The long-term objective of this project is to contribute to the quality of clinical psychopharmacology research by providing improved methodology in- areas of experimental design, data analysis, assessment and classification.
Specific aims i nclude data analytic support for established clinical investigators and collaborative participation in design, administration, data collection, and analysis with younger investigators; refinement of methodology for empirical classification research with special focus on the heterogeneity of schizophrenia; and pursuit of solutions to several statistical methodological problems of current concern in clinical psychopharmacology research. The statistical problems include interim analyses in ongoing clinical trials, multi-stage sampling designs, criteria for confirming the superiority of a combination drug over each of its components, appropriate corrections for multiple tests of significance, and heterogeneity of variance in association with unequal cell frequencies in a non-orthogonal experimental design. Corrections for baseline differences using simple delta scores, percentage change scores, or analysis of covariance will be evaluated under different conditions with the aim of recommending when each may be most useful in assessing treatment response.
Showing the most recent 10 out of 40 publications