Literature Review: The Physiology of Loneliness

A. C. Applebee and A. D. Masters

The Numbers That Should Have Changed Everything

In 2010, Julianne Holt-Lunstad and her colleagues at Brigham Young University published a meta-analysis in PLoS Medicine that should have rewritten public health policy overnight. They synthesized data from 148 studies involving 308,849 participants tracked over an average of 7.5 years. The finding: people with stronger social relationships had a 50% increased likelihood of survival compared to those with weaker social connections. Flipped around, social isolation increased mortality risk by 26%. Loneliness -- the subjective feeling of being disconnected -- increased it by 26%. Living alone pushed it to 32%.

To put that in context, the effect size was comparable to quitting smoking. Holt-Lunstad's team made the comparison that inadequate social connection carried a mortality risk roughly equivalent to smoking 15 cigarettes a day. That comparison caught fire in the press and has been repeated in nearly every subsequent article on loneliness. It is worth noting that this is a rough equivalence drawn from comparing effect sizes across different meta-analyses, not a precise biochemical calculation. The point was never mathematical exactness -- it was to convey scale. Loneliness is not a soft problem. It kills at rates we associate with the hardest public health threats we know.

Five years later, Holt-Lunstad published a second meta-analysis in Perspectives on Psychological Science (2015), this time looking at 70 studies with over 3.4 million participants. This one separated social isolation, loneliness, and living alone as distinct risk factors. All three independently predicted early death. Social isolation increased all-cause mortality by 29%. Loneliness by 26%. Living alone by 32%. The consistency across both meta-analyses, across hundreds of studies, across