Evaluation of the home and community-based services brain injury Medicaid Waiver Programme in Colorado

To evaluate Colorado's Traumatic Brain Injury (TBI) Waiver programme in terms of functional, community integration and other key psychosocial outcomes.

Results of telephone surveys of 66 TBI survivors who had received services through the Colorado Medicaid programme were compared with those of a matched sample not receiving such services.

Participants were identified by Colorado's Medicaid programme and a control group was randomly selected from the population-based Colorado Traumatic Brain Injury Registry and Follow-up System. Groups were matched on Glasgow Coma Scale score at injury, age, gender,whether inpatient rehabilitation had been received and the number of years post-injury when the follow-up interview was conducted. All participants completed an extensive follow-up interview, which included CHART, the Sickness Impact Profile, the Satisfaction with Life Scale, the SF-12 and questions on symptoms and service use.

For four of the evaluated outcomes-relating to mental health problems (SF-12) and alcohol use-the Medicaid Waiver group had significantly fewer problems than the control group. In eight areas, the Waiver group demonstrated poorer outcomes than the control group. The Waiver group scored significantly lower on most measures of societal participation (measured by CHART), needed more help with IADLS (instrumental activities of daily living) and were less likely to be competitively employed. Additionally, Waiver recipients had significantly more case management, physical therapy and group home services and more second rehabilitation admissions than the control group. No significant differences were found for the remaining 62 outcomes that were evaluated.

Although there were significant differences between the two groups in a few important areas, it is impossible to conclude that any outcome differences--or lack of differences---are attributable to the Waiver programme alone, since many important potential group differences could not be adequately controlled for by the design. Recommendations are offered for future research to address these limitations.