Pregnancy and child

Study fails to prove effects of stress on fertility

“Stress can double the risk of infertility for women,” The Daily Mail reports. However, the evidence provided by the latest study is not as clear cut as the reporting implies.

This research recruited around 400 US couples who were trying to conceive. The women gave two samples of saliva: one when they enroled in the study and another just after their first period during the study.

The researchers looked at whether levels of two stress hormones measured in the saliva – cortisol and an enzyme produced in response to adrenaline levels (alpha amylase) – affected the likelihood of getting pregnant over the course of a year. Women were also asked to complete daily “stress journals”.

Most couples in this study (87%) successfully conceived over the 12 months.

According to results, women with the highest third of alpha amylase levels had borderline decreased odds of becoming pregnant compared to women with levels in the lowest third.

They were also twice as likely to not conceive over the 12 months.

There are several limitations to this study, however. Measuring stress hormones on only two occasions provides, arguably, a very limited assessment of daily stress, as does the use of “stress journals”.

The analysis may have had a detrimental effect in itself, with participation in the study possibly increasing stress and anxiety levels about becoming pregnant. This factor could mean these couples are not representative of the general population trying to conceive.

Overall, this is another study suggesting the possible association between stress and chances of conceiving, but it doesn't provide conclusive answers.

Where did the story come from?

The study was carried out by researchers from The Ohio State University College of Medicine, Division of Intramural Population Health Research at the Eunice Kennedy Shriver National Institute of Child Health and Human Development, and Texas A&M Health Science Center (all in the US). This study was supported by the Intramural Research Program of the Eunice Kennedy Shriver National Institute of Child Health and Human Development.

The study was published in the peer reviewed medical journal Human Reproduction.

The UK media’s reporting of this story was generally accurate, although more of an effort could have been made to highlight the numerous limitations of the study.

They have also used the US definition of infertility reported in this study, which uses a one-year timescale, rather than the UK definition, which is classed as “failure to conceive after frequent unprotected sexual intercourse for one to two years in couples in the reproductive age group”.

What kind of research was this?

This was a cohort study investigating whether higher stress levels impacted infertility.

Stress was assessed using levels of “stress hormones” in the saliva, which are said to be stimulated when a person feels continually tense and anxious.

The study followed a total of around 400 couples who were trying to conceive.

The researchers state that numerous pieces of previous research had demonstrated a link between stress and fertility. However, the direction of that relationship has been unclear – that is, whether increased stress leads to lower chances of conceiving, or whether failure to conceive increases stress.

The researchers aimed to objectively measures the body’s response to stress. When we are continually anxious, stress hormones are produced in two ways:

  • the hypothalamic-pituitary system leads to an increase in the levels of cortisol 
  • the adrenal glands, positioned at the top of the kidneys, release noradrenaline into the bloodstream, which results in the large salivary gland in the cheek (the parotid gland) releasing the enzyme alpha-amylase

Therefore, the researchers measured levels of both cortisol and alpha-amylase in the saliva, to try and objectively measure stress levels.

What did the research involve?

This research included 501 couples enrolled on an ongoing US cohort study between 2005 and 2009. The study enrolled people who were discontinuing use of any contraceptives, with the aim of getting pregnant. They said that in the absence of any established methods of recruiting a sample of people intending to get pregnant, they made use of a fish/hunting license registry in 16 counties in Michigan and Texas. They recruited people who met the following criteria:

  • where a woman aged 18-40 years was currently not pregnant and was either married or in a committed relationship with a male partner aged over 18
  • the woman had a self-reported menstrual cycle length of 21-42 days
  • the woman had not used hormonal birth control injections in the last 12 months (this was due to uncertainty over the time required for a return to normal fertility)
  • the couple had never been told by a healthcare provider that they could not get pregnant without medical help
  • the couple said they were actively trying to get pregnant and had not been using contraception for at least two months when the study began

When an eligible couple agreed to participate, the man and woman were interviewed separately at home, and were trained in the use of daily journals, fertility monitors and pregnancy tests. The woman collected a sample of her saliva first thing in the morning on two occasions:

  • on the morning following enrolment into the study
  • on the morning following her first period in the study

Levels of salivary cortisol and alpha-amylase were measured in the laboratory.

The women were asked to fill out a daily journal, which included the question: “Please tell us your overall stress level each day”. The possible answers were:

  • 1 = almost no stress
  • 2 = relatively little stress
  • 3 = a moderate amount of stress
  • 4 = a lot of stress

The couples were followed for up to 12 months – or if pregnancy occurred, they were followed up throughout pregnancy. The main outcome the researchers were interested in was the time it took to conceive, as defined by a positive home pregnancy test.

They also looked at the specific probabilities of becoming pregnant during the peak six-day window around ovulation, as indicated by the fertility monitor.

