Job loss linked to stillbirth and miscarriage

Job loss

Published 22 November 2023

Data based on more than 8,000 UK pregnancies show odds of non-live birth almost double for parents-to-be who become unemployed.

The risk of miscarriage and stillbirth almost doubles if a woman or her partner lose their job during pregnancy, according to the first analysis (1) of its kind. This is compared with those who are not made involuntarily redundant after they conceive.

Based on data from more than 8,000 pregnancies and nearly 5,000 women in the UK, the findings suggest that some pregnancy losses can be prevented, according to the authors.

The results show the odds of a non-live birth are more than 20% for pregnancies where one or both parents-to-be is made redundant whereas the figure is 12% if no job loss occurs.

The authors urge policymakers to improve measures to protect labour markets and social security systems to safeguard pregnant women and their partners.

Economic downturns, stressful life events and intense working hours have all been studied previously in the context of pregnancy loss. However, there has been little analysis of whether job loss disrupts the course of pregnancy, according to the authors.

This knowledge gap needs addressing, they say, given that job loss has ‘far-reaching consequences for women’ and their families.

On this basis, the study set out to provide evidence on the link between involuntary termination of employment and pregnancy disruption (ie miscarriage and stillbirth), starting from the time of conception and ending with pregnancy outcome.

The data was drawn from Understanding Society (2), the UK Household Longitudinal Study (UKHLS) which has a sample size of 40,000 representative UK households (approximately 100,000 individuals). Conducted annually from 2009 to 2022, the UKHLS features self-reported information from women.

The sample used from the UKHLS for this present study represented 8,142 pregnancies among 4,942 women (and partners where relevant). Predicted outcomes were pregnancy loss or live birth, and the risk of pregnancy loss from either partner’s involuntary job loss because of redundancy or dismissal.

The authors write that an advantage of the UKHLS is that it records the length in months of any gestation that ends prematurely in a non-live birth; and the timing of a job loss also in months. As such, they were able to combine this information to determine if the exposure to job loss occurred during the pregnancy and at what stage.

Statistical analysis was carried out using 3 models. For model 1, the authors estimated a baseline to predict the risk of a non-live birth. This included the primary exposure variable and covariates for women’s demographic characteristics and background (age, ethnicity, and parents’ social class), the experience of a prior pregnancy loss, and year and month fixed effects.

Model 2 assessed highest educational attainment, marital status, and health additional to the demographic characteristics used in the first model; and model 3 included socio-economic status of the partner and household income in addition to the characteristics used in the first 2 models.

Pregnancy loss was defined as a pregnancy that did not result in a live birth, miscarriage as a loss before the 24th week of pregnancy, and stillbirth as a loss after the 24th week. Study exclusions from the main analysis included pregnancy termination, ongoing pregnancy at time of interview, and job loss that involved factors such as a contract termination and illness.

Results showed that around 1 in 10 (n=985; 11%) of the total pregnancies (n=8142) resulted in miscarriage or stillbirth; and job loss occurred among less than 1% (n=136) of pregnancies.

Of the 136 pregnancies where job loss has taken place, pregnancy loss occurred almost a quarter (24% overall; miscarriage=23.5%; stillbirth 0.7%) vs the 12% of pregnancies where there was no job loss. This represented an increased risk for the job loss group (odds ratio (OR)=1.99, 95% CI: 1.32-2.99).

The authors say that this association between pregnancy disruption and job loss ‘remained robust’ for model 2 (OR=1.83, 95% CI: 1.22-2.75) and model3 (OR=1.81, 95% CI: 1.20-2.73).

Strengths of the study included the fact the authors could distinguish between involuntary and anticipated causes of job loss. Among the limitations were the small sample size and the fact participants self-reported the date/outcome of conception.

So, what mechanisms are at work that cause job loss to lead to pregnancy disruption? Even though this was not an aim of the study, the authors cite evidence that stress results in the production of hormones (eg cortisol) which are known to increase the risk of miscarriage. Moreover, a reduction in available income could restrict access/compliance to prenatal care.

In conclusion, the authors suggest their analysis be replicated in different countries to assess the validity of the findings from a global viewpoint. This would also identify the extent to which measures against unemployment help reduce the impact of job loss on pregnancy outcomes.

1 Di Nallo A, Selin Köksal S. Job loss during pregnancy and the risk of miscarriage and stillbirth. Human Reprod, vol 38(11) Nov 2023; 2259–2266;

Get notified of new articles with our ESHRE newsletter.

Sign up and never miss an update