Intergenerational education and premature mortality in Norway, 1990-2019: a nationwide cohort study.

Abstract

Although education is recognised as a determinant of health, how intergenerational patterns of education relate to adult mortality remains poorly understood. The aim of this study is to examine the joint association of parental and offspring education with premature mortality at ages 30-70 years. In this nationwide cohort study, we individually linked registry data provided by Statistics Norway and the Norwegian Institute of Public Health during 1990-2019. We included data on Norwegian residents who contributed person-time data. Education was classified using the International Standard Classification of Education and categorised as low (less than upper secondary), middle (upper secondary), or high (tertiary). Cox models estimated hazard ratios (HRs) with parental × offspring interactions (reference: high offspring-high parental group [ie, those with high education themselves, born to parents with high education]); cumulative mortality from age 30 years to age 70 years was estimated using Kaplan-Meier analysis. We analysed data from 2 849 735 (59%) of 4 840 784 Norwegian residents followed up from Jan 1, 1990, to Dec 31, 2019 (54·7 million person-years; 162 462 deaths). Cumulative mortality by age 70 years was 26·38% [95% CI 26·13-26·63] in male individuals and 16·03% [95% CI 15·81-16·27] in female individuals in the low offspring-low maternal group (n=489 126), versus 9·77% [9·20-10·37] and 6·86% [6·35-7·40] in the high offspring-high maternal group (n=238 232). Estimates for paternal education groups were similar but slightly attenuated. Relative risks were highest among the low offspring education group, particularly when mothers were highly educated (n=23 920; HR 3·16 [95% CI 2·82-3·55] for female individuals and 3·57 [3·26-3·89] for male individuals). Upwardly mobile individuals with high offspring education but low maternal education had substantially lower risks than the stable low offspring-low parental group, although risks did not fully converge with the high offspring-high maternal group (n=219 471; HR 1·05 [95% CI 0·98-1·12] in female individuals and 1·24 [1·17-1·30] in male individuals). Estimates for paternal education followed a similar pattern. Educational trajectories across generations are associated with premature mortality. Downward mobility was less common than upward mobility and was associated with increased mortality risk, whereas upward mobility was associated with reduced risk. None.