The over-arching goal of the UT Houston Substance Abuse Research-Medication Development Center is to develop safe, effective medications for cocaine dependence. The current Center proposal focuses on elucidating behavioral and neurochemical mechanisms of cocaine dependence and using these findings to guide clinical trials for promising medications. The Advanced Clinical Design and Statistical Analysis Core (ACDSAC) will serve as a critical mechanism for promoting this goal by providing statistical/methodological support for NIDA-funded clinical trials in drug dependence taking place at University of Texas Health Science Center at Houston-Medication Development Center as well as consultation to other NIDA-funded centers. The ACDSAC will accomplish this by: 1) providing support for the design and analysis of conventional clinical trials; 2) advancing utilization of flexible/adaptive study designs for the development of more efficient clinical trials; 3) advancing statistical methodologies for the analysis of clinical trials with a focus on Bayesian modeling and the use of structural equation modeling methodologies for evaluation of mediational hypotheses of drug action. 4) serving as a liaison to the statistical and design divisions of the Center for Clinical Research and Evidence-Based Medicine as well as the Center for Clinical Translational Science, both of the University of Texas Health Science Center at Houston; 5) providing training to post-doctoral staff, fellows and students in study design and advanced quantitative methods; The Advanced Clinical Design and Statistical Analysis Core will advance the overall goal of the UT Houston Medication Development Center by providing advanced novel statistical design and analysis support for medication development clinical trials.

The dearth of acceptable and effective treatments for substance dependence coupled with the heterogeneous nature of the patient population and the accruing wealth of neurobiological evidence concerning this problem argue that innovative design and analyses should be implemented to maximize the likelihood of identifying successful treatments in the most efficient way possible.