Table 3.
Risk factors for acquiring Group B Streptococcus by capsular type. Logistic regression model predicting incidence of a capsular type or capsular type grouping, using all other data as reference. 129 female and 128 male freshmen college students living in a single dormitory.
Risk factor | Beta | Std Error | OR | (95% CI) | P value |
---|---|---|---|---|---|
Capsular type la or lb | |||||
Log frequency of sex acts* | 0.172 | 0.220 | 1.2 | (0.77,1.83) | 0.43 |
Number of sex partners previous year | 0.079 | 0.149 | 1.1 | (0.81,1.45) | 0.59 |
Eats fish 2 or more times per week | 1.982 | 0.489 | 7.3 | (2.78,18.90) | <0.001 |
Frequency eats in the dormitory | −0.220 | 0.277 | 0.80 | (0.47,1.38) | 0.43 |
Male gender | −0.100 | 0.509 | 0.90 | (0.33,2.45) | 0.84 |
Capsular type V | |||||
Log frequency of sex acts* | 0.560 | 0.198 | 1.8 | (1.19,2.58) | 0.001 |
Number of sex partners previous year | −0.244 | 0.202 | 0.8 | (0.53,1.16) | 0.23 |
Eats fish 2 or more times per week | 0.538 | 0.659 | 1.7 | (0.47,6.23) | 0.41 |
Frequency eats in the dormitory | 0.786 | 0.520 | 2.2 | (0.79,6.07) | 0.13 |
Male gender | −0.313 | 0.495 | 0.73 | (0.38,1.93) | 0.53 |
All other capsular types | |||||
Log frequency of sex acts* | 0.362 | 0.184 | 1.4 | (1.00,2.06) | 0.05 |
Number of sex partners previous year | 0.328 | 0.117 | 1.4 | (1.1,1.75) | 0.005 |
Eats fish 2 or more times per week | −0.134 | 0.786 | 0.88 | (0.19,4.08) | 0.87 |
Frequency eats in the dormitory | −0.504 | 0.253 | 0.6 | (0.37,0.99) | 0.05 |
Male gender | −0.024 | 0.508 | 0.9 | (0.36,2.64) | 0.96 |
All capsular types, combined | |||||
Log frequency of sex acts* | 0.383 | 0.119 | 1.5 | (1.16,1.85) | 0.001 |
Number of sex partners previous year | 0.142 | 0.082 | 1.2 | (0.98,1.36) | 0.08 |
Eats fish 2 or more times per week | 1.113 | 0.354 | 3.0 | (1.51,6.10) | 0.002 |
Frequency eats in the dormitory | −0.136 | 0.183 | 0.9 | (0.61,1.25) | 0.46 |
Male gender | −0.085 | 0.301 | 0.9 | (0.51,1.66) | 0.78 |
Risk periods calculated assuming no infection with multiple capsular types, so is a slight under-estimate
Log base 2 of maximum (frequency oral sex, vaginal sex), interpreted as the ratio of the odds of GBS incidence for a doubling in the number of sex acts. For example, an OR of 1.7 is interpreted as a 70% increase in odds of GBS with a doubling in the number of sex acts.