Table 2 Methods sequence

1. Sampling and preparation

 i. 10 practices selected

 ii. Practices paired by specialty and size

 iii. One practice from each pair randomly assigned to CMgt condition

 iv. Interview guide constructed

2. Data collection: baseline (pre-intervention)

 i. Semi-structured interviews with care managers and practice members

 ii. Observation (30–120 min) during visit

 iii. Practice summary report generated after visit

 iv. Summary report member checked

 v. Interviews transcribed, cleaned and formatted in Atlas.ti

3. Data collection: interim

 i. Each care manager interviewed three times between baseline and follow-up

4. Data collection: follow-up

 i. Same process as Baseline data collection (9 months post intervention start)

 ii. Outcomes data collected for RE-AIM (16 months post intervention start)

5. Analysis: macrocognition

 i. Development of macrocognition-coding guide (a priori categories)

 ii. Initial coding by team members, coding calibration, and then coding completion

 iii. Quotation outputs generated by practice by code

 iv. Independent evidence table constructed over several team meetings

 v. Team met to reconcile all evidence tables and themes

 vi. Team members independently rated practices on how well and often they engaged in each macrocognition process

 vii. Team members independently assigned each practice an overall implementation score

 viii. Team met to reconcile macrocognition and implementation scores

6. Analysis: RE-AIM

 i. Data (quantitative) for reach, effectiveness, adoption and maintenance of RE-AIM analyzed by practice

 ii. Data for implementation part of RE-AIM created by independent ratings and reconciled by qualitative team members

7. Analysis: Outcomes

 i. Overall themes related to use of macrocognition processes

 ii. Care management implementation success (RE-AIM) by practice

 iii. Use of macrocognition processes by practice

 iv. Comparing implementation success with use of macrocognition processes by practice