As we come to the next presidential election, there is no doubt going to be a resurgence of talk about the war in Iraq and threats of terrorism. A recent study from Alison Holman and colleagues at UC Irvine suggests that people who were really stressed out by 9/11, and have ongoing worry, may be literally taking the threat to heart. The study of 2592 people showed that people who reported “ongoing worry about terrorism post 9/11″ showed a lot more physician diagnosed heart problems two and three years after the attacks than those who didn’t worry as much (risk ratios were 4.67 two years after and 3.22 three years after).
What is so nice about this study is that the authors had baseline data regarding health problems and anxiety before the attacks, and so here we get a rare, prospective look at the effect of our nation’s stress response and what that means in concrete, physical health related terms.
What is interesting, though, is that the media coverage (e.g. NYT coverage) has directly linked the government response (terror alerts) and the heart health of the population. While this may be true, the study did not assess exposure to terror alerts, or look at any mediating factors related to ongoing exposure.
The other major issue with the paper is that the authors repeatedly refer to “reports of physician diagnosed illness,” but what they are measuring is SELF-REPORTED by patients. Therefore, while the authors have done everything possible to control for confounds, and have really done an excellent job given what they had to work with, we still don’t know if what is being measured is heart disease, or whether it is anxiety that causes people to think they have heart problems (and which doctors may dignify by telling patients that they have a “racing heart”, “heart palpitations”, whatever). To their credit, the authors do point out that their measure is validated against the National Health Interview Survey (which is validated against medical charts), but I’d like to know what goes into that validation procedure.
Length of main text: 4 journal pages
Readability: 7.5/10