Association between age at first sexual intercourse and all-cause mortality in adults: two nationwide cohort studies in the USA and UK

Age at first sexual intercourse (AFS) is a life-course marker reflecting an individual’s early social environment, psychological maturity, and behavioral trajectories—one that has been linked to enduring health outcomes across the lifespan. Previous studies have established associations between earlier AFS and adverse health risks, such as sexually transmitted diseases and psychological distress,1,2 while delayed AFS may also hinder the development of emotional, cognitive, and interpersonal skills.3 However, direct evidence linking extreme AFS to all-cause mortality remains scarce. The present study posits that earlier AFS may be associated with high-risk sexual behaviors and mental health issues, whereas delayed AFS may correlate with social isolation—both of which could indirectly impact long-term mortality. Considering this framework, the current study uses data from 2 nationwide cohorts in the USA and UK to characterize the nonlinear association between AFS and all-cause mortality, as well as whether there are gender differences. This retrospective cohort study analyzed data from 2 nationwide cohorts: the National Health and Nutrition Examination Survey (NHANES, 1999-2016, n = 29 343) and the UK Biobank (UKB, 2006-2010, n = 310 619). The exposure was AFS, collected via self-report (minimum age of 9 years in NHANES and 12 years in UKB). Participants with missing, refused, or unknown AFS data were excluded, accounting for 19 840 individuals (21.55%) in NHANES and 67 062 individuals (13.35%) in UKB. The outcome was all-cause mortality: NHANES participants were followed up to 2017 (mean follow-up of 10.8 years), while UKB participants were followed up to 2021 (mean follow-up of 12.4 years). Statistical analyses included descriptive statistics and group comparisons (Rao-Scott χ2 test for NHANES and χ2 test for UKB). Cox proportional hazards regression with restricted cubic splines was used to assess the nonlinear association between AFS and all-cause mortality. For critical confounders (gender, education, and race/ethnicity), participants with missing data were excluded. For other confounders, missing continuous data were imputed using the mean, and missing categorical data were imputed using the mode. Study flowcharts were presented in Figure S1. Analyses were performed using R 4.3.2, with significance set at P < .05.