Multinomial Logistic Regression on Contraception Use
| LARC OR | 95%CI | P-value | Non-LARC OR | 95%CI | P-value | |
|---|---|---|---|---|---|---|
| Insurance/Race | ||||||
| Commercial/non-Hispanic White | -- | -- | -- | -- | -- | -- |
| Commercial/non-Hispanic Black | 0.97 | (0.93-1.02) | .575 | 1.00 | (0.96-1.03) | .887 |
| Commercial/Hispanic | 1.19 | (1.10-1.29) | .030 | 0.93 | (0.87-0.99) | .279 |
| Commercial/Other race | 0.43 | (0.40-0.46) | < .001 | 0.58 | (0.55-0.61) | < .001 |
| Medicaid/non-Hispanic White | 1.05 | (0.95-1.17) | .636 | 1.02 | (0.94-1.10) | 0.835 |
| Medicaid/non-Hispanic Black | 1.25 | (1.19-1.32) | < .001 | 1.57 | (1.52-1.63) | < .001 |
| Medicaid/Hispanic | 2.23 | (2.07-2.41) | < .001 | 1.26 | (1.17-1.36) | .001 |
| Medicaid/Other race | 1.78 | (1.61-1.97) | < .001 | 1.13 | (1.02-1.25) | .215 |
| Self-pay and Other insurance/non-Hispanic White | 0.70 | (0.65-0.76) | < .001 | 0.71 | (0.66-0.75) | < .001 |
| Self-pay and Other insurance/non-Hispanic Black | 0.94 | (0.89-1.00) | .346 | 1.02 | (0.98-1.07) | .646 |
| Self-pay and Other insurance/Hispanic | 1.38 | (1.30-1.47) | < .001 | 1.80 | (1.72-1.88) | < .001 |
| Self-pay and Other insurance/Other race | 0.45 | (0.38-0.52) | < .001 | 0.70 | (0.63-0.77) | < .001 |
| Age | 0.97 | (0.96-0.97) | < 0.001 | 0.93 | (0.93-0.94) | < .001 |
| Married | 1.44 | (1.38-1.51) | < 0.001 | 0.99 | (0.95-1.03) | .834 |
| Area-level socioeconomic status | ||||||
| Median household income in thousands | 1.00 | (0.99-1.00) | 0.001 | 1.00 | (1.00-1.00) | .023 |
| Percentage of patients with poverty | 1.00 | (1.00-1.00) | 0.392 | 1.01 | (1.00-1.01) | < .001 |
| Percentage of patients with bachelor’s degree | 1.00 | (1.00-1.00) | 0.906 | 1.00 | (1.00-1.00) | .053 |
| Unemployment rate | 1.02 | (1.01-1.02) | 0.009 | 1.00 | (0.99-1.00) | .908 |
Note. No contraception, ORs show the odds of using LRAC/Non-LARC for difference insurances and race/ethnicity groups compared to the non-Hispanic White women with commercial insurance. Other controlled variables: age and area-level proxy including median household income, percentage of patients with poverty, percentage of patients with bachelor’s degree, and unemployment rate.