Back

Childhood predictors of dementia and mortality: 78 year follow-up of the 1946 British Birth Cohort.

King-Robson, J.; Nicholas, J. M.; James, S.-N.; Groves, J. W.; Chaturvedi, N.; Richards, M.; Schott, J. M.

2026-03-09 epidemiology
10.64898/2026.03.03.26347490 medRxiv
Show abstract

BackgroundEarly-life origins of late-life health are unclear. It is uncertain when in life risk-factors for dementia begin to exert effects. We examined childhood predictors of dementia and mortality in the worlds longest continuously-running birth cohort. MethodsProspective cohort study, 78-year follow-up. Exposures assessed from birth-age 16. Dementia/death identified from national registries. Cox regression identified childhood predictors. FindingsOf 5,046 participants, 121 (2{middle dot}4%) were diagnosed with dementia and 1,243 (24{middle dot}6%) died by age 78. Independent childhood predictors of dementia were low cognition (HR 2{middle dot}06, 95% CI 1{middle dot}31-3{middle dot}24, P=0{middle dot}002), pollution exposure (HR 2{middle dot}05, 1{middle dot}15-3{middle dot}65, P=0{middle dot}02), left-handedness/ambidexterity (HR 1{middle dot}93, 1{middle dot}15-3{middle dot}23, P=0{middle dot}01), sleep disturbance (HR 1{middle dot}78, 1{middle dot}00-3{middle dot}17, P=0{middle dot}049), nocturnal enuresis (HR 1{middle dot}52, 1{middle dot}06-2{middle dot}17, P=0{middle dot}02), and emotional problems (HR 1{middle dot}38, 1{middle dot}12-1{middle dot}69, P=0{middle dot}003). APOE L4 (HR 3{middle dot}38, 1{middle dot}67-6{middle dot}85, P=0{middle dot}001), and BMI (HR 1{middle dot}47, 1{middle dot}17-1{middle dot}84, P<0{middle dot}001) associated with dementia only in females. Independent childhood predictors of mortality were BMI (HR 1{middle dot}07, 1{middle dot}01-1{middle dot}14, P=0{middle dot}02), lower socioeconomic position (HR 1{middle dot}25, 1{middle dot}09-1{middle dot}42, P=0{middle dot}001), sleep disturbance (HR 1{middle dot}27, 1{middle dot}05-1{middle dot}53, P=0{middle dot}02), nocturnal enuresis (HR 1{middle dot}32, 1{middle dot}04-1{middle dot}68, P=0{middle dot}02) emotional problems (HR males:1{middle dot}09, 1{middle dot}00-1{middle dot}18, P=0{middle dot}046, females:1{middle dot}25, 1{middle dot}14-1{middle dot}36, P<0{middle dot}0001), and conduct problems (HR 1{middle dot}13, 1{middle dot}06-1{middle dot}20, P=0{middle dot}0002). Female sex (HR 0{middle dot}67, 0{middle dot}59-0{middle dot}75, P<0{middle dot}001), education (HR 0{middle dot}76, 0{middle dot}67 to 0{middle dot}87, P<0{middle dot}001), and breastfeeding (HR 0{middle dot}93/3-months, 0{middle dot}89-0{middle dot}98, P=0{middle dot}002) associated with lower mortality. InterpretationChildhood factors predict a significant proportion of dementia and early mortality risk. While some associations are genetically determined or may relate to early brain development, the majority are potentially modifiable. FundingMedical Research Council, Alzheimers Research UK, Alzheimers Association, NIHR, UK Dementia Research Institute.

Matching journals

The top 12 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.