Table 3.

Estimates, standard errors, and variables associated with postoperative diplopia in multivariate logistic regression analysis using the 70% dataset.

	Estimate	SE	aOR	95% CI
Intercept	− 3.15	0.98	–	–
Age	0.02	0.01	1.02	0.99–1.05
Globe/other soft tissue repair	− 0.87	0.67	0.42	0.11–1.56
Preoperative periorbital ecchymosis/swelling	− 1.25	0.47	0.29	0.11–0.72
Medial wall fracture	0.69	0.46	2.00	0.82–4.87
Preoperative enophthalmos	0.77	0.47	2.16	0.86–5.42
Fracture defect > 2 cm2 or displacement > 3 mm	1.27	0.60	3.55	1.11–11.38
Orbital fracture repair > 2 weeks after injury	0.54	0.47	1.71	0.68–4.28
C-statistic	0.804
Hosmer–Lemeshow	0.199

Significant 95% confidence intervals are bolded.

Reference groups were as follows: no globe or other soft tissue repair, no preoperative periorbital ecchymosis/swelling, no medial wall fracture, no preoperative enophthalmos, mildly severe fractures (< 2 cm² defect or < 3 mm displacement), orbital fracture repair < 2 weeks after injury.

SE standard error, aOR adjusted odds ratio, CI confidence interval, C-statistic concordance statistic or the area under the receiver operating curve.