The advice shows up in almost every guide to healthy aging: stay social, keep making friends, and your body and mind will benefit. It’s a comforting message, and it isn’t baseless. Older adults with more social ties do tend to show better outcomes on measures like inflammation and blood pressure. But a large new analysis of one of the most detailed long-running datasets on American aging suggests the arrow connecting friendship and health may run more strongly in the opposite direction than most people assume.
What the study actually did
Psychologists William Chopik, Hyewon Yang and Richard Lucas used data from the Health and Retirement Study, a nationally representative panel that has followed the same Americans for decades, to track 23,248 middle-aged and older adults over sixteen years. At repeated points across that span, participants reported on their friendships, how many they had, how often they were in touch, and how good the relationships felt, alongside measures of their own health: self-rated physical health, depressive symptoms, and cognitive functioning.
Rather than simply correlating friendship and health at a single point in time, the researchers used statistical models built specifically to test order: which one, on average, tends to shift first, health or friendship. That kind of model is considerably more informative than a one-time survey showing that sociable older adults also tend to be healthier ones, because it can separate a person’s stable baseline (some people are just consistently more sociable, or consistently healthier, across their whole adult life) from what actually moves within their own life from one wave to the next. It’s the second kind of movement, not the stable baseline, that speaks to which factor tends to lead and which tends to follow.
The Artful Age
A weekly letter on aging well, family across generations, and the creative life after the kids leave home.
The Health and Retirement Study surveys Americans aged fifty and older, so the sample skews middle-aged at its youngest and well into the seventies and eighties by its oldest waves, tracked repeatedly as the same people age through the panel. The finding isn’t limited to the oldest respondents specifically, but it lands squarely on the population usually addressed by the “friendship keeps you healthy” advice: people well past midlife, for whom both health and friendship networks are already starting to shift.
The advice has always run one direction: get out, make friends, and your health will follow. Sixteen years of data suggest the arrow more often points the other way, that staying well enough is what lets the friendships hold.
Health moved first, more often than friendship did
Across the sixteen-year span, the researchers found more consistent evidence for health predicting later friendship characteristics than for friendship predicting later health. In other words, a person’s physical health, mood, or cognitive standing in one wave was a better predictor of their friendship network several years later than their friendship network was of their subsequent health.
The study’s authors were careful to note that the effects running in either direction were modest in size. This isn’t a finding that friendship is irrelevant to health, only that the far more commonly repeated causal story, make friends and your health will follow, has noticeably weaker support in this dataset than its mirror image.
What plausibly explains the direction
The study doesn’t hand readers a single mechanism, and it would be overreaching to invent one on its behalf. But the direction it found lines up with something fairly intuitive once it’s pointed out: health has a way of shaping who is available to have friends. Chronic pain, hearing loss, reduced mobility, low energy from depression, all of these make it harder to initiate a coffee, accept an invitation, or keep a standing weekly call going, regardless of how much a person values the relationship.
Friendship, from this angle, becomes easier or harder to sustain depending on how much capacity health leaves for it, more than it’s something a person simply chooses more or less of.
What this doesn’t mean
This isn’t an argument against friendship, or a reason to stop encouraging older adults to stay connected. What it complicates is the specific advice to treat friendship as a lever for health, as though making more friends functions like a supplement an older adult could add to improve their numbers. The sixteen-year data points toward a different sequence: protecting and managing one’s own health may do more to protect the friendships that follow than the other way around. That’s a subtler message than “get out more,” and a harder one to put on a poster, but it’s closer to what this particular dataset actually supports.
What it might change for adult children paying attention to an aging parent
For anyone whose real concern isn’t the research literature but a specific aging parent who seems to be seeing friends less than they used to, this reordering suggests a different first question. The instinct is often to nudge a parent toward more social activity directly: join a club, call an old friend, get out of the house. This study suggests it may be worth checking the health side of the equation first, whether pain, hearing, mobility, sleep, or mood are quietly making it harder to keep up with the friendships that are already there, rather than assuming the friendships themselves are what need fixing. Addressing the harder-to-see health friction may do more for a parent’s social world than any direct push toward socializing would.
This isn’t a diagnosis, and it’s no substitute for an actual conversation with a doctor about what’s changed. But it’s a reasonable reframe: if a parent’s world seems to be shrinking, the friendships may be a symptom worth noticing rather than the first thing to fix.