The Quarter That Became Two Numbers: Redefining Isolation in Older Adults

In 2020, the National Academies of Sciences, Engineering, and Medicine reported that roughly one quarter of adults aged 65 and older in the United States are socially isolated. The figure has since become the default statistic in policy documents, grant applications, and news coverage on aging and connection. It sounds precise. It is worth asking what, exactly, it was measuring.

The National Academies’ consensus report used social isolation to mean an objective, structural condition: a small network, infrequent contact, few social roles. That is consistent with how researchers have long distinguished isolation from loneliness — the former a property of a person’s network that can be counted, the latter a subjective state that can only be reported. John Cacioppo’s 2008 account of loneliness as an aversive internal signal, evolved to motivate reconnection, rests on exactly this distinction: isolation is what a network looks like from outside; loneliness is what disconnection feels like from inside. The two overlap only partially. A person can have a small network and not feel lonely. A person embedded in a large one can feel intensely alone.

The National Academies’ quarter-of-older-adults figure was, by design, about the first thing, not the second. That distinction held for two years as a matter of definition without much consequence for how the number got used. Then, in 2022, the American Heart Association’s scientific statement on isolation and cardiovascular and brain health did something the title alone signals: it treated “objective” and “perceived” social isolation as two separate variables requiring separate measurement, rather than a single condition with two names. That is a small change in wording and a real change in what counts as evidence.

What the split actually changed

The practical consequence shows up in the risk estimates the AHA statement reports. Isolation and loneliness together were associated with roughly a 30% increased risk of heart attack, stroke, or death from either — specifically a 29% increased risk of heart attack or death from heart disease, and a 32% increased risk of stroke. Those figures come from studies that measured isolation and loneliness as related but distinct exposures, not as interchangeable proxies for a single “disconnection” construct.

This mirrors what Julianne Holt-Lunstad’s 2015 meta-analysis in Perspectives on Psychological Science had already shown at the level of general mortality: social isolation carried an odds ratio of 1.29, loneliness 1.26, and living alone 1.32, for early mortality — three numbers, not one, from constructs that correlate with each other but do not reduce to each other. Living alone is the easiest of the three to measure and the least like the others conceptually; it is a household fact, not a network property or a feeling, yet it carried a comparable and in this analysis slightly larger effect size. A 2010 meta-analysis by the same author, covering 308,849 participants across 148 studies, had already established that stronger social relationships broadly defined were associated with a 50% increased likelihood of survival — a figure large enough that it invited exactly the kind of conflation the later work tried to undo, by folding several distinct measures of “relationship strength” into one number.

The AHA statement’s contribution was to make the separation a requirement rather than a footnote, at least for cardiovascular and brain health research going forward. It also stated plainly what the separation exposes: an absence of intervention evidence. If isolation and loneliness are different exposures with different biological pathways, an intervention that reduces one (say, structural — moving someone into a group living arrangement) is not established to reduce the other (a persistent subjective sense of not being known). The 2020 National Academies report and its 2020 clinical commentary both called for routine assessment of isolation in health care settings; neither could specify, and the AHA statement does not resolve, which instrument that assessment should use, because objective network size and perceived isolation are typically captured by different tools altogether.

Why the AARP figure survives the split better than most

Not every older-adult prevalence figure is vulnerable to this problem. AARP’s 2018 national survey of 3,020 adults aged 45 and older, which found that one in three respondents were lonely, used the 20-item UCLA Loneliness Scale — a standard academic instrument rather than a bespoke survey question. That matters here specifically because the UCLA scale measures the subjective, perceived side of the construct cleanly; it is not attempting to also capture network size, so it is not exposed to the same definitional ambiguity that complicates the National Academies’ isolation figure. The AARP survey’s most useful finding is not the topline third but the gradient underneath it: 33% of respondents who had spoken with a neighbor were lonely, against 61% of those who never had. That is a structural predictor (contact with neighbors) tracked against a subjective outcome (loneliness), reported as two separate things — which is precisely the discipline the AHA statement later asked cardiovascular research to adopt more broadly.

A 2023 review in BMC Public Health surveying the state of loneliness and social isolation research names inconsistent measurement as one of the field’s central barriers to comparison. The isolation-versus-loneliness split described here is one instance of that broader problem, but it is a consequential one, because it sits directly underneath the statistic — one quarter of older adults are isolated — that most policy discussion treats as settled and singular.

What this means for the number everyone cites

None of this means the National Academies’ estimate was wrong. It means the estimate answers a narrower question than its casual use implies. “One quarter of older adults are socially isolated” is a defensible claim about network structure. It is not a claim about how many older adults feel lonely, and it is not a claim about which quarter faces the cardiovascular risk the AHA statement documents, because that risk attaches somewhat differently to the objective and perceived versions of the exposure. Using the figure to argue for interventions aimed at loneliness — companionship visits, phone check-ins — assumes a link between structural isolation and subjective loneliness that the same body of research treats as only partial.

A study designed to close this gap would need to measure both constructs longitudinally in the same older-adult sample, with validated instruments for each, and track cardiovascular and mortality outcomes against both independently rather than a combined score. That is a more expensive study than most social-connection research has been willing to fund. Until it exists, the honest version of the widely cited figure is not “one quarter of older adults are isolated,” but “one quarter met one particular structural definition of isolation, in one national assessment, and that definition is not the same thing several more recent statements are asking researchers to measure.”