Table 2. Rare Events Logistic Regression Models Predicting Attempts to Deceive Physicians
CharacteristicsAttempting to Deceive a Physician for Pharmaceuticals
Any ReasonFor AbuseTo Sell
b (SE)ORb (SE)ORb (SE)ORb (SE)ORb (SE)ORb (SE)OR
Sex (1 = male)−0.04 (0.18)0.96−0.07 (0.19)0.93−0.08 (0.18)0.93−0.09 (0.19)0.91−0.17 (0.19)0.85−0.23 (0.17)0.80
Race*
    African American0.19 (0.29)1.210.47 (0.30)1.600.00 (0.32)1.000.30 (0.34)1.350.30 (0.42)1.350.62 (0.43)1.85
    Other0.79 (0.36)2.190.75 (0.41)2.120.90 (0.37)2.470.88 (0.42)2.401.38 (0.44)3.991.24 (0.51)3.46
Age−0.02 (0.03)0.98−0.03 (0.03)0.97−0.02 (0.04)0.98−0.03 (0.03)0.97−0.02 (0.05)0.98−0.05 (0.04)0.96
    Student athlete (1 = yes)0.66 (0.39)1.940.78 (0.39)2.180.65 (0.41)1.920.76 (0.41)2.150.91 (0.51)2.481.04 (0.48)2.83
    Fraternity/sorority (1 = yes)0.15 (0.27)1.16−0.11 (0.21)0.890.11 (0.28)1.11−0.15 (0.21)0.860.55 (0.39)1.730.44 (0.42)1.55
Employed§
    Part time0.43 (0.25)1.540.48 (0.25)1.610.47 (0.27)1.60.53 (0.27)1.700.63 (0.35)1.870.62 (0.37)1.87
    Full time1.15 (0.45)3.161.12 (0.49)3.061.29 (0.46)3.631.28 (0.5)3.601.74 (0.54)5.681.68 (0.63)5.34
    Sexual orientation (1 = LGBT)1.32 (0.4)3.751.29 (0.39)3.631.45 (0.41)4.281.45 (0.39)4.261.64 (0.49)5.171.56 (0.46)4.74
Family income−0.56 (0.23)0.57−0.57 (0.22)0.57−0.64 (0.24)0.53−0.64 (0.23)0.53−0.62 (0.31)0.54−0.58 (0.3)0.56
Family income20.06 (0.02)1.060.06 (0.02)1.060.07 (0.02)1.070.06 (0.02)1.070.06 (0.03)1.070.05 (0.03)1.05
Alcohol use0.17 (0.09)1.190.15 (0.09)1.160.10 (0.12)1.10
Marijuana use0.00 (0.07)1.00−0.03 (0.07)0.970.09 (0.09)1.10
Pharmaceutical misuse0.34 (0.07)1.400.35 (0.07)1.420.38 (0.10)1.46
Constant−2.22−2.88−2.19−2.74−3.14−3.49
Model χ241.3164.6545.5462.0250.4160.43
Pseudo-R20.0660.1140.0800.1210.1300.177
Model c (area under ROC curve)0.6520.7100.6670.7190.6980.761
  • * White is the reference category for race.

  • P < .05.

  • P < .10.

  • § Not employed is the reference category for employment.

  • LGBT, lesbian/gay/bisexual/transgender; OR, odds ratio; ROC, receiver operating characteristic; SE, standard error.