When OGTT-derived physiologic variables were included with HbA1c, FPG, and family history of diabetes in a stepwise multivariable regression analysis, only HbA1c and ln insulin secretion/insulin resistance predicted the development of diabetes (r = 0.49; P < 0.0001).
When information on use of diuretics, the use of ACE inhibitors/ARBs, or the type of diabetes was added to the multivariable regression analyses, only the type of diabetes was a significant covariate for the association between urinary KIM-1, NAG, NGAL, cystatin C, and H-FABP with albuminuria.
In the multivariable linear regression only antiviral use (within 48 hours) was associated with the number of QALDs lost, and only in confirmed H1N1v cases (p = 0.084).
According to multivariable logistic regression, only age group and surveillance site remained statistically significant (Table 2).
In a multivariable logistic regression, only HER2 expression, increasing tumour size, and PLR were significantly associated with an elevated NLR.
Following multivariable logistic regression, only age at diagnosis, postmenopausal status, Chinese ethnicity, and tumour size were independently and significantly associated with mastectomy (Table 3).
After multivariable logistic regression, only type of recruiting centre (P<0.01), receiving adjuvant therapy or not (P<0.001), and type of patient recruited (new or already on follow-up; P<0.01) showed a significant association with undertreatment.
In multivariable logistic regression only black race, male sex, SBP, and NIHSS were independent predictors of critical care needs with p <0.1 (Table 2); DBP and history of hypertension were not included because of collinearity with SBP.> For score development, the best model included race, sex, SBP with cut points at 160 mm Hg and 200 mm Hg, and NIHSS with cut points at 7 and 13.
Variables that were significant at p < 0.10 were entered into the multivariable regression models but were only included in the final parsimonious models if p < 0.05.
