Skip to main content
. 2020 Feb 3;20:40. doi: 10.1186/s12877-020-1449-8

Table 3.

Univariate and multivariate analysis for variables significantly associated with postoperative delirium

Variables n Delirium (n = 29) Crude RR (95% CI) p-value Adjusted RR (95% CI) p-value
Pre-existing dementia, n (%)
 No 207 13 (6.3) 1 1
 Yes 42 16 (38.1) 6.07 (3.16–11.65) < 0.001 3.95 (1.91–8.17) < 0.001
Age ≥ 75 years, n (%)
 No 124 6 (4.8) 1 1
 Yes 125 23 (18.4) 3.80 (1.60–9.03) 0.001 2.54 (1.11–5.80) 0.027
Preoperative hematocrit <30%, n (%)
 No 214 21 (9.8) 1 1
 Yes 35 8 (22.9) 2.33 (1.12–4.84) 0.042 1.53 (0.79–2.96) 0.204
Preoperative opioids, n (%)
 No 213 18 (8.5) 1 1
 Yes 36 11 (30.6) 3.62 (1.87–7.01) 0.001 1.80 (0.83–3.91) 0.139
Preoperative benzodiazepines, n (%)
 No 212 20 (9.4) 1 1
 Yes 37 9 (24.3) 2.58 (1.27–5.22) 0.022 1.41 (0.66–3.01) 0.370
Preoperative infection, n (%)
 No 225 23 (10.2) 1 1
 Yes 24 6 (25.0) 2.45 (1.11–5.41) 0.044 1.47 (0.68–3.19) 0.329
Barthel Index score ≤ 70, n (%)
 No 198 16 (8.1) 1 1
 Yes 51 13 (25.5) 3.15 (1.62–6.13) 0.001 1.14 (0.54–2.40) 0.734

A p-value < 0.05 indicates statistical significance

Abbreviations: RR risk ratio, CI confidence interval