Return to 2001 Abstacts List

*260. Measuring Utilities for Mental Health Outcomes Among Individuals with Co-Occurring Substance Use Disorders and Schizophrenia

LJ Roberts, VA Puget Sound Health Care System (CESATE); B Beck, VA Puget Sound Health Care System (CESATE); K Sloan, VA Puget Sound Health Care System (CESATE); P McKnight, VA Puget Sound Health Care System (CESATE); L Lenert, USCD and San Diego Health Care System; DR Kivlahan, VA Puget Sound Health Care System (CESATE)

Objectives: This study investigates the feasibility of accurately measuring mental health outcome utilities among individuals with co-occurring substance use disorders (SUD) and schizophrenia (SZ). This effort is consistent with VHA efforts to identify cost-effective, patient-centered care.

Methods: Using a multi-media computerized assessment (Lehnert et al), utilities were measured with a visual analog scale (VAS) and standard gamble (SG). In the VAS, subjects rated health state vignettes for global quality of life on a vertical analog scale with 0.0 equal to death and 1.0 as perfect health. In the SG method, participants rated the risk of death they would be willing to accept to live without the health condition (e.g., moderately severe symptoms of schizophrenia for the remainder of their lives. Completion of the utility measurement was defined as providing 10 utility values (VAS and SG for each of the following health states: mild; mild + side effects; moderate; moderate + side effects; and own current health state).

Results: To date, 31 male participants receiving treatment for co-occurring SUD and SZ have completed the assessment. Prior to this experience, 48% of participants had never used computers. Time to complete the assessment ranged from 25 to 80 minutes, median = 40. Mean VAS scores for the aforementioned health states were: 60,45,45,36,52, respectively; and mean SG scores were 27,45,32,30,29, respectively. Sixty-one percent of the participants had at least one logically inconsistent response, and of those, 55% were repaired following prompting. Program admission alcohol severity scores (AUDIT X = 23, SD = 12), drug severity scores (DAST X = 7, SD = 3) and drug use in the previous 30 days (32%) were not significantly associated with utility scores or occurrence of logical errors. Psychopathology, as rated by the BPRS total score, ranged from 30 to 91 (X = 59, SD = 17) and was not correlated with occurrence of logical errors (all p's < .05).

Conclusions: All recruited participants were able to complete the computerized measurement of utilities. Mean scores for the VAS decreased as health states worsened. Mean scores for the SG were less consistent. Consistent with rates from studies involving SZ participants without SUD, more than half of the participants made logical errors, and among those, more than half attempted to repair those logical errors. SUD and psychiatric symptom severity was independent of making logical errors.

Impact: Substance use disorders are a significant problem among veterans with schizophrenia, and are associated with poor treatment adherence, high institutionalization costs, poor psychosocial functioning and worse treatment outcome. Given the difficulty and expense of treating this comorbid population, it is important to identify methods to evaluate cost-effective treatments that lead to outcomes that are valued by the patients. This study supports the feasibility of accurately measuring health state utilities among individuals with co-occurring schizophrenia (SZ) and substance use disorders (SUD).