Table 4.
Logistic regression models for opioid misuse risk (COMM 1) (n = 122).
| Predictor Variables | Model 1. Pain severity and interference | Model 2. Add depression | Model 3. Add pain location | Model 4.Add depression * pain location |
|---|---|---|---|---|
| Variables | OR (95% CI) | OR (95% CI) | OR (95% CI) | OR (95% CI) |
| Total average pain severity 2 | 0.88 (0.60–1.29) | 0.92 (0.61–1.37) | 0.87 (0.58–1.31) | 0.81 (0.52–1.26) |
| Total average pain interference 2 | 2.25 (1.45–3.48) | 1.91 (1.21–3.01) | 1.95 (1.23–3.09) | 2.09 (1.26–3.47) |
| Depression (Yes vs. No) | 3.32 (1.20–9.16) | 3.32 (1.19–9.23) | ||
| Pain Location 3 | ||||
| CPOL | 1.00 | - | ||
| CLBP | 0.64 (0.23–1.77) | - | ||
| Depression * Pain Location 4 | ||||
| CPOL: Depression (Yes vs. No) | 10.57 (2.21–50.49) | |||
| CLBP: Depression (Yes vs. No) | 1.04 (0.25–4.41) | |||
| Chi-square change (df, p-value) | 29.58 (2, <0.0001) | 5.48 (1, 0.019) | 0.74 (1, 0.390) | 5.00 (1, 0.025) |
| Nagelkerke R-square | 0.320 | 0.372 | 0.378 | 0.423 |
Note: OR = odds ratio; CI = confidence interval; 1 COMM = Chronic opioid Misuse Measure; 2 Brief Pain Inventory (Average Severity and interference: 0–10), odds ratio represents the change in odds of opioid misuse risk given one unit increase in average pain severity or interference; 3 CPOL: Chronic pain other location, CLBP: Chronic low back pain; 4 Wald Chi-square p = 0.031.