Skip to main content
. 2013 Jan;131(1):e37–e44. doi: 10.1542/peds.2012-1665

TABLE 3.

Multivariable Analysis (Multinomial Logistic Regression) of Initial Evaluation Location Impact on Preoperative Imaging With Adjustment for Other Variables

Preoperative Imaging Ultrasound Only CT Only CT and Ultrasound
OR 95% CI P OR 95% CI P OR 95% CI P
Initial evaluation
 Community 0.20 0.07–0.58 .003 4.37 1.70–11.19 .002 0.99 0.34–2.98 .99
Age, y 0.88 0.76–1.02 .08 1.03 0.90–1.18 .66 0.88 0.76–1.03 .11
Female gender 1.53 0.52–4.50 .44 1.30 0.47–3.59 .61 4.51 1.47–13.82 .008
Race/ethnicity
 White 2.63 0.59–11.71 .20 5.39 1.31–22.19 .02 4.48 0.75–26.66 .10
 African American 3.29 0.48–22.50 .22 3.26 0.50–21.08 .22 4.42 0.48–40.58 .19
 Other Ref. Ref. Ref.
BMI percentile 0.98 0.96–0.99 .01 0.99 0.97–1.01 .16 0.98 0.96–1.00 .03
Symptom duration, days 1.16 0.74–1.84 .52 1.44 0.93–2.24 .11 1.81 1.15–2.86 .01
WBC count, K/mm3 0.96 0.87–1.06 .39 0.98 0.89–1.07 .65 0.87 0.78–0.97 .01

ORs and 95% CI for imaging (ultrasound only, CT scan only, ultrasound and CT) compared with no imaging (reference [Ref.]) are listed. Reference categories are children’s hospital for site of initial evaluation, male for gender, and other for race/ethnicity. OR is change per unit specified for age, BMI percentile, symptom duration, and WBC count.