“Repeat abortions linked to premature birth,” BBC News has reported, with the Daily Mail website adding that multiple abortions could cause “life-threatening problems in later pregnancy”.
The story is based on Finnish research that looked at the effect of induced abortions on a subsequent first birth.
The study found that women who had three or more abortions were at a slightly increased risk of giving birth prematurely and of having a baby with subsequent low birth weight.
While the Daily Mail’s claims that multiple abortions could lead to “life-threatening problems” is technically correct, its tone is needlessly alarmist. In the “three or more abortions” group of women, only 0.9% of cases had complications that led to the death of the baby (compared with 0.48% in the “no abortions” group).
Furthermore, this type of observational study can’t prove direct cause and effect and it's possible that other factors affected the results. For example, previous research has found that women who have repeat abortions are also more likely to come from poorer backgrounds, which is an independent risk factor for premature birth (and higher mortality rates) in babies.
Still, as the researchers rightly pointed out, it is important to stress that while abortions are generally safe, they do carry a small risk of both short- and long-term complications.
The best way of avoiding an unwanted pregnancy is to use an effective method of contraception.
Where did the story come from?
The study was carried out by researchers from several academic institutions in Finland and Sweden. It was funded by the National Institute for Health and Welfare and the Academy of Finland. The study was published in the peer-reviewed medical journal Human Reproduction.
The study was covered relatively fairly in the papers. However the BBC’s report that “the more abortions a woman has before her first child, the more likely she is to give birth prematurely” was perhaps misleading. It could suggest that just one abortion might increase the risk of premature birth, when the study found no statistically significant evidence for this.
The BBC does deserve praise for placing the increased risks in their proper context and including the following quote from the lead researcher: "The increased risk is very small, particularly after only one or even two abortions, and women should not be alarmed by our findings."
Conversely, while the print edition of the Daily Mail covered the story accurately, the online version was arguably sensationalist by using the phrase "life-threatening problems" in a banner headline.
What kind of research was this?
This was an observational study that looked at whether induced abortions had any health effects on a woman’s subsequent first birth.
The authors pointed out that while some research has found an association between induced abortions and risk of premature birth, previous studies have had conflicting results and more data is needed.
What did the research involve?
Researchers used information on all first-time mothers in Finland who had given birth to a baby between 1996 and 2008 (except those who had had twin or multiple births) from a Finnish national birth register.
The register also contains information on:
- the mothers’ backgrounds
- care during pregnancy and delivery
- the babies’ health up to the age of seven days
They linked this data to the country’s abortion register for the period 1983 to 2008.
The register is based on obligatory information from doctors and includes information on:
- the reasons for induced abortions
- the abortion method
- when during the pregnancy the abortion was carried out
- the woman’s socioeconomic background and her reproductive health
The researchers then identified each mother’s first birth and whether she had had any abortions. Mothers were classified according to information from the abortion register, by the number of induced abortions before the first birth (none, two, three or more).
The researchers then calculated several birth outcomes according to the mother’s previous history of induced abortions.
The outcomes they looked at were:
- very low birth weight (<1,500g)
- low birth weight (<2,500g)
- very premature birth (<28 weeks)
- premature birth (<37 weeks)
- low one-minute Agpar scores (the Agpar score is a test designed to assess the newborn’s immediate state of health after birth)
- perinatal death (defined as the death of a baby from 22 weeks of pregnancy to seven days after birth)
Using information from the medical register, they adjusted their results to take account of established confounders that could also increase the risk of premature birth or low birth weight, such as:
- maternal age
- marital status
- socioeconomic position
- where the women lived
- whether they smoking during pregnancy
- any previous history of ectopic pregnancies or miscarriages
What were the basic results?
The researchers found that of the 300,858 Finnish mothers in the study, between 1996 and 2008 31,083 (10.3%) had had one abortion, 4,417 (1.5%) had two and 942 (0.3%) had three or more induced abortions before a first birth (excluding twins and triplets). Most abortions were carried out surgically and before 12 weeks of pregnancy. Most were undertaken for “social” reasons (in other words, the woman did not want to go forward with the pregnancy rather than there being medical reasons as to why an abortion would be recommended).
Below are the main findings, given after adjustment for possible confounders:
Compared with women who had not had an abortion, women who had had three or more abortions had a small, but statistically significant increased risk of:
- giving birth prematurely (odds ratio [OR] 1.35 95%, confidence interval [CI] 1.07 to 1.71)
- having a low birth-weight baby (OR 1.43, 95%, CI 1.12 to 1.84)
- having a very low birth-weight baby (OR 2.25, 95% CI 1.43 to 3.52)
There was a “dose response relationship” between the number of abortions a woman had and the number of very premature births. This means that generally the risk of having a very premature baby increased as the number of abortions increased, but only the risk after having a second abortion was statistically significant.
How did the researchers interpret the results?
The researchers pointed out that the increased risk of adverse outcomes found in women who had had abortions was very small and sometimes marginal. In an accompanying press release they pointed out that for every 1,000 women, three who have had no abortions will have a baby born under 28 weeks of pregnancy. This rises to four in women who have had one abortion; six in women who have had two abortions and 11 in women who have had three or more.
They said it is possible that the increased risk of premature birth and low birth weight associated with repeat abortions could be due to infection, especially given the stronger association between a risk of very premature birth and two or three abortions. Although the risks are slight, they suggested that health education should contain information about the potential health hazards of repeat abortions including very premature birth and low birth weight in subsequent pregnancies.
The results of this large study are a little worrying, although as the authors pointed out, the study can’t prove that repeat abortions carry risks for a subsequent pregnancy. It is possible that measured and unmeasured confounders might have affected the results, although the authors tried to take these into account. In particular, women who have repeat abortions are more likely to come from lower socioeconomic groups, which is an independent risk factor for premature birth and higher perinatal mortality.
Women should not be alarmed by these findings, but as the authors argued, the potential health problems associated with repeat abortions should perhaps be taught in health education. Effective contraception remains the recommended method of avoiding unwanted pregnancies.
Analysis by Bazian
Edited by NHS Website
Links to the headlines
BBC News, 30 August 2012
The Guardian, 30 August 2012
Daily Mail, 30 August 2012
Links to the science
Human Reproduction. Published online August 29 2012