Outcomes - Ohio Valley Center for Brain Injury Prevention and Rehabilitation

Home

About Us

Substance Abuse

Agitation

Resource and
Service Coordination

Outcomes

Internet and TBI

Related Publications

Contacts

Site Map

Search

Outcomes
Substance Abuse

The Cost Effectiveness of Early Intervention for Substance Abuse

Full Report
Between one third and one half of individuals sustaining a traumatic brain injury are intoxicated at the time of the injury, and almost two-thirds of adolescents and adults admitted to brain injury rehabilitation programs have a history of substance abuse. Case management for the treatment of substance abuse following brain injury has been piloted in several states based on a model of treatment developed at Ohio State University. While recent reports suggest that persons two years post-injury increase their substance use, evaluations of the OSU treatment model suggest that clients started early post-injury are more likely to drop out of treatment.

The purpose of the current study was to evaluate the cost effectiveness for persons with traumatic brain injuries of early versus late treatment of substance abuse using a case management model. Subjects were drawn from a multi-center study funded by the Center for Disease Control and Prevention, with treated groups drawn from a program at Ohio State University and comparison group subjects from the Rehabilitation Institute of Chicago. The early group was comprised of individuals who initiated treatment within one year post-injury; the late group began treatment two to five years post-injury. Outcomes after nine months of treatment were compared between groups, as well as with a comparison group of subjects with substance abuse problems, matched for time post-injury, who did not receive case management. Cost per client (based on staff contact hours) of changes in substance use, community integration, and subjective well-being were compared between early and late groups.

The power of the analyses was weakened by the loss of subjects at the 9-month follow-up. As a result, only a few significant differences were found despite most means being in the hypothesized direction. Additionally, for the entire study sample including the subsample analyzed here, there was a tendency for alcohol use to increase from admission to 9 months, which we speculated may in part be an artifact of subjects' greater comfort with the program, and thus greater honesty, at follow-up interview.

Because the power of this design appeared to be a limiting factor, we conducted post hoc comparisons of differences at admission only, those preserving subjects who were lost to follow-up at the 9-month assessment. For the 65 treated subjects referred 1 year (n=38) or 2-5 years (n=27) post-injury, there were significant differences in amount of alcohol consumed at admission (p<.05; the later group's score was 83% higher than the early group), as well as the amount of other drugs used (p<.005; the later group's score was 166% higher than the early group). There were no significant differences between the community integration and life satisfaction scores for the two groups at admission.
While this study did not support greater cost effectiveness during the first 9 months of treatment, there was considerable evidence suggestive of advantages gained from early for later care. To summarize these findings:

Clients referred for treatment 2 to 5 years post-injury were twice as likely to drop out by 9 months.
There was a trend for those treated later post-injury to increase reported use from admission to 9 months, while the early treated group did not change (alcohol) or reduced use (other drugs).
Post hoc analyses showed that upon admission the earlier referred group was consuming significantly less alcohol and other drugs than those referred later.
There was significant improvement over time in community integration, with a trend for those treated earlier post-injury to improve more than those treated later.

Further studies, with more powerful designs, should be pursued.

