Dataset

Predictors of low-grade inflammation estimated by multivariable logistic regression analysis.

public-library-of-science,

2014

DOI:10.1371/journal.pone.0088196.t003, Dimensions: 1076179,

Affiliations

Organisations

  1. (1) Aarhus University Hospital, grid.154185.c, Central Denmark Region

Countries

Denmark

Continents

Europe

Description

Predictors of low-grade inflammation (CRP>3 mg/L) in multivariable logistic regression analysis. Obesity: Body Mass Index (BMI). Waist circumference, larger vs. smaller than WHO cut-offs: 80 cm for women and 94 cm for men. The analysis was stratified by sex and, among women, by menopausal status and use of combined oral contraception (OC). Predictors in all four groups included: current smoking (yes/no), age (ten-year increment), meat and fish consumption, and physical activity at work and at leisure. For the analysis of waist circumference, BMI was excluded from the analysis. The calculation of pack years was not possible for 745 participants. Thus, the analysis for pack years was made separately and includes only 14,966 participants. Among women, additionally predictors were childbirth; and among premenopausal women not using OC, the use of progesterone-only pill (POP) and hormone-releasing intrauterine devices (IUD). Among premenopausal women not taking OC, the effects of POPs were tested after excluding users of IUD. Also, the effects of IUD were tested after excluding users of POPs. For the analysis of cumulative tobacco consumption, defined as pack years, a total of 745 participants were excluded. Odds ratio (OR) including 95% confidence interval (CI), p-value (p), numbers (n), not available (NA).

Funders

Research Categories

Main Subject Area

Fields of Research

Licence

CC BY 4.0

External sources

Access at Figshare