This talk addresses estimating and testing treatment effects with multi-variate outcome in randomized control trials where imperfect diagnostic devices are used to assign subjects to treatment and control group(s). Sometimes, we may have more expensive diagnostic devices to assess the status of subjects accurately, yielding partially validated data. The talk focuses on pre-post design and proposes two new methods for estimating and testing treatment effects, with or without partially validated data. The methods are compared with each other and with a traditional method that ignores the imperfection of the diagnostic device. In particular, the likelihood-based approach shows a significant advantage in terms of power, coverage probability and in reduction of required sample size. The application of the results are illustrated with data from electroencephalogram (EEG) recordings of alcoholic and non-alcoholic subjects.