Minimization--a method of obtaining comparable groups in clinical trials.

The practicability and efficiency of a method known as minimization in minimizing the differences between treatment and control groups in clinical trials was investigated in a trial with an anorexiant. Eight factors which might have influenced outcome were identified and used as variates in the minimization programme. The subcategories of these variates totalled 21. The minimization procedure, which is discussed in detail, was carried out on 41 obese patients; an even distribution of these 21 subcategories among treatment and control groups resulted. Minimization proved to be a simple and effective method for minimizing the differences between treatment and control groups in a clinical trial.