Announcement on benefits of walking to improve pregnancy outcomes relies heavily on secondary analysis

Does the news release adequately quantify the benefits of the treatment/test/product/procedure?

Not Satisfactory

The news release mentions two benefits of physical activity with regards to “fecundability” (the ability to become pregnant):

1. Walking at least 10 minutes at a time improved fecundability in overweight woman (defined as a body mass index, or BMI, > 25)
2. Vigorous physical activity (>4 hours/week) was also associated with improved chances of becoming pregnant

So the study put into context the amount of physical activity associated with an increase in fertility rates but it didn’t tell us how much more likely vigorous walkers were to conceive compared to those who didn’t walk, or those who walked only a little.

Does the news release seem to grasp the quality of the evidence?

Not Satisfactory

It’s important to realize that the benefits mentioned in this study were found by way of a “secondary analysis” of data from a much larger, multi-center trial in which the primary research question had to do with aspirin use and pregnancy. A secondary analysis is simply using existing data to retrospectively explore an alternate question that’s distinct from the original research.

The news release does allude to this being a secondary analysis but doesn’t explain that such analyses usually lack the statistical power of findings related to the primary research question.

Importantly, the 1,214 study subjects had a history of 1-2 miscarriages, and the release does well in mentioning that findings from this unique sub-group may not generalize to all women trying to conceive. Another limitation is also included:

That physical activity is related to other behaviors and lifestyle factors, and women who are more physically active may be different from women who are less active in many ways.

Although the research attempted to adjust for these differences (called “confounding variables”) they could not do so completely. This was noted in the study and could have been pointed out in the release as an important limitation. (For more information on why reporting on variables is important see our post and video on a diet soda study that had some of the same issues.)

Another limitation that was not mentioned in the release is how physical activity — which is a key measurement in this study — was measured. It relied upon activity questionnaires completed by the women. These are not always reliable and can limit the significance of the findings.

Does the news release compare the new approach with existing alternatives?

Not Satisfactory

What other types of physical activity were measured? Many women might be left wondering if other forms of exercise are just as good as walking. It might have helped to clarify this (and if the study did not address this, then make that clear to readers). It would also help readers to note what the “dose” of exercise was that was associated with a positive effect on fertility.

There is a statement included from the lead author that could confuse people:

One of our main findings is that there was no overall relationship between most types of physical activity and the likelihood of becoming pregnant for women who had already had one or two pregnancy losses [except for walking, which was associated with higher likelihood of becoming pregnant among women who were overweight and obese.

That’s a lot of caveats! And many women — overweight or not, with previous miscarriages or not — might be left wondering if their form of exercise helps or hinders their chances at conception.

Does the news release establish the true novelty of the approach?

Not Satisfactory

Although the study authors provides a nice summary of other research in which the results do and do not agree with their study, this is not included in the news release and would have been a helpful addition.