This was a descriptive study of N. meningitidis carriage, with cross-sectional data obtained at the four sampling occasions and with longitudinal data for students participating more than once. Logistic regression was used to adjust for possible confounders in the analysis of associations for risk factors and carriage. The effect of individuals participating more than once was balanced by using logistic regression with generalised estimating equations and with unstructured correlation structure between sampling occasions. Both unadjusted and adjusted analyses were performed to evaluate the variables' independent associations with carriage. All risk factors for carriage were analysed as categorical variables; however, age was analysed both as a continuous variable using odds ratio (OR) per year and in three categories (≤22, 23–29, ≥30 years). In addition, the frequencies of a number of habits (cigarette smoking, use of Swedish snus, alcohol drinking and attending parties, pubs or clubs) were analysed as linear trend (OR per category) in relation to carriage. The statistical calculations were performed using version 17 of STATA (STATACorp LLC, College Station, TX, USA). Two-sided P values of <0.05 were considered to be statistically significant.
Free full text: Click here