The impact of early life factors on cognitive function in old age: The Hordaland Health Study (HUSK).
Skogen JC., Øverland S., Smith AD., Mykletun A., Stewart R.
BACKGROUND: Previous studies have shown that adverse conditions during fetal and early life are associated with lower performance on neurocognitive tests in childhood, adolescence and adult life. There is, however, a paucity in studies investigating these associations into old age. The aim was to investigate the impact of early life factors on cognitive function in old age by taking advantage of the potential for a linkage between a community survey and historical birth records. METHODS: A historical cohort study employing a linkage between a community survey of people aged 72-74 years with the participants' birth records (n=346). Early life factors included anthropometric measures taken at birth, birth complications, parental socioeconomic status, and maternal health status. The main outcome was a z-scored composite cognitive score, based on test scores from Kendrick Object Learning Test, Trail Making Test A, a modified version of the Digit Symbol Test, Block Design, a modified version of Mini-Mental State Examination and an abridged version of the Controlled Oral Word Association Test (COWAT). The separate cognitive tests were also individually analysed in relation to measures identified at birth. RESULTS: Higher parental socioeconomic status (SES; based on father's occupation) was associated with a higher value on the composite cognitive score (by 0.25 SD, p=0.0146) and higher Digit Symbol and Trail Making Test A performance. Higher head circumference at birth was associated with higher COWAT and Trail Making Test A performance. Both higher parental SES and head circumference at birth predicted cognitive function in old age independently of each other. There were no other consistent associations. CONCLUSIONS: In general we found little evidence for a substantial role of early life factors on late-life cognitive function. However, there was some evidence for an association with parental SES status and head circumference on certain cognitive domains.