Table 2.

The predictAL-10 and predictAL-9 modelsa to predict the onset of hazardous alcohol drinking at 12 months

Risk factorsIncidence of hazardous alcohol drinkingPredictAL-10b (N= 2264)PredictAL-9c (N= 2278) (excluding childhood sexual abuse from the model)
NoYes (%)ORORd95% CIPORORe95% CIP-value
Constant0.00080.00090.0001 to 0.0071<0.0010.00110.00120.0001 to 0.0079<0.001
Province
  Granada (Reference)4997 (1.38)1.01.01.01.0
  Zaragoza44716 (3.46)2.021.940.49 to 8.370.3331.801.730.49 to 6.700.379
  Madrid4064 (0.98)0.720.690.15 to 3.560.6900.660.640.15 to 2.930.585
  Logroño (La Rioja)42929 (6.33)7.126.842.05 to 24.790.0026.466.222.11 to 19.790.001
  Majorca25813 (4.80)5.325.111.11 to 25.620.0374.744.571.08 to 20.840.040
  Las Palmas1857 (3.65)3.163.030.61 to 16.280.1703.092.960.62 to 15.410.168
Sex
  Female (Reference)162228 (1.70)1.01.01.01.0
  Male60248 (7.38)3.203.071.29 to 7.910.0122.822.721.62 to 6.830.022
  Age (range 18–75 years)0.9930.9530.972 to 1.0150.5390.9890.9520.969 to 1.0100.316
  AUDIT-C2.512.411.63 to 3.85<0.0012.372.281.57 to 3.59<0.001
  AUDIT-C*Agef0.9910.9510.984 to 0.9990.0450.9930.9560.985 to 1.0000.068
Cigarette consumption per day
  Non-smoking (Reference)175641 (2.28)1.01.01.01.0
  <1018112 (6.22)2.392.291.21 to 4.730.0122.422.331.20 to 4.880.014
  10–2020011 (5.21)1.281.230.51 to 3.180.6001.381.330.57 to 3.310.475
  >208712 (12.1)3.483.341.31 to 9.270.0133.593.461.33 to 9.680.012
Financial strain
  Living comfortably (Reference)1833 (1.61)1.01.01.01.0
  Doing all right163456 (3.31)1.941.860.48 to 7.820.3512.071.990.50 to 8.510.313
  Finding it difficult or very difficult40517 (4.03)4.194.020.98 to 17.840.0534.664.491.08 to 20.030.039
Ever treated for alcohol problems
  No (Reference)220473 (3.21)1.01.01.01.0
  Yes123 (20.0)11.7711.301.98 to 70.050.00713.1912.702.61 to 66.630.002
Childhood sexual abuse
  No (never) (Reference)216170 (3.14)1.01.0
  Yes (rarely, sometimes, often, frequently)496 (10.9)5.074.871.71 to 15.090.003
  • a Multilevel logistic regression with health centre as a random component and weighting for the inverse probability of remaining in the follow-up to 12 months.

  • b Discriminative validity: c-index: 0.886 (95% CI = 0.854 to 0.918) and effect size (Hedges’ g): 1.694 (95% CI = 1.460 to 1.928).

  • c Discriminative validity: c-index: 0.880 (95% CI = 0.847 to 0.913) and effect size (Hedges’ g): 1.658 (95% CI = 1.425 to 1.892).

  • d Overfitting estimate: Copas’ shrinkage factor = 0.960.

  • e Overfitting estimate: Copas’ shrinkage factor = 0.963.

  • f Likelihood ratio test for the interaction: χ2 (degree of freedom:1) = 5.84; P = 0.0157. AUDIT = Alcohol Use Disorders Identification Test.