Is fecundability associated with month of birth? An analysis of 19th and early 20th century family reconstitution data from The Netherlands

The relationship between fecundability and month of birth was investigated in a cohort of 1526 women who married between 1802 and 1929, using only women whose first marriage occurred before the age of 35 years. On the basis of their time to pregnancy (TTP, calculated as time between wedding and first birth minus gestational length), women were categorized into two groups: fecunds (TTP up to 12 months or prenuptial conceptions, n = 1348) and subfecunds (TTP >18 months, n = 118). By use of logistic regression, cosinor functions with a period of 1 year or 6 months and variable shift and amplitude were fitted through the monthly odds of subfecunds versus fecunds. The best fitting curve was unimodal, with a zenith in September (P = 0.13 for H0: no differences). Exclusion of childless women (n = 36, minimum follow-up 5 years) from the subfecunds led to a similar curve (P < 0.01), while childless women, as compared with fecunds, showed a birth distribution that was best represented with a bimodal curve with zeniths in January and July (P = 0.06). This study provides evidence for the existence of differences in fecundability by month of birth. The cause of this relationship is unclear, but may lie in a melatonin-dependent circannual variability of the quality of the oocyte.