| Abstract: | A resource allocation model was developed to assist in planning services for the severely mentally ill. Given program objectives and resource constraints, the model employs linear programming methods to find optimal strategies for assigning clients to services. It can also forecast client movement and resource consumption. To field test the model it was applied to a Community Support Program in a predominately rural state. Model results suggest the need for targeting resources to specific client subpopulations.H. Stephen Leff is with Management Research Associates, North Charles Foundation and Harvard Medical School. Stephen C. Graves is with the Alfred P. Sloan School of Management, Massachusetts Institute of Technology. Judith Natkins is with Management Research Associates, North Charles Foundation and Psychology Department, Boston University. Jeffrey Bryan is also with Management Research Associates, North Charles Foundation as well as the Massachusetts School of Professional Psychology.This research was supported by NIMH Grant No. 1 R01-MH33581-02, a grant supplement from NIMH's Community Support Program, and the Whitaker Health Sciences Fund. The authors would like to give thanks to P. Clark Santos and Maqbool Dada for their help with the computational component of this project, to Mindy Schimmel, Michael Senger, and Marion Bloch for help in reviewing the literature for the transition probability estimation, to Jacqueline Rosenberg, Ronald Manderscheid, Larry Kessler and Rosalyn Bass of NIMH for their support and encouragement, and to Midstate's CSP director and staff for their commitment and collaboration. |
