Table 4 Representative examples of data that were used in building the case studies. In this table, the way in which raw data was indexed to the components of the program model is shown. Where relevant, how this data was linked to the key elements of the CAFF model is also shown in order to demonstrate the way in which inferences were drawn between the original raw data, the program model, and, ultimately, the framework

Facilitation:

Facilitator encouraged and created space for these change cues to be raised and discussed by group.

Physician engagement/change talk/change planning [8]

A high degree of social interaction in case 1. Data captured from original AGFS transcripts

Reflecting and sharing of practices around using intravenous anesthesia:

“I find that the distinction between deep and awake has become blurred with TIVA. I just turn down the TIVA and just extubate them and they’re almost awake but they’re not as deep as for gas extubation. I usually call it deep but it’s not really a deep extubation in the truest sense… it kind of is halfway in between.”

A change cue from another participant:

“Our biggest problem is post-op pain”

This change cue led to a discussion between the participants about strategies for improving pain management without worsening nausea and vomiting.

Facilitation:

Facilitator encouraged and created space for this discussion to occur.

Physician engagement/change talk/change planning [8]

Case 2A

Data captured from original AGFS transcripts

A participant expressed their desire to improve their prescribing habits:

“…Let us say 3 calls that I get at night are requests, this patient wants a sleeping pill. I might be successful in 1 in 3 in convincing the patient that no they do not really need this. Or convincing the nurse, no they do not really need this. Maybe I’ll be successful with 1 on 3. But its’ worth trying.”

This change cue led to a discussion around the need for the group to develop consensus on de-prescribing the compounds in question.

Recipients and context

Case 2D:

The physician group in this hospital was young and they believed their group make-up and relationships with consultants at that site influenced their prescribing habits.

Data from original AGFS transcripts

“We are a relatively younger group compared to some... And I think a lot of us try to utilize… restraint in antipsychotic prescriptions. We have had good teams involved with a lot of our patients and we try and avoid a lot of that.”

Data representation

Style of report

Case 2D: Participants raised concerns about the complexity and cognitive load in their AF reports.

Data from original AGFS transcripts

“But when you first look at the document… it is very difficult to understand a lot of it, because there’s a lot of information.”

Examples of data extracted from interviews with CPLP Staff

Representative quotations

Relationship building and question choice

Recipients, innovation, and context

When asked about how the project originated and nature of the innovation (case 1):

CPLP facilitator: “Many of these docs had been involved in [our first] project, but there were some younger recent grads who were more involved in choosing questions—NAME was seen as a strong leader, group had bought into his vision of education in general and the idea of audit and feedback. Maybe the group generally has a culture of monitoring and data collection—… not a big sell to this group – they were keen, thoughtful, interested, and academic in their thinking….they had the advantage of having already been through the process. Lots of groundwork done by our team working with the clinic staff, daycare staff, data analyst, etc... lots of face time from PLP on site—seemed to be helpful with buy in.”

Relationship Building & Question Choice

(this group was not directly involved in co-creation of question)

Physician Engagement/Change Talk/Change Planning [8]

When asked about group dynamic (case 2D):

Facilitator: “Poor participation