TABLE 5.
Multiple Logistic Regression Model for Prediction of Treatment Statusa
| Variableb | B | SE | OR | 95% CI for OR |
|
|---|---|---|---|---|---|
| Lower | Upper | ||||
| Age | −0.02 | 0.01 | 0.98 | 0.96 | 1.01 |
| Age of onset** | −0.09 | 0.02 | 0.92 | 0.88 | 0.95 |
| Total Medical Comorbidities | 0.02 | 0.09 | 1.02 | 0.86 | 1.22 |
| Total Psychiatric Comorbidities** | 0.16 | 0.06 | 1.17 | 1.05 | 1.31 |
| Checking | −0.06 | 0.04 | 0.95 | 0.95 | 1.02 |
| Hoarding | 0.01 | 0.04 | 1.01 | 0.94 | 1.09 |
| Neutralizing | 0.04 | 0.03 | 1.04 | 0.98 | 1.11 |
| Ordering | −0.01 | 0.05 | 0.99 | 0.90 | 1.09 |
| Washing | −0.01 | 0.03 | 0.99 | 0.93 | 1.06 |
| Obsessing** | 0.19 | 0.04 | 1.21 | 1.13 | 1.31 |
| Cannabis use frequency** | −0.06 | 0.02 | 0.94 | 0.90 | 0.98 |
| %Medicinal (vs. recreational) use | −0.00 | 0.00 | 1.00 | 0.99 | 1.00 |
*
p < .05
**=
p < .005
a
Treatment status coded as a binary outcome (0=not currently in treatment, 1=currently in treatment)
b
Variables with p<.1 on independent t-tests were forwarded stepwise into the logistic regression model