Skip to main content
. Author manuscript; available in PMC: 2014 Oct 1.
Published in final edited form as: Am J Geriatr Psychiatry. 2013 May 6;21(10):946–956. doi: 10.1016/j.jagp.2013.01.069

Table 2.

Preoperative Predictors of Delirium in Final Model

Effect Label Wald Chi-square
Maximum Likelihood
df χ2 P Estimate Standard Error Odds Ratio (95% CI)
Gender Female vs. Male 1 9.51 .002 0.59 0.19 1.81 (1.24–2.63)
Preoperative TICS score <30 vs. ≥ 35 2 19.32 <.0001 1.18 0.27 3.25 (1.90–5.54)
30–35 vs. ≥ 35 0.71 0.23 2.03 (1.30–3.18)
History of central nervous system disorder Yes vs. No 1 4.03 .045 0.38 0.19 1.47 (1.01–2.13)
Surgical risk III vs. I–II 1 8.42 .004 0.67 0.23 1.96 (1.24–3.08)
Surgery type Orthopedic/Neurological vs. Other 1 8.99 .003 0.59 0.20 1.81 (1.23–2.66)

TICS = telephone interview of cognitive status. N = 541.

Logistic regression was performed. The goodness of fit test of the final model was satisfied (P = .838) and C statistic of the prediction model is .687 (95% CI = .642–.733). Area under the curve by 10-fold cross-validation is .650.