Table 4.
Hierarchical regression analysis assessing the influence of subject, tooth and site characteristics in the variability of probing depth (PD) reduction
| Intercept | ΔPDBaseline-6months | |
|---|---|---|
| Empty model β±SE |
Multivariate model β±SE |
|
| 2.816±0.084 | 3.622±0.264 | |
| Patient (Level 3) | 0.246±0.063 (17)‡ | 0.203±0.051 (-17%)† |
| Tooth (Level 2) | 0.885±0.040 (61)‡ | 0.711±0.033 (-20%)† |
| Site (Level 1) | 0.325±0.006 (22)‡ | 0.304±0.006 (-9%)† |
| Total variance | 1.456 | 1.218 |
| -2 LL | 14790.139 | 10193.559* |
‡
Proportion of variability in ΔPD provided by the multilevel analysis at the subject, tooth, and site levels.
†
Change in the proportion of variability in ΔCAL at the subject, tooth, and site levels when explicatory factors were incorporated in the analysis
*
Variation in -2 LL was significant (P<0.001)