Adverse childhood circumstances and cognitive function in middle-aged and older Chinese adults: Lower level or faster decline?

Abstract

We examine the long-term relationship between childhood circumstances and cognitive aging. In particular, we differentiate the level of cognitive deficit from the rate of cognitive decline. Applying a linear mixed-effect model to three waves of China Health and Retirement Longitudinal Surveys (CHARLS 2011, 2013, 2015) and matching cognitive outcomes to CHARLS Life History Survey (2014), we find that key domains of childhood circumstances, including family socioeconomic status (SES), neighborhood cohesion, friendship, and health conditions, are significantly associated with both the level of cognitive deficit and the rate of decline. In contrast, childhood neighborhood safety only affects the level of cognitive deficit. Childhood relationship with mother only affects the rate of cognitive decline. The effects of adverse childhood circumstances are generally larger on level of cognitive deficit than on rate of cognitive decline. Moreover, education plays a more important role in mediating the relationships compared to other later-life factors. These findings suggest that exposure to disadvantaged childhood circumstances can exacerbate cognitive deficit as well as cognitive decline over time, which may be partially ameliorated by educational attainment.

Keywords: Childhood circumstances; Cognitive aging; Education; Life course factors.

Fig. 1

Course of cognitive aging with diverse childhood circumstances. Notes: Panel A–J illustrate the diverse course of cognitive aging from age 45 to age 80 with different childhood circumstances, including parental education, parental work status, childhood relationships, childhood neighborhood social environment, and childhood health conditions. The X axis denotes the respondents’ age when their cognitions were assessed. As cognition function is measured longitudinally in CHARLS, each individual may contribute more than one observation to the trend, and his/her cognitive function may reflect in more than one age group depending on the exact time of cognitive assessment. The plotted points in each panel, thus, represent the average level of cognitive deficit, for the ones with particular ages when the cognitive test was conducted, and with particular childhood circumstances. Cognitive deficit is defined as the reversed summary score of five cognitive tests, with higher value indicating greater cognitive deficit; age is specified as the age when the cognitive tests were conducted. All the regression lines are fitted using local linear smoothing.

Fig. 2

The effects of childhood circumstances on cognitive aging among people with lower and higher education. Panel A. Childhood circumstances and level of cognitive deficit by education.Panel B. Childhood circumstances and rate of cognitive decline by education. Notes: Coefficient plots of the childhood circumstances on level of cognitive deficit (Panel A) and rate of decline (Panel B) among people with lower (primary school or below) and higher education level (middle school or above). The cross-equation test is conducted respectively to examine the statistical difference between the coefficients in two linear regressions. P-value is calculated based on Chow/Wald test, showing at the rightmost side of each panel (Only significant results are illustrated, whereas other estimates are available upon request. Statistical significance: ***p < 0.001, **p < 0.01, *p < 0.05).