In Iran, religion was highly protective, whether religion was measured as the rate of mosque attendance or as whether the individual thought of themselves as a religious person.
In Brazil, going to religious services and personal religiosity were both highly protective. Bizarrely, however, religious affiliation was not. That might be because being Protestant was linked to greater risk, and Catholicism to lower risk. Put the two together, and it may balance out.
In Estonia, suicides were lower in those who were affiliated to a religion, and those who said they were religious. They were also a bit lower in those who
In India, there wasn't much effect of religion at all - a bit lower in those who go to religious services at least occasionally.
Vietnam was similar. Those who went to religious services yearly were less likely to have attempted suicide, but no other measure of religion had any effect.
In Sri Lanka, going to religious services had no protective effect, but subjective religiosity did.
In South Africa, those who go to Church were no less likely to attempt suicide. In fact, those who said they were religious were actually nearly three times more likely to attempt suicide, and those who were affiliated to a religion were an incredible six times more likely!