Table 2.
The Relationship between Predisposing, Enabling, and Illness Variables and Linkage with Primary Medical Care over Two Years for Adults with Addictions (n=392*)
| Hazard Ratio (Adjusted) (95% Confidence Interval)** | Predicted Probability of Linkage at Twelve Months§ (Unadjusted for Other Factors) | ||
|---|---|---|---|
| Older age (years) | 1.01 (0.99–1.03) | ≥35 years | 0.48 |
| <35 years | 0.41 | ||
| Female sex | 1.67 (1.23–2.28) | Female | 0.52 |
| Male | 0.42 | ||
| Minority race/ethnicity | 1.32 (0.98–1.78) | Minority | 0.48 |
| White | 0.39 | ||
| Married | 1.49 (0.95–2.36) | Married | 0.51 |
| Not married | 0.44 | ||
| Incarceration (past five years) | 0.73 (0.56–0.96) | Incarceration | 0.38 |
| No incarceration | 0.52 | ||
| Family or friends support abstinence | 1.60 (1.18–2.18) | Support | 0.49 |
| Do not support | 0.36 | ||
| Medical visit (episodic, past six months)++ | 1.48 (1.10–1.97) | One or more | 0.53 |
| None | 0.42 | ||
| Smoking (current) | 1.41 (0.92–2.14) | Smoker | 0.47 |
| Nonsmoker | 0.30 | ||
| Substance problem recognition*** | 1.04 (0.99–1.10) | ≥34 | 0.44 |
| <34 | 0.46 | ||
| Belief that medical treatment is important | 1.12 (0.81–1.56) | Important | 0.46 |
| Not important | 0.42 | ||
| SF-36 PCS∧ | 1.00 (0.98–1.01) | ≥48.6 | 0.41 |
| <48.6 | 0.49 | ||
| CES-D score+ | 1.01 (0.99–1.02) | >16 | 0.44 |
| ≤16 | 0.47 | ||
| Suicide attempt ever | 0.95 (0.67–1.35) | Ever | 0.49 |
| Never | 0.44 | ||
| Psychiatric medication ever | 1.14 (0.81–1.60) | Ever prescribed | 0.48 |
| Never prescribed | 0.44 | ||
| Insurance during follow-up | 1.63 (1.19–2.22) | Insurance | 0.54†† |
| No insurance | 0.41 | ||
| Randomization to enhanced primary care referral | 1.78 (1.36–2.34) | Enhanced referral | 0.52 |
| Usual care referral | 0.37 | ||
All variables in the model were collected at the baseline research interview except insurance during follow-up, which was assessed at 6, 12, 18, and 24 months for each preceding 6-month period and entered as a time-variable predictor (see text).
Cut-point for continuous variables was the median value, except for the CES- D for which the clinically meaningful 16 or greater cutoff was used. See text for predicted probability of linkage by insurance during follow-up.
Eight of four hundred subjects with follow-up in the study had missing data on one or more of the variables listed and are therefore excluded from these proportional hazards regression analyses.
Adjusted for all other variables listed in the table.
Recognition subscale of the SOCRATES (Stage of Change Readiness and Treatment Eagerness Scale).
SF-36 PCS=Short Form Health Survey Physical Component Summary.
CES-D=Center for Epidemiologic Studies Depression.
At least one visit to a medical clinic or physician, not primary care, in the past six months.
Based on the Kaplan-Meier product limit estimator.
Predicted probability of linkage at six months associated with insurance during that six-month follow-up period.