The Science of Religion Part II: Mental Health

Religious practice is overwhelmingly beneficial for well-being

As you might have heard, we’re in the middle of a mental health crisis of somewhat gargantuan proportions. The mental health of young Americans, especially girls, has plummeted so dramatically that something like 60% of all college students are now diagnosed with psychological disorders. More than 200,000 people perish each year in the United States from “deaths of despair,” including suicide, alcoholism, and drug addiction or overdose. These are gold-rush times for professional counseling and therapy, but astute observers wonder whether these treatments are enough — or even if a therapy-drenched culture might sometimes contribute to the problem. One study, for example, recently found that therapy-based trainings made teenagers worse off and damaged their relationships with their parents.

This isn’t to knock therapy, but to point out that mental health is a big topic these days, and no one seems to really know how to solve the problem. The crisis isn’t hitting everyone equally, though. All things being equal, people who practice a religious tradition enjoy better mental health than those who don’t. Of course, there are all kinds of caveats to this statement. For instance, being religious might actually increase the risk of depression in multireligious societies like Japan. Some research suggests that religious practice seems to protect against depression but not necessarily anxiety.

But overall, religious practice is pretty consistently related to fewer mental health disorders. In my last post on the scientific study of religion, I surveyed the growing evidence that religion is helpful for getting in-group cooperation off the ground. This time, I’ll focus on the even more well-established relationship between religiosity and mental health, which in many cases spans different countries, populations, and religious groups, from Christians to Muslims and Hindus.

Small Effect, Big Consequences

The relationship between religion and mental health isn’t especially strong, statistically speaking — usually about r = 0.1 across studies, meaning that religiousness only explains about 1% of total variance in mental health.¹ But at a population level, even small effects can have big consequences. For example, a recent analysis suggested that the mental health costs of declining religious service attendance explained a full 40% of the increase in suicide rates among Americans between 1999 and 2014.

This particular (grim) finding highlights a couple key points about religion and mental health:

1. It’s public service attendance, such as going to church or temple, that boosts mental well-being.

It’s not not mere religious affiliation (belonging nominally to a religious group) or private belief, which often doesn’t track with mental health at all.

The fact that private beliefs on their own don’t help much with mental health may have to do with what psychologists call “religious coping,” or using spirituality and religion for dealing with stress. Positive religious coping means relying on a secure connection with God or other spiritual supports, while negative religious coping might mean blaming God for one’s  problems or ruminating on sin. Unsurprisingly, negative religious coping is strongly linked with poorer mental (and physical) health.

Belief in God, then, can nourish mental health or sap it, depending on how we approach it. But religious practice — participating in the rituals of one’s religious community — more consistently encourages well-being.

2. Religiosity raises the mental-health floor, not the ceiling.

Faith is far better at helping us skirt around our worst mental-health pits than it is at adding bonus points to our well-being. In particular, religiosity seems really, really good at protecting people against suicide. In enormous longitudinal study of nearly 90,000 subjects,ª researchers led by Harvard epidemiologist Tyler VanderWeele found that women who regularly attended religious services were more than five times less likely to die by suicide at any given time than those who never attended. Another very large (20,000-subject) American study of both men and women found similar results: frequent religious attenders were 77% less likely to die by suicide at any given time than non-attenders.

That’s a far cry from a 1% boost in overall mental health.

Interestingly, the researchers found that Catholics who attended weekly were about 20 times less likely to commit suicide at any given time than non-attenders, whereas Protestant weekly attenders were only three times less likely to kill themselves. Of course, the protective effect of Protestant church attendance was still very good — any medical or therapeutic intervention that reliably reduced suicide rates by three times would be front-page news, especially if pharmaceutical companies could monetize it² — but Catholicism seemed even more potent.

This Catholic advantage supports the findings of Émile Durkheim, one of the founders of modern sociology. In 19th century, Durkheim showed that European Catholics were much less likely to kill themselves than Protestants. In his view, Catholic communities provided more social support and robust norms than Protestant communities, which were looser and allowed for — indeed, encouraged — more personal autonomy.

Autonomy can be a great thing, but Durkheim worried that too much of it leads to anomie, a despairing sense of “normlessness” that can cause people to give up on living. VanderWeele’s 21st-century findings suggest that Durkheim was onto something.

