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

Miscarriage is a rarely discussed, almost taboo topic. Couples confronted with pregnancy loss are usually unprepared and often struggle alone with harrowing bereavement and unwarranted guilt. People reportedly think that miscarriage is quite rare, affecting 5% of pregnancies, but the rate is actually 3-4 times higher. Moreover, most losses occur before pregnancy is even clinically recognizable.

Public awareness of miscarriage

Jonah Bardos and colleagues recently presented results from a national survey that assessed public understanding and emotional effects of miscarriage. Over 1000 men and women responded, with 15% reporting that they or a partner had experienced at least one miscarriage. Yet most respondents greatly underestimated its occurrence, believing that around 5% of pregnancies are affected. Stressful events and lifting heavy objects were cited as commonest causes, but one in four respondents blamed certain contraceptives. Of those that experienced miscarriage, almost half had guilt feelings, and a third perceived it as losing a child.

This survey highlighted the scarcity of information about frequency and causes of miscarriage. Yet 80% of respondents wanted to know why it happened, even if nothing could have been done to stop it. If a cause was identified, far fewer people blamed themselves. Most worryingly, many who experienced miscarriage felt guilty and isolated. Clearly, those affected need better information and counseling. Yet, to cite the Word Health Organization’s Reproductive Health Library: “Unfortunately, hospital staff are often not sensitized to provide support for the emotional distress associated with a miscarriage.”

Timing of pregnancy loss

Most pregnancy losses  —  spontaneous abortions in medical terminology  —  pass unnoticed. Clinically recognized cases are just the tail end of an elimination process that begins rapidly before slowing markedly. The standard WHO definition of miscarriage is any loss during the first 23 weeks after the last menstrual period (LMP). Any later losses are stillbirths. Ovulation typically occurs about two weeks after LMP, so miscarriages actually occur throughout the first 21 weeks after conception. But spontaneous abortions during the first month are not usually undetected. In practice, miscarriages are clinically recognized only between weeks 4 and 21, with most recorded cases occurring before week 12. Studies indicate that clinically recognizable miscarriage in weeks 4-21 occurs in 15-20% of human pregnancies.

All developmental stages are tiny during the first 4 weeks after conception, and between fertilization and implantation they are virtually invisible. A human egg is about as big as the period at the end of this sentence, and after conception it does not increase in size as it travels down the oviduct because additional resources are supplied only when the placenta is established. During its first 9 days the fertilized egg slowly divides to form a hollow ball of cells (blastocyst) that will implant in the womb’s inner lining. Even after implantation, the embryo remains quite small for the first 3 weeks, although it is detectable with ultrasound after about a week. We know very little about early pregnancy loss because careful monitoring, including hormone assays, is essential to distinguish early pregnancy loss from irregular menstruation.

Detecting early pregnancy loss

In a pioneering prospective study published in 1980, conception was identified by detecting the pregnancy-specific hormone human chorionic gonadotropin (hCG) in urine samples. 152 conceptions were identified, of which 87 progressed beyond mid-pregnancy and ended in live birth in all but two cases. So 43% were lost between conception and mid-pregnancy. But only 10% of conceptions ended with clinically recognized miscarriages; for the other 33% the sole evidence was hCG (“chemical pregnancy”). Moreover, because hCG is detectable only after implantation, any losses during the 9 days between conception and implantation escaped detection.

Figure adapted from Chard (1981).
Pie chart showing losses during successive phases of human pregnancy.
Source: Figure adapted from Chard (1981).

In 1981, obstetrician Tim Chard combined various lines of evidence to yield a composite picture of losses after conception. He concluded that 7 of 10 conceptions fail during the first half of pregnancy. He estimated a loss of 30% between conception and implantation. Agreeing with the 1980 study, Chard identified a further loss of 30% between implantation and the end of the first month after conception, when miscarriage becomes routinely recognizable. Chard indicated a frequency of about 10% for clinically recognizable miscarriage.

Aiming to estimate survival from conception to birth more rigorously, in a 1990 paper geneticist Charles Boklage fitted theoretical curves to available data. Such estimates yield an essential baseline for assessing the effectiveness of in-vitro fertilization (IVF). Boklage’s analysis indicated that about 75% of natural single conceptions are unlikely to survive beyond 6 weeks. He identified a consistent pattern: The loss rate was clearly greater during the first few weeks than over the rest of pregnancy. He therefore identified two distinct “eras” of prenatal development, with the transition between them occurring after the first 5-6 weeks, after which loss rates slowed markedly. He estimated losses of about 31% between conception and implantation and another 39% during the week following implantation, giving a somewhat higher overall rate than estimated by Chard. Incidentally, data from four clinics indicated that more IVF embryos  —  about 85-90% rather than 70%  —  were lost before pregnancy was clinically recognized. During the second era, beginning with development of the major body organs, any remaining embryos were found to have a 90% probability of surviving to birth with both natural conception and IVF.

Figure adapted from Boklage (1990).
Graph showing survival probabilities from conception to birth in human pregnancy.
Source: Figure adapted from Boklage (1990).

