TABLE 1.
Association between elevated plasma C-reactive protein concentrations and preterm delivery* according to multivariate conditional logistic regression analysis, Project Viva, Massachusetts, 1999–2003
C-reactive protein concentration | All cases (n = 117)
|
Spontaneous delivery (n = 55 cases)
|
Indicated delivery (n = 61 cases)
|
|||
---|---|---|---|---|---|---|
Odds ratio | 95% confidence interval | Odds ratio | 95% confidence interval | Odds ratio | 95% confidence interval | |
≥8 vs. <8 mg/liter† | ||||||
Adjusted for gestational age at blood collection‡ | 2.76 | 1.23, 6.21 | 4.70 | 0.99, 22.37 | 1.64 | 0.58, 4.59 |
Multivariate adjusted§ | 2.55 | 1.05, 6.20 | 4.64 | 0.94, 22.96 | 1.42 | 0.44, 4.61 |
≥12 vs. <12 mg/liter† | ||||||
Adjusted for gestational age at blood collection‡ | 2.95 | 1.05, 8.33 | 7.50 | 0.89, 62.95 | 1.29 | 0.34, 4.88 |
Multivariate adjusted§ | 3.19 | 0.98, 10.32 | 7.24 | 0.85, 61.53 | 1.39 | 0.28, 7.02 |
For one case, information on type of delivery was missing.
A threshold value of 8 mg/liter corresponded to the 90th percentile, and 12 mg/liter corresponded to the 94th percentile in the control group.
Controlled for matching factors of age, race/ethnicity, and cigarette smoking.
Controlled for matching factors of age, race/ethnicity, and cigarette smoking; additionally adjusted for gestational age at blood collection and prepregnancy body mass index.