Table 3.

Multinomial logistic regression analyses predicting mammography at follow-up with cancer-related information seeking from nonmedical and medical sources.

Sources of cancer-related information seeking	Adherence to mammography (Once in the past 12 months vs. None)
	Unadjusted RRR	95% CI	Adjusted RRR	95% CI
Nonmedical	0.90	0.48 – 1.68	1.14	0.38 – 3.45
Medical	0.70	0.29 – 1.70	0.40	0.11 – 1.54
	Overutilization of mammography (Two or more times in the past 12 months vs. None)
	Unadjusted RRR	95% CI	Adjusted RRR	95% CI
Nonmedical	0.86	0.46 – 1.68	0.89	0.29 – 2.78
Medical	0.82	0.33 – 2.01	0.50	0.12 – 2.01
	Overutilization of mammography (Two or more times in the past 12 months vs. Once)
	Unadjusted RRR	95% CI	Adjusted RRR	95% CI
Nonmedical	0.96	0.73 – 1.25	0.78	0.53 – 1.15
Medical	1.17	0.81 – 1.68	1.24	0.74 – 2.05

Notes. RRR = Relative risk ratio; 95% CI = 95% confidence interval. Adjusted RRRs controlled for baseline confounder variables (age, education, ethnicity, marital status, concern about recurrence, cancer stage, health status, surgery, radiation therapy, systemic therapy, following doctors’ recommendations for tests, and Lerman Cancer Worry scale).