Objective: To review the way in which collaborative research has been conducted under the National Institute on Disability and Rehabilitation Research (NIDRR)-funded Model Spinal Cord Injury Systems (MSCIS) Program, changes made in that process, and significant outcomes. Methods: A comparison of changes by NIDRR in the way collaborative research was competed and funded in the 1995 and 2001 competitions. A review of outcomes of the 1995 collaborative projects was based on queries to lead centers. Results: Collaborative research through the model SCI systems has been conducted and continues to be conducted through 2 main venues: The National Spinal Cord Injury Statistical Center (NSCISC) database, which has provided data for a number of collaborative studies, and specifically funded proposals for collaborative research. In the 1995 competition for NIDRR funding, collaborative research proposals were submitted as part of the Model SCI Systems competitive applications. In the 2001 competition, collaborative research was parceled out and a separate competition held. There have been a number of publications stemming from the 1995 competition; some of the data from these projects are still being explored and used for manuscripts. The outcomes for the 2001 competition will not be known for several years. Conclusions: Collaborative research has the advantage of generating larger numbers more quickly than any 1 center can typically generate, producing a more broadly based sample and, therefore, generalizable result, and facilitating the use of expertise not always available in a single center. Collaborative research activities have been among the most productive aspects of the Model Systems program; the change in the way this component is competed in the most recent competition is yet to be evaluated in terms of its efficacy compared with methods used for funding collaborative research in past competitions.