TABLE 2.
Association of plasma and RBC zinc concentrations with 30-d death or urgent readmission
| Variable | Parameter estimate (SE) | P |
|---|---|---|
| Plasma zinc, unadjusted | ||
| Intercept | −0.12 (0.62) | 0.8508 |
| (Plasma zinc)0.5 | −0.14 (0.07) | 0.0518 |
| Plasma zinc, adjusted | ||
| Intercept | −1.65 (0.79) | 0.0362 |
| (LACE score)2 | 0.01 (0) | 0.0006 |
| (Plasma zinc)0.5 | −0.08 (0.08) | 0.3097 |
| RBC zinc, unadjusted | ||
| Intercept | −2.05 (0.45) | 0 |
| (RBC zinc)2 | 1.23 × 10−5 (0) | 0.0707 |
| RBC zinc, adjusted | ||
| Intercept | −3.82 (0.92) | 0 |
| (LACE score)2 | 0.01 (0) | 0.0001 |
| (RBC zinc)2 | 0.01 (0) | 0.0844 |
This table presents parameter estimates (with SE) and P values of terms in 4 logistic regression models with the 30-d death or urgent readmission as the outcome. Negative parameter estimates indicate an increased outcome risk as values decrease. Adjusted models include the LACE score [21]. Analyses were limited to people who survived their index hospitalization (n = 241). Plasma and RBC zinc concentrations were measured in micromoles per liter. LACE, length of stay, acuity of admission, comorbidities, emergency department use; .