Baseline characteristics of the studied population
| Total sample (N=826) | Delirium (N=123) | No delirium (N=703) | P value | Corrected P value* | |
| Age, mean (SD) | 63.7 (15.16) | 76.9 (13.15) | 61.3 (14.27) | <0.001 | N.A. |
| Males, n (%) | 544 (65.9%) | 69 (56.1%) | 475 (67.6%) | 0.01 | <0.001 |
| Triage colour, n (%)† | <0.001 | 0.02 | |||
| Green | 409 (49.5%) | 19 (15.4%) | 390 (55.6%) | ||
| Yellow | 297 (36%) | 56 (45.5%) | 241 (34.3%) | ||
| Red | 119 (14.4%) | 48 (39%) | 71 (10.1%) | ||
| Institutionalised, n (%)† | 54 (6.5%) | 32 (26%) | 22 (3.1%) | <0.001 | 0.01 |
| Stroke/TIA, n (%)‡ | 37 (4.5%) | 24 (3.4%) | 13 (10.6%) | <0.001 | 0.01 |
| Parkinson, n (%)‡ | 9 (1.1%) | 5 (4.1%) | 4 (0.6%) | 0.001 | <0.001 |
| Alzheimer’s disease, n (%)‡ | 9 (1.1%) | 6 (4.9%) | 3 (0.4%) | <0.001 | <0.001 |
| Other dementias, n (%)§ | 54 (6.5%) | 43 (35%) | 11 (1.6%) | <0.001 | <0.001 |
| Epilepsy, n (%)¶ | 25 (3%) | 12 (9.8%) | 13 (1.9%) | <0.001 | <0.001 |
| Other neurological diseases, n (%)‡ | 62 (7.5%) | 20 (16.3%) | 42 (6%) | <0.001 | <0.001 |
| Depression, n (%)§ | 31 (3.8%) | 12 (9.8%) | 19 (2.7%) | <0.001 | <0.001 |
| Number of chronic drugs, median (IQR)¶ | 2 (IQR 0–5) | 5 (IQR 3–8) | 1 (IQR 0–4) | <0.001 | 0.025 |
| Anticholinergic Burden score, median (IQR)** | 0 (IQR 0–0) | 1 (IQR 0–2) | 0 (IQR 0–0) | <0.001 | 0.01 |
| Psychoactive drugs, n (%)¶ | 78 (9.4%) | 32 (26.2%) | 46 (6.6%) | <0.001 | <0.001 |
| Sedative-hypnotics | 50 (6.1%) | 20 (16.4%) | 30 (4.3%) | <0.001 | <0.001 |
| Antidepressants | 60 (7.3%) | 22 (18%) | 38 (5.4%) | <0.001 | <0.001 |
| Antipsychotics | 39 (4.7%) | 24 (19.7%) | 15 (2.1%) | <0.001 | <0.001 |
| Memantine | 1 (0.1%) | 1 (0.8%) | 0 (0%) | 0.02 | 0.01 |
| Acetylcholinesterase inhibitors | 5 (0.6%) | 4 (3.3%) | 1 (0.1%) | <0.001 | <0.001 |
| Dopaminergic drugs | 12 (1.5%) | 7 (5.7%) | 5 (0.7%) | <0.001 | <0.001 |
*Benjamini-Hochberg correction. Benjamini-Hochberg corrected p values were calculated by computing p(k)=m/k for all sorted p values (p=sorted p values; m=total number of sorted p values; k=corresponding rank). Then p values were adjusted from the bigger to the smallest. If the value p(k)=m/k was bigger than the previous adjusted p value, the corresponding corrected p value was equal to the previous one since numbers should not increase.
†Four missing.
‡Six missing.
§Five missing.
¶Eight missing.
**Fourteen missing.
n, number; TIA, transient ischaemic attack.