Table 3.
Results of logistic regression analysis of the associations of HIV stigma, substance use stigma with primary and secondary outcomes.
| Unadjusted logistic regression | Fully adjusted regression propensity score logistic regression* | |||||
|---|---|---|---|---|---|---|
| Crude OR (95% CI) | zvalue | p-Value | Adjusted OR (95% CI) | z-value | p-Value | |
| Primary outcomes | ||||||
| Poor access to general healthcare in past year | ||||||
| High HIV stigma (≥22) | 1.15 (0.63, 2.10) | 0.46 | 0.64 | 1.08 (0.70, 1.66) | 0.34 | 0.74 |
| High SU stigma (≥59) | 1.85 (1.02, 3.39) | 2.03 | 0.04 | 2.31 (1.50, 3.5 7) | 3.78 | <0.001 |
| Recent general outpatient care in past 3 months | ||||||
| High HIV stigma (≥22) | 1.02 (0.51, 2.02) | 0.06 | 0.95 | 0.72 (0.45, 1.14) | −1.39 | 0.16 |
| High SU stigma (≥59) | 0.59 (0.29, 1.16) | −1.51 | 0.13 | 0.66 (0.40, 1.08) | −1.65 | 0.10 |
| Secondary outcomes | ||||||
| Recent any inpatient care in past 3 months | ||||||
| High HIV stigma (≥22) | 2.07 (0.64, 7.23) | 1.20 | 0.23 | 0.32 (0.14, 0.65) | −3.02 | 0.003 |
| High SU stigma (≥59) | 4.02 (1.18, 18.35) | 2.06 | 0.04 | 1.42 (0.61, 3.37) | 0.81 | 0.42 |
| Recent ED visit in past 3 months | ||||||
| High HIV stigma (≥22) | 1.93 (0.41, 10.03) | 0.85 | 0.40 | 0.74 (0.22, 2.32) | −0.51 | 0.61 |
| High SU stigma (≥59) | 2.87 (0.6, 20.38) | 1.24 | 0.21 | 0.89 (0.31, 2.44) | −0.22 | 0.83 |
| Recent HIV care in past 3 months | ||||||
| High HIV stigma (≥22) | 0.71 (0.36, 1.40) | −0.99 | 0.32 | 0.73 (0.45, 1.19) | −1.27 | 0.20 |
| High SU stigma (≥59) | 1.08 (0.55, 2.10) | 0.22 | 0.83 | 0.95 (0.59, 1.54) | −0.21 | 0.84 |
| Recent substance use and/or mental health care in past 3 months | ||||||
| High HIV stigma (≥22) | 1.39 (0.71, 2.74) | 0.96 | 0.34 | 0.92 (0.58, 1.45) | −0.36 | 0.72 |
| High SU stigma (≥59) | 1.21 (0.62, 2.37) | 0.56 | 0.58 | 0.86 (0.54, 1.35) | −0.67 | 0.51 |
The comparison group are those with low HIV and low SU stigma scores, respectively.
Covariates were age, gender (female vs. male), time since HIV diagnosis, an education level (≤9th vs. >9th grade), employed regularly, depressive symptoms (CES-D≥16 vs. <16), social support ≥ median vs. <median), alcohol dependence (AUDIT, 1 year, yes vs. no), unhealthy alcohol use (past 30 days), recent substance use (past 30 days), HIV symptom index (past 4 weeks), pain interference (past 4 weeks), and transactional sex (past 12 months). The fully adjusted models for HIV stigma and substance use stigma included the respective other form of stigma as covariate.