Table 3.
Model fitting results of the multilevel analyses for the domains communication with ophthalmologist, communication with nurses, and communication about medication and for the global rating of ophthalmologist, nurses and hospitals (standard errors added in parentheses)
| Outcome variable | intercept | β age1 | β gender1 | β education1 | β health1 | β% drops | Var patients | Var hospital | ICC2 |
| Domain ophthalmologist (N = 3942) | |||||||||
| Model 1 | 3.640 (0.014)* | - | - | - | - | - | 0.246 (0.006)* | 0.006 (0.002)* | 0.02 |
| Model 2 | 3.687 (0.031)* | -0.005 (0.016) | 0.017 (0.016) | -0.031 (0.017) | -0.102 (0.017)* | 0.000 (0.000) | 0.243 (0.006)* | 0.007 (0.002)* | 0.03 |
| Domain nurses (N = 3925) | |||||||||
| Model 1 | 3.726 (0.009)* | - | - | - | - | - | 0.196 (0.004)* | 0.002 (0.001) | n.c. |
| Model 2 | 3.767 (0.024)* | -0.020 (0.014) | 0.005 (0.015) | -0.016 (0.015) | -0.089 (0.015)* | 0.000 (0.000) | 0.194 (0.004)* | 0.002 (0.001) | n.c |
| Domain medication (N = 3887) | |||||||||
| Model 1 | 2.670 (0.033)* | - | - | - | - | - | 1.246 (0.028)* | 0.038 (0.011)* | 0.03 |
| Model 2 | 3.015 (0.070)* | -0.214 (0.036)* | -0.035 (0.037) | -0.166 (0.039)* | -0.020 (0.039) | -0.002 (0.001)* | 1.232 (0.028)* | 0.031 (0.010)* | 0.02 |
| Rating ophthalmologist (N = 3940) | |||||||||
| Model 1 | 8.791 (0.039)* | - | - | - | - | - | 2.124 (0.048)* | 0.046 (0.015)* | 0.02 |
| Model 2 | 8.894 (0.088)* | 0.104 (0.047)* | 0.013 (0.048) | -0.106 (0.051) | -0.308 (0.051)* | 0.000 (0.001) | 2.101 (0.048)* | 0.045 (0.015)* | 0.02 |
| Rating nurses (N = 3910) | |||||||||
| Model 1 | 8.916 (0.027)* | - | - | - | - | - | 1.466 (0.033)* | 0.016 (0.007)* | 0.01 |
| Model 2 | 8.943 (0.065)* | 0.084 (0.039)* | 0.004 (0.040) | -0.086 (0.042) | -0.219 (0.043)* | 0.001 (0.001) | 1.455 (0.033)* | 0.011 (0.006) | n.c |
| Rating hospital (N = 3970) | |||||||||
| Model 1 | 8.775 (0.036)* | - | - | - | - | - | 1.730 (0.039)* | 0.041 (0.013)* | 0.02 |
| Model 2 | 8.987 (0.080)* | 0.064 (0.042)* | -0.040 (0.043) | -0.127 (0.045)* | -0.306 (0.046)* | -0.001 (0.001) | 1.709 (0.039)* | 0.037 (0.013)* | 0.02 |
*p < 0.05
1Reference group age = younger than 75; reference group gender = males; reference group education = low education; reference group health = good health
2ICC (intra-class correlation) = Var hospital/(Var patients + Var hospital)
n.c. = not calculated. The variance explained by the hospital level is not significant and, therefore, the ICC is not calculated.