When Is the Best Time to Give Birth?

Why it may matter when a baby is born

Posted Sep 14, 2015

Original cartoon by Alex Martin
Source: Original cartoon by Alex Martin

Medical intervention in human birth has become so widespread that spontaneous delivery is an increasingly rare event. Yet convincing evidence reveals an underlying 24-hour biorhythm, reflecting a general pattern among mammals, including nonhuman primates. This is perhaps no more than a carryover from ancestors adapted for birth while inactive. But the basic rhythm may remain biologically important, such that we ignore it at our peril.

Birth in nonhuman primates

In two pioneering papers published in 1972 and 1973, the sadly departed primatologist Alison Jolly (1937-2014) reported that births most commonly occur during the inactive phase of a new mother’s daily cycle. This pattern is widespread among mammals generally. Favoured explanations are that it reduces the risk of attracting predators or (for social species) avoids mothers being left behind by a moving group. Several exceptions to the general rule are known, notably for certain day-active monkeys and apes that reportedly give birth around the clock. Remember, though, that there is a built-in bias because we mainly observe monkeys and apes during daytime.

Time of human birth

Human births can occur at any time, but an underlying pattern is nonetheless discernible. Nowadays, meaningful information is elusive because medical intervention is so commonplace. But early reports on birth timing yield information collected before intensive management became widespread. In fact, Belgian polymath Adolphe Quételet—best known for his Body Mass Index (ratio between body weight and height), still widely used today as an obesity indicator—was among the first to report a pattern in human birth hour. His lengthy treatise published in 1869 reporting both year-round and daily rhythms in birth timing is far less widely known. His diagram (illustrating a general trend rather than actual data) indicates peak birth frequency around midnight and a trough at midday.

Diagram redrawn from Quételet 1869. Portrait of Adolphe Quételet by Joseph-Arnold Demannez from Wikimedia Commons (public domain)
Diagram of human birth hour published by Belgian polymath Adolphe Quételet (inset)
Source: Diagram redrawn from Quételet 1869. Portrait of Adolphe Quételet by Joseph-Arnold Demannez from Wikimedia Commons (public domain)

Considerable time elapsed before further reports appeared. In a 1933 review of more than 350,000 births in Switzerland for 1926-1930, pediatrician Edouard Jenny reported that although births occurred around the clock, they were most frequent in the early morning (2:00–5:00 a.m.) and least common in the afternoon and early evening (1:00–7:00 p.m.). Variation in hourly birth frequency was marked; the early morning peak was some 40% higher than afternoon levels. Other publications, notably by Enid Charles (1953) and Irwin Kaiser and Franz Halberg (1962), subsequently confirmed this general pattern. One crucial finding reported by Kaiser and Franz Halberg, among others, is that average labour duration is significantly longer with afternoon/evening births than with early morning births.

In fact, it eventually emerged that the clearest pattern of hourly variation is shown by the onset of labour, not birth time. With all mothers, including first-timers, labour onset peaks sharply around 2:00 a.m., as was clearly evident from Enid Charles’ study and from many subsequent investigations. In 1972, Michael Smolensky and colleagues reviewed data for the timing of over 2 million natural births and more than 30,000 induced births, along with over 200,000 labour onset times. Frequency of labour onset showed pronounced variation, with a clear maximum at about 2:00 a.m. and a minimum just before noon. By contrast, variation in birth frequency over the day was less marked, with a peak at 4:00–6:00 a.m. and a trough at 3:00–5:00 p.m. Times of birth induction were starkly different, occurring predominantly during daylight hours, with a peak at 10:00 a.m.

Figure redrawn from Smolensky et al. (1972)
Plot of occurrence of labour and birth around the clock
Source: Figure redrawn from Smolensky et al. (1972)

Physiological factors influencing birth timing

Scant attention has been devoted to physiological processes underlying birth timing. A prominent exception is a Ph.D. thesis produced by gynecologist Maria Honnebier, based on a comparative study of pregnancy and labour in women and rhesus macaques. Round-the-clock patterns of variation were found with several physiological indicators, including pregnancy-associated hormone levels and contractures (mild muscle twitches in the womb wall). In both rhesus macaques and women, contractures progressively give way to full contractions as birth approaches, with the hormone oxytocin playing a key role. (See my blog posting Oxytocin — The Multitasking Love Hormone on May 12, 2015.) For rhesus macaques, Honnebier identified a clear 24-hour rhythm in blood levels of oxytocin and showed that contractions occur when the concentration peaks.

Honnebier also conducted pulse tests with oxytocin in macaques to assess responsiveness of the womb to the hormone. The womb wall was maximally reactive during nighttime hours. Similar pulse tests conducted with nine women between 19 and 30 weeks of pregnancy yielded the same result. So there is a physiological basis for a 24-hour rhythm in the frequency of spontaneous births.

