The Survival of Spouses Marrying Into Longevity-Enriched Families

TitleThe Survival of Spouses Marrying Into Longevity-Enriched Families
Publication TypeJournal Article
Year of Publication2017
AuthorsPedersen, Jacob K., Irma T. Elo, Nicole Schupf, Thomas T. Perls, Eric Stallard, Anatoliy I. Yashin, and Kaare Christensen
JournalThe Journals of Gerontology Series A: Biological Sciences and Medical Sciences
ISBN Number1758-535X (Electronic)1079-5006 (Linking)
Accession NumberPMID: 27540092
AbstractBackground: Studies of longevity-enriched families are an important tool to gain insight into the mechanisms of exceptionally long and healthy lives. In the Long Life Family Study, the spouses of the members of the longevity-enriched families are often used as a control group. These spouses could be expected to have better health than the background population due to shared family environment with the longevity-enriched family members and due to assortative mating.Methods: A Danish cohort study of 5,363 offspring of long-lived siblings, born 1917–1982, and 4,498 “first spouses” of these offspring. For each offspring and spouse, 10 controls were drawn from a 5% random sample of the Danish population matched on birth year and sex. Mortality was assessed for ages 20–69 years during 1968–2013 based on prospectively collected registry data.Results: During the 45-year follow-up period, 437 offspring deaths and 502 offspring spouse deaths were observed. Compared with the background population, the hazard ratio for male offspring was 0.44 (95% confidence interval [CI]: 0.38–0.50) and for female offspring it was 0.57 (95% CI: 0.49–0.66). For male spouses, the hazard ratio was 0.66 (95% CI: 0.59–0.74), whereas for female spouses it was 0.64 (95% CI: 0.54–0.76). Sensitivity analyses in restricted samples gave similar results.Conclusion: The mortality for ages 20–69 years of spouses marrying into longevity-enriched families is substantially lower than the mortality in the background population, although long-lived siblings participation bias may have contributed to the difference. This finding has implications for the use of spouses as controls in healthy aging and longevity studies, as environmental and/or genetic overmatching may occur.