at the two points in time showed significant decreases in overall severity ratings and in anxiety, belligerence, somatic concerns, and agitation. A few symptoms, not present in the child sample, increased over time (substance abuse, suicidal ideation). The current rate of disaster-related post-traumatic stress disorder (PTSD) was 7%, down from a postflood rate of 32%. There were no differences by age group in current psychological status; however, women evidenced more PTSD-related symptoms than did men. All current PTSD cases were women. Comparisons with similar subjects from a nonexposed community showed no differences.

CONCLUSIONS

The findings indicated that the children studied, although having shown earlier effects, had "recovered" from the event by the time of long-term follow-up.

Vulnerability-stress factors in development of posttraumatic stress disorder

Log-linear chi 2 analyses were conducted to examine potential interactions between the presence of precrime axis I psychiatric diagnoses and differential levels of crime stress in association with posttraumatic stress disorder (PTSD) in a community sample of 295 female crime victims. High crime stress was defined as crime that included either perceived life threat, actual injury, or completed rape. Crime stress level was significantly associated with PTSD after the crime. Thirty-five percent of subjects with high crime stress exposure met criteria for PTSD, as opposed to 13% of those with low crime stress exposure. Precrime diagnosis was not associated with high crime stress exposure, indicating that this is not a vulnerability factor for exposure to crime characteristics associated with increased rates of PTSD. There were no significant independent associations between precrime axis I diagnoses and PTSD after the crime. However, a significant interaction was observed among crime stress level, precrime depression, and PTSD such that the rate of PTSD was substantially higher in association with precrime depression only in the high crime stress exposure group. Major findings are consistent with previous results implicating trauma exposure as the primary factor in development of PTSD. However, the results indicate that precrime depression may constitute a vulnerability factor for development of PTSD under conditions of high crime stress exposure.