Cross-sectional and within-subject seasonality and regularity of hospitalizations: A population study in mood disorders and schizophrenia
Abstract
Background Seasonal peaks in hospitalizations for mood disorders and schizophrenia are well recognized and often replicated. The within-subject tendency to experience illness episodes in the same season, that is, seasonal course, is much less established, as certain individuals may temporarily meet criteria for seasonal course purely by chance.
Aims In this population, prospective cohort study, we investigated whether between and within-subject seasonal patterns of hospitalizations occurred more frequently than would be expected by chance. Methods Using a compulsory, standardized national register of hospitalizations, we analyzed all admissions for mood disorders and schizophrenia in the Czech Republic between 1994 and 2013.
We used bootstrap tests to compare the observed numbers of (a) participants with seasonal/regular course and (b) hospitalizations in individual months against empirical distributions obtained by simulations. Results Among 87 184 participants, we found uneven distribution of hospitalizations, with hospitalization peaks for depression in April and November (X-2(11) = 363.66, P < .001), for mania in August (X-2(11) = 50.36, P < .001) and for schizophrenia in June (X-2(11) = 70.34, P < .001).
Significantly more participants than would be expected by chance, had two subsequent rehospitalizations in the same 90 days in different years (7.36%, bootstrap P < .01) or after a regular, but non-seasonal interval (6.07%, bootstrap P < .001). The proportion of participants with two consecutive hospitalizations in the same season was below chance level (7.06%).
Conclusions Psychiatric hospitalizations were unevenly distributed throughout the year (cross-sectional seasonality), with evidence for regularity, but not seasonality of hospitalizations within subjects. Our data do not support the validity of seasonal pattern specifier.
Season may be a general risk factor, which increases the risk of hospitalizations across psychiatric participants.