A search for associations between weather and the onset of human parturition

Weather data were examined for association with hospital records indicating the time at which pregnant women at term first experienced labor contractions (onsets). There is a considerable advantage, compared with mortality and morbidity, to using this response to infer weather influences. Three tests were used. First, the time series of daily onsets was analyzed to determine if characteristics known to exist in time series of weather data — variability and persistence — were evident in daily onsets. Second, the frequency distributions of nine weather variables were stratified, mostly by terciles, and onset means calculated for each of these divisions. Response means much different from average were then used to specify the nature of weather at such times. Third, the weather data were organized as weather types — pre- and post-cold frontal and general cold frontal — and onsets at these times were compared with those at non-frontal times. The time series characteristics were not found, but the other analysis revealed subsets of days on which onsets were above average, and in some cases the results were statistically significant. One such subset consists of winter days with low pressure, temperature markedly lower than the day before, and high wind speeds. On such days onsets were 34% above average. They were also above average during the 48 hours before and after cold front passage, and especially so in the 12 hours prior to the front. These findings constitute a weak, but statistically significant indication that human parturition is influenced by weather. Follow-up studies are urged.