The association between self-rated health and high-sensitivity C-reactive protein level: A cross-sectional and 5-year longitudinal study

Abstract

Background: Although self-rated health (SRH) independently predicts mortality, the biological background of this association remains unexplained. This study aimed to examine the association between SRH and serum high-sensitivity C-reactive protein (hsCRP) level. Methods: Subjects were 899 participants aged 35-69 years (237 men and 662 women) in the Daiko Study, part of the Japan Multi-Institutional Collaborative Cohort Study. They were enrolled from 2008 to 2010. Of the subjects, 666 participated in a second survey 5 years later. Lifestyle factors and SRH were assessed using a self-administered questionnaire. Serum hsCRP level was measured using a latex-enhanced immunonephelometric assay. The association between SRH and serum hsCRP level was evaluated using a general linear model with covariates. We further longitudinally investigated whether higher serum hsCRP level at baseline predicts poor SRH after 5 years using an unconditional logistic regression model. Results: A higher serum hsCRP level was significantly associated with poor SRH at baseline after adjusting for covariates (p for trend = 0.023). The age- and sex-adjusted odds ratio and 95% confidence interval (CI) for poor SRH after 5 years was 1.45 (95% CI: 0.76-2.78) for the highest tertile compared with the lowest tertile of serum hsCRP level at baseline with a significant linear trend (p for trend = 0.033), although the risk increase disappeared after adjustment for other covariates. Conclusions: The present study demonstrated that poor SRH is cross-sectionally associated with higher serum hsCRP level. However, the longitudinal data did not support the relationship between serum hsCRP level at baseline and future SRH. Further longitudinal studies that include data on mortality and multiple inflammatory markers are warranted to elucidate the possible role of low-grade inflammation in the association between SRH and mortality risk.