Table 2.
Univariate and adjusted logistic regression models predicting odds of delirium in the hip fracture cohort (n = 279)
| Univariate | Adjusted* | |||
|---|---|---|---|---|
| Odds ratio (95% CI) | P | Odds ratio (95% CI) | P | |
| MMP-2 | 1.025 (1.009; 1.041) | 0.002 | 1.006 (0.988; 1.025) | 0.51 |
| MMP-3 | 1.001 (0.999; 1.002) | 0.11 | 1.000 (0.999; 1.001) | 0.98 |
| MMP-10 | 1.011 (1.002; 1.021) | 0.02 | 1.002 (0.993; 1.011) | 0.65 |
| MMP-12 | 0.977 (0.924; 1.032) | 0.40 | 0.948 (0.883; 1.016) | 0.13 |
| TIMP-1 | 1.013 (1.005; 1.021) | 0.001 | 1.003 (0.992; 1.014) | 0.59 |
| TIMP-2 | 1.026 (1.013; 1.039) | 8.5*10−5 | 1.015 (0.998; 1.031) | 0.08 |
| TIMP-3 | 1.164 (1.029; 1.318) | 0.02 | 1.020 (0.861; 1.205) | 0.82 |
| TIMP-4 | 0.550 (0.335; 0.902) | 0.02 | 0.424 (0.224; 0.802) | 0.008 |
*Adjusted for sex, age, and presence/absence of dementia (IQCODE ≥ 3.44 = dementia (n = 118), IQCODE < 3.44 = no dementia (n = 161). In all models, higher age and the presence of dementia were significant predictors of delirium.
CI confidence interval, IQCODE Informant Questionnaire on Cognitive Decline in the Elderly, MMP matrix metalloproteinase, TIMP tissue inhibitor of matrix metalloproteinase