Most of us treat choosing a partner as a romantic decision. Occasionally a practical one — money, geography, whether they want children.
Almost nobody treats it as a medical one.
They should. The person you build a life with will co-regulate your nervous system or dysregulate it. They will shape your stress hormones, your immune profile, the quality of your sleep, and — measurably — the speed at which your body ages. Some of their gut bacteria will become yours, though as we’ll see, only under one specific condition.
And the evidence that this matters is not soft. It is some of the most replicated data in epidemiology.
The trouble is that it sits next to a great deal of nonsense. So this piece does two things: it lays out what the research actually shows, and then — in a section most articles skip — it tells you where the evidence is weaker than you have been led to believe.
How relationships affect health and longevity: the strongest evidence
The largest synthesis available pooled 90 prospective cohort studies covering 2,205,199 adults. In the general population, social isolation was associated with a 32% higher risk of dying of any cause (pooled effect 1.32, 95% CI 1.26–1.39). Loneliness — the felt experience rather than the objective state — carried a 14% higher risk (1.14, 95% CI 1.08–1.20).
Isolation also raised cardiovascular mortality risk by 34% and cancer mortality by 24%.[1]
Read from the other direction, an earlier meta-analysis of 148 studies and 308,849 participants found that people with stronger social relationships had roughly a 50% greater likelihood of survival across the follow-up period (odds ratio 1.50, 95% CI 1.42–1.59).[2]
Note the phrasing there, because it gets mangled constantly: that is a 50% greater likelihood of survival, not a 50% reduction in death. The distinction matters if you care about being accurate.
It isn’t only how long. It’s how fast.
There are now blood tests that estimate the pace of your biological aging rather than the total.
GrimAge, a DNA-methylation composite, predicts time-to-death, time-to-coronary-heart-disease and time-to-cancer more strongly than any epigenetic clock before it.[3] DunedinPoAm takes a different approach — it was built by tracking eighteen biomarkers across twelve years in 954 people all born in the same two years, producing something closer to a speedometer than an odometer.[4]
When researchers applied these to 3,647 adults aged 50–100 in the Health and Retirement Study, the people without a partner and without close friends showed an older GrimAge and a faster pace of aging. Among those who did have relationships, lower support — from a spouse, a child, other family, or friends — tracked with older biological age.[5]
One honest caveat, and it’s the kind that usually gets left out: in the models that also adjusted for income and lifestyle, only support from other family members and friends remained significant. The marital-specific finding is not reported as surviving that adjustment.
So the correct sentence is “people without a partner or close friends showed faster biological aging, and some of those associations held after controlling for income and lifestyle” — not “your partner changes how fast you age, independent of lifestyle.”
The strangest finding in this literature
Researchers integrated microbiome sampling into the Wisconsin Longitudinal Study — 94 spouse pairs and 83 sibling pairs. Spouses had more similar gut microbiota, and more bacterial species in common, than siblings did. Siblings, meanwhile, were no more similar to each other than to unrelated strangers. The effect held after accounting for diet.
So far, unsurprising: people who share a kitchen share bacteria.
Except that is not what drove it.
The difference between married couples and unrelated strangers was driven entirely by the couples who described their relationship as close. Couples who described theirs as only “somewhat close” were no more microbially similar to each other than two strangers picked at random.
And married people carried greater microbial diversity and richness than people living alone — with the greatest diversity of all among couples reporting close relationships.[14]
Proximity was not the variable. Closeness was.
I want to be careful here, because this is exactly the kind of finding that gets oversold. It is 94 couples. It is correlational — the paper’s own title says correlate, not cause. And nobody has shown that microbial convergence is what makes close relationships protective.
But sit with the shape of it. Two people can share an address, a bed, a fridge and a bathroom for a decade, and if the relationship is merely cordial, their gut ecosystems stay about as different as those of strangers.
Something about closeness itself — not cohabitation, closeness — appears to travel.
The pattern shows up in other body sites too. Cohabiting partners’ skin microbiomes are similar enough that a machine-learning model correctly paired them 86% of the time from swabs alone, though body site and individual identity remained stronger signals than partnership.[15] And couples’ salivary microbiota similarity tracks with how often, and how recently, they kiss — a controlled experiment using marker bacteria put the transfer at roughly 80 million bacteria per ten-second intimate kiss.[16]
That last one is a good party fact and a small study — 21 couples. Worth noting that a single kiss did not measurably shift the community. It takes frequency.
Where the evidence is weaker than you have been told
This is the section most articles on this subject don’t write. It is also the reason you can trust the rest of this one.
The “married people live 5–12 years longer” claim
You will see this everywhere: married women live five to twelve years longer, married men seven to seventeen. People usually attribute it to the Harvard Study of Adult Development.
I could not trace those figures to any peer-reviewed survival analysis. They appear in institutional summaries and secondary write-ups, without hazard models, life-table derivation, or confidence intervals. That doesn’t make them false—it makes them uncitable, which isn’t enough for a claim about how long you are going to live.
What the Harvard Study genuinely shows is directional and still remarkable: the quality of people’s close relationships in midlife predicted their physical health and cognitive function decades later, more strongly than cholesterol or social class.[6] That is the claim worth making. The specific year counts are not.
