How to have better relationships, according to the longest study of happiness

A study begun in 1938 keeps returning one answer about what makes a good life — and a handful of modern experiments show, in minutes rather than years, how to act on it.

Sunlight crossing a quiet woodland path beneath orange autumn leaves

What the longest study of happiness keeps finding

In 1938, researchers at Harvard began following two very different groups of young men — 268 Harvard sophomores and 456 boys from Boston's poorest neighborhoods — intending to learn what makes a life go well. The study never stopped. It outlived its founders, absorbed the men's medical records and marriages and careers, and now follows roughly 1,300 of their children. It is the longest-running study of adult life ever conducted, and its current director, psychiatrist Robert Waldinger, summarizes almost nine decades of data in one unfashionable sentence: the good life is built on good relationships.

The precision behind that sentence is what makes it worth keeping. In his 2015 TED talk on the study, Waldinger reported that how satisfied the men were in their relationships at age 50 was a better predictor of their physical health at 80 than their midlife cholesterol levels. Harvard's own editorial summary of the research, written with Waldinger and the study's former director George Vaillant, puts it just as plainly: close relationships, more than money or fame, are what keep people happy across a lifetime — and what delay mental and physical decline.

This is not a self-help sentiment. It is the study's most durable empirical result, and it reframes the question. How to have better relationships is not a soft-skills topic adjacent to health and happiness; on this evidence, it is the main event.

“Good relationships keep us happier and healthier. Period.”

Robert Waldinger · TED, 2015

[1][2]

Connection is closer to a vital sign than a luxury

The Harvard finding would matter less if it stood alone. It does not. In 2010, Julianne Holt-Lunstad and colleagues pooled 148 studies covering 308,849 people followed for an average of seven and a half years. People with stronger social relationships had a 50% greater likelihood of survival over the follow-up period (an odds ratio of 1.50) — an effect the authors found comparable to quitting smoking, and larger than the effects associated with obesity or physical inactivity.

That meta-analysis has since crossed from journals into public policy. The 2023 U.S. Surgeon General's advisory on loneliness — which reports that about half of American adults experience it — draws on the same data to conclude that the mortality impact of social disconnection is comparable to smoking up to 15 cigarettes a day. Waldinger, in the Harvard Gazette, is blunter still: “Loneliness kills. It's as powerful as smoking or alcoholism.”

None of this means a rich social life is medicine, and no honest writer should promise that it is. What the evidence does support is a change of posture: tending your relationships belongs on the same shelf as sleep and movement — a practice, not an afterthought for whatever energy remains at the end of the week.

[2][3][4]

It is the quality of the bond, not the size of the crowd

The Harvard study is careful on one point that popular summaries flatten: what protects people is not the number of names in their address book but the warmth and security of their closest ties. One elegant demonstration comes from Waldinger and Marc Schulz's 2010 study of 47 married couples in their eighties, who completed daily diaries for eight days. Physical pain usually drags mood down with it — but for the happily partnered, that link was buffered. On high-pain days, their happiness held steady. A good marriage did not remove the pain; it changed what the pain could reach.

The direction of the field has surprised even its stewards. George Vaillant, who directed the study for over three decades, told the Harvard Gazette that when the research began, nobody much cared about empathy or attachment — yet by the end, the data pointed one way: “the key to healthy aging is relationships, relationships, relationships.” What the numbers reward is not a wide circle but a secure one: someone who will answer at two in the morning, and the settled knowledge that they would.

Vaillant watched hundreds of these lives unfold from youth to old age and condensed the whole archive into five words. It is the shortest conclusion in this Journal, and the hardest-earned.

“Happiness is love. Full stop.”

George Vaillant · Triumphs of Experience, 2012

[2][5]

How you argue matters — and so does how you celebrate

If quality is what counts, what does quality look like in daily behavior? John Gottman and Robert Levenson followed couples for 14 years and found a telling asymmetry: negative behavior during conflict — contempt, defensiveness, escalation — predicted divorces that happened early, in roughly the first seven years, while the absence of positive affect predicted the divorces that came later. Their model classified who divorced in their sample with 93% accuracy — a within-sample result, not a crystal ball, but a sharp portrait of the two ways a bond fails: it can burn, or it can quietly cool.

The cooling half of that portrait points somewhere unexpected: the moments when nothing is wrong. Shelly Gable and colleagues studied what they call capitalization — sharing good news — and found that telling someone about a good event boosted daily positive affect beyond the effect of the event itself. More striking, how the listener responded mattered enormously. Partners who responded actively and constructively — enthusiastic, engaged, asking questions — had relationships with higher intimacy and higher daily marital satisfaction. The quiet heresy of the couples literature: the health of a relationship may be written less in how you fight than in how you receive each other's good days.

Gable's framework names four ways to answer good news, and reading them is an uncomfortable mirror. There is the enthusiastic, question-asking response; the quiet, understated “that's nice”; the response that finds the cloud (“more responsibility, though”); and the response that changes the subject entirely. Only the first — active and constructive — tracked with thriving relationships in these studies. The other three are rarely malicious. They are what distraction sounds like, delivered to someone mid-joy.

The practice, then, is almost embarrassingly concrete. When someone you love arrives with good news, put the phone face down and turn toward it. Ask the second question. Let the moment be about them for a full minute longer than feels natural. It is a small courtesy with an evidence base.

[6][7]

The 21-minute repair

Improving a relationship sounds like a renovation. Sometimes it is closer to adjusting a hinge. Eli Finkel and colleagues followed 120 couples across two years and gave half of them a writing exercise: three times a year, for seven minutes, each partner wrote about their most recent conflict from the perspective of a neutral third party who wanted the best for everyone — and how they might carry that perspective into future disagreements. Twenty-one minutes of writing per year, in total.

