Establishment and Use of a Consumer-Operated Services Data Repository
The purpose of this Center research project is to create a Consumer-Operated Services (COS) Data Repository using data collected from state-of-the-science randomized controlled trial (RCT) studies of interventions delivered by people in mental health recovery. After the Repository has been assembled, the original investigators from each study will then collaborate on a meta-analysis of study results. The project will thus determine whether the empirical evidence for COS justifies grading it at the highest level of the evidence grading ladder.
The project will also make Repository data available to other researchers, particularly consumer researchers, and will mentor less-experienced researchers who are interested in learning how to work with large, complex data bases. This will occur through awarding 15 scholarships of $2,000 for a period of two years, and by providing statistical consultation and support to researchers who work with these data. To accomplish these goals, UIC Center staff will work in collaboration with the Principal Investigators of the original studies, including Drs. Jean Campbell, Jeffrey Draine, Caroline Kaufmann, Robert Paulson, and Phyllis Solomon; and an expert in the field of biostatistics and meta-analysis, Dr. Steven Banks.
Solomon and Draine's study randomly assigned clients of a community mental health center to consumer- versus non-consumer delivered intensive case management. Paulson's research randomly assigned community mental health center clients to 1 of 3 conditions: 1) an Assertive Community Treatment (ACT) team staffed entirely by people in recovery; 2) an ACT team staffed by non-consumers; and 3) treatment as usual. Kaufmann's study randomly assigned participants to a consumer self-help vocational program working in partnership with professional providers versus a "services as usual" condition. The CMHS-funded Consumer-Operated Services Program (COSP) multi-site study directed by Campbell included 4 drop-in centers, 2 mutual support programs, and 2 educational/advocacy programs. In Cook's study of mobile crisis teams, participants were randomly assigned to either mobile crisis outreach services delivered by consumer provider dyads or services from non-consumer provider dyads.
A cross walk indicates considerable overlap in study data. Thus, outcomes that can be examined in this research include: quality of life; recovery/wellness; employment status and earnings; satisfaction with services; psychiatric symptomatology; medication adherence; number of hospitalizations and length of time hospitalized; homelessness; incarcerations and arrests; use of alcohol; and use of recreational drugs. In addition, variables that are conceptually the same, but assessed with different measures can be converted to standardized metrics for comparison purposes (e.g., z-score or d-index). Access to original data will allow the project to employ a random-effects prospective meta-analysis with thorough sensitivity analyses, an approach that combines research synthesis methods with the hypothesis-based deductive approach of RCTs. An advisory committee composed of consumer and non-consumer researchers and advocates will provide input into the study and the awarding of scholarships.
