Pregnancy and child

Does putting the clocks forward make IVF more likely to fail?

"Miscarriages for women on IVF 'double when the clocks go forward because the loss of an hour in bed puts more stress on a mother-to-be's body'," reports the Mail on Sunday about a study of more than 1,500 cycles of IVF treatment in the US.

In a similar system to British Summer Time (where clocks go forward on the last Sunday of March), American clocks are set forward on the second Sunday of March – this is known as Daylight Saving Time (DST).

The rationale behind both systems is that we trade an hour of daylight in the early morning for an extra hour in the evening.

The researchers looked at whether the shift into DST during the women's treatment was associated with their likelihood of getting pregnant or their risk of miscarriage. They found women were more likely to have a miscarriage if the clocks went forward within 21 days of an embryo being implanted, compared to other times.

Critics of DST-style systems argue that the body's biological clock continues to be synched to the timings of light and darkness over the course of a 24 hour day and not the "adjusted time" on a clock; leaving us all slightly jet-lagged. The authors of this study propose this may disrupt delicate biological processes such as pregnancy.

While this sounds worrying, it is worth bearing in mind that this study only looked at a small group of women at one clinic, and cause and effect has not been proven.

Where did the story come from?

The study was carried out by researchers from Boston University School of Medicine and IVF New England, Massachusetts, in the US. The study did not receive any funding and the researchers report no conflict of interest.

The study was published in the peer-reviewed Chronobiology International and is open-access meaning it is free to read online.

The UK media generally reported the story accurately, though there were a number of errors. The Daily Mirror inaccurately claims the "loss of an extra hour of sleep was found to cause additional stress and anxiety to women in the very early stages of pregnancy". However, the length of sleep was not actually investigated and neither were the anxiety or stress levels of women, merely the time of year and relation to the clocks changing.

The Times' headline does not make clear that the link was only investigated in women having IVF, and not all pregnancies.

What kind of research was this?

This was a retrospective cohort study aiming to look back in time at the relationship between the timing of Daylight Saving Time (DST) and the success of in vitro fertilisation (IVF).

DST is the term used in the US for when the clocks go forward an hour in the middle of March (where an hour is "lost") or backwards in autumn (where an hour is "gained").

It's similar to the UK's British Summer Time (BST) scheme, except BST begins at the end of March.

A cohort study is good at looking at links between factors such as DST and outcomes such as pregnancy loss, however cannot prove cause and effect.

This is because factors other than DST may differ between the groups being compared and affect the outcomes.

While researchers can take such differences into account, they need to have data available on these factors to do this.

This may not be the case in a retrospective study, as researchers can only look back at data collected at the time and may not have all the data they need.

What did the research involve?

The pregnancies of 1,654 women who were treated at a fertility centre in New England, US, were included in the sample. All women had undergone IVF using their own freshly collected eggs rather than eggs from a donor or eggs that had been frozen.

The IVF cycles were split into groups depending on timing of DST compared to when embryo transfer occurred:

  • Group 1 – DST occurred between the first day of stimulation of the ovaries to produce mature eggs and the day of embryo transfer to the womb.
  • Group 2 – DST occurred within 21 days following embryo transfer to the womb.
  • Control group – DST happened at least 10 weeks after the first day of ovarian stimulation.

The IVF cycles were also put in categories according to whether they occurred in the spring or autumn. Researchers looked at whether each cycle resulted in a pregnancy where the embryo was growing inside the womb as normal and also whether a spontaneous pregnancy loss (miscarriage) occurred.

The researchers looked at other confounding variables that may have affected the results, including body mass index (BMI), smoking, and cycle factors such as hormone levels and number of embryos transferred at the time of IVF. Where a factor looked like it might be having an effect the researchers took this into account in their analyses.

They also looked at whether the woman's age or whether they had previously experienced pregnancy loss affected the results.

What were the basic results?

They found that rate of pregnancies did not differ between the three groups in spring (between 40.1% and 43.9%) nor did they differ between the groups in autumn (between 40.5% and 45.3%). The overall pregnancy rates in spring and autumn were also very similar (41.4% versus 42.2%).

Between the three groups, there were no differences in pregnancy loss in the IVF cycles carried out in the autumn (that is, when clocks were going back). However, for the IVF cycles carried out in spring:

  • Group 2, when DST occurred within 21 days after embryo transfer to the womb, had a significantly higher rate of pregnancy loss than Group 1, when DST occurred between day 1 of ovarian stimulation and embryo transfer, at 24.3% compared to 10.2%.
  • Group 2 had a significantly higher rate of pregnancy loss than controls (when DST occurred a long time after ovarian cycle stimulation and embryo transfer) at 24.3% compared to 12.5%.

Women who had a history of pregnancy loss experienced a higher rate of pregnancy loss than those with no such history. Among women who had a history of pregnancy loss:

  • Group 2 had a higher rate of loss than Group 1 or Controls in spring (60% compared to 32.4% and 22.4%).
  • No real difference was found between the three groups in autumn.

Among women who had not experienced previous pregnancy loss, no difference was found between the three groups in spring or autumn.

The timing of DST still showed significant link to pregnancy loss after researchers took account of age, number of previous pregnancies, and previous pregnancy loss.

BMI and smoking did not differ between groups and neither did factors such as hormone levels, therefore were less likely to have influenced differences between groups.

How did the researchers interpret the results?

The researchers concluded that "in our present work, transition into daylight savings time in the Spring has an association with spontaneous pregnancy loss specifically in patients who have a history of prior loss."

They added that "a higher rate of pregnancy loss occurred when Spring DST took place within the 21 days after embryo transfer when compared to patients who experienced DST before embryo transfer (Group 1), or well outside the window of the entire cycle (Control). We found that only patients with a prior history of spontaneous pregnancy loss were susceptible to the transient stress of Spring DST."

However, they do conclude: "It is important to note that an association does not necessarily imply causality. Further investigation by prospective study is required before assuming that this association is causal, clinically significant, or necessitates intervention".


This study has identified a link between the clocks going forward in spring and pregnancy loss for women who have had IVF embryos implanted in the past 21 days. This link seemed to be particularly pronounced in women who had experienced pregnancy loss before.

However, there are some important limitations to bear in mind:

  • The women were all taken from one clinic in America, and were mostly white, so it is hard to generalise results to other populations, including women in a UK setting. Further studies in women from multiple clinics and in different countries are needed to check whether the link is seen in these settings.
  • Although the study included more than 1,500 cycles of IVF, once the researchers started splitting women up into subgroups for analysis (for example, those having IVF in spring, and who had a previous pregnancy loss) the numbers were smaller. Larger studies are needed so we can be more confident in the results.
  • As this was a retrospective study that looked back at medical records, the researchers only had access to data that was recorded at the time. This may not have included important factors that they may have wanted to take into account, for example life events occurring at the time, diet, family and other socio-economic factors. A prospective study, looking forward, would be needed to ensure the researchers had all the relevant data.
  • The underlying biological mechanism for this apparent link was not studied, so it cannot be said what it was that may have contributed to the miscarriages – but the researchers suggested that it may have been stress from disrupting the body's rhythm. However, the level of stress was not actually looked at.
  • Only women having IVF treatment were included in the research. For women who are getting pregnant naturally, it is not known if DST is linked to pregnancy or pregnancy loss.

Based on the evidence provided by this study, if you are undergoing IVF treatment, the timing of the treatment should probably not be of concern.

Current evidence suggests that maintaining a healthy weight and avoiding alcoholsmoking and caffeine during treatment may improve your chances of having a baby with IVF. 

NHS Attribution