Table 4 

Multivariate Models with Covariates for Alcohol Consumption 

MONTREAL (r^{2 }= 0.62) 

Effect 
Estimate 
Standard Error 
tstatistic 
pvalue 
Intercept 
0.59 
0.13 
4.59 
<.0001 
recessive haplotype 2  MPDZ 
1.02 
0.24 
4.17 
<.0001 
rs701492  GAD1 (AA) 
0.70 
0.10 
6.94 
<.0001 
rs701492  GAD1 (AB) 
0.03 
0.07 
0.42 
0.6714 
rs701492  GAD1 (BB) 
referent 

Age 
0.01 
0.00 
4.50 
<.0001 
Alcohol dependence in past year 
1.04 
0.10 
10.73 
<.0001 
Alcohol abuse in past year 
0.73 
0.10 
7.67 
<.0001 
Family history of depression (1st degree relative) 
0.50 
0.09 
5.45 
<.0001 
Familial depression and alcohol dependence* 
0.85 
0.14 
6.09 
<.0001 
Used medication other than antidepressants in last month 
0.26 
0.07 
3.91 
0.0001 
SYDNEY (r^{2 }= 0.62) 

Effect 
Estimate 
Standard Error 
tstatistic 
pvalue 
Intercept 
0.24 
0.06 
4.12 
<.0001 
CHRM5 (recessive) 
2.10 
0.35 
5.95 
<.0001 
GABRB2 (dominant) 
3.03 
0.43 
7.00 
<.0001 
PPP1R1B (recessive) 
2.97 
0.32 
9.22 
<.0001 
Alcohol dependence in past year 
0.58 
0.11 
5.06 
<.0001 
Alcohol abuse in past year 
0.58 
0.11 
5.14 
<.0001 


Shown are the coefficients associated with the final multivariate models with covariates for alcohol consumption as outcome. Final models were chosen using a backward selection process with an exit criterion of p > 0.001. * Familial depression is defined as major depression in the subject and in a firstdegree relative. 

Tabakoff et al. BMC Biology 2009 7:70 doi:10.1186/17417007770 