The final outcome was “clinical infertility”. This was defined by the researchers in this study as a failure to achieve pregnancy, despite 12 months of regular, appropriately timed unprotected intercourse.

Outcomes were adjusted for the following confounders:

  • woman’s age
  • income
  • ethnicity
  • use of cigarettes
  • alcohol intake
  • caffeine intake
  • difference in age from her partner

What were the basic results?

Of the 501 couples who enroled in the study, 100 (20%) withdrew; this was mostly due to a lack of interest in participating. Of the 401 (80%) women who completed the study, 347 (87%) became pregnant and 54 (13%) did not. Among those 401 women, 373 women (93%) had complete saliva data for this analysis.

There was no significant change in the levels of stress hormones measured in the saliva at the study’s enrolment and the second measure, taken after the woman’s first period.

There were no differences between those who did and did not become pregnant as to the number of times intercourse was had during the fertility window, or in their levels of salivary stress hormones.

There was also no difference in the level of daily stress reported by women each month.

When the researchers studied the likelihood of becoming pregnant and time to pregnancy, they did find some associations with levels of salivary hormones. They found that women with salivary alpha-amylase levels in the highest third were around a third less likely to fall pregnant, compared with women with levels in the lowest third.

However, this was only of borderline statistical significance (odds ratio (OR) of pregnancy 0.71, 95% confidence interval (CI) 0.51 to 1.00).

Across each menstrual cycle, there was a general trend for women in the highest third of stress levels to be less likely to be pregnant than women with the lowest levels, though results were not statistically significant at all time points.

The researchers also found that women with salivary alpha-amylase levels in the highest third were twice as likely as women with levels in the lowest third to have not conceived by the end of the 12 months. As such, they meet the US definition of clinical infertility used in this study (relative risk (RR) 2.07, 95% CI 1.04 to 4.11). 

There were no significant differences observed between women in the middle third of salivary alpha-amylase levels compared to women with the lowest levels.

There was no significant associations observed between likelihood of pregnancy and salivary cortisol levels.

How did the researchers interpret the results?

The researchers say this was “the first US study to demonstrate a prospective association between salivary stress biomarkers and time to pregnancy, and the first in the world to observe an association with infertility”. 


This study found that higher levels of stress, as measured by a woman’s salivary alpha-amylase levels, were associated with a longer time to pregnancy. There was also a link between alpha-amylase levels and an increased chance of a couple failing to conceive within 12 months of regular, unprotected sex.

This study benefits from a relatively large sample of couples recruited from the general population. Previous studies of this type tended to include couples recruited from fertility clinics, meaning they were not representative of the population at large.

Stress hormones were measured only on two occasions – at enrolment and after their first period. By assessing women early on, they could be less stressed about whether or not they would actually conceive than they may have been if assessed several months after trying. This may help the researchers in trying to examine the temporal nature of the relationship – whether stress levels influence the chances of becoming pregnant, or the reverse scenario, that failing to become pregnant influences stress levels. However, it still cannot very reliably tell us how or if stress and successful conception are related. It is likely to be a combination of both factors. 

If repeated saliva samples had been taken on numerous occasions throughout the time the woman was trying to conceive, this may have given different results. 

Other limitations to consider include the fact that:

  • there was only a very crude measure of the woman’s daily reported stress levels. This did not reveal any difference between women who did and did not conceive. However, just asking someone to place on a scale their overall levels of stress each day is also not likely to give a very good indication of their overall psychological health and wellbeing
  • 87% of women succeeded in becoming pregnant. Only 13% (54 women) did not. Analyses examining the chances of not becoming pregnant according to stress hormone levels samples small numbers of women, which increases the possibility of chance findings
  • there was reportedly no significant difference between overall levels of salivary stress hormones in women who did and did not become pregnant
  • The main outcome the researchers were interested in was if salivary stress hormones were associated with the likelihood of becoming pregnant. Women with levels in the highest third of salivary alpha-amylase had a lower chance of becoming pregnant than women in the lowest third – but this was only of borderline statistical significance. There was also no association at all with levels of the stress hormone cortisol
  • Though this study was as natural as possible, the fact that couples were asked to complete journals, monitor their peak fertility in each cycle and knew they were taking part in a study that was assessing whether they became pregnant, could all potentially cause stress and anxiety. Therefore, this study may not be representative of all couples in the general population trying to conceive

Overall, this is another study suggesting the possible link between stress and chances of conceiving. However, it doesn't provide conclusive answers, and those trying to conceive should not be concerned by the results.

Ultimately, getting stressed about the notion that stress could make it more difficult to conceive is self-defeating.

If you do find yourself struggling with stress, visit the NHS Choices Moodzone, which contains a range of useful articles about how to combat feelings of stress and anxiety.

NHS Attribution