Relationships & health: what the research actually says

The case for spending time on relationships is one of the better-evidenced findings in public health. The numbers below are large and they come from the studies most often cited in the field, not press summaries.

The Holt-Lunstad meta-analysis (2010)

In 2010, Julianne Holt-Lunstad, Timothy Smith, and Bradley Layton published a meta-analysis in PLOS Medicine that synthesized 148 studies covering 308,849 participants [1]. The studies followed people over an average of 7.5 years and tracked mortality.

The headline finding: people with stronger social relationships had a 50% greater likelihood of survival over the follow-up period than people with weaker social relationships.

To put that effect size in context, the authors found it to be:

• Comparable to quitting smoking.
• Larger than many well-known mortality risk factors, such as obesity.
• Larger than physical inactivity.

That puts social connection alongside risk factors that public health already takes very seriously.

The Harvard Study of Adult Development

The Harvard Study of Adult Development is one of the longest-running studies of adult life ever conducted. It began in 1938, has since expanded to follow the children of its original participants, and is currently directed by Dr. Robert Waldinger [2].

The study has tracked physical and mental health outcomes alongside detailed records of relationships, careers, and life events. Its central finding has been remarkably consistent across decades:

The clearest message that we get from this 75-year study is this: Good relationships keep us happier and healthier. Period. (Robert Waldinger, 2015 TED talk [2])

Specifically, the study has found that:

• The participants who were most satisfied in their relationships at age 50 were the healthiest at age 80. When the researchers looked back at what they knew about the men at 50, it wasn't their middle-age cholesterol levels that best predicted how they would grow old; it was how satisfied they were in their relationships [2].
• Among participants in their 80s, those in securely attached relationships, who felt they could count on their partner, kept their memories sharp for longer [2].
• Waldinger has summarized the study's view of isolation bluntly: people who are more isolated than they want to be are less happy, their health declines earlier in midlife, and they live shorter lives [2].

Separately from the Harvard study, a 2015 meta-analysis by Holt-Lunstad and colleagues looked at loneliness and isolation directly. After controlling for other factors, loneliness was associated with a 26% increased likelihood of death over the study periods, social isolation with 29%, and living alone with 32% [4].

The most commonly proposed mechanism is stress regulation: close relationships appear to help buffer the body's chronic stress response, and that buffering plausibly compounds across decades.

The U.S. Surgeon General's advisory (2023)

In May 2023, U.S. Surgeon General Vivek Murthy issued an advisory titled Our Epidemic of Loneliness and Isolation [3]. The advisory pulls together the evidence base and makes specific public-health claims:

• Recent surveys found that approximately half of U.S. adults report experiencing measurable levels of loneliness.
• The rate of loneliness among young adults rose every year from 1976 to 2019.
• Poor social relationships, social isolation, and loneliness are associated with a 29% increase in risk of heart disease and a 32% increase in risk of stroke.
• Chronic loneliness and social isolation are associated with an approximately 50% increase in risk of dementia in older adults.

The advisory frames loneliness explicitly as a public-health concern, comparing the mortality risk of insufficient social connection to smoking up to 15 cigarettes a day.

What "stronger relationships" actually means

The studies vary in how they operationalize "social relationships," but the components most consistently associated with better outcomes are:

• Number of close ties. Not strangers and acquaintances. People you can call when something is wrong.
• Perceived support. The subjective feeling that someone has your back.
• Frequency of meaningful contact. Not Instagram likes; conversations, meals, walks.
• Integration into a community. A workplace, a sports team, a church, a friend group, a neighborhood.

The research doesn't hand you a magic number of friends. What it points to is quality and integration, not just headcount: in the 2010 meta-analysis, multidimensional measures of social integration predicted survival more strongly than simple yes/no measures like marital status or living alone [1].

Where remembering comes in

"Quality" and "perceived support" stay abstract until you ask what actually produces them. One concrete answer, from a growing line of research, is being remembered. In 2019, Devin Ray and colleagues found that when someone forgets things about you, you tend to read it as a signal that you don't matter much to them. People are usually understanding, and the impact of a single lapse is typically small, but it is measurable [5]. Forgetting isn't neutral; it quietly lowers felt closeness.

The more useful finding is the flip side. A 2025 study by Anca Pintea and Devin Ray found that deliberately bringing up a specific detail someone shared earlier makes them feel more valued, and amplifies your other attempts to show you care. The catch: participants underused it, even though it worked [6]. That gap, between how much remembered detail matters and how little we use it, is the kind of small friction a memory tool can remove. (More in the science of making people feel remembered.)

What the evidence does not say

It's easy to overclaim from this kind of literature. To stay honest:

• Most of these are observational studies. They show correlation, controlled for many variables, not strict causation. People with stronger relationships may have other shared characteristics.
• The mortality effect sizes are population-level. Your individual outcome depends on many factors. The framing "stronger relationships = better outcomes" is well-supported as a population trend, not as a deterministic personal forecast.
• "Social media use" and "social connection" are not the same construct. The advisory is careful about this; popular summaries often aren't.

Even with those caveats, the convergent evidence across meta-analysis, longitudinal study, and public-health advisory is unusually strong by social-science standards.

The practical takeaway

Most of us already know that close relationships matter. The research mainly tells us how much they matter, and that the effect is on the same scale as the major lifestyle interventions we already take seriously.

If you'd reorganize your week to exercise more or eat better, you should also be reorganizing it to maintain the people you care about. The mechanics are familiar: write things down, follow up, remember birthdays, show up. None of those require an app. All of them are easier with one. And new people count too: if you meet them at conferences or meetups, a simple capture-and-tag routine keeps them from evaporating by Monday.

References

1. Holt-Lunstad, J., Smith, T. B., & Layton, J. B. (2010). Social Relationships and Mortality Risk: A Meta-analytic Review. PLOS Medicine, 7(7), e1000316. PLOS Medicine (open access)
2. Harvard Study of Adult Development. Directed by Robert Waldinger. Official site; overview in the Harvard Gazette, "Over nearly 80 years, Harvard study has been showing how to live a healthy and happy life" (2017). The "Period." quote is from Waldinger's 2015 TED talk, "What makes a good life? Lessons from the longest study on happiness."
3. Office of the U.S. Surgeon General. (2023). Our Epidemic of Loneliness and Isolation: The U.S. Surgeon General's Advisory on the Healing Effects of Social Connection and Community. PDF, HHS.
4. Holt-Lunstad, J., Smith, T. B., Baker, M., Harris, T., & Stephenson, D. (2015). Loneliness and social isolation as risk factors for mortality: A meta-analytic review. Perspectives on Psychological Science, 10(2), 227-237. SAGE
5. Ray, D. G., Gomillion, S., Pintea, A. I., & Hamlin, I. (2019). On being forgotten: Memory and forgetting serve as signals of interpersonal importance. Journal of Personality and Social Psychology, 116(2), 259-276. PubMed
6. Pintea, A. I., & Ray, D. G. (2025). Deliberate memory display can enhance conveyed value. British Journal of Psychology, 116(3), 617-635. DOI (open access)