Body height in young adult men and risk of dementia later in adult life

Abstract

This study examined the relationship between body height and dementia and explored the impact of intelligence level, educational attainment, early life environment and familial factors. A total of 666,333 men, 70,608 brothers, and 7388 twin brothers born 1939-1959 and examined at the conscript board were followed in Danish nationwide registers (1969-2016). Cox regression models were applied to analyze the association between body height and dementia. Within-brothers and within-twin pair analyses were conducted to explore the role of shared familial factors including partly shared genetics. In total, 10,599 men were diagnosed with dementia. The association between one z-score difference in body height and dementia (HR: 0.90, 95% CI: 0.89;0.90) was inverse and weakened slightly after adjustment for intelligence test scores and educational level. The associations persisted in within-brother analysis and revealed a stronger, but less precise, point estimate than the cohort analysis of brothers. The twin analysis showed similar, but imprecise estimates.

Keywords: body height; cognitive reserve; cohort study; dementia; epidemiology; global health; human; sibling design.

Figure 1.. Selection of the study population.

Figure 1—figure supplement 1.. Time line of data collection from each of the registers and the total follow-up time.

Foot note: *Anatomic Therapeutical Chemical (ATC) code for dementia (N06D) is also used for patients with Parkinson’s disease, but the treatment is only indicated if the patients have comorbid dementia. **Dementia was identified by international classification of disease (ICD) version 8 codes (290.00–290.99) until the end of 1994 and by ICD10 codes (F00.0-F03.9; G30.0-G30.9) from 1995 and thereafter.

Figure 2.. Hazard Ratios (HRs) and corresponding 95% confidence intervals (95% CIs) for the association between z-score of body height at the entry to adulthood as a cubic spline with four knots and dementia.

A z-score of 0 is the reference value. Stratified by birth cohort and adjusted for conscript board district, educational level and intelligence test scores. Age included as underlying scale of the model. The analysis included the total population of men. Identify cohort-specific values of one z-score in Supplementary file 2.

Figure 2—figure supplement 1.. Distribution of mean numeric difference in body height in cm between brothers (A) and twins (B).

Note: to hide micro data, the value 26 includes all values above for brother and the value 17 includes all values above for twins. In cases where there were more than two brothers with a shared mother, the difference in body height between all brothers were included. For example, for a groups of three brothers, the differences between brother one and two, brother one and three, and brother two and three, respectively, were included.