Skip to main content
. Author manuscript; available in PMC: 2013 May 3.
Published in final edited form as: Arch Pediatr Adolesc Med. 2011 Jan;165(1):61–67. doi: 10.1001/archpediatrics.2010.250

Table 2. Multivariate Analyses.

High-risk predictors for Lyme disease among patients with facial palsy for the three models.

Characteristic Odds Ratio 95% Confidence Intervals P value
Clinical model (AUC* = 0.71)

Headache 4.4 2.2 to 7.5 <0.0001
Fever 3.3 1.6 to 7.1 0.0017

Epidemiological model (AUC=0.84)

Lyme season** 25 8.6 to 107 <0.0001
High-risk location*** 20 7.4 to 68 <0.0001

Contextualized Model (AUC=0.89)

Fever 3.9 1.5 to 11 0.0071
Headache 2.7 1.3 to 5.8 0.0095
Lyme season** 25 8.3 to 113 <0.0001
High-risk location*** 18 6.5 to 69 <0.0001
*

AUC: Area under Receiver Operator Characteristic Curve

**

Lyme season = June to November

***

High risk location: 3 year average Lyme incidence > 4/100,000 in county of residence.