Positive association between constipation and mild cognitive impairment in elders: A cross-sectional study

Abstract

This study aimed to examine the association between constipation and mild cognitive impairment (MCI); and further elucidate the possible mechanisms involved. A cross-sectional study was conducted among community-dwelling elders (N = 789) in Nanning, China. Trained research staffs collected detailed information through questionnaires and physical examinations. A Bayesian network model was used to explore the hypothesized causal path. Synergistic effects of constipation with infrequent fruit consumption, inactive physical exercise, or history of stroke were observed in the risks of MCI occurrence. The Bayesian network model analyses showed 3 hypothesized causal-association paths leading to MCI occurrence. Among these, constipation, history of stroke, and years of schooling were directly related to the occurrence of MCI. Years of schooling indirectly affected MCI through infrequent fruit consumption and constipation; or through inactive physical exercises and history of stroke. This study demonstrates a direct association between constipation and increased risks of MCI.

Figure 1.

The respective interactions of constipation with other important influencing factors. The respective interactions of constipation with other important influencing factors included history of stroke (A), inactive physical exercise (B), inactive social activities (C), and infrequent fruit consumption (D). *P < .01, ^#P < .05.

Figure 2.

Bayesian network causal inference of MCI and influencing factors. Bayesian network approach explored the causal-association diagram between MCI and exposure variables, and the importance of exposure variables. The higher magnitude the value is, more important the variable. Meanwhile, the importance of variables was visualized via the gradual changes of color of nodes.