Table 3. Interaction analysis showing evidence of interaction between DRB1 * 15 and other allele groups influencing the risk of MS.
Logstic regression, statistical interaction | Attributable proportion, biologic interaction | |||||||
Interaction term | OR DRB1 * 15 | OR interacting allele | OR (interaction term, 95% C.I.) | nominal p-value (interaction term) | Bonferroni corrected p-value | AP (95% C.I.) | nominal p-value | Bonferroni corrected p-value |
DRB1*15×C*12# | 3.84 | 1.6 | 0.37 (0.21–0.64) | 0.00035 | 0.0028 | 0.58 (0.31–0.84)* | 2.57×10-05 | 0.00021 |
DRB1*15×DRB1*01# | 3.74 | 0.97 | 0.54 (0.36–0.82) | 0.0041 | 0.033 | 0.47 (0.27–0.66)* | 4.00×10-06 | 3.20×10-05 |
DRB1*15×A*02 | 3.54 | 0.66 | 1.01 (0.76–1.35) | 0.92 | 1.00 | 0.23 (0.067–0.40)* | 0.0058 | 0.046 |
DRB1*15×B*18# | 3.66 | 1.97 | 0.54 (0.32–0.91) | 0.019 | 0.15 | −0.23 (−0.68–0.23) | 0.33 | 1.00 |
DRB1*15×A*10 | 3.7 | 1.79 | 0.63 (0.39–1.00) | 0.049 | 0.39 | −0.049 (−0.43–0.34) | 0.80 | 1.00 |
DRB1*15×B*12 | 3.35 | 0.61 | 1.25 (0.87–1.79) | 0.23 | 1.00 | 0.12 (−0.11–0.35)* | 0.29 | 1.00 |
DRB1*15×DRB1*07 | 3.38 | 0.64 | 1.24 (0.78–1.98) | 0.36 | 1.00 | 0.1 (−0.21–0.41)* | 0.53 | 1.00 |
DRB1*15×B*14 | 3.63 | 2.02 | 1.00 (0.45–3.00) | 1.00 | 1.00 | 0.36 (−0.13–0.84) | 0.15 | 1.00 |
= risk is coded as absence of this allele group since it is nominally protective.
= Model has significantly better fit than one with only the two variables by themselves (no interaction term).