Field of Science

Do migraines cause religion?

Stop sniggering there at the back. I'm serious!

This is a study on nearly 25,000 Norwegians conducted over 10 years (the Nord-Trøndelag Health Survey (HUNT)). The purpose of HUNT is to track changes in health in the whole population over several decades - and one of the things they measured in the latest survey round (HUNT 3, conducted 2006-2008) was religious attendance.

After controlling for other factors related to headache (age, gender, educational level and chronic musculoskeletal complaints), they found that those who had headaches in 1995 were more likely to be frequent attenders at religious services in 2006 (by frequent religious attenders, they mean in the Norwegian sense - i.e. at least once a month!).

Now, this was entirely down to migraines (rather than regular headaches), and the effect was quite large. Those who had migraines more than 7 days a month were 50% more likely to be frequent religious service attenders 10 years later than those who were migraine free.

What was really interesting was that there was no relationship to current headache levels. So it's headache in the past, rather than current headache, that's associated with current religious service attendance.

And it was something particular about religion, too. There was no relationship with visiting concerts, cinema and/or theatre.

Now, they didn't measure religious service attendance in the earlier surveys. So there's a little bit of ambiguity here. But on the whole, it's pretty good data and it fits with some theoretical expectations.

The study's authors point out that other research has found that patients with chronic pain often report that they turn to religion.

Indeed, in the HUNT study itself they have previously found that regular churchgoers had lower blood pressure. And one of the symptoms of raised blood pressure is... you guessed it - headache!

I'm not entirely convinced, however. In particular, why does current headache not link to religious service attendance?

I'd like to see how headache changed among individuals between 1995 and 2006. Some people stopped having headaches - was that linked to frequent religious service attendance?

If not, then why would people with headaches turn to religion?
Erling Tronvik, Torgeir Sørensen, Mattias Linde, Lars Bendtsen, Ville Artto, Katarina Laurell, Mikko Kallela, John-Anker Zwart, & Knut Hagen (2014). The relationship between headache and
religious attendance (the Nord-Trøndelag health study- HUNT) The Journal of Headache and Pain, 15 (1)

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


  1. I have recently been reading some articles by Michael Persinger which link temporal lobe symptoms to religious and mystical experiences. I am wondering if there is a connection between migraine headaches and temporal lobe symptoms, and if migraines might trigger religious/mystical experiences? If this were the case, perhaps this might lead to an increased interest in religious attendance. I have not read any research that confirms this, so I'm just speculating, but it would be interesting to know.

    1. There does seem to be some link between temporal lobe disturbance and mystical experience (including brain damage and epilepsy in the temporal lobes). It seems to create a feeling of 'non-location'. But I doubt this is related to migraines. Migraines do cause visual disturbance, and that can trigger mystical visions (Hildegard von Bingen is the most famous migraine suffering mystic), but it's a different effect to the temporal lobe thing.

  2. I think one useful taxonomy of atheists is those who have empathy for religion and those who don’t. I am a former believer — thus not an atheists. And to add to the data, I am susceptible to migraine Headaches. I wonder if migraine suffering atheists tend to be more empathetic or sympathetic to religion, while “all-religion-is-bad” atheists tend to be “natural” atheists (never having been an adult believer].

    Stats question: When I think of cause and effect, I imagine very long regression formula with tons of coeficients in front of each factor (influence in the cause). So if migraines incline a person to religion (or something causes both migraines and religiosity), I’d like to know how large that influence is (the coefficient) in relation to other coefficients. So, though there may be an influence of the migraine connection, I wonder how strong the influence compared to others. Is there a stat for such a strength of relevance? Which addresses Jayarava's common anecdotal objection -- besides, Jayarava IS religious (a devout Buddhist), though he may think his religion is not really a religion.

    Personally, I don’t think it is the pain that makes them religious, but whatever the brain condition that causes migraines (or chronic pain] set up a religious mind too. Probably like Scott is asking, I think it may be a part of the brain or some neural circuit affected by the disorder and not the perceived pain. Your thoughts?

    BTW, I love the Britishisms:

    (1) “snigger” — had to look that up. Sure enough, a Britishism for the American snicker. You could probably get in trouble using your version in this country, much like I have to be careful to pronounce clearly when I tell people I love “foreign films”.

    [2] “this was entirely down to migraines” — ? “due to” ?

    1. So, in this case the cause and effect is deduced simply because the migraines came before the religion. They didn't look at what other factors predicted later religion - they just controlled for factors that can explain headache. But the effect was pretty big - migraine sufferers were 50% more likely to frequent religious attenders.

      Top ho!

  3. Religion causes everybody headaches!

  4. I've had migraines since 1981. If anything, religion makes them worse.

  5. So, as people give their anecdotal data, it makes us wonder about the strength of the effect. In medicine, we differentiate between statistical significance and clinical significance. I wonder if we should have a similar viewpoint in society. If 51% of migraine sufferers are religious while the 49% non-migraine sufferers are non-religious, is this really socially significant -- probably not, even if it is statistically significant. Thus my question about the huge regression formula with very, very small coefficients.

    Further, I wonder exactly what sort of religiosity migraines are associated with -- here it is church attendance which does not tell us much about many other different religious behaviors -- but it is a good start.

  6. The study appears to have had a tonne of variables and the connection looks to have been identified post-hoc rather than being hypothesised in advance: it sounds to me like significance hunting.

    1. Fair point. Sometimes it seems to me that just about all studies in psychology and social science should really be labelled 'hypothesis generating' rather than bona fide theory testing. It would help if researchers published their protocols in advance of running the study.

    2. Unfortunately you are probably correct. Advance publishing of protocols would be great but its really unlikely to happen. The online null result databanks are a good first step.

  7. What age does religion form? Many brain pathologies including migraines and various mental problems, emerge in the pre-teen years. Time to do a prospective cohort.


Markup Key:
- <b>bold</b> = bold
- <i>italic</i> = italic
- <a href="">FoS</a> = FoS