The telomere story
Telomeres — the protective caps on chromosomes that shorten as cells divide — are the most popular biological explanation for how relationships “get into the body.”
Some of the findings are real. In 321 middle-aged adults, those married or living with a partner had longer leukocyte telomeres than the widowed, divorced, separated, or never-married, independent of diet, alcohol, exercise, smoking, social support, poverty, and obesity.[7]
But the popular version leaves out this. The largest and best-designed study in this area followed 10,058 Swedish twins prospectively for a median of 9.8 years. Living singly was associated with shorter telomeres and with a 21% higher risk of cardiovascular disease. The researchers then tested whether telomere length explained that risk.
It barely moved the number. The hazard ratio went from 1.21 to 1.20 after adjustment. In the authors’ own words: “LTL appeared to mediate little of the association between marital status and CVD.”[8]
So telomeres travel alongside the effect. They do not carry it. Anyone telling you that your relationship shortens your telomeres and that is why you get sick is telling you a story the best evidence does not support.
The effect sizes are also small — in that first study, a ratio of 1.69 against 1.59 — and telomere measurement is a technically demanding assay rather than a settled readout.
I am including all of this because you deserve to know which parts of a claim are load-bearing. The mortality and epigenetic data are strong. The telomere mechanism is not.
And yet — one telomere finding is worth your attention
In a probability sample of 3,526 adults aged 50 and over, researchers measured salivary telomere length and compared it with marital history.
You would expect the shortest telomeres in people who had been alone their whole lives.
That is not what they found.
The shortest telomeres were in the separated and divorced — shorter than in the continuously married, and shorter than in the never-married. And adults who were currently married but had a history of divorce still showed shorter telomeres than the continuously married or never-married.
This held after adjusting for demographics, socioeconomic variables, neuroticism, smoking, body mass, traumatic life events and other stressors.[9]
Whatever that is measuring, it is not simply being alone.
It is rupture.
Given everything above, hold that loosely as mechanism. But hold it seriously as a pattern — because it points somewhere genuinely useful.
So what should you actually screen for?
Most people screen for someone who does not trigger them.
That target is wrong, and it is also unreachable. The person you love will touch your oldest wounds — willingly or not, knowingly or not. That is not a risk of intimacy. It is the mechanism of it.
Ed Tronick spent decades filming mothers and infants frame by frame, coding every second of interaction. The finding that should have changed how we all talk about love is this: in perfectly healthy pairs, the proportion of time they are genuinely coordinated is small. Mismatch is not the exception. It is the water they swim in.[10]
What separated the pairs that thrived was not fewer ruptures. In Tronick’s words, it was “frequent reparations of interactive errors.” What predicted poor outcomes was “sustained periods of interactive failure.”[11]
Not the rupture. The repair rate.
There is a mechanistic reason this matters more than compatibility. When a partner physically regulates your threat response — the classic demonstration is hand-holding during anticipated discomfort — the size of that effect is dose-dependent on the quality of the relationship.[12] A hand attached to someone with capacity does something measurable. A hand attached to someone without it does considerably less.
You cannot co-regulate with capacity that is not there. That is not a moral judgment about them. It is closer to a stock level.
What to do on Tuesday
Write your terms down. Not a mood board — a list, in three parts.
What you need. Five things you would notice the absence of within a month.
What you will not do again. Three, specific and behavioral. Not “I won’t be with someone emotionally unavailable” — that is a diagnosis you can argue yourself out of at 2 am. Try “I won’t wait more than one conversation for a straight answer about what we are.” You want a sentence a stranger could score.
The repair clause. How soon you need first contact after a rupture. What repair has to include for you — acknowledgment of impact, changed behavior, touch, space first? What you will reliably do on your side. And the sign that this is drift rather than a bump.
Then say them early, out loud, and watch whether actions match words.
There is a reason saying them works beyond simple bravery. When people put a feeling into words, amygdala activity drops and the right ventrolateral prefrontal cortex comes online — the two move in opposite directions.[13] Naming is not just expressive. It is regulatory. Saying the thing calms the system that is afraid to say it.
Which means the anxiety is highest before the sentence exists, and drops once it does.
And if they tell you honestly that they do not have it — some will, generously — the correct response is not devastation. It is thank you for telling me.
They did not withhold it. They did not have it.
I made the list I use for this — what I need, what I won’t do again, and the repair clause. It’s free.
Frequently asked questions
Does being in a relationship actually make you live longer?
The association is robust: across 148 studies covering more than 300,000 people, those with stronger social relationships had about a 50% greater likelihood of survival. What the data does not support is a specific number of extra years — claims that marriage adds five, twelve or seventeen years are not traceable to a peer-reviewed survival analysis. The direction is well established; the precise figures are not.
Is loneliness really as bad for you as smoking?
That comparison is popular and imprecise. What the evidence supports: across 90 cohort studies and 2.2 million adults, loneliness was associated with a 14% higher all-cause mortality risk, and social isolation with 32%. Those are meaningful, well-replicated numbers, and they are better stated directly than converted into a cigarette equivalence.
