Myers D H, Davies P
Psychol Med. 1978 Aug;8(3):433-40. doi: 10.1017/s003329170001610x.
Monthly hospital admission rates (HA) for mania were classified by sex, type of admission (first, or re-admission) and by country (England and Wales or Scotland). Of the 8 classes thus created, all but one showed a statistically significant annual cycle with a peak in Summer and trough in Winter. There was no significant difference in phase or amplitude between male and female cycles. A linear increase in yearly re-admission rates was found during the 8 years of the English and the 11 years of the Scottish data. Current month's mean daily temperature (lag0) and last month's (lag1) mean day-length and mean daily hours of sunshine correlated better with admission rate than did the values for other months. In a multiple regression analysis temperature made the other 2 climatic variables redundant in accounting for variation in HA.
按性别、入院类型（首次入院或再次入院）以及国家（英格兰和威尔士或苏格兰）对躁狂症的每月住院率（HA）进行了分类。在由此产生的8个类别中，除了一个类别外，所有类别均显示出具有统计学意义的年度周期，夏季达到峰值，冬季降至谷底。男性和女性周期在相位或幅度上没有显著差异。在英格兰的8年和苏格兰的11年数据中，每年的再次入院率呈线性上升。当月的平均日气温（滞后0）、上月的（滞后1）平均日照时长和平均每日日照小时数与入院率的相关性比其他月份的值更好。在多元回归分析中，温度使得其他两个气候变量在解释HA的变化时变得多余。
