Pregnancy and child

Air pollution associated with low birthweight

"Study links low birthweight to air pollution and traffic," The Guardian reports.

A new EU-wide study has found a strong link between exposure to air pollution in pregnant women and low birthweight babies. Low birthweight can increase the risk of the child developing a chronic disease. The study found that pregnant women living in areas with higher levels of pollution, usually associated with traffic density, had a correspondingly higher risk of having a low birthweight baby.

Crucially, the study found that the risk persisted even when pollution levels met or were below European air quality limits.

This was a large, well-conducted European study and its findings are concerning. It should be noted that it did not measure the women's exposure to air pollution directly. Instead, it recorded where the women lived. 

There is little that pregnant women can do to limit their exposure to air pollution. Hopefully, the findings of this study will encourage policymakers to make further efforts to reduce air pollution in Europe.

Where did the story come from?

The study was carried out by researchers from European research institutes. It was funded by the European Union.

The study was published in the peer-reviewed medical journal The Lancet Respiratory Medicine.

It was covered reasonably well in the papers, with several including comments from independent UK experts.

However, The Independent’s claim that the risk from air pollution is as great as smoking during pregnancy is potentially misleading.

As the authors point out, at the individual level, smoking has a greater effect than air pollution on birthweight. The collective risk comes from the fact that more women are exposed to air pollution than smoking during pregnancy.

What kind of research was this?

This was a prospective cohort study that assessed the effect of maternal exposure to air pollution during pregnancy on:

  • low birthweight at term (less than 2,500g after 37 weeks of pregnancy),
  • birthweight
  • the baby’s head circumference (important because of the potential effect on brain development)

The research was part of the European Study of Cohorts for Air Pollution Effects (ESCAPE), in which the association between exposure to outdoor air pollution and health is being investigated. 

It is being co-ordinated by the University of Utrecht in the Netherlands.

Cohort studies are often used to examine the link between lifestyle factors (in this case, exposure to air pollution) and later health outcomes (birthweight) because they can follow very large groups of people.

The main limitation of the study design is that many factors can affect the risk of low birthweight and it is difficult to rule out the possibility that factors other than the one being studied are influencing any link seen.

While researchers can take steps to reduce the impact of these factors on their analyses there may be other factors (confounders) that are unaccounted for that influence the results.

The researchers point out that previous research has linked air pollution to premature birth (birth at less than 37 weeks of pregnancy), low birthweight, congenital malformation and other adverse effects. Low birthweight is associated with wheezing and asthma during childhood and poor lung function in adults.

The researchers were particularly interested in air pollution from fine particulate matter (PM), which is found in traffic fumes and industrial air pollutants.

PM consists of a mixture of tiny particles and droplets of liquid. As PM is so small it has the ability to bypass the body’s defences against foreign bodies and can cause damage to the heart and lungs.

What did the research involve?

Using data from ESCAPE, the researchers pooled data from 14 cohort studies in 12 European countries: Norway, Sweden, Denmark, Lithuania, England, Netherlands, Germany, France, Hungary, Italy, Spain and Greece. 

The study involved 74,178 women who lived in the study area and who had singleton babies between February 1994 and June 2011, and for whom home addresses during pregnancy, infant birthweight and gestational age and sex were available.

Measurements of air pollution were carried out between October 2008 and February 2011 in several sites within each area. 

Concentrations of nitrogen oxides (gases produced by industrial processes) and different sizes of particulate matter (PM) were estimated at the women’s home addresses, using a recognised method of measuring pollution called land-use regression (LUR).

LUR involves creating a statistical predictive model based on multiple samples taken in a specific geographical area over a period of time.

Traffic density (number of vehicles daily) on the nearest road and total traffic load on all major roads within 100m of residence were also recorded.

Because of financial constraints in some EU countries, sampling of particulate matter was not done everywhere and data is missing from some of the centres involved in the research. 

Also, there was little data on nitrogen dioxide from some air monitoring networks. 

Information about each baby’s gestational age, birthweight, head circumference, sex and mode of delivery was obtained from birth records and questionnaires.

