Switch from depression to mania--a record study over decades between 1920 and 1982.

A representative sample of 908 hospital records covering admissions between 1920 and 1982 for depression was analyzed in order to assess the switch rate to hypomania/mania. The results are the following: (1) Over the decades of this century there has been a substantial increase in hospital admissions in Zurich for both depression and mania, but the ratio remained constant. (2) Due to this increase the clinicians can observe more spontaneous switches from depression to mania, which favors the assumption of a causal relationship when treatment is applied. (3) 64 of the 908 patients (7.0%) admitted for depression switched to hypomania or mania. Hypomania was observed in 48 cases (5.3%) and mania in 16 cases (1.7%). (4) The analysis of predisposing factors to a switch has resulted in a simple finding. Bipolar patients (including schizomanics) have an 8-fold higher switch rate (28.9%) than the unipolars (3.7%). The switchers are equally distributed over the two sexes and do not differ in the frequency of a family history of affective psychoses, schizophrenia, schizoaffective disorders, or suicide. (5) Bipolarity correlates positively with 'higher number of previous episodes', with 'readmitted' and with 'switch'. Therefore, studies selecting readmissions [Lewis and Winokur 1982] overrepresent switchers purposely. (6) A loglinear analysis together with some univariate strategies show that over the decades (from 1920 to 1982) there was no significant increase in switches of unipolar or bipolar patients. In conclusion, there is no evidence for a treatment-induced switch. This result is in line with Prien et al. [1973] and with Lewis and Winokur [1982].