Loneliness and Skipping Family Gatherings Raise Heart Disease Risk Over 25 Years

For decades, cardiologists have focused on cholesterol, blood pressure, smoking, and diabetes as the levers that determine who will suffer a heart attack or stroke. A new study adds a quieter but potentially powerful factor to that list: the shape and texture of our social lives. In an analysis spanning a quarter of a century, researchers followed nearly 25,000 middle-aged adults in the Swedish city of Malmö and found that specific dimensions of social connection—loneliness, solidarity with relatives, living alone, and participation in cultural and family activities—were measurably linked to the risk of developing cardiovascular disease. The work, published in BMC Medicine, is among the first to dissect social connection into its many components simultaneously and to test how those components interact with one another over such a long follow-up period.

The study drew on the Malmö Diet and Cancer cohort, one of Scandinavia’s largest population-based research resources. Between baseline examinations in the 1990s, 24,863 participants with a mean age of 57.7 years—60.9 percent of them women—completed a self-administered questionnaire covering twenty-five distinct measures of social connection. Rather than treating social life as a single fuzzy variable, the investigators organized these measures into three conceptual components. The quality component captured how relationships feel, including loneliness and the sense of solidarity with relatives, colleagues, a society or club, and the neighborhood. The functional component measured the support people receive, such as help in times of illness or practical assistance. The structural component counted the concrete architecture of social life: who lives with whom, how often people attend gatherings, and which leisure and cultural activities they take part in.

Incident cardiovascular disease was identified through Sweden’s comprehensive national registers, which link each resident’s unique identification number to hospital discharge and cause-of-death records. Over twenty-five years of follow-up, the researchers documented 5,825 new cases of cardiovascular disease. Using Cox proportional hazards models—a statistical framework that estimates how a given characteristic changes the rate of disease onset over time while adjusting for other factors—they calculated hazard ratios and 95 percent confidence intervals for each of the twenty-five social connection measures. The team also applied false discovery rate correction to account for the many comparisons performed, a safeguard against spurious findings that many earlier studies in this field have lacked.

The results paint a nuanced picture. Within the quality component, people who reported feeling lonely often had a 19 percent higher risk of cardiovascular disease than those who never felt lonely, with a hazard ratio of 1.19. A weaker but similar signal emerged for solidarity with relatives: participants reporting no solidarity with family members had an 18 percent higher risk compared with those reporting very strong solidarity. Intriguingly, the quality of ties to colleagues, to a society or club, and to neighbors showed no association with cardiovascular outcomes at all. The benefits of connection, it seems, are not distributed evenly across every relationship in a person’s life; the emotional bonds within the family appear to carry particular weight for cardiovascular health.

The functional component, which asked whether people could count on practical or emotional support when needed, produced no significant associations with cardiovascular disease risk. This null result is itself informative. It suggests that the objective availability of help may matter less for the heart than the subjective experience of belonging—or its absence. Epidemiologists have long debated whether the health effects of social connection flow through stress physiology, health behaviors, or access to care, and the pattern seen here hints that the felt quality of relationships may be more consequential than their instrumental utility.

The structural measures delivered some of the study’s most striking findings. Regular attendance at the theatre or cinema was associated with a 13 percent lower risk of cardiovascular disease, with a hazard ratio of 0.87. Visits to art exhibitions conferred a modest 8 percent reduction, nightclub or dance hall attendance a 7 percent reduction, and family gatherings a 7 percent reduction. On the other side of the ledger, living alone was associated with a 16 percent higher risk. These associations held after adjustment for conventional risk factors, suggesting that the activities themselves—or the social worlds they represent—carry information about cardiovascular risk that standard clinical measures do not capture.

Perhaps the most clinically significant finding emerged when the researchers tested interactions between measures. The elevated risk associated with loneliness was not uniform across the population. Among participants who did not attend family gatherings, loneliness was linked to a 25 percent higher risk of cardiovascular disease, with a hazard ratio of 1.25 and a false discovery rate-adjusted interaction p-value of 0.015. In other words, lonely individuals who also lacked the routine, recurring contact that family gatherings provide formed a distinct high-risk group. This interaction suggests that the two factors are not simply additive; the absence of structured family contact appears to amplify the cardiovascular toll of loneliness, identifying a population that standard screening might otherwise overlook.

The biological plausibility of these associations rests on well-studied pathways. Chronic loneliness and social disconnection are known to activate the hypothalamic-pituitary-adrenal axis and the sympathetic nervous system, sustaining elevated levels of cortisol and catecholamines that promote hypertension, insulin resistance, and inflammatory processes central to atherosclerosis. Social isolation is also consistently associated with poorer health behaviors—lower physical activity, less adherence to medication, higher rates of smoking and unhealthy eating—and with reduced likelihood of seeking timely medical care. Cultural activities such as theatre visits may operate through additional channels, including cognitive stimulation, positive affect, and stress reduction, mechanisms that have been explored in the growing literature on arts and health.

The study’s strengths are considerable. Its sheer scale—nearly 25,000 participants and almost 6,000 cardiovascular events over 25 years—provides statistical power that few studies of social connection can match. The use of national registers minimizes loss to follow-up and avoids the self-report bias that plagues outcome ascertainment. The simultaneous evaluation of twenty-five measures across three conceptual domains, combined with formal interaction testing and false discovery rate correction, addresses the fragmentation that has characterized earlier research, which typically examined isolation or loneliness in isolation from the broader architecture of social life. The population-based design of the Malmö Diet and Cancer cohort, with written informed consent from all participants and approval from the Ethical Committee at the Medical Faculty of Lund University, further strengthens the generalizability of the findings within comparable settings.

Certain limitations nonetheless warrant caution. The social connection measures were self-reported at a single baseline assessment, and social lives change over decades; some participants may have become lonelier or more isolated after enrollment. Residual confounding by socioeconomic position, mental health, or subclinical disease cannot be excluded, and the cohort, drawn from a Swedish urban population in the 1990s, may not translate directly to other cultural contexts where family structures and leisure patterns differ. The associations, while statistically robust, are modest in magnitude compared with established risk factors such as smoking or hypertension. Even so, the identification of lonely individuals who do not attend family gatherings as a high-risk group offers a concrete, testable target for prevention. If future trials confirm that strengthening family contact or reducing loneliness lowers cardiovascular risk, the humble family gathering may earn a place alongside diet and exercise in the cardiologist’s prescription—a reminder that the heart keeps score not only of what we eat, but of who we see.

Subject of Research: The association between multiple dimensions of social connection and long-term cardiovascular disease risk

Article Title: Social connections and risk of cardiovascular disease: a 25-year population-based cohort study

Article References: Du, Y., Hu, X., Lindström, M., Borné, Y., & Sonestedt, E. (2026). Social connections and risk of cardiovascular disease: a 25-year population-based cohort study. BMC Medicine. https://doi.org/10.1186/s12916-026-05230-y

Image Credits: AI Generated

DOI: 10.1186/s12916-026-05230-y

Keywords: cardiovascular disease, loneliness, social connection, cohort study, epidemiology, family gatherings, living alone, solidarity, cultural activities, Malmö Diet and Cancer cohort, hazard ratio, public health