TABLE 3.

—Self-Reported Mammography Use In Past 2 Years From Baseline (1993–1994) to Follow-Up (1997–1997): Black Women in the North Carolina Breast Cancer Screening Program Intervention Counties

				Difference of Differences,a %
Exposure	Baseline, %	Follow-Up, %	Increase, %	Estimate	95% CI	P
Awareness
Yes (n = 142)	49	72	23	9	(–2, 21)	.10
No (n = 244)	36	50	14
Any advice
Yes (n = 55)	49	76	27	12	(–3, 27)	.13
No (n = 332)	39	55	16
Advice from lay health advisor
Yes (n = 95)	46	72	26	11	(–2, 23)	.11
No (n = 292)	39	54	15
1 or more program materials
Yes (n = 232)	45	65	20	5	(–6, 16)	.33
No (n = 155)	34	48	14
Do the Right Thing logo
Yes (n = 113)	42	67	25	12	(0, 24)	.05
No (n = 274)	41	54	13
Any phantom material
Yes (n = 88)	49	68	19	2	(–11, 15)	.76
No (n = 299)	38	55	17

Note. See text for complete descriptions of indicators. As a result of rounding, estimates may not sum in an exact manner.

^aUnadjusted estimates and confidence intervals (CIs) obtained from linear risk models with generalized estimating equations.