Table 2

Univariable associations in Australian Cattle Dogs for congenital hereditary sensorineural deafness (CHSD) from multilevel logistic models and animal models
Variable No. of dogs tested* % normal hearing % unilateral deafness % bilateral deafnessa Multilevel logistic model** Logistic animal model***
Odds ratio (95% CI)† P-value Odds ratio (95% CI)† P-value
Sex 0.044 0.059
 Male 381 91.1% 5.8% 3.1% Reference category Reference category
 Female 417 86.8% 9.1% 4.1% 1.7 (1.0 to 2.8) 0.044 1.6 (1.0 to 2.6) 0.057
 Not recorded 12 75.0% 16.7% 8.3%
Colour***** 0.785 0.916
 Red 327 88.7% 8.0% 3.4% Reference category Reference category
 Blue 137 87.6% 7.3% 5.1% 1.2 (0.6 to 2.7) 0.589 1.1 (0.5 to 2.3) 0.824
 Blue black tan 319 88.7% 7.5% 3.8% 0.8 (0.4 to 1.6) 0.563 0.9 (0.5 to 1.7) 0.775
 Blue tan 18 88.9% 11.1% 0.0% 0.7 (0.1 to 4.1) 0.723 0.6 (0.1 to 3.1) 0.571
 Red tan 1 100.0% 0.0% 0.0%
 Blue black 6 100.0% 0.0% 0.0%
 Blue mottled 1 100.0% 0.0% 0.0%
 Red mottled 1 100.0% 0.0% 0.0%
Base colour 0.882 0.885
 Blue 481 88.6% 7.5% 4.0% Reference category Reference category
 Red 329 88.8% 7.9% 3.3% 1.0 (0.6 to 1.9) 0.882 1.0 (0.6 to 1.8) 0.890
Blue black tan 0.447 0.768
 No 491 88.6% 7.7% 3.7% Reference category Reference category
 Yes 319 88.7% 7.5% 3.8% 0.8 (0.4 to 1.5) 0.447 0.9 (0.5 to 1.6) 0.775
Blue black
 No 804 88.6% 7.7% 3.7% NM NM
 Yes 6 100.0% 0.0% 0.0%
Blue tan 0.688 0.951
 No 792 88.6% 7.6% 3.8% Reference category Reference category
 Yes 18 88.9% 11.1% 0.0% 1.4 (0.2 to 8.5) 0.688 1.0 (0.2 to 4.7) 0.954
Red tan
 No 809 88.6% 7.7% 3.7% NM 2 NM 2
 Yes 1 100.0% 0.0% 0.0%
Speckled 0.609 0.726
 No 423 88.7% 7.3% 4.0% Reference category Reference category
 Yes 387 88.6% 8.0% 3.4% 1.2 (0.7 to 2.1) 0.609 1.1 (0.6 to 1.9) 0.733
Mottled 0.507 0.589
 No 805 88.7% 7.7% 3.6% Reference category Reference category
 Yes 5 80.0% 0.0% 20.0% 2.4 (0.2 to 33.0) 0.507 1.9 (0.2 to 20.8) 0.593
Mask 0.001 ****
 No 185 85.4% 12.4% 2.2% Reference category Reference category
 Left 66 83.3% 10.6% 6.1% 1.0 (0.4 to 2.4) 0.992 1.0 (0.4 to 2.3) 0.988
 Right 58 81.0% 12.1% 6.9% 1.4 (0.6 to 3.3) 0.455 1.2 (0.5 to 2.9) 0.603
 Bilateral 229 95.6% 3.1% 1.3% 0.2 (0.1 to 0.5) 0.001 0.2 (0.1 to 0.5) <0.001
 Not recorded 272 87.9% 6.6% 5.5%
Head spot 0.767 0.911
 No 439 88.8% 8.7% 2.5% Reference category Reference category
 Yes 99 89.9% 6.1% 4.0% 0.9 (0.4 to 2.0) 0.767 1.0 (0.5 to 2.0) 0.916
 Not recorded 272 87.9% 6.6% 5.5%
Pigmented body spot 0.034 0.043
 None 430 87.4% 9.3% 3.3% Reference category Reference category
 At least one 104 95.2% 3.8% 1.0% 0.3 (0.1 to 0.9) 0.034 0.4 (0.1 to 1.0) 0.041
 Not recorded 276 88.0% 6.5% 5.4%
White head/body patches 0.852 0.933
 No 508 89.0% 8.3% 2.8% Reference category Reference category
 Yes 27 88.9% 7.4% 3.7% 1.1 (0.3 to 4.7) 0.852 0.9 (0.3 to 3.5) 0.937
 Not recorded 275 88.0% 6.5% 5.5%
Sire CHSD status 0.085 0.205
 Tested normal 502 91.4% 6.4% 2.2% Reference category Reference category
 Unilateral deaf left ear 10 70.0% 20.0% 10.0% 5.9 (0.8 to 44.0) 0.085 3.3 (0.5 to 21.1) 0.200
 Unknown 298 84.6% 9.4% 6.0%
Dam CHSD status 0.321 0.402
 Normal 430 88.6% 8.4% 3.0% Reference category Reference category
 Unilateral deaf right ear 3 66.7% 0.0% 33.3% 4.8 (0.2 to 105.5) 0.321 3.4 (0.2 to 59.7) 0.400
 Unknown 377 88.9% 6.9% 4.2%

The odds ratios are for having at least one deaf ear rather than normal hearing.

*No. indicates number.

** Multilevel logistic models with dog fitted within litter within kennel of birth. Bolded P-values are overall likelihood ratio test P-values; unbolded P-values are Wald P-values for each coefficient.

*** Logistic animal (‘threshold’) models using the full pedigree (ie all pedigree data that were available), with dog fitted as a random effect with a numerator relationship matrix. Bolded P-values are overall Wald P-values; unbolded P-values are Wald P-values for each coefficient.

**** Overall p-value for mask in logistic animal model not calculated.

***** Last 5 categories were pooled for analyses.

† 95% CI indicates 95% confidence interval.

EX Dogs excluded because 0 dogs or 1 dog in group had hearing loss.

NM Variable not modelled as no dogs in group had hearing loss.

NM 2 Variable not modelled as only one dog in red tan group.

a Some rows do not sum to 100.0% due to rounding.

Sommerlad et al.

Sommerlad et al. BMC Veterinary Research 2012 8:202   doi:10.1186/1746-6148-8-202

Open Data