Feasibility and impact of implementing a private care system’s diabetes quality improvement intervention in the safety net: a cluster-randomized trial

Table 3 Results of segmented regression analyses, early implementation effects (Controls = late implementation clinics)

	Estimate	Standard Error	p value
Percent actively prescribed statin and ACE/ARB, among patients indicated for both
Difference in slope of the trend between control and intervention groups prior to the intervention	0.040	0.1208	0.744
Difference between control and intervention groups in change in level following the intervention	0.398	0.9742	0.685
Difference between control and intervention groups in change in slope of the trend from pre- to post-intervention	1.102	0.1706	<0.001
Percent actively prescribed statins, among patients indicated for a statin only
Difference in slope of the trend between control and intervention groups prior to the intervention	0.103	0.241	0.673
Difference between control and intervention groups in change in level following the intervention	4.225	1.987	0.040
Difference between control and intervention groups in change in slope of the trend from pre- to post-intervention	0.491	0.318	0.131

Full models included a constant term, a term to model the pre-intervention linear trend slope, a variable for study group (intervention vs. control), terms to estimate change in level from pre- to post-intervention and change in slope of the trend, interaction terms (group by pre-intervention trend, group by change in level and group by change in trend), percent female, percent aged 18–39, gender by age interaction, and a first-order autoregressive parameter

ISSN: 1748-5908