. Author manuscript; available in PMC: 2016 Jul 1.

Published in final edited form as: Acad Radiol. 2015 Apr 4;22(7):846–852. doi: 10.1016/j.acra.2015.02.011

Table 4.

Availability of a Combination of Imaging and Imaging-Guided Biopsy Services by Breast Cancer Surveillance Consortium Facility Characteristic

		Both any ultrasound and any MRI	Both screening ultrasound and screening MRI	Any ultrasound and any MRI and guided biopsy services
Number of facilities providing non-missing data on use of this technology		102	102	101
Facility Characteristics	N
Profit / Not for profit status
For profit	23	11 (47.8%; 29.2, 67.1)^a	2 (8.7%; 1.8, 32.7)	4 (18.2%; 5.5, 46.1)
Not for profit	59	22 (37.3%; 25.0, 51.5)	6 (10.2%; 3.0, 29.6)	20 (33.9%; 23.3, 46.4)
Unknown	20	9 (45.0%; 25.0, 66.7)	2 (10.0%; 2.2, 35.1)	9 (45.0%; 25.0, 66.7)
P value^*		0.46	0.89	0.32
Academic medical center status
Yes	8	6 (75.0%; 31.6, 95.1)	2 (25.0%; 8.2, 55.6)	6 (75.0%; 31.6, 95.1)
No	94	36 (38.3%; 29.1, 48.4)^b	8 (8.5%; 3.9, 17.6)	27 (29.0%; 22.4, 36.7)
P value		0.11	0.022	0.041
Facility type
Multi-specialty breast center	25	19 (76.0%; 45.9, 92.2)	7 (28.0%; 12.7, 50.9)	19 (76.0%; 45.9, 92.2)
Full diagnostic radiology practice	64	23 (35.9%; 24.9, 48.7)^c	3 (4.7%; 1.6, 12.9)	14 (22.2%; 13.7, 33.9)
Breast imaging only	4	0/4 (0%)	0/4 (0%)	0/4 (0%)
Non-radiology practice	9	0/9 (0%)	0/9 (0%)	0/9 (0%)
P value^†		0.023	0.0009	0.0047

Results shown here are the number of facilities that reported providing the imaging service in question (columns) within each category of the facility characteristic (row). Percentages are based on observed frequencies divided by the denominators shown in the column marked “N”, except for the following columns:

Denominator is 22 for “for profit” facilities offering both any ultrasound and any MRI
Denominator is 93 for non-academic medical centers offering both any ultrasound and any MRI
Denominator is 63 for full diagnostic radiology practices offering both any ultrasound and any MRI

The 95% confidence intervals are derived via delta method from post-estimation predicted probability estimates obtained from unadjusted GEE models of the outcome regressed on the corresponding facility characteristic. Models accommodate the non-independence of facilities belonging to the same practice.

All p values are from joint Wald tests of parameters estimated by GEE modeling for each descriptive variable.

P-values shown for profit status were obtained from models restricted to facilities with non-missing profit status information.

^†

P-values shown for facility type were obtained from models restricted to multi-specialty breast centers and full diagnostic radiology practices, except in the case of the provision of any ultrasound, for which modeling was restricted to full diagnostic, breast imaging only, and non-radiology practices.