In a paper recently published on European Sociological Review together with Marco Cozzani and Moris Triventi we analyse how prenatal stress affects prematurity and low birth weight and how this effect varies by parental socio-economic status.
A key problem that researchers who study the effect of stress on pregnancy outcome is that there are a number of maternal traits and behaviors that might determine both the level of stress experienced and pregnancy outcomes, making the observed associations between stress and birth outcome possibly spurious (i.e. false). To address this problem (that in social science is technically labelled ”endogeneity”) we use a natural experiment and consider the 2004 Madrid train bombings as an exogenous source of stress.
In our paper we employ a statistical technique, a difference-in-differences estimation with a synthetic control approach, that enable us to address the endogeneity problem discussed above and identify the causal effect of stress due to the exposure to the bombing on later pregnancy outcomes
The study is, however, also based on in-depth knowledge and experience of the phenomenon investigated. It happens that in 2004 I lived in Madrid and I actually live at about 800 m. from the Atocha station where the terrorist attack took place. I have direct experience and a clear memory of the sequences events and more importantly of the feelings and overall emotional climate in the city in the hours and days following the terrorist. A the beginning I experienced a mix of sudden fear and deep sadness. I experience the feeling of being helpless and fragile and the scaring thought shared by all friends that any of us could have also been on that train. And breath-taking anxiety once I suddenly realized that some friends did usually take those trains. I think that everybody in Madrid had the paralyzing though that a friend or relative or an acquaintances could had been involved in the attacks. And the phone line collapsed for quite some time after the bombing, the information was contradictory, it was not possible to communicate on the cell phone. Later the shock and fear gave space to angriness against a government who had involved Spain in the Iraq war and was lying point to ETA as the author of the terrorist attack. Two days after the attack there were spontaneous demonstrations in front of the PP party headquarter and all around the city, with a mixture of strange euphoria, angriness and sadness (the photo heading this blog was taken in front of the PP party headquarter on the 13th of March and you might even recognise someone…).
In the aftermath I wrote a report article with a chronical of those days in Madrid for Italian magazine Carmilla. Here the link:
All this to say that the Madrid bombing in 2004 is not a natural experiment that we have found sitting on the sofa and waiting to stumble into a proper causal identification design. I have actually lived the impact and emotional turmoil that the terrorist had on the city very closely.
*Spoiler: In our study We find that children in utero exposed to the Madrid bombing had a higher risk of prematurity and low birth weight, and that this detrimental effect is consistently concentrated among low-SES offspring. We interpret the latter pattern as consistent with a model of compensatory advantage