JAMA Netw Open. 2026 Jul 1;9(7):e2623042. doi: 10.1001/jamanetworkopen.2026.23042.
ABSTRACT
IMPORTANCE: During the COVID-19 pandemic, the major burden of morbidity and mortality was in older age groups. While within-household transmission is well documented, the wider role of social networks for transmission to the older population is less well understood.
OBJECTIVE: To estimate whether children’s birthdays were associated with grandparents’ SARS-CoV-2 infection risk.
DESIGN, SETTING, AND PARTICIPANTS: This nationwide, register-based cohort study included grandparents residing in Denmark with 1 to 5 grandchildren aged 0 to 15 years during the period between February 19, 2020, and February 28, 2022. The timing of birthdays of grandchildren was used as a natural experiment to examine whether birthday-related opportunity for family gatherings was associated with the risk of SARS-CoV-2 infection in grandparents. Family structures were mapped at the individual level. Data were analyzed from June 2024 to December 2026.
EXPOSURES: Birthdays of grandchildren turning 1 to 16 years. A birthday risk window, the period when positive test results could be ascribed to a celebration, was defined as the 7-day time window from 2 to 8 days after the grandchild’s birthday.
MAIN OUTCOMES AND MEASURES: The primary outcome was a positive result on a SARS-CoV-2 test (by polymerase chain reaction or antigen testing) recorded in the national surveillance system. Hazard ratios comparing the hazard of having positive test result for SARS-CoV-2 (outcome) in grandparents in birthday-windows vs grandparents who at the same moment were in nonbirthday periods, were estimated using Cox proportional hazards regression with birthdays as time-varying covariates.
RESULTS: Among 1 106 493 grandparents (median [IQR] age, 67 [60-73] years; 54% female), grandparents had 9.9% (95% CI, 7.9%-12.0%) higher hazard of SARS-CoV-2 infection during birthday risk windows. The hazard varied over time and by variant and was not seen during the period when Alpha dominated. The risk was greater for birthdays of preschool-aged grandchildren (adjusted hazard ratio, 1.15; 95% CI, 1.12-1.18) than during birthdays of school-aged grandchildren (adjusted hazard ratio, 1.07; 95% CI, 1.04-1.09).
CONCLUSIONS AND RELEVANCE: This cohort study of grandparents found that grandchild birthdays, despite official advice not to gather during part of the pandemic, were associated with increased SARS-CoV-2 infection risk among grandparents during much of the pandemic, with magnitude of risk varying over time and by viral variant. These results emphasize the role of culturally important events in intergenerational transmission dynamics, supporting future targeted risk mitigation around small family events.
PMID:42440315 | DOI:10.1001/jamanetworkopen.2026.23042
