Table 5.

BMD₁₀ for glomerulosclerosis in female rat (data from Larson, 1979, dose group 1 excluded)

Model	Total number of parameters in the model	AIC	Goodness of fit test P-value	Scaled residual near BMR	BMD10 (mg/kg/d)	BMD10-HED (mg/kg/d)
Gamma-Restricted	3	73.996	0.218	−0.25	0.0857	0.0219
Gamma	3	74.198	0.594	0.01	0.0043	0.0011
Logistic	2	78.734	0.009	1.19	0.2102	0.0537
LogLogistic-Restricted	3	72.206	0.848	−0.02	0.0495	0.0126
LogLogistic	3	74.083	0.679	0.01	0.0259	0.0066
LogProbit-Restricted	3	73.859	0.215	−0.15	0.1235	0.0315
LogProbit	3	74.093	0.671	0.01	0.0288	0.0073
Multistage-Restricted	2	73.996	0.218	−0.25	0.0857	0.0219
Multistage	2	73.915	0.969	0.00	0.0552	0.0141
Multistage-Cancer 1	2	73.996	0.218	−0.25	0.0857	0.0219
Multistage-Cancer 2	3	73.996	0.218	−0.25	0.0857	0.0219
Multistage-Cancer 3	4	73.996	0.218	−0.25	0.0857	0.0219
Probit	2	79.168	0.016	1.28	0.2378	0.0607
Weibull-Restricted	3	73.996	0.218	−0.25	0.0857	0.0219
Weibull	3	73.996	0.218	−0.25	0.0857	0.0219
Quantal-Linear	3	73.996	0.218	−0.25	0.0857	0.0219
Minimal BMD						0.001
Maximal BMD						0.032
Ratio max/min						29
BMD from best AIC model						0.0126
Geometric mean of acceptable models BMD						Not applicable

The benchmark response is 10 %. AIC: Akaïke’s Information Criterion. BMD₁₀: Benchmark dose for 10 % excess risk. Results in bold are acceptable models according to these criteria: total number of parameters does not exceed the number of groups modeled (n = 4); and p-value > 0.1; and scaled residual < │2│ and no computation error or unrealistic graphical BMD₁₀. If ratio max/min is less than 3, then geometric mean of acceptable models is selected, else best AIC model is selected. “na”: not applicable