Table 2.

Interrupted Time Series Model Estimates for Difference in Colorectal Cancer Screening Completion by Educational Attainment

	% Difference*	p
Primary analysis: All clinics
Baseline level	9.6	<0 .001
Trend prior to TopCare	−0.04	0.002
TopCare effect	−0.68	< 0.001
Change in trend after TopCare	0.0008	0.48
Stratified analysis: All clinics except community health center with pre-existing patient navigation
Baseline level	8.93	< 0.001
Trend prior to TopCare	−0.01	0.20
TopCare effect	−0.69	< 0.001
Change in trend after TopCare	0.0003	0.77
Stratified analysis: Community health center with pre-existing patient navigation
Baseline level	4.99	< 0.001
Trend prior to TopCare	−0.18	< 0.001
TopCare effect	0.16	0.74
Change in trend after TopCare	0.004	0.08

*Model estimate represents difference in % of eligible patients who completed colorectal cancer screening in high minus low educational attainment group (negative values represent decreasing disparity)

All models also account for autocorrelation