A new paper by a trio of academics bolsters the case that ‘deaths of despair’ stem in part from weakening social ties including a decline in religious participation.

It shows that mortality from these causes—drug overdoses, alcohol-related illness and suicides—among middle-aged white people in the US stopped falling around 1990, well before the dramatic rise in opioid use. The authors, Tyler Giles of Wellesley, Daniel Hungerman of Notre Dame and Tamar Oostrom of Ohio State, instead looked at religious services for an underlying cause.

They found that states with more religious participation had fewer deaths of despair, and that the faster religious attendance fell in a state, the more such deaths rose.

An earlier paper published in JAMA Psychiatry in 2020, showed that of 110,000 health workers, those who went to religious services were less likely to die from these causes.