Do birth rhythms matter?

Anthropologists Cristina Bernis and Carlos Varea examined the influence of increasing medicalization on hourly birth frequency in a 2012 paper reviewing over 25,000 deliveries at a university maternity hospital in Spain. Their main finding was that the nighttime predominance of births recorded in earlier studies had being replaced by prevalence of daytime deliveries. But the original pattern was still recognizable for births without intervention. Births following intervention showed two daytime peaks: in the morning for multiple births, breech presentations, pre-term births and Caesareans, and in the afternoon for vaginal deliveries with different interventions. Bernis and Varea suggested that increasingly common birth interventions may be contributing to a general increase in prematurity and low birth weight. One negative outcome is reducted likelihood of breastfeeding.

The European findings reported by Bernis and Varea have now been overwhelmingly confirmed by a massive survey of U.S. data conducted by demographers T.J. Mathews and Sally Curtin (Division of Vital Statistics, Center for Disease Control & Prevention). Their survey benefited from widespread adoption of a revised birth certificate for 2013, including birth hour along with delivery method and location. Using data from the new certificate from 41 states and the District of Columbia, Mathews and Curtin covered 90% of U.S. births in 2013 (over 3,500,000). Three distinct patterns in birth timing emerged for: Caesarean sections, induced vaginal deliveries, and vaginal deliveries without induction. Caesareans and vaginal deliveries in hospitals all occurred mainly during daytime, with two clear peaks for Caesareans at 8:00–9:00 a.m. and 12:00–1:00 p.m. Although most vaginal births in hospital happened during daytime, they were more likely to occur during late evening and early morning, especially if not induced. By contrast, births outside hospitals (less than 2%) were most likely in the early morning (1:00–5:00 a.m.).

Figure redrawn from Mathews & Curtin (2015)
Birth timing inside and outside of hospitals for over 3,500,000 U.S. births in 2013
Source: Figure redrawn from Mathews & Curtin (2015)

In sum, both in Europe and in the U.S.A., increased medical intervention has clearly been associated with a shift in birth timing, with the peak now occurring in daytime rather than during early morning hours. Does this modification of a natural rhythm have any negative impact on health of mothers and babies? One might expect that a more natural timing of human birth would necessarily be beneficial. Honnebier and Nathanieltz suggested, for instance, that induction of labor with oxytocin might best be performed during nighttime, when the womb’s sensitivity to the hormone is maximal. Yet it is also possible that benefits of medical intervention outweigh any disadvantages of deviation from the natural rhythm. Once again, what we need is more research. It is vital to know, for example, whether birth hour is related to frequency of premature birth, incidence of postnatal depression or difficulties in initiating breastfeeding—all of which have been climbing for unknown reasons.


Backe, B. (1991) A circadian variation in the observed duration of labor: Possible causes and implications. Acta Obstetricia et Gynecologica Scandinavica 70:465-468.

Bernis, C. & Varea, C. (2012) Hour of birth and birth assistance: From a primate to a medicalized pattern? American Journal of Human Biology 24:14-21

Charles, E. (1953) The hour of birth: A study of the distribution of times of onset of labour and delivery throughout the 24-hour period. British Journal of Preventive & Social Medicine 7:43-59.

Honnebier, M.B.O.M. (1994) The role of the circadian system during pregnancy and labor in monkey and man. Acta Obstetricia et Gynecologica Scandinavica 73:85-88.

Honnebier, M.B.O.M. & Nathanielsz, P.W. (1994) Primate parturition and the role of maternal circadian system. European Journal of Obstetrics & Gynaecology and Reproductive Biology 55:193-203.

Jenny, E. (1933) Tagesperiodische Einflüsse auf Geburt und Tod. Schweizerische medizinische Wochenschrift 63:15-17.

Jolly, A. (1972) Hour of birth in primates and man. Folia Primatologica 18:108-121.

Jolly, A. (1973) Primate birth hour. International Zoo Yearbook 13:391-397.

Kaiser, I.H. & Halberg, F. (1962) Circadian periodic aspects of birth. Annals of the New York Academy of Sciences 98:1056-1068.

Mathews, T.J. & Curtin, S.C. (2015) Morning, noon, or night? Birth certificate data for 2013. NCHS Data Brief No. 200:1-7.

Quételet, A. (1869) Physique Sociale ou Essai sur le Développement des Facultés de l’Homme. Paris: J.-B. Baillière et Fils.

Smolensky, M.H., Halberg, F. & Sargent, F. (1972) Chronobiology of the life sequence. pp. 281-318 in: Advances in Climatic Physiology (ed, Itoh, S.). New York: Springer-Verlag.