Impact of Untreated Trauma on PSMD Public Treatment Costs

Impact of Untreated Trauma on Public Treatment Costs of
Persons with Severe Mental Disorder

Co-Principal Investigators: Mary Jane Alexander, Ph.D. Kristine Jones, Ph.D., Jay Yoe, Ph.D., Steve Turner, Ph.D.

PROJECT GOALS

Project aims are to:

Measure the public treatment costs of PSMD who have trauma histories, and
Determine what factors influence these costs.

RESEARCH ACTIVITIES AND RESULTS

Method: Study data files were constructed for a cohort (N=1056) that consisted of members of a class action suit against the State of Maine who were comprehensively assessed, and were Medicaid eligible from 1/96-1/98. Data files consist of Comprehensive Assessment data, Medicaid billable services, including pharmacy claims and State Hospital admissions. State hospital costs have been constructed.

The members of the study cohort of 1056 people met these criteria: They were 18-64 years of age on January 1, 1996, had a Medicaid identifier between 1996 and 1998, were inpatient at a Maine State Psychiatric Hospital, were class members in a suit against the State of Maine, and had completed a Comprehensive Assessment interview that included questions about physical and sexual abuse. The study data file included Medicaid billable services and manually collected information about the cohort’s State Hospital admissions for the major state hospital usage. State hospital financial information including census, operating costs and physical plant organization were reviewed to develop costs for State Hospital admissions.

Results: Cohort members with a trauma history (T) were more likely than those with no trauma history (NT) to report histories of substance abuse, to have been arrested, in jail or have other contact with the legal system. They were less likely to report use of anti-psychotic medications, to have a legal guardian, or to live in community based quasi-institutional settings such as group homes. They were more likely to report use of psychotropic medications that are not anti-psychotics, as well as counseling, intensive outpatient, partial hospital, residential treatment and detox services than the NT group. These results suggest that the T group experiences complex impairment that brings them into contact with multiple public systems whereas the NT is more traditionally psychiatrically impaired. Overall there was no difference in total costs between the two groups, but costs were differentially distributed. The T group had two times as many general hospital inpatient stays (but 20% lower costs), and had higher expenditures for physical health, substance abuse rehabilitation for head trauma, and “hybrid” services, a MA category that includes mental and physical health services. The NT group had higher expenditures for psychiatric medications and mental health services.

POLICY IMPLICATIONS

This study describes public sector costs in a large service system prior to the introduction of trauma sensitive or specific services. It provides virtually the only information available for policy makers to understand the cost implications of not recognizing the impact of trauma on PSMD. The data set constructed for this project considers more than mental health services, although costs of substance abuse services, criminal justice contacts and housing that is not directly financed by Medicaid are not included. The characteristics of the cohort indicate that the T group would have higher costs in each of these areas. The data structure does allow us to do further, rigorous analysis of two groups with substantial psychiatric disability who show clear differences in utilization patterns. The data set is limited to high users, making it more useful in planning for public sector services than for other purposes.

INCLUSION OF GENDER/ MINORITY GROUPS

	American Indian or Alaskan Native	Asian or Pacific Islander	Black, not of Hispanic Origin	Hispanic	White, not of Hispanic Origin	Other of Unknown Origin	TOTAL
Female					468		468
Male					588		588
Unknown
TOTAL					1056		1056

PLANS

A multivariate two-part model will be constructed that takes into account the large numbers of non users of services and that controls for factors besides trauma history that might impact costs. We will examine costs over time, taking into account the point at which the Maine BDS instituted trauma sensitive services. We also plan to refine the trauma indicator, although we are limited by the Comprehensive Assessment survey, which captures only self-reported Trauma exposure, not sequelae. We plan to submit a manuscript for publication to a mental health services research journal.

Presentations:

Center Policy Advisory Board meeting, 2000 - preliminary analysis.

American Public Health Association, 2000.

Alexander MJ, Jones K. Impact of untreated trauma on the public treatment costs of persons with severe mental disorder in the State of Maine. Workshop presented at the 5th Annual Trauma Conference. Maine Department of Behavioral and Development Services. Augusta, ME, May 2, 2003.

Updated: June 2003