“Stress and tension does not prevent women undergoing infertility treatment from becoming pregnant,” reported The Daily Telegraph .
This news story is based on a review of previous studies, which investigated whether anxiety or depression affect the chances of becoming pregnant after a single cycle of fertility treatment, such as IVF. The review identified 14 studies in 3,583 women from 10 different countries, and combined their results to investigate this question.
The results showed that women who became pregnant after the treatment cycle did not differ significantly in levels of anxiety or depression before their treatment than women who did not become pregnant.
This well-conducted review provides a reliable summary of the existing studies on this topic. The researchers’ conclusions were appropriate, and the review should reassure women and doctors that the emotional distress of fertility problems or other life events should not damage the chance of becoming pregnant through fertility treatment.
Where did the story come from?
The study was carried out by researchers from Cardiff University and the University of Thessaloniki, Greece. The research did not receive funding. The study was published in the peer-reviewed British Medical Journal . The BBC, Telegraph and Mail accurately reflected the findings of this review.
What kind of research was this?
A systematic review, which searches the global literature on a particular subject, is the best way of identifying all relevant studies that have investigated whether a particular exposure (in this case, emotional distress) is associated with an outcome (in this case, pregnancy after fertility treatment). The difficulty with this sort of review is that the studies included are likely to have differed in some ways. For example, the populations studied and the methods and technology used may have differed between studies. In particular, emotional distress is a very subjective experience.
To account for this, studies should ideally use validated methods for assessing emotional distress. The reviewers did assess whether this was the case in the studies they included, and found that most of the studies used validated methods. In addition, in order to assess whether emotional distress could affect the outcome of fertility treatment, it would be important to measure emotional distress before the start of the treatment. To ensure that this was the case, the review only included studies that did this.
What did the research involve?
The researchers searched medical databases from 1985 to 2010 and hand-searched relevant publications and reports of fertility conferences to identify potential studies. They were interested in studies that investigated whether a woman’s level of emotional distress (anxiety or depression) before fertility treatment affected her chances of becoming pregnant. For inclusion in the review, studies had to assess the outcome of one cycle of assisted reproductive technology (in vitro fertilisation, intracytoplasmic sperm injection or gamete intra-fallopian transfer).
To be included, the studies had to have data available on pre-treatment anxiety or depression for women who became pregnant and women who did not. For their search, the researchers did not specify that the studies had to use particular methods for assessing anxiety or depression, but they did assess whether a reliable validated tool had been used. The researchers said that for those studies that used multiple measures for assessing emotional distress, they prioritised the assessments of “state anxiety”, which reflects a person’s current emotional state and is sensitive to “anticipatory” emotions (tension or worry, for example). The review used data on depression for studies that did not measure anxiety.
The researchers also looked at whether the pregnant and non-pregnant groups in each study differed in other factors that could affect the women’s chances of pregnancy, such as age, previous use of assisted reproductive technology, previous births and duration of infertility. They gave each study an overall quality rating based on a standard rating system.
The researchers say that they looked at outcomes after only a single cycle of treatment to prevent variations in the number of treatment cycles and duration of treatment from affecting the results. Researchers classified studies according to how they defined a successful pregnancy: a positive pregnancy test, β-human chorionic gonadotrophin urine or blood test within 21 days of embryo transfer, positive ultrasound scan or live birth.
Independent researchers assessed the studies’ eligibility, quality and extracted data. The main outcome measure was the mean (average) difference in pre-treatment anxiety and depression scores between the group of women who became pregnant and the group who did not.
What were the basic results?
Fourteen cohort studies met the researchers’ eligibility criteria. The studies included 3,583 women undergoing a cycle of fertility treatment in 10 countries. The average age of women was 29.7-36.8 years, and the average duration of infertility was 2.6-7.8 years.
Three studies included only women who had never used an assisted reproductive technology before, and the other 11 studies included a mix of women who had or hadn’t previously used this method of reproduction. The studies collected data between 1992 and 2006. The most commonly used measure of emotional distress was the validated Spielberger state-trait anxiety inventory. In almost half the studies, distress was assessed in the month before the treatment cycle began. In 11 studies, 80% of the participants completed follow-up. Three studies included groups of pregnant and non-pregnant women who were similar in all four key factors that could affect the chance of pregnancy (age, previous use of assisted reproductive technology, previous births and duration of infertility). Six studies included groups that were similar in at least two of these factors. Overall, six studies were considered to be of high quality, three of average quality and five of low quality.
The researchers found that pre-treatment emotional distress was not associated with pregnancy outcome after a single cycle of assisted reproductive technology. The pooled results of all 14 studies demonstrated that women who became pregnant did not have significantly different average pre-treatment anxiety and depression scores from women who did not become pregnant. Statistical tests showed that the included studies did not show significant variation in their results.
Analyses of whether results differed in different subgroups of women showed that previous use of assisted reproductive technology had no effect. Neither did the characteristics of the non-pregnant group (whether it excluded women who did not respond to ovarian stimulation or whose embryos were not fertilised), nor did the timing of the emotional assessment. Studies of different quality ratings also did not appear to vary in their results. However, the researchers reported that they found some evidence of publication bias (in other words, studies reporting certain results may not have been published). An analysis that predicted the effect these unpublished studies might have had on the results still showed no difference in pre-treatment anxiety or depression scores between pregnant and non-pregnant groups.
How did the researchers interpret the results?
The researchers concluded that the findings of their systematic review and meta-analysis “should reassure women and doctors that emotional distress caused by fertility problems or other life events will not compromise the chance of becoming pregnant [through fertility treatment]”.
This well-conducted systematic review and meta-analysis has several strengths:
- The study population was relatively large, including 14 studies and 3,583 women.
- The researchers ensured that the included studies had measured emotional distress before fertility treatment began, which means the levels of distress observed are more likely to have preceded pregnancy.
- Most studies had used validated assessment questionnaires and inventories to reliably assess anxiety and depression.
- The researchers chose to assess pregnancy outcomes after only one treatment cycle to prevent their results being affected by differences in the number of fertility treatment cycles given and cycle duration.
- Their statistical assessment of variability between the studies demonstrated that the results of the studies did not differ significantly, and therefore were more suitable for pooling.
However, there are a couple of points to note:
- As the researchers say, there was some evidence of publication bias, and other studies relevant to this question may not have been available. Had they been included, the results may have been different.
- The studies that were included did not all report how women were selected to participate, and therefore it is not clear whether the population in these studies was representative of women receiving fertility treatment as a whole.
- Not all of the studies had pregnant and non-pregnant groups that were balanced for potential confounding factors, which could have affected results (age, previous use of assisted reproductive technology, previous births and duration of infertility). These findings would be made more robust if this were the case.
- Only one out of the 14 studies assessed the outcome of live birth. The rest looked at positive pregnancy test results and positive scans. Therefore, the outcome of the pregnancies in the studies, and whether they resulted in the birth of a healthy baby, is unknown.
Overall, this review provides a reliable summary of the existing studies on this question. Based on this, pre-treatment emotional distress does not appear to reduce a woman’s chances of successful pregnancy through fertility treatment. It is worth noting that these results cannot tell us whether emotional distress has any effect on chances of natural conceptions.
Analysis by Bazian
Edited by NHS Website
Links to the headlines
BBC News, 25 February 2011
The Daily Telegraph, 25 February 2011
Daily Mail, 25 February 2011
Links to the science
British Medical Journal 2011; 342: d223