Table 2. Mediation Analysis Estimations for Racial Disparity in Prostate MRI Receipt.
| Mediator | Log odds (95% CI) | Relative effects (95% CI) | ||
|---|---|---|---|---|
| Comparison of race identified as Black vs White a | ||||
| Direct effects | −0.07 (−0.15 to −0.01) | 0.19 (0.02 to 0.36) | ||
| Indirect effects | −0.28 (−0.33 to −0.22) | 0.81 (0.64 to 0.98) | ||
| Dual Medicaid eligibility | −0.04 (−0.05 to −0.03) | 0.11 (0.07 to 0.16) | ||
| High-poverty area | −0.07 (−0.1 to −0.04) | 0.19 (0.11 to 0.28) | ||
| Regional use of prostate cancer imaging | 0.001 (0 to 0.003) | −0.003 (−0.008 to 0.001) | ||
| SEER registry | −0.08 (−0.12 to −0.05) | 0.24 (0.14 to 0.32) | ||
| ICE group | −0.07 (−0.11 to −0.03) | 0.19 (0.10 to 0.29) | ||
| Total | −0.34 (−0.43 to −0.27) | NA | ||
| Comparison of race identified as Black vs other a , b | ||||
| Direct effects | −0.03 (−0.08 to 0.006) | 0.08 (−0.02 to 0.26) | ||
| Indirect effects | −0.32 (−0.41 to −0.22) | 0.92 (0.74 to 1.02) | ||
| High-poverty area | −0.05 (−0.098 to −0.014) | 0.15 (0.04 to 0.28) | ||
| Regional use of prostate cancer imaging | −0.002 (−0.007 to 0) | 0.006 (0 to 0.018) | ||
| SEER registry | −0.19 (−0.28 to −0.10) | 0.54 (0.34 to 0.74) | ||
| ICE group | −0.07 (−0.13 to −0.04) | 0.22 (0.10 to 0.36) | ||
| Total | −0.35 (−0.45 to −0.25) | NA | ||
Abbreviations: ICE, Index of Concentration at the Extreme; MRI, magnetic resonance imaging; NA, not applicable; SEER, Surveillance Epidemiology, and End Results.
a
Models adjusted for modified Cancer of the Prostate Risk Assessment score.
b
Dual Medicaid eligibility was not an eligible mediator for this comparison.