Background and preliminary studies

Since publication of the "White Paper" in 1988 by the National Head Injury Foundation 1 (now the Brain Injury Association), there has been an increased interest in the impact of prior and subsequent substance abuse on outcome following traumatic brain injury. It has been estimated that between one third and one half of individuals sustaining a traumatic brain injury are intoxicated at the time of the injury, and almost two-thirds of adolescents and adults admitted to brain injury rehabilitation programs have a history of substance abuse. 2 While the results of studies investigating the effects of intoxication at the time of the injury on eventual outcome have been mixed, studies have consistently indicated that a prior history of substance abuse is linked to negative outcomes. 2 Individuals who chronically abuse alcohol or other drugs have increased mortality, increased likelihood for mass lesion, and greater impairment on neuropsychological batteries at both one month and one year post-injury. 3,4 A history of substance abuse has been found to play a role in the prediction of work and school re-entry. 5,6 For those who have a history of substance abuse, prevention of the resumption of use appears to be imperative if the best possible rehabilitation outcomes are to be achieved. 7 A few models for the secondary prevention or treatment of substance abuse following brain injury have emerged in response to this concern. 8-11 Almost all models have emphasized the development of behaviors incompatible with substance abuse and have recognized that readiness for change is an important factor in determining whether the program will be effective. While some of the more traditional programs have excluded individuals who are not ready to change their behaviors, other programs have incorporated a "Stages of Change" model into their treatment methodology. 10-12 This theory emphasizes the need to work with individuals in ways that are consistent with their current awareness of the problem, and to focus on improving this awareness and their desire to change through the introduction of dissonance between substance use and the achievement of their goals. Programs for the treatment of substance abuse following brain injury also have emphasized the role of the family in generalization to the home environment, and the use of other resources to assist in generalization to the community. 7-11 In outpatient settings, a case manager is often considered to be critical to the process, to assist in providing access to community resources and supports for the family and the patient. 9
The use of case management for the treatment of substance abuse may date back to skid row mission workers and public health nurses, though the past ten years have been marked by a significant increase in the application of this approach. 13 Willenbring suggested that the increased attention given case management in alcohol and other drug treatment is due to its apparent applicability to groups that have been less responsive to conventional treatment, including those with more severe dependence, co-existing medical or psychiatric conditions, severe disability in multiple areas of life functioning, greater chronicity, or poorer socio-economic resources. 14 Additionally, the growing specialization of the alcohol and other drug treatment field has increased fragmentation in the system of care, a problem for which case management is an obvious remedy. In particular, case management addresses access to appropriate substance abuse treatment services, continuity between treatment phases, and coordination among multiple social service providers. While case management may hold promise for more complex and/or recalcitrant groups of persons with substance abuse problems, proof of the efficacy of this approach is only in its infancy. 15 Case management for the treatment of substance abuse following brain injury has been piloted in several states based on a model of treatment developed at Ohio State University.
Corrigan, Lamb-Hart and Rust described a pilot program initiated at Ohio State University in 1991 to address the problems of substance abuse among persons who had experienced traumatic brain injury. 16 Called the TBI Network, this model of treatment is community-based, using an interdisciplinary staff with specialized knowledge to coordinate services and facilitate treatment by local professionals. This model has subsequently been replicated at other treatment sites, including at Wright State University via their RRTC on Drugs and Disability. Client eligibility criteria are: (1) incurred a brain injury resulting in persistent functional impairment; (2) have a substance use disorder or are at high risk of same; and (3) are living in the community or making plans for imminent reintegration. Case management is provided to make linkages with existing substance abuse providers, and to sustain those linkages over the prolonged course of community integration. Network staff use case consultation to assist substance abuse program staff to understand the unique strengths and weaknesses of an individual, adapt services and treatment plans to the individual's abilities, and trouble-shoot problems as they arise.

The treatment philosophy of the program does not limit service coordination to substance abuse resources, as staff assist in "wrapping around" other resources and services that individuals require to stabilize their health, financial and social situation. The TBI Network model explicitly includes coordination with the vocational rehabilitation system, as well as the provision of in-house vocational services. The TBI Network staff includes an Employment Services Coordinator who, in addition to traditional job search skills training, job development, job placement and individual and group counseling, provides services usually associated with supported employment, e.g., socialization of consumers in the work place, consumer advocacy, development and maintenance of natural supports at worksites, education of co-workers, managers, and employers, and long-term follow-along. A basic tenet of the TBI Network model is that successful vocational outcomes provide longer term stability for a lifestyle incompatible with substance abuse. Programmatic outcomes of the TBI Network are operationalized in terms of both substance use and productivity.

