Table 3. Predictive Value of Baseline Clinical Parameters for HbA1c Reduction by Sitagliptin Using Linear Mixed Effect Model (Type III Tests of Fixed Effects) (n = 3,201) .
| Effect | Numerator of degree of freedom | Denominator of degree of freedom | F value | Pr > F |
|---|---|---|---|---|
| Treatment period | 3 | 8,076 | 839.64 | < 0.001 |
| Treatment period × body mass index (kg/m2) | 3 | 8,076 | 0.91 | 0.4330 |
| Age (years) | 1 | 3,071 | 9.29 | 0.0023 |
| Gender (female/male) | 1 | 3,071 | 0.07 | 0.7871 |
| Body mass index (kg/m2) | 1 | 3,071 | 4.44 | 0.0353 |
| Estimated duration of diabetes (years) | 1 | 3,071 | 2.42 | 0.1202 |
| Baseline HbA1c levels (%) (NGSP) | 3 | 3,071 | 1,512.15 | < 0.001 |
| eGFR (mL/min/1.73 m2) | 1 | 3,071 | 1.94 | 0.1640 |
| Type of diabetes drugs | 2 | 3,071 | 7.90 | 0.0004 |
Age (< 65 vs. ≥ 65), gender (male vs. female), duration of type 2 diabetes (< 10 years vs. ≥ 10 years), BMI (< 25 vs. ≥ 25), HbA1c levels (6, vs. 6 - 6.9 vs. 7.0 - 7.9 vs. ≥ 8), eGFR (60, vs. ≥ 60) and type of anti-diabetes drugs (sitagliptin monotherapy vs. sitagliptin plus SU vs. sitagliptin plus OHA and/or insulin without SU) were entered as covariate. eGFR: estimated glomerular filtration rate; NGSP: National Glycohemoglobin Standardization Program.