People with strong social relationships more often outlive their more isolated peers, consistently across dozens of large studies. The associations with heart disease, stroke and dementia have been demonstrated repeatedly, although these are associations rather than proven causes. Reverse causality plays a role but does not fully explain the relationship.
Social connectedness is one of the best-documented factors in ageing research: people with strong social relationships demonstrably outlive their more isolated peers, and that signal is consistent across dozens of large studies. The associations with heart disease, stroke and dementia have also been shown repeatedly, although these are associations, not proven causes. Reverse causality plays a role too -- illness also makes people lonelier -- but that does not fully explain away the relationship.
In practical terms, this means: it is not about the number of contacts, but about quality and the sense of connectedness. What works in practice is contact that is regular, reciprocal and meaningful. That can be a small network, as long as it does not feel empty. Actively maintaining existing relationships carries more weight than collecting new ones, and being physically together generally holds more significance in the research than digital contact, although the two are not interchangeable for everyone.
If you want to make a change: the step that gives the most back is to choose one relationship that has faded and deliberately rebuild it, before you start anything new. New social contact through shared activities (sport, volunteering, a regular group) turns out to be more robust than one-off arrangements, probably because the structure naturally brings its own maintenance with it.
Overview across multiple factors (3 research records, 5 sources). The strength of evidence varies by component -- read the answer for the nuance.