Association of childhood meat consumption with late-life cognitive function: mediation by adult education and moderation by parental literacy.

Abstract

Background: Results on the associations of childhood nutrition with late-life cognitive function are inconsistent, and little is known about whether family educational resources modify this association or whether adult education lies on the life-course pathway. We examined the associations of childhood meat consumption (CMC), used as an indicator of childhood nutrition, with late-life cognitive function and mild cognitive impairment (MCI). Moreover, we further examined whether these associations varied by sex, age, parental literacy, and socioeconomic status, and whether they were mediated by adult education.

Methods: Guangzhou Biobank Cohort Study (GBCS) participants, who were 50 years old and above with Guangzhou household registration and recruited from 2006 to 2008 were included. CMC was reported as never/about once a year, about once a month, about once a week, or almost daily. Cognitive function was assessed using the Mini-Mental State Examination (MMSE), its four domains, and the Delayed 10-Word Recall Test (DWRT). Information on parental literacy and adult education was also collected. Multivariate linear, logistic regression and mediation analyses were done.

Results: In 8,865 participants aged ≥50 years, after adjustment for sex, age, and parental literacy, compared with participants who ate meat never or about once a year during childhood, those who ate meat about once a month, once a week and almost daily had higher MMSE scores [β (95% CI): 0.33 (0.17-0.49), 0.83 (0.68-0.99) and 1.13 (0.93-1.32)] and lower odds of MCI [OR (95% CI): 0.84 (0.71-1.00), 0.55 (0.45-0.66) and 0.31 (0.22-0.42)], respectively. Associations were strongest for orientation. Frequent CMC (≥1/week) was more strongly associated with MMSE score [β (95% CI): 0.73 (0.63-0.84)] among participants with literate parents and lower odds of MCI [OR (95% CI): 0.45 (0.38-0.54)] than among those with illiterate parents. Adult education mediated 40% [95% CI: 34.1%-47.3%] of the association between CMC and MMSE. Consistent results were found using DWRT.

Conclusion: More frequent CMC was associated with better late-life cognitive function and lower MCI prevalence, particularly among participants with literate parents. Adult education accounted for a substantial proportion of the association. These results provide new evidence that improving early-life nutrition and promoting educational attainment during early and mid-life may contribute to delaying and preventing cognitive impairment in later adulthood.