Miscarriage in non-human primates

A key question regarding human pregnancy loss, with only 25% of conceptions ending in live birth, is whether high loss rates are unique to our species or whether other primates  —  and perhaps other mammals  —  show comparable levels. This topic has received scant attention, but experience with breeding laboratory mammals such as rodents and rabbits indicates relatively little loss once fertilization has occurred. Similar conclusions can be drawn for at least some primates. For instance, my field observations of seasonally breeding mouse lemurs in Madagascar indicated that most females conceived during the first round of matings and carried their pregnancies through to birth.

Figure adapted from Beehner et al. (2006).
Graphical representation of fecal oestrogens for successful pregnancy and losses in plains baboons.
Source: Figure adapted from Beehner et al. (2006).

It is, however, possible that higher rates of pregnancy loss emerged with the evolution of higher primates or perhaps of apes. Information is scarce, but in 2006 Jacinta Beehner and colleagues published valuable data from five groups of wild baboons in Kenya. Over a 5-year period, they collected fecal samples to assay levels of various steroid hormones (oestrogens, progestins, glucocorticoids). Those hormones were examined across 188 pregnancies with a combination of longitudinal and cross-sectional data. Results confirmed that external indicators permit reliable early detection of pregnancy, so an additional 450 pregnancies without hormonal data could be taken into account. An overall estimate of just under 14% for the probability of loss between conception and birth emerged. This is only a fifth of the level estimated for human pregnancies.

Causes of pregnancy loss

Substantial losses between conception and mid-pregnancy have been interpreted as a natural mechanism that selectively eliminates developmentally defective human embryos. Studies have indicated that chromosomal abnormalities are prominent at all stages of miscarriage. At least half of failures during the first three months of pregnancy can be attributed to aberrant chromosomes. Indeed, the proportion may be even higher for losses during the first month after conception, when any natural mechanism for selecting against abnormal embryos would be expected to kick in. But most available information comes from studies of clinically recognizable pregnancies. In 1990, clinical geneticist Bernd Eiben and colleagues published findings obtained by investigating placenta samples from 750 miscarriages. They identified various kinds of chromosomal abnormalities in half of them.

So couples faced with pregnancy loss generally have no reason whatsoever to blame themselves, although certain lifestyle choices such as drug abuse may exert some influence. But that does not mean that nothing can be done to avoid unnecessary pregnancy losses. A recent article in The New York Times indicated that manicurists seemingly suffer a higher incidence of miscarriage. Any such cases warrant urgent investigation and appropriate action.

But here's the thing: Several different studies have inferred that, with repeated coitus during an unprotected menstrual cycle, there is only a 1-in-4 chance of conception. But 3 of 4 conceptions fail in early pregnancy. So perhaps a woman will conceive in practically every unprotected cycle but carries only 1-in-4 through to term. Unusually high early pregnancy losses may be the price our species pays for exceptional developmental complexity.


Bardos, J., Hercz, D., Friedenthal, J., Missmer, S.A. & Williams, Z. (2015) A national survey on public perceptions of miscarriage. Obstetrics & Gynecology DOI: 10.1097/AOG.0000000000000859.

Beehner, J.C., Nguyen, N., Wango, E.O., Alberts, S.C. & Altmann, J. (2006) The endocrinology of pregnancy and fetal loss in wild baboons. Hormones & Behavior 49:688-699.

Boklage, C.E. (1990) Survival probability of human conceptions from fertilization to term. International Journal of Fertility 35:75-94.

Chard, T. (1991) Frequency of implantation and early-pregnancy loss in natural cycles. Baillière's Clinical Obstetrics & Gynaecology 5:179-189.

Eiben, B., Bartels, I., Bähr-Porsch, S., Borgmann, S., Gatz, G., Gellert, G., Goebel, R., Hammans, W., Hentemann, M., Osmers, R., Rauskolb, R. & Hansmann, I. (1990) Cytogenetic analysis of 750 spontaneous abortions with the direct-preparation method of chorionic villi and its implications for studying genetic causes of pregnancy wastage. American Journal of Human Genetics 47:656-663.

Martin, R.D. (2013) How We Do It: The Evolution of Sex, Childbirth and Parenting. New York: Basic Books.

Miller, J.F., Wlliamson, E., Glue, J., Gordon, Y.B., Grudzinskas, J.G. & Sykes, A. (1980) Fetal loss after implantation. Lancet 316:554-556.

New York Times May 8, 2015  —  Article by Sarah Naslin Nir: “Perfect Nails, Poisoned Workers.”


Wilcox, A.J., Baird, D.D. & Weinberg, C.R. (1999) Time of implantation of the conceptus and loss of pregnancy. New England Journal of Medicine 340:1796-1799.

You are reading

How We Do It

Twins With Two Daddies

Strange cases of split paternity raise questions about human conception.

The Rising Tide of Caesarean Births

Can the worldwide surge in surgical delivery of human babies be reversed?

Going, Going, Gone? Human Sperm Counts Are Plunging

Recent findings confirm declining sperm counts linked to male health problems.