Skip to main content
. Author manuscript; available in PMC: 2019 Dec 1.
Published in final edited form as: Psychometrika. 2018 Apr 2;83(4):991–1006. doi: 10.1007/s11336-018-9612-2

Table 5.

Summary of mediation/confounding effect estimations for ethnic disparity in depression PROMIS score.

Mediator/confounder Linear models Nonparametric models
IE (95% CI) RE (%) IE (95% CI) RE (%)
Days from diagnosis 0.01 (−0.02, 0.04) −0.4 (−1.7. 1.0) −0.01 (−0.04,0.02) 0.2 (−1.0, 1.7)
Income −0.24 (−0.6, 0.13) 9.1 (−5.5, 23.7) −0.15 (−0.29, −0.04) 6.2 (1.6, 13.8)
Education −0.34 (−0.58, −0.10) 13.6(1.7, 25.5) −0.51 (−0.77, −0.27) 21.5 (10.6, 39.6)
Age at diagnosis −0.35 (−0.53,−0.18) 14.1 (4.1, 24.2) −0.34 (−0.50, −0.19) 16.7 (9.2, 24.2)
Social support −0.52 (−1.02, −0.03) 19.5 (1.4, 37.6) −0.71 (−1.12, −0.24) 29.1 (13.5, 46.6)
Spirituality 0.74 (0.31, 1.17) −31.4 (−68, −8.7) 0.72 (0.39, 1.08) −31.9 (−68.6, −12.5)
Employment 0.04 (−0.09, 0.17) −1.5 (−7.1, 4.1) 0.00 (−0.02, 0.02) 0.0 (−1.0, 0.9)
US born −.62 (−1.16, −0.09) 25.3 (−3.3, 54) −0.38 (−0.76, −0.07) 15.9 (2.8, 36.2)
Chemotherapy −0.0 (−0.10, 0.06) 0.9 (−2.3, 4.1) 0.00 (−0.01,0.03) −0.2 (−1.0, 0.4)
Insurance 0.01 (−0.25, 0.23) 0.3 (−9.4, 10.0) −0.04 (−0.15,0.03) 1.7 (−1.5, 6.6)
Joint effect −0.21 (−0.66, 0.24) 7.7 (−9.9, 25.3) −0.18 (−0.34, −0.04) 7.5 (1.6, 16.4)
Direct effect −1.33 (−2.32, −0.33) 50.2(16.6, 83.9) −0.84 (−1.46, −0.28) 34.4 (15.9, 55.3)
Total effect −1.33 (−1.44, −3.76) −2.45 (−3.40, −1.42)

IE indirect effect, RE relative effect. “joint effect” refers to the joint indirect effect from employment, insurance, and income. Non-Hispanic whites are the reference group.