Fornasin Alessio, Rizzi Laura, Fonseca Giovanni
University of Udine, Udine, Italy.
Am J Hum Biol. 2025 Aug;37(8):e70120. doi: 10.1002/ajhb.70120.
The aim of this study is to evaluate the influences exerted by temperature on the circadian rhythm of births.
In the past births occurred mainly between midnight and dawn, while today births tend to be less frequent during the night hours. Today, almost all births are hospitalized and, therefore, they may adapt to the organizational requirements of the health care structures and staff. This piece of research regards births in Udine, a city in north-eastern Italy, recorded at the beginning of the 19th century. The data on births come from the French civil register. Weather data come from very detailed daily collection with three measurements per day. From the statistical point of view, we apply methods developed for circular data. In order to highlight relationships between time of birth and explanatory variables, we estimate MANOVA (multivariate analysis of variance) models and perform a statistical test for comparison between groups.
The test against the homogeneity of the hour-of-birth distributions across the 4 seasons is significant (p < 0.01). One-to-one comparison of the distributions is performed via Watson's two-sample test on data for each couple of seasons: the Summer-Autumn comparison is the only non-significant test of homogeneity. Moreover, we consider daily temperatures in the bivariate model in several different ways and transformations. The daily temperature effect is more significant if considered as the difference between the evening temperature of the day before the birth and the mean daily temperature of the same day, over the nine-year period 1807-1815. Finally, based on this division of the births, Watson's two-sample homogeneity test for the distribution of the hour of birth is significant (p < 0.05).
We found that the circadian rhythm of births is influenced by temperature, with an anticipation of the time of birth on warmer days. To our knowledge, our results are the first evidence of the effects of daily temperature on the time of birth.
本研究旨在评估温度对出生昼夜节律的影响。
过去分娩主要发生在午夜至黎明之间，而如今夜间分娩频率趋于降低。如今，几乎所有分娩都在医院进行，因此，它们可能会适应医疗保健机构和工作人员的组织要求。本研究涉及意大利东北部城市乌迪内19世纪初记录的分娩情况。分娩数据来自法国民事登记册。气象数据来自非常详细的每日收集，每天测量三次。从统计学角度来看，我们应用针对圆形数据开发的方法。为了突出出生时间与解释变量之间的关系，我们估计多变量方差分析（MANOVA）模型并进行组间比较的统计检验。
对四个季节出生时间分布的同质性检验具有显著性（p < 0.01）。通过对每对季节的数据进行沃森双样本检验来进行分布的一对一比较：夏季 - 秋季比较是唯一不显著的同质性检验。此外，我们以几种不同方式和变换在双变量模型中考虑每日温度。如果将每日温度效应视为1807 - 1815年九年期间出生前一天晚上的温度与同一天平均每日温度之间的差异，则更为显著。最后，基于这种对分娩的划分，出生时间分布的沃森双样本同质性检验具有显著性（p < 0.05）。
我们发现出生的昼夜节律受温度影响，温暖日子出生时间会提前。据我们所知，我们的结果是每日温度对出生时间影响的首个证据。