The health outcomes in the babies they looked at were:

  • low birthweight at term (weight <2500g at birth after 37 weeks of gestation),
  • term birthweight
  • head circumference at birth

The researchers accounted for changes of home address during pregnancy when the date of moving and the new address were available.

An exception was made for traffic density, which was analysed only for women who did not change home address during pregnancy.

Detailed information about individual women was obtained during the pregnancy through interviews and self-administered questionnaires in most cohorts. 

This data included factors (confounders) that could affect birthweight such as:

  • gestational age (usually measured from the start of the last menstrual period and delivery)
  • sex
  • number of other children
  • maternal height
  • weight before pregnancy
  • average number of cigarettes smoked per day during second term of the pregnancy
  • maternal age
  • maternal education
  • season of conception (January-March, April-June, July-September or October-December)

Using statistical methods the researchers created several different models looking at the association between exposure to air pollution and low birthweight at term, birthweight generally and head circumference. They adjusted their findings for the individual factors described above, such as smoking.

They calculated the percentage of cases of low birthweight that would be prevented within the population if PM concentrations were reduced to 10 micrograms per cubic metre of air (10µg/m³) or less – which is the World Health Organization’s maximum level for good air quality. 

What were the basic results?

The researchers found that all air pollutants, particularly fine particulate matter (PM with a diameter of 2.5 micrometres or less), and traffic density, increased the risk of term low birthweight and reduced average head circumference at birth, after they had accounted for confounders such as maternal smoking.

The researchers estimated that if levels of PM 2.5 were reduced to 10µg/m³, around one in five (22%) cases of low birthweight among term deliveries could be prevented.

The detailed findings were as follows.

  • For every increase of 5 micrograms per cubic metre (5µg/m³) in exposure to fine particulate matter during pregnancy, the risk of low birthweight at term rose by 18% (adjusted odds ratio [OR] 1.18, 95% confidence interval [CI] 1.06-1.33).
  • This increased risk was recorded at levels below the existing EU annual air quality recommendations of a PM 2.5 limit of 25µg/m³ (OR for 5μg/m³ increase in participants exposed to concentrations of less than 20μg/m³ 1.41, 95% CI 1.20-1.65).
  • Larger particles, nitrogen dioxide and traffic density were also associated with a higher risk of low birth weight at term.
  • If levels of PM 2.5 were reduced to 10μg/m³ during pregnancy, 22% of cases of low birthweight could be prevented.

How did the researchers interpret the results?

The researchers say their findings suggest that in-utero exposure to ambient air pollution in European urban areas could explain a substantial proportion of cases of low birthweight at term.

They say the mechanisms by which pollution might affect foetal growth are unknown but it could affect hormones important in pregnancy, the growth and function of the placenta (which provides nutrients to the foetus) or cause oxidative stress (damage to cells caused by a disruption in cell signalling).

According to the lead researcher Dr Marie Pedersen, from the Centre for Research in Environmental Epidemiology in Barcelona, Spain: "The widespread exposure of pregnant women worldwide to urban ambient air pollution at similar or even higher concentrations than those assessed in our study provides a clear message to policymakers to improve the quality of the air we all share."

Writing in an accompanying commentary, Professor Jonathan Grigg from Queen Mary, University of London, says: "Overall, maternal exposure to traffic-derived particulate matter probably increases vulnerability of their offspring to a wide range of respiratory disorders in both infancy and later life… Dissemination of [these] results to the wider public could, therefore, further increase the pressure on policymakers to reduce exposure of urban populations to particulate matter.”


The strengths of this study lie in its standardised assessment of exposure to pollution, the detailed information it had on potential confounders and its large population spread through a wide geographical area.

However, it did not directly measure the women’s exposure to pollution but used estimates based on where they lived. 

As the authors point out, there is always the possibility that pollution measures were misclassified, and also that confounders – both measured and unmeasured – affected the results.

Individually, there is little we can do about reducing air pollution in urban environments. 

But, hopefully, this research, along with similar studies, will help to persuade politicians, policymakers and planners to make more of an effort to create "greener" urban environments – a goal that is unlikely to be met overnight.

NHS Attribution