On the periodicity of manic-depressive insanity, by Eliot Slater (1938): translated excerpts and commentary.

Since the classic descriptions of the course of bipolar and recurrent depressive forms of manic-depressive illness by Emil Kraepelin a century ago, it has been considered a truism that the rate of cycling increases, and wellness intervals shorten, with rising counts of recurrences, particularly early in the natural history of the illness. Less well known is that the analysis of this phenomenon is vulnerable to a computational artifact first described by Eliot Slater, based on his reanalysis of data from manic-depressive patients first evaluated by Kraepelin at the Munich Psychiatric Institute. Slater realized that there is an increasingly disproportionate representation of faster-cycling patients in sub-samples involving higher cycle-counts in pooled samples of subjects. More accurate results require analyzing illness-course either within individuals, or in groups matched for episode-counts. This artifact is pervasive in the older and modern research literature, but still not widely recognized. Since Slater's 1938 report in German is not well known, we provide an abbreviated English translation with commentary and additional reanalysis to highlight the phenomenon that might be termed 'Slater's Fallacy'.