In the design of the TBI Network, early intervention has been assumed to be an essential ingredient for addressing substance abuse after brain injury, particularly with individuals and families soon after injury. Several recent studies have reported increases in substance use among persons two or more years post-injury. 17-19 Close working relationships with providers of acute and post-acute healthcare services have been developed, allowing for extensive outreach efforts focused on persons receiving acute rehabilitation. Corrigan, Rust and Lamb-Hart reported that 42% of referrals to the program were from inpatient rehabilitation or follow-up clinics, with almost all of these clients being recently injured. 17 However, they also found that the only distinguishing characteristics between clients who dropped out of treatment (despite extensive efforts to sustain their involvement) were time post-injury (more recent being more likely) and source of referral (inpatient brain injury unit and follow-up clinic referrals being more likely). These findings broached the question whether earlier involvement was indeed more cost effective.
The purpose of the current study is to evaluate the cost effectiveness for persons with traumatic brain injuries of early versus late treatment of substance abuse using a case management model. The early group will be those individuals who initiate treatment within one year post-injury; the late group will be those initiating treatment two to five years post-injury. Outcomes after nine months of treatment will be compared between groups, as well as with a control sample of subjects with identified substance abuse problems, matched for time post-injury, who did not receive case management. Cost per client (based on staff contact hours) of changes in substance use, community integration, and subjective well-being will be compared between early and late groups. Substance use, community integration, and subjective well-being after nine months of treatment will be compared between controls, matched according to time post-injury, and earlier versus later treated groups. Subjects will be drawn from a multi-center study funded by the Center for Disease Control and Prevention, with treated and control groups drawn from treatment programs at Ohio State University, Wright State University and Rehabilitation Institute of Chicago.

Objectives

The hypotheses of the study are based on the assumption that case management services for substance abuse following traumatic brain injury are effective regardless when initiated; however, earlier versus later intervention will be more cost effective.

After 9 months of treatment, subjects admitted in their first year after injury will (a) use fewer substances, (b) have better community integration, and (c) experience better subjective well-being than a comparison group with traumatic brain injury and substance abuse problems who has not received treatment.
After 9 months of treatment, subjects admitted in 2 to 5 years after injury will (a) use fewer substances, (b) have better community integration, and (c) experience better subjective well-being than a comparison group with traumatic brain injury and substance abuse problems who has not received treatment.
Compared to treated subjects who initiate services 2 to 5 years post-injury, treated subjects who initiate treatment within the first year post-injury will show greater improvement in (a) use of substances, (b) community integration, and (c) life satisfaction.
Compared to treated subjects who initiate services 2 to 5 years post-injury, treated subjects who initiate treatment within the first year post-injury will use fewer hours of case management. This hypothesis presumes that clients treated early require fewer services.

Method

Subjects. All subjects from one of the two treatment sites (OSU) and those from the comparison site (RIC) who met the following inclusion criteria were included in these analyses: (1) brain injury of traumatic origin in the last year or between 2 and 5 years previous, (2) age greater than or equal to 18 years, and (3) post-injury substance abuse or dependence. For the treatment group, exclusion criteria were unwillingness to participate in a case management program and active, unmanaged psychosis. The eventual sample was as follows:
1 year post 2-5 years post
Treated (OSU) 38 27
Comparison (RIC) 26 25
Of these 116 subjects meeting criteria, 34 (29%) dropped-out before the 9-month assessment and could not be included in analyses requiring data from that assessment time. Subjects with 9-month follow-ups were as follows:
1 year post 2-5 years post
Treated (OSU) 28 13
Comparison (RIC) 21 20
Subjects in treatment who were two to five years post-injury were more likely to drop-out than any of the other subgroups.

Procedures and instrumentation.The analyses were conducted on a sample of patients drawn from a larger study. All program admissions were monitored by project staff and screened for study eligibility. DSM-IV criteria for alcohol and drug abuse and dependence were used. All clients who met the eligibility criteria at each of the sites were invited to participate in the study. An initial assessment battery was completed upon enrollment and again nine months later. Case characteristics and service delivery information was recorded prospectively in a consistent fashion across sites. A standard protocol was developed and all clinical staff and research assistants received joint training to assure consistency of assessment procedures. Outcome measures for the current study were the alcohol and drug composite scores from the Addiction Severity Index (ASI) 20 , the Community Integration Questionnaire (CIQ) 21 and the Satisfaction with Life Scale (SWLS) 22 . Treatment services were recorded contemporaneously by treatment staff in 15-minute increments. As the cost of a case management approach is overwhelmingly personnel, the total amount of services delivered in hours were used for efficiency calculations.

