Skip to main content
. 2019 Jul 27;46(6):847–857. doi: 10.1007/s10488-019-00958-7

Table 7.

Three-level regression model for days with contact with a health care professional

Elasticity Std. Err. p-value N
Cluster 0 0.3589 0.0250 < 0.01 11,563
Cluster 1 0.3105 0.0133 < 0.01 21,291
Cluster 2 0.3700 0.0141 < 0.01 26,647
Cluster 3 0.4648 0.0176 < 0.01 86,234
Cluster 4 0.4957 0.0208 < 0.01 114,109
Cluster 5 0.4676 0.0186 < 0.01 48,274
Cluster 6 0.6198 0.0278 < 0.01 22,451
Cluster 7 0.6180 0.0282 < 0.01 54,671
Cluster 8 0.6181 0.0321 < 0.01 48,131
Cluster 10 0.6758 0.0378 < 0.01 27,553
Cluster 11 0.6298 0.0264 < 0.01 65,760
Cluster 12 0.6288 0.0293 < 0.01 59,551
Cluster 13 0.6208 0.0311 < 0.01 36,559
Cluster 14 0.3827 0.0235 < 0.01 21,724
Cluster 15 0.3663 0.0233 < 0.01 5,960
Cluster 16 0.6381 0.0430 < 0.01 11,816
Cluster 17 0.8204 0.0793 < 0.01 15,895
Cluster 18 0.5004 0.0219 < 0.01 115,125
Cluster 19 0.4613 0.0233 < 0.01 125,115
Cluster 20 0.4571 0.0269 < 0.01 50,384
Cluster 21 0.4232 0.0322 < 0.01 19,835

Days with contact with a health care professional is the dependent variable. Length of cluster-episode is the explanatory variable. Levels are cluster-episodes—patient—provider. Elasticity = (proportional change in days with contact with a health care professional)/(proportional change in length of cluster-episode)