Skip to main content

Table 3 Goodness of fit statistics for negative binomial regression and generalized estimating equation models applied to cases ascertained by juvenile diabetes algorithms

From: Control charts for chronic disease surveillance: testing algorithm sensitivity to changes in data coding

Algorithm Model Fit Measures Number of RCS knots
Incidence
  Residual Deviance Residual DOF Residual Deviance/ Residual DOF  
 1: 1 + H or 1 + P 175 169 1.04 5
 1: 1 + H or 2 + P 178 170 1.05 4
 1: 1 + H or 3 + P 182 171 1.06 3
 1: 1 + H or 4 + P 192 170 1.13 4
 2: 1 + H or 1 + P 172 165 1.04 5
 2: 1 + H or 2 + P 173 166 1.04 4
 3: 1 + H or 1 + P 167 161 1.04 5
 3: 1 + H or 2 + P 168 162 1.04 4
 1: 1 + P 175 169 1.04 5
 1: 2 + P 192 170 1.13 4
 2: 1 + P 171 165 1.04 5
 2: 2 + P 182 166 1.10 4
 2: 3 + P 186 167 1.11 3
 2: 4 + P 192 167 1.15 3
 2: 5 + P 205 167 1.23 3
Prevalence
  Marginal R2  
 1: 1 + H or 1 + P 0.94 5
 1: 1 + H or 2 + P 0.97 5
 1: 1 + H or 3 + P 0.96 5
 1: 1 + H or 4 + P 0.90 5
 2: 1 + H or 1 + P 0.92 5
 2: 1 + H or 2 + P 0.97 5
 3: 1 + H or 1 + P 0.91 5
 3: 1 + H or 2 + P 0.97 5
 1: 1 + P 0.94 5
 1: 2 + P 0.97 5
 1: 3 + P 0.96 5
 1: 4 + P 0.83 5
 2: 1 + P 0.92 5
 2: 2 + P 0.98 5
 2: 3 + P 0.97 5
 2: 4 + P 0.97 5
 2: 5 + P 0.96 5
  1. DOF Degrees of freedom