Analyses. Each hypothesis was tested using multivariate analysis of variance (MANOVA) to compare dependent variables between treatment and control groups (hypotheses 1 and 2) and early versus late treatment (hypotheses 3 and 4). Dependent variables were the Alcohol Use Composite index of the ASI, 20 the Drug Composite Index, the CIQ Total score 21 and the total score from the SWLS. 22 The latter two scores were Rasch transformed. A p<.05 level of significance was used for each MANOVA.

Findings

Regarding Hypothesis 1, both treated and comparison groups increased their use of substances, though the differences were not significant. The difference between treated and comparison groups were not significant, either; though the treated group was consuming more alcohol and other drugs at admission. Both treated and comparison groups improved significantly (p<.05) in community integration from admission to 9 months, but the two groups did not differ significantly despite the improvement for the treated group being almost 3 times that for the untreated. Life satisfaction did not show significant improvement over time or between group differences; again, despite the treated group improving approximately 20% while the untreated group declined very slightly.

Regarding Hypothesis 2, which addressed subjects 2-5 years post-injury, both treated and comparison groups increased their use of substances, though the differences were not significant. The treated and comparison groups differed significantly in their overall use, with the treated group consuming more alcohol and other drugs. The treated and comparison groups did not improve significantly in community integration from admission to 9 months, and the two groups did not differ significantly. Life satisfaction did not show significant improvement over time or between group differences; though again, the treated group improved by approximately 20% while the untreated group declined slightly.

Results of Hypothesis 3, which actually compared outcomes for the early versus later treated groups, showed no significant differences between groups or over time. However, the increase in substance use noted for the entire sample was only evident in the group referred 2-5 years post-injury (a 60% increase in alcohol use and a 35% increase for other drugs). The group referred within in the first year post-injury showed no change in alcohol use at 9 months and a decline of 30% for other drugs. For community integration, there was significant improvement over time, but the difference between groups was not significant despite those treated 2-5 years post-injury showing a 3% increase in their CIQ score while the early treated group improved by 16%. Similarly, life satisfaction improved significantly from admission to 9 months, but the differences between groups was not significant. The early treated group showed a 25% increase in SWLS score, while the later treated group showed a 27% increase.

Finally, Hypothesis 4 results showed no differences in the amount of case management services utilized by the early versus later treated groups, with the late group using about 15% fewer hours than the early group. Given the lack of significant differences in effectiveness, greater cost effectiveness of early intervention would not be supported.
Because the power of this design appeared to be a limiting factor, we conducted post hoc comparisons of differences at admission only, those preserving subjects who were lost to follow-up at the 9-month assessment. For the 65 treated subjects referred 1 year (n=38) or 2-5 years (n=27) post-injury, there were significant differences in amount of alcohol consumed at admission (p<.05; the later group's score was 83% higher than the early group), as well as the amount of other drugs used (p<.005; the later group's score was 166% higher than the early group). There were no significant differences between the community integration and life satisfaction scores for the two groups at admission.
While this study did not support greater cost effectiveness during the first 9 months of treatment, there was considerable evidence suggestive of advantages gained from early for later care. To summarize these findings:

Clients referred for treatment 2 to 5 years post-injury were twice as likely to drop out by 9 months.
There was a trend for those treated later post-injury to increase reported use from admission to 9 months, while the early treated group did not change (alcohol) or reduced use (other drugs).
Post hoc analyses showed that upon admission the earlier referred group was consuming significantly less alcohol and other drugs than those referred later.
There was significant improvement over time in community integration, with a trend for those treated earlier post-injury to improve more than those treated later.

