The Harvard Study of Adult Development has followed participants for decades and has repeatedly drawn attention to the importance of close relationships. Its findings are often reduced to a single rule that relationships are the strongest predictor of a good life.
The real evidence is more useful and more limited. Relationship quality has been associated with later health and wellbeing in the study, but an observational project cannot isolate social connection from every other influence across a lifetime.
This article is for general information and does not provide medical or mental-health advice.
What the Harvard work found
A Harvard Gazette account of the study reports associations between satisfaction with close relationships in midlife and later measures of health and happiness. The original cohorts were not representative of the whole population, and the project later expanded its participants and questions.
These findings support taking relationships seriously as part of health. They do not demonstrate that connection outranks exercise, diet, education, medical care, or socioeconomic conditions for every person.
Dementia findings remain observational
A separate large ARIC cohort study found that stronger midlife social relationships were associated with a lower incidence of dementia over follow-up. The authors described a potentially protective association and called for further research.
That study did not randomly assign relationships and cannot prove that improving a social score prevents dementia. Social connection may be one relevant factor among many, while health, cognition, opportunity, and personality can also affect a person’s relationships.
Editor’s note: This article was substantially revised on September 14, 2026, to remove a ranking unsupported by the evidence and distinguish association from dementia prevention.
