. Author manuscript; available in PMC: 2018 Sep 1.

Published in final edited form as: Med Care. 2017 Sep;55(Suppl 9 2):S43–S49. doi: 10.1097/MLR.0000000000000746

Table 4.

Unadjusted and adjusted models for associations between perceived race-based discrimination and contraceptive use

Covariates	Any prescription vs. non-prescription/no method³	Multinomial regression model^b
	Any prescription vs. non-prescription/no method³	Hormonal vs. non-prescription/no method	Sterilization vs. non-prescription/no method	IUD or implants vs. non-prescription/no method

	OR (95%CI)	RRR (95%CI)	RRR (95%CI)	RRR (95%CI)
UNADJUSTED MODELS
Perceived race-based discrimination	0.57 (0.38, 0.87)^d	0.52 (0.30, 0.92)^f	0.81 (0.50, 1.33)	0.35 (0.18, 0.67)^g

ADJUSTED MODELS^c
Perceived race-based discrimination	0.65 (0.42, 1.00)^e	0.63 (0.35, 1.12)	0.92 (0.54, 1.57)	0.40 (0.20, 0.79)^h
Race/Ethnicity
NH Black vs. NH White	0.71 (0.53, 0.96)	0.76 (0.53, 1.11)	0.63 (0.43, 0.92)	0.74 (0.51, 1.09)
Hispanic vs. NH White	0.55 (0.38, 0.80)	0.65 (0.41, 1.02)	0.49 (0.30, 0.80)	0.48 (0.29, 0.80)
NH Other vs. NH White	0.65 (0.40, 1.05)	0.57 (0.30, 1.10)	0.56 (0.30, 1.04)	0.87 (0.48, 1.58)
Age	0.99 (0.97, 1.02)	0.95 (0.92, 0.98)	1.10 (1.06, 1.13)	0.95 (0.92, 0.98)
Single/DSW	0.62 (0.47, 0.82)	0.80 (0.57, 1.12)	0.50 (0.35, 0.71)	0.59 (0.41, 0.84)
Parous	1.63 (1.23, 2.16)	0.89 (0.64, 1.24)	3.70 (2.45, 5.60)	1.92 (1.34, 2.75)
Income <$40,000	1.13 (0.85, 1.50)	1.17 (0.83, 1.66)	1.06 (0.75, 1.51)	1.15 (0.81, 1.65)
Non VA insurance	1.23 (0.94, 1.60)	1.29 (0.94, 1.78)	1.14 (0.82, 1.59)	1.27 (0.91, 1.78)

Vs., versus; OR, odds ratio; CI, confidence interval; RRR, relative-risk ratio; NH, non-Hispanic; DSW, divorced, separated, widowed; VA, Veterans Affairs.

N = 1,325 due to missing data. Estimates in bold statistically significant at p-value<0.05.

Logistic regression model with the outcome defined as prescription method vs. non-prescription/no method.

Multinomial regression model with 4 levels of contraceptive use: non-prescription/no method, hormonal, sterilization, IUD or implants. Non-prescription/no method was used as the reference group.

Adjusted for race, age, marital status, parous, income and insurance. Patient covariates were selected for inclusion in adjusted models based on whether they were associated with perceived race-based discrimination or either contraceptive outcome in bivariate analyses at the p<0.15 level.

P-value = 0.008;

P-value = 0.048;

P-value = 0.02;

P-value = 0.002;

P-value = 0.008.