Further studies, with more powerful designs, should be pursued.

References

1. National Head Injury Foundation Substance Abuse Task Force. White paper. Southborough, MA: National Head Injury Foundation, 1988.
2. Corrigan JD. Substance abuse as a mediating factor in outcome from traumatic brain injury. Arch Phys Med Rehabil 1995;76:302-309.
3. Ruff RM, Marshall LF, Klauber MR, et al. Alcohol abuse and neurological outcome of the severely head injured. J Head Trauma Rehab 1990;5:21-31.
4. Dikmen SS, Donovan DM, Loberg T, Machamer JE, Temkin NR. Alcohol use and its effects on neuropsychological outcome in head injury. Neuropsychol 1993;7(3):296-305.
5. Ip RY, Dornan J, Schentag C. Traumatic brain injury: Factors predicting return to work or school. Brain Inj 1995;9:517-532.
6. Girard D, Brown J, Burnett-Stolnack M, et al. The relationship of neuropsychological status and productive outcomes following traumatic brain injury. Brain Inj 1996;10:663-676.
7. Langley MJ. Preventing post-injury alcohol-related problems: A behavioral approach. In McMahon BT, Shaw LR, eds. Work Worth Doing: Advances in Brain Injury Rehabilitation. Paul M Deutsch Press, Inc., Orlando, FL, 1991.
8. Langley MJ, Kiley DJ. Prevention of substance abuse in persons with neurological disabilities. Neurorehabilitation 1992;2(1):52-64.
9. Sparadeo FR, Strauss D, Kapsalis KB. Substance abuse, brain injury, and family adjustment. Neurorehabilitation 1992;2(1):65-73.
10. Sparadeo FR. Substance use: A critical training issue for staff in brain injury rehabilitation. In Durgin D, Schmidt N, and Fryer LJ, eds. Staff Development and Clinical Intervention in Brain Injury Rehabilitation. Gaithersburg, MD: Aspen Publishers, Inc., 1993;189-209.
11. Blackerby WF, Baumgarten A. A model treatment program for the head-injured substance abuser: Preliminary findings. J Head Trauma Rehabil 1990;5(3):47-59.
12. Prochaska JO, DiClemente CC, Norcross JC. In search of how people change: Applications to addictive behaviors. Am Psychol 1992;47:1102-1114.
13. Ridgely MS. Practical issues in the application of case management to substance abuse treatment. J Case Manag 1994;3:132-138.
14. Willenbring ML. Case management applications in substance use disorders. J Case Manag 1994;3:150-157.
15. Ashery RS. Case management for substance abusers: More issues than answers. J Case Manag 1994;3:179-184.
16. Corrigan JD, Lamb-Hart GL, Rust B. A programme of intervention for substance abuse following traumatic brain injury. Brain Inj 1995;9(3): 221-236.
17. Corrigan JD, Rust E, Lamb-Hart GL. The nature and extent of substance abuse problems among persons with traumatic brain injuries. J Head Trauma Rehabil 1995;10(3):29-45.
18. Kreutzer, JS, Witol, AD, and Marwitz, JH. Alcohol and drug use among young persons with traumatic brain injury. J Learn Disabil 1996;29(6):643-651.
19. Corrigan JD, Smith-Knapp K, and Granger C. Outcomes in the first five years following traumatic brain injury. Arch Phys Med Rehabil 1997;in review.
20. McLellan, A.T., et al., Addiction Severity Index Administration Manual, Fifth Edition, . 1990, Penn-VA Center for Studies of Addiction: Philadelphia.
21. Willer B, Linn R, Allen K. Community integration and barriers to integration for individuals with brain injury. In Finlayson MAJ, Garner S, eds. Brain Injury Rehabilitation: Clinical Considerations. Baltimore, MD: Williams & Wilkins, 1993:355-375.
22. Pavot W, Diener E. Review of the Satisfaction With Life Scale. Psychol Assess 1993; 5:164-72.