Researchers in Denmark carried out a study between 1996 and 2002, with the results being published in September 2007. Using a questionnaire, they interviewed more than 92,000 women about their exercise patterns during pregnancy. The women's responses were documented through computer assisted telephone interviews.

The authors of the study concluded that there appears to be an association between exercise in early pregnancy and the risk of subsequent miscarriage. They do highlight that the results should be treated with caution as the method they used to obtain the information could have led to bias which then affected the study results.

The telephone interviews were carried out using computer assisted technology. The women were asked "Now that you are pregnant do you engage in any kind of exercise?". If they answered yes to this question, they were then asked:

"What kind of exercise do you engage in?"
"How many times a week do you engage in [this exercise]?"
"How many minutes a time do you engage in [this exercise]?"
"Do you engage in other kinds of exercise?"

If they answered yes to the final question, they then looped through the series of questions again to allow them to answer for all types of exercise they did. The researchers categorized the answers into six categories: high impact exercise (jogging, ball games, and racket sports), low impact exercise (aerobic
for pregnant women, aerobic, dance, and walking/hiking), workout or fitness training, bike riding or horseback riding, swimming, and other. When asking questions about other factors for analysis of the results they included:

• mother's age

• previous miscarriages

• education level & employment status

• coffee consumption during pregnancy

• smoking during pregnancy

• amount of physical exertion at work

• history of an eating disorder

• body mass index before pregnancy

• history of fertility treatment

• history of chronic disease

• number of pregnancies and previous live births

Miscarriage was defined as the death of an infant before 22 weeks gestation (not including terminations). They compared the miscarriage rates of women who exercised with those who did not. The findings indicate that the risk of miscarriage triples in women who exercise strenuously when compared with those who do not exercise.

There are several issues worth considering before assuming that exercise increases the risk of miscarriage.

Previous history of exercise not taken into account

The researchers do not appear to have asked the women whether they had exercised prior to their pregnancy. We are unable to determine whether strenuous exercise in itself is a risk factor for miscarriage or whether it is only a risk factor if the woman has exerted herself more in pregnancy than she normally would.

Bias due to self disclosure about smoking and coffee drinking

The women were all asked to report whether they smoked or drank coffee during pregnancy. Although the women were not being interviewed face to face, the social judgments made towards women who smoke during pregnancy could have led women who did smoke to not disclose this information. The same is true, although possibly to a lesser degree, with coffee consumption since it is now well known amongst pregnant women that drinking excess coffee can impact the health of a developing baby and may increase the miscarriage rate.

Assumptions about exercise and health leading to bias

It is not clear what the women were told the study was examining. If they were told the study was looking at the links between exercise and miscarriage, there is a possibility that they assumed it was looking at whether exercise DECREASED the risk of miscarriage. If this is the case, women may have overstated how much exercise they did during pregnancy. We would assume that women are more likely to say they did more exercise than they really did, rather than say they were sedentary, since generally exercise is considered a beneficial activity.

Miscarriage itself may affect recall about exercise patterns

Nearly 77% of the women who miscarried were interviewed after the miscarriage had taken place. The miscarriage itself may have affected how the women recalled the amount of exercise they did during pregnancy. Bearing in mind that grief affects our recall of events, and guilt is often a factor in miscarriage, it is possible that the women overstated the amount of exercise they did.

Healthy pregnancies may lead to more women stopping exercise

The authors point out another source of bias. It is well documented that women with healthy pregnancies are more likely to experience nausea in the first trimester, when compared with women who will subsequently miscarry. If women experience nausea, they may be more likely to stop strenuous exercise in early pregnancy because they feel ill. This would mean that rather than the exercise being a cause of the miscarriage, it was possible because the woman was going to miscarry anyway.

Number of minutes of exercise is different from intensity of exercise

The authors asked women in the interviews how many minutes of exercise they did. However, they did not ask the women what levels of intensity they used when exercising. Two women doing an aerobic workout may do it at very different levels. Walking and hiking were included in low impact exercise for example, but there is a significant difference between walking to pick the kids up from school and hiking in the forest. Without knowledge of intensity of exercise it is difficult to draw any conclusions simply from the number of minutes that a woman exercised.

Miscarriage rates are inconsistent with "typical" miscarriage rates

In the study only 3.4% of the women miscarried. This is significantly lower than the number of pregnancies typically expected to miscarry which is usually quoted as between 15-20%. It is not clear as to why the miscarriage rate was so low in this study population but this could have been a confounding factor that impacted the results.

The authors acknowledge the many areas of potential bias in the study and state themselves that it is not possible to draw any conclusions or make any recommendations based on their findings.  Interestingly though, many of the news reports that followed the publishing of this paper used headlines that appeared to draw conclusions.

Inquirer: Exercise during pregnancy can lead to miscarriage

The Australian: Heavy exercise can up miscarriage risk

News (news.com.au): Exercise can cause miscarriage

The Guardian: Exercise can increase risk of miscarriage

It is worth thinking about how these headlines affect women and how they behave during their pregnancy as a result of reading them. The newspaper articles do not evaluate the research or look at some of the questionable areas. It is unlikely that the reporters have even looked at the research itself. In most cases, they receive a brief article from a news agency such as Reuters, which is then replicated across the world in each news publication.

Many news articles concluded with: But the findings do argue in favor of a review of exercise guidelines for pregnant women, the authors suggest.

What the researchers actually said in their conclusion was: In spite of the findings of this study, we do, however, think that it is too early to draw any public health inferences on this basis. Many positive effects of exercise are well established, and the findings of this study need to be replicated.

You can read the full text of the research here:

http://www.blackwell-synergy.com/doi/pdf/10.1111/j.1471
-0528.2007.01496.x

For more information on training with Childbirth International, take a look at our website, or contact us.

Childbirth International offers training programs for Birth Doulas, Postpartum Doulas and Childbirth Educators. From September 2008 we will also be launching a Breastfeeding Counseling course.