In the couples who did nothing, marital quality declined across the two years — the typical, well-documented drift. In the couples who wrote, the decline was eliminated. The fights themselves did not disappear; what changed was the meaning the partners could make of them. Perspective, the study suggests, is not a temperament you are issued at birth but a material you can work deliberately — seven quiet minutes at a time, three times a year.

[8]

Gratitude only binds when it is said out loud

Most of us feel more appreciation than we ever deliver. The research suggests that gap is expensive. Nathaniel Lambert and colleagues ran three studies (with samples of 137, 218, and 75 people) and found that expressing gratitude to a partner — beyond merely feeling it — increased the expresser's own perception of the relationship's communal strength. Saying it changes the sayer.

It changes the receiver too. Amie Gordon's multimethod studies at Berkeley found that people who felt appreciated by their partners became more appreciative and responsive in return, were more committed — and were less likely to have broken up nine months later. Appreciation, expressed and felt, behaves like a current running both ways: each partner's gratitude makes the other's easier.

The exercise below is the study version, not an app feature: note what someone did for you and why it mattered. Then — this is the part the research insists on — tell them.

[9][10]

Try it now

Three thanks worth delivering

Note three things someone did for you lately, and why they mattered. Your notes stay in this browser — but the studies above suggest the real effect begins when you say them to the person.

A quiet practice, kept small on purpose

Read together, the studies sketch a practice modest enough to survive real life. None of it requires a retreat, a workbook, or a personality transplant — only a standing appointment with the people who matter.

A single daily cue helps the appointment keep itself. Some readers use one of RiseHush's quote reminders as that cue — a line from Marcus Aurelius arriving at six in the evening is a surprisingly effective prompt to send one before the day closes. The words do the remembering; you do the reaching. And if you want the wider evidence that tiny gestures land far better than we predict, the companion piece to this article — on the science of small acts — takes up exactly that question.

  • Receive good news well. Once a day, respond to someone's small win with genuine, questioning enthusiasm.
  • Say one felt gratitude out loud. Specific, named, delivered — not merely journaled.
  • Reappraise one conflict in writing. Seven minutes, from a neutral well-wisher's view, a few times a year.
  • Protect one ritual of contact. A standing call, a shared walk, a weekly table that does not move.

[1][7][8][9]

What is the most important factor in a lasting relationship, according to research?

Warmth and security of the bond, not shared hobbies or the absence of conflict. The Harvard Study of Adult Development found relationship satisfaction at 50 predicted health at 80 better than cholesterol, and Gottman's 14-year study suggests that missing positive affect — not just fighting — quietly predicts later separation in the couples they followed.

How long does it take to actually improve a relationship?

Less than folklore suggests. In Finkel's two-year study of 120 couples, a writing exercise totalling 21 minutes per year — reappraising conflicts from a neutral third party's view — eliminated the typical decline in marital quality. Small, repeated practices outperformed grand gestures.

Do I need a romantic partner to get the health benefits of connection?

No. Holt-Lunstad's meta-analysis of 148 studies measured social relationships broadly — friends, family, community — and found stronger ties linked to a 50% greater likelihood of survival. Close friendships and even everyday acquaintances carry measurable weight.

What is active-constructive responding, and why does it matter?

It is responding to someone's good news with genuine enthusiasm and interest — asking questions, letting the moment expand. In Gable's studies, this style of response predicted higher intimacy and daily relationship satisfaction; how partners received good news mattered remarkably compared with how they handled bad news.

  1. Waldinger, TED, 2015 — relationship satisfaction at age 50 predicted physical health at 80 better than midlife cholesterol levels
  2. Mineo (feat. Waldinger & Vaillant), Harvard Gazette, 2017 — close relationships, more than money or fame, keep people happy and delay mental and physical decline
  3. Holt-Lunstad, Smith & Layton, PLOS Medicine, 2010 — meta-analysis of 148 studies (308,849 people): stronger social relationships linked to 50% greater likelihood of survival (OR 1.50)
  4. Murthy, U.S. Surgeon General advisory, 2023 — mortality impact of social disconnection comparable to smoking up to 15 cigarettes a day; about half of U.S. adults report loneliness
  5. Waldinger & Schulz, Psychology and Aging, 2010 — in 47 octogenarian couples keeping 8-day diaries, marital satisfaction buffered the daily link between pain and negative mood
  6. Gottman & Levenson, Journal of Marriage and Family, 2000 — negative conflict predicted early divorce, absent positive affect later divorce; 93% classification accuracy in their sample
  7. Gable, Reis, Impett & Asher, Journal of Personality and Social Psychology, 2004 — sharing good news boosted daily positive affect beyond the event; active-constructive responses predicted intimacy and daily marital satisfaction
  8. Finkel, Slotter, Luchies, Walton & Gross, Psychological Science, 2013 — a 21-minute-per-year conflict-reappraisal writing exercise eliminated the typical two-year decline in marital quality among 120 couples
  9. Lambert, Clark, Durtschi, Fincham & Graham, Psychological Science, 2010 — expressing gratitude to a partner, beyond feeling it, increased the expresser's perceived communal strength (three studies, Ns = 137, 218, 75)
  10. Gordon, Impett, Kogan, Oveis & Keltner, Journal of Personality and Social Psychology, 2012 — partners who felt appreciated were more responsive and committed, and less likely to have broken up nine months later