Can a bad relationship be worse than being single?
The evidence points that way, though carefully. In a sample of over 3,500 older adults, the shortest telomeres were found in the separated and divorced — shorter than in the continuously married and shorter than in the never-married. Relationship disruption, rather than solitude, tracked with the effect.
What should you actually look for in a long-term partner?
Repair capacity. Conflict is guaranteed; what varies is whether someone comes back, how quickly, and whether they can acknowledge impact without collapsing or attacking. Decades of frame-by-frame research on close pairs found that healthy ones are out of sync much of the time — the marker of a good relationship is the rate of repair, not the absence of rupture.
References
- Wang, F.; Gao, Y.; Han, Z.; et al. A Systematic Review and Meta-Analysis of 90 Cohort Studies of Social Isolation, Loneliness and Mortality. Nature Human Behaviour 2023, 7 (8), 1307–1319. DOI: 10.1038/s41562-023-01617-6
- Holt-Lunstad, J.; Smith, T. B.; Layton, J. B. Social Relationships and Mortality Risk: A Meta-Analytic Review. PLoS Medicine 2010, 7 (7), e1000316. DOI: 10.1371/journal.pmed.1000316
- Lu, A. T.; Quach, A.; Wilson, J. G.; et al. DNA Methylation GrimAge Strongly Predicts Lifespan and Healthspan. Aging (Albany NY) 2019, 11 (2), 303–327. DOI: 10.18632/aging.101684
- Belsky, D. W.; Caspi, A.; Arseneault, L.; et al. Quantification of the Pace of Biological Aging in Humans through a Blood Test, the DunedinPoAm DNA Methylation Algorithm. eLife 2020, 9, e54870. DOI: 10.7554/eLife.54870
- Rentscher, K. E.; Klopack, E. T.; Crimmins, E. M.; et al. Social Relationships and Epigenetic Aging in Older Adulthood: Results from the Health and Retirement Study. Brain, Behavior, and Immunity 2023, 114, 349–359. DOI: 10.1016/j.bbi.2023.09.001
- Waldinger, R. J.; Schulz, M. S. The Good Life: Lessons from the World’s Longest Scientific Study of Happiness; Simon & Schuster: New York, 2023.
- Mainous, A. G.; Everett, C. J.; Diaz, V. A.; Baker, R.; Mangino, M.; Codd, V.; Samani, N. J. Leukocyte Telomere Length and Marital Status among Middle-Aged Adults. Age and Ageing 2011, 40 (1), 73–78. DOI: 10.1093/ageing/afq118
- Chen, R.; Zhan, Y.; Pedersen, N.; Fall, K.; Valdimarsdóttir, U. A.; Hägg, S.; Fang, F. Marital Status, Telomere Length and Cardiovascular Disease Risk in a Swedish Prospective Cohort. Heart 2020, 106 (4), 267–272. DOI: 10.1136/heartjnl-2019-315629
- Whisman, M. A.; Robustelli, B. L.; Sbarra, D. A. Marital Disruption Is Associated with Shorter Salivary Telomere Length in a Probability Sample of Older Adults. Social Science & Medicine 2016, 157, 60–67. DOI: 10.1016/j.socscimed.2016.03.029
- Tronick, E. Z.; Cohn, J. F. Infant-Mother Face-to-Face Interaction: Age and Gender Differences in Coordination and the Occurrence of Miscoordination. Child Development 1989, 60 (1), 85–92. PMID: 2702877
- Tronick, E. Z. Emotions and Emotional Communication in Infants. American Psychologist 1989, 44 (2), 112–119. DOI: 10.1037/0003-066X.44.2.112
- Coan, J. A.; Schaefer, H. S.; Davidson, R. J. Lending a Hand: Social Regulation of the Neural Response to Threat. Psychological Science 2006, 17 (12), 1032–1039. DOI: 10.1111/j.1467-9280.2006.01832.x
- Lieberman, M. D.; Eisenberger, N. I.; Crockett, M. J.; Tom, S. M.; Pfeifer, J. H.; Way, B. M. Putting Feelings Into Words: Affect Labeling Disrupts Amygdala Activity in Response to Affective Stimuli. Psychological Science 2007, 18 (5), 421–428. DOI: 10.1111/j.1467-9280.2007.01916.x
- Dill-McFarland, K. A.; Tang, Z.-Z.; Kemis, J. H.; Kerby, R. L.; Chen, G.; Palloni, A.; Sorenson, T.; Rey, F. E.; Herd, P. Close Social Relationships Correlate with Human Gut Microbiota Composition. Scientific Reports 2019, 9 (1), 703. DOI: 10.1038/s41598-018-37298-9
- Ross, A. A.; Doxey, A. C.; Neufeld, J. D. The Skin Microbiome of Cohabiting Couples. mSystems 2017, 2 (4), e00043-17. DOI: 10.1128/mSystems.00043-17
- Kort, R.; Caspers, M.; van de Graaf, A.; van Egmond, W.; Keijser, B.; Roeselers, G. Shaping the Oral Microbiota through Intimate Kissing. Microbiome 2014, 2, 41. DOI: 10.1186/2049-2618-2-41
