Table 1.

Theoretical constructs, measures, data sources, and data collection timing

Construct	Measures	Source	Timing
Implementation support	Frequency, duration, and mode of PF contacts	PF contact logs	I
	Frequency, duration, and mode of academic detailing contact	Contact, webinar logs	I
	Attendance at regional collaborative meetings	Attendance logs	I
Practice capacity for QI	Change Process Capacity Questionnaire	KI survey	B, E, F
	Adaptive Reserve Questionnaire	Provider/staff survey	B, E, F
Organizational readiness	Organizational Readiness for Change Questionnaire (ORIC)	Provider/staff survey	B
Implementation Policies and Practices (IPPs)	Key Driver Implementation Scales	PF Ratings	I
	Implementation barriers, facilitators, and IPPs	KI interview	B, E
Implementation climate	Implementation climate questionnaire	Provider/staff survey	E, F
Implementation effectiveness	ABCS measures/clinical measures	HIE	B, I, E, F
	Acceptability of implementation support	KI interview	E
Innovation effectiveness	Patient outcomes (communication, shared decision-making)	Patient survey	B, E
	Patient outcomes (healthcare utilization and mortality)	Claims data	B, E, F
	Practice outcomes (e.g., financial benefits)	KI Interview	E, F
Inner context	Practice characteristics, patient population, EMR capabilities	KI survey, PF contact logs	B, E,F
Outer context	External policies and incentives, market conditions	KI survey, KI interview, PF contact logs	B, E, F

Because readiness is conceived as an organization-level construct, we will test whether sufficient inter-rater reliability and inter-rater agreement exist to aggregate individual responses to the practice level [22–26]. If tests do not justify aggregation, we will use a measure of intra-practice variability in readiness rather than a practice-level mean in our analysis [15, 16].

Source: PF practice facilitator, KI key informant; timing: B baseline, I intervention, E end of intervention, F 6 and 12 months post-intervention