Effect of religious rules on time of conception in Romania from 1905 to 2001

STUDY QUESTION: Does the interdiction of sexual intercourse during Nativity and Lent fasting periods have any effect on when babies are conceived in Romania, in the 20th century? SUMMARY ANSWER: Based on date of birth records from the 20th century, Lent had a greater effect than the Nativity fast on conception within the Eastern Orthodox (ORTHD) population. WHAT IS KNOWN ALREADY: Seasonality of births (and therefore of conception) is affected by geographical factors (latitude, weather, day-length). Other demographic, economic and socio-cultural characteristics (education, ethnicity, religion) have been proved to have an influence on conception. STUDY DESIGN, SIZE, DURATION: The analyzed data consists of registered daily birth records for a long time series (35 429 points = 365 (days/year) × 97 years + 24 leap years), with 24 947 061 births in Romania over the period 1905-2001. The data were obtained from the 1992 and 2002 censuses. PARTICIPANTS/MATERIALS, SETTING, METHODS: Based on the reported birth date of each person, the estimated date of conception is computed using a standard gestation period of 280 days. The population was grouped into two categories (ORTHD and Non-Orthodox (NORTHD)) based on religious affiliation. Data analysis is performed in the same manner for both groups. Preliminary data analyses regarding seasonal variations in conception are considered first. Econometric models are applied and tested. The dependent variable in these models is the calculated date of conception, while the independent variables are: (i) religious affiliation; (ii) dates of Nativity and Lent fasts (the latter varies slightly from year to year); (iii) rural versus urban residence; (iv) length of day-light; (v) non-working days and (vi) trend. The models are tested for validity using analysis of variance while the regression coefficients are tested by the Student t-test. MAIN RESULTS AND THE ROLE OF CHANCE: All models are statistically valid (P < 0.01); all regression coefficients for the ORTHD group are valid (P < 0.01, except for rurality between 1990 and 2001, with P < 0.05). The data analysis indicates smaller standard error bars on the parameters for the ORTHD group as compared with the NORTHD group. The conclusion is that religious affiliation is an important factor in date of conception. LIMITATIONS, REASONS FOR CAUTION: The data do not refer to all births during the analyzed period, but only to those persons still alive at the 1992 and 2002 censuses. The date of conception was estimated assuming 280 days for gestation, which is a medically accepted time interval but will undoubtedly vary. However, the primary independent variables (Lent and Nativity fast at 48 and 40 days, respectively) are long enough to overlap the uncertainty in the conception date following the sexual intercourse event. We also must assume that the religious affiliation of the parents is well defined, based on the information given by their offspring at census time, and is the same for both parents. WIDER IMPLICATIONS OF THE FINDINGS: Our findings are consistent with other studies, which show differences between religious groups on date of conception, although we reach different conclusions regarding the influence of weather on fertility in Romania.