Smaller Effects for Other Measures

Compared with suicide, though, religiosity’s effects on overall life satisfaction are quite a bit smaller. One survey of more than 50,000 British respondents found that frequent religious attendance predicted only about 3% to 4% higher scores on self-reported scales of mental health and well-being.

These values correspond remarkably well to those reported by another team, who analyzed a very large longitudinal study of adolescents in the US, finding that frequent attendance at religious services improved life satisfaction and positive affect by around 3% to 4%.³ In both studies, these effects were extremely statistically significant, but comparatively wee.

Another study of different European countries found that Catholic affiliation alone was mildly (very mildly — on the order of about 0.1%) predictive of subjective well-being, whereas Protestant affiliation was not. But church attendance of any kind (Protestant, Catholic, or Orthodox) had a much stronger beneficial effect. Someone who attended religious services every day could expect to be about 4.5% more satisfied with life than someone who never attended.

(This is a silly way to talk about well-being, I’ll admit. But social scientists are often obliged to break life up in clunky and unnatural ways if they want any data to talk about at all.)

This finding dovetails with what we’ve already seen — attendance is more important for mental health than mere affiliation or belief. But even so, 4.5% is a pretty small bump compared with other variables. Someone in the highest income bracket was likely to be 16% more satisfied with life than someone in the lowest bracket. In other words, income was about four times better at making people feel good than religious service attendance, all things considered.⁴

Interestingly, in that same study, “other Christian” affiliation predicted significantly worse well-being. This affiliation probably referred to Eastern Orthodox Christians, the second-largest Christian communion after Catholicism. In a World Values Survey study, another team of researchers backed up this finding by reporting that Orthodox Christians were the least happy and least satisfied of all religious groups.

Don’t be too quick to blame Orthodoxy, though. This finding may reflect the hangover of Communism, since most Orthodox Christians live in Eastern Europe and former Soviet Bloc countries. That is, Orthodox folks probably aren’t less happy than others because their religion is faulty, but because they live in societies still recovering from 70 years of oppressive, totalitarian government and scorched social trust.

Religion Makes You Less Delinquent

Religious belief and practice also benefit well-being in less direct ways. For instance, they reduce the risk of self-destructive behaviors.

In a large study of American adolescents and young adults, researchers found that religious service attendance predicted small (2%-3%) boosts in mental health and protection against depression. (You might be getting bored with this kind of statistic by now, but that only means that these findings are pretty reliable across different studies.) But the protective effects of service attendance against delinquent behaviors were much stronger.

For example, those who attended religious services at least once per week were 15% less likely to smoke cigarettes than those who never attended, and they were 33% less likely to do illegal drugs. Frequent attenders also had a much lower risk (35%) of having had sex early in life (before age 16).

Other studies have found that religiosity on average protects against alcohol abuse, teen delinquency, and participation in violence such as street fighting. Perhaps contrary to stereotype, in politically fraught East Jerusalem, Arab boys and young men with strong religious faith are less likely to partake in political violence than their less-religious peers.

Not surprisingly, many studies find that reduced alcohol use statistically mediate, or explain, the protective effects of religiosity against violence. That is, religious teens and adults get into fewer fights because they just drink less.

Similarly to mental health and depression, personal or private beliefs don’t do the heavy lifting here.

In a study of nearly 18,000 American teens, respondents who were sincerely devoted to their religious faith — including attending public services — were at significantly less risk for drug abuse and committing violence. But those who were “privately religious,” affirming religious beliefs without performing outward rituals or ceremonies, didn’t enjoy nearly the same protection.

As always, religion isn’t a panacea. One meta-analysis of more than 40 different research articles found that religiousness was significantly associated with less interpersonal and sexual violence — but the effect didn’t extend to domestic violence. It looks like religions haven’t quite figured out how to get intimate partners to mistreating each other (at least in the worst cases).

Religion and Self-Control

Another way that religiosity can protect against delinquency and other related problems, such as criminal behavior, is through enhanced self-control.

A longitudinal study by psychologist Steven Pirutinsky examined data from adolescent serious criminal offenders to see how changes in religiosity affected subjects’ risk of reoffending. Pirutinsky found that changes in self-reported religiosity at one timepoint predicted less criminal offending at future timepoints, and that this effect was mediated by gains in self-control.

It’s worth noting that Pirutinsky’s study included objective reoffender data. It wasn’t just that some offenders conveniently reported a sudden religious conversion and then, hoping to make themselves look good to their parole officers, claimed to do less criminal stuff. The religiosity and self-control scales were self-report, but data on who went on to commit more crimes were third-party records.

Other studies support the concept that religion is good for self-control. One recent study found that, although most teens tended to become less religious over the course of adolescence, those who grew more religious or even lost their religiosity more slowly showed improvements in self-regulation later down the line. Another found that college students who were more religious were also less impulsive.

A large literature review of published studies found that, although laboratory manipulations of religiosity had little to no effect on self-control in the moment, real-world participation in religious rituals and institutions did improve self-control over the long term. Once again, it was religious practice and socialization, not interior belief, that did the heavy lifting.

In the next post, we’ll talk about the cognitive psychology of why rituals build self-control. But now, we’ll wrap up by asking the question:

Is Religion Only Beneficial in Religious Societies?

A lot of the studies of religion and mental health come from the United States, which, while less religious than, say, Nigeria, is still more devout than nearly all of Europe. So could the outlier status of American religiosity (an outlier, at least, among developed nations) be driving the effects we’ve been looking at?

Many skeptical researchers have asked this entirely reasonable question because, well, social norms matter. If your society overwhelmingly approves of your religious beliefs, people will probably treat you with more respect, which will redound to your well-being. But if you’re a despised outsider, your mental health will probably suffer.

Accordingly, researchers have found that the positive mental health effects of religion are stronger in more fervently religious countries, whereas they’re smaller or, some researchers claim, nonexistent in secular societies like Germany or the UK.

So is it true? Are the benefits of religion due to “cultural congruence,” or the fit between your faith and your social environment? Is it really just a matter of conformity to social norms?

The answer turns out to be yes and no. But mostly no.

Correlation between religiosity and mental health is strongest in countries where being religious is the norm, particularly in the developing world. But it’s nevertheless still found even in rich, developed countries where religious belief is on the wane.

One recent study recruited more than a hundred different analysts to crunch data from an international dataset with more than 10,000 respondents. This “many analysts” approach was intended to circumvent problems with individual bias among researchers. In the end, all but 3 of the analyst teams found that, indeed, religion benefited subjective well-being, while around two-thirds of the teams found that it partly depended on how religious the surrounding society was.

So being religious is particularly beneficial if you live in a religious society. This is no big surprise when you consider that being a cultural minority is often stressful and unpleasant. If you’re a devout Christian in über-secular Boston or Stockholm, you might not enjoy as much life satisfaction as a recent convert in the Deep South, all of whose neighbors are congratulating her. But the protective effects against suicide and other catastrophic mental health outcomes would largely remain.

Floor. Ceiling. Faith

The data are clear: religiosity — especially public ritual practice, such as Sunday church services or Friday night shabbat — is overwhelmingly good for mental health. The effects for overall mental health, life satisfaction, and happiness aren’t huge, but they’re meaningful. But the protection against suicide and, to a lesser extent, other self-destructive and antisocial behaviors is far stronger and more robust.

Religion only raises the ceiling a little, but it raises the floor a lot.

And the floor matters. In a society where suicides have recently tripled among preteen girls, where hundreds of thousands of people of all ages actively push themselves off the mortal coil each year, a medical intervention that resulted in a threefold (or twentyfold) reduction in suicide would be hailed as a miracle cure.

And of course, these findings flout the critical talking point that religion is a form of mental illness. Nobody really believes this idea anymore, but the research I’ve surveyed here is part of the reason why. The 2000s- and 2010s-era anti-religion crusaders made bold, empirically tractable claims. In particular, they argued that religion is a sort of bug in the human software, an extraneous, harmful tic that we’re better off without. The demographic data on suicide flatly contradict at least the second part of this hypothesis.

In the next post: How humans are built for ritual, and how ritual gives us self-control.

1. The coefficient (r) is the measure of pure correlation between two variables. The square of this coefficient (r²) is the amount of variance in one variable explained by the other variable.

2. Cynical, but true.

3. A longitudinal study is one that collects data over the course of months or years to see how different variables influence each other over time. A correlational study is one that only takes data from a single slice in time. Longitudinal studies are way better than correlational studies at identifying cause-and-effect relationships in the social sciences, and the guys who run them do not let you forget it.

4. Appropriately, another study found that the benefits of religious practice for subjective well-being are substitutable for money. People with higher incomes or who lived in wealthy countries benefited less, relatively speaking, from attending religious services than poorer people. Interestingly, though, the effects of religious faith, while smaller, were not affected by income.