STATISTICAL APPROACHES TO TRIAL DURATIONS IN EPISODIC AFFECTIVE-ILLNESS

In light of the high variability in illness characteristics and patter ns among patients with bipolar illness, parallel group designs present severe methodologic difficulties. Crossover, off-on-off-on (B-A-B-A), and other individualized designs may be a useful substitute, but no c onsensus exists about how to estimate the individual trial durations r equired in these instances. Several methods for determining optimum tr ial lengths in crossover designs are presented, illustrated, and discu ssed. These include: chi-square (chi(2)) for th, expected versus obser ved number of either episodes or days well; exceeding two standard dev iations for average duration of episodes or euthymic intervals; or the Sequential probability Ratio Test (SPRT), which detects when mean val ues differ from prior statistical expectations. Each method was applie d to three demonstration cases using data from actual clinical trials of three patients with different patterns of recurrent affective illne ss. Each method detected changes in illness severity, although differe nt tests appeared to be sensitive to differing cycle patterns in the p atients illustrated. We suggest that these types of analyses and other s can be used as indicator statistics to augment global impressions an d clinical judgment, and to assist in determining individualized trial durations, both in formal clinical trials and in clinical treatment s ettings. Once individual responsivity is confirmed with an appropriate interplay of trial design and statistical analysis, the percentage re sponse in a given population can then be compared to other agents or i n other populations. Moreover, meta-analytic techniques based on addit ion of z scores from individuals' effect sizes can then be used to ass ess overall significance of a drug effect in a given population or sub population. The need for further development of appropriate and altern ate study designs and analysis methods for bipolar illness is highligh ted. Approaches to estimating required trial durations in individuals with different cycle frequencies in crossover and B-A-B-A designs cons titute one element of that exploration. (C) 1998 Elsevier Science Irel and Ltd.