Decision Science in Rehabilitation

Specific aims of the project:

1. Conduct a seminar on informatics system design in rehabilitation, regularly attended by a core interdisciplinary group of junior and senior investigators in Lincoln, with planned visits by distant collaborators in accordance with the seminar’s topical progression . The purpose of this seminar will be to a) develop a group of collaborators with extensive experience in translating clinical experience and data analysis into informatics system design for rehabilitation settings; b) exchange data, expertise and techniques among the project collaborators; c) compare results of analyses of the Lincoln data bases with modeling analyses performed by the distant collaborators, to determine similarities and sources of differences; d) provide guidance and oversight to the Lincoln-based data mining and analytic activities of the project; e) interpret the results of the data analyses, identify key decision points and processes revealed by the analyses, and articulate the implications for clinical data management and decision support system design.

2. Perform a battery of complex modeling analyses on an existing psychiatric rehabilitation database previously constructed by the PI. The purpose of these analyses will be to a) identify measures of personal functioning and circumstances that predict outcome; b) determine the pathways of cognitive and behavioral change that lead to recovery and related dimensions of clinical outcome; c) translate these findings into implications for clinical informatics design.

3. Perform complementary modeling analyses on a second rehabilitation database previously constructed by a project collaborator. These analyses will provide comparisons of patterns and processes in psychiatric rehabilitation with a physiatric setting (strokes, head injury, spinal injury). This will provide initial information on the degree to which patterns and processes are similar in physical and psychiatric rehabilitation, an early step in informatics system design.

4. Extend the existing psychiatric rehabilitation database to include new settings and a broader range of consumer characteristics. Like the complementary analyses of collaborators’ databases, extension of the existing database will help determine the needed specificity of clinical informatics systems to settings and subpopulations. Developments in the Nebraska mental health system are creating new opportunities to study recovery from severe mental illness (SMI) in a range of community settings. The new database will track individuals through inpatient- and community-based rehabilitation. This part of the project also anticipates the next stage of the research program, and creates new data for future analyses.

This project is innovative in bringing basic and clinical scientists together for the purpose of enhancing decision making and outcome in psychiatric rehabilitation. The project will pioneer application of advanced analytical techniques, e.g., hierarchical linear modeling, formal concept analysis, neural network analysis, Bayesian networks and latent growth modeling, to clinical rehabilitation data. The project is also innovative in using the findings of modeling analyses to guide and inform informatics technology, specifically, development of advanced clinical data management and decision support systems.