Suicide, age and poison

At the end of the 19th Century, the ground-breaking sociologist Émile Durkheim made an important discovery: across Europe, Protestant regions had a higher suicide rate that Catholic regions. This, he said, was because Catholicism created more integrated societies. In today’s parlance, Catholicism generates more social capital.

Since then many studies reinforced this theory, showing that Catholicism, and indeed religion in general, seems to protect against suicide. Unfortunately, almost all these studies have been flawed – most often because they looked at average suicide rates and average religious beliefs across particular societies. They didn’t look at the individual characteristics of those people who commit suicide.

Three new studies have addressed this problem. Each of them them takes advantage of new data to explore in some detail the link between religion and reduced suicide.

I’m going to write them up in separate posts, as they all tell different aspects of the story. What they do have in common, though, is that they all show that that the story isn’t quite as straightforward as Durkheim believed!

Matthias Egger, at the University of Bern in Switzerland, has cleverly linked census data to death records – not at all as straightforward as you might imagine. What that gives, for the first time, is a large database with reliable records of individual’s religious affiliation in the last few years before they took their life.

What they found was that, as Durkeheim found when looking at Swiss data a century earlier, Catholics had the lowest suicide rate and Protestants higher. What’s more, Egger found that the unaffiliated had the highest of all.

But there was more to this story than meets the eye. One thing that jumped out was that the gap was much bigger for older people. At ages 35-44, there was essentially no difference. The gap grows gradually with age: in the oldest group (aged 85-94), Protestants are twice as likely as Catholics to commit suicide, and the unaffiliated four times as likely.

And Egger found something else. Strangely enough, the effect was particularly strong for death by poisoning.

That’s a perplexing result, until you remember that Switzerland is one of the few countries where assisted suicide is legal (so long as the motive is not selfish). There are several societies in Switzerland that provide assisted dying, with the usual method being an injection of barbiturates.

On the death record, that’s recorded as a death by poisoning.

So what we’re seeing here is driven in large part by differences in attitudes towards assisted dying. Elderly Catholics, who see suicide as a sin, prefer a natural death. Elderly Protestants and, especially, the unaffiliated, have a different view.

The unaffiliated see assisted dying as an acceptable way to deal with terminal, debilitating, and often painful illness. As a result, the suicide rate among the very old is higher.

That’s not to say that Durkheim was wrong about religion. Social integration is important in reducing suicide, and that may well have contributed to the differences seen. Egger found that married people, and those living with others, also had lower suicide rates. But these data couldn’t show that religion affected social integration.

But what Egger has shown is that the relationship between religion and suicide is more complex than sometimes assumed. Just how complex is the topic of the next post, which looks at suicide rates in different cultures around the world.

Spoerri A, Zwahlen M, Bopp M, Gutzwiller F, Egger M, & for the Swiss National Cohort Study (2010). Religion and assisted and non-assisted suicide in Switzerland: National Cohort Study. International journal of epidemiology PMID: 20841328

This article by Tom Rees was first published on Epiphenom. It is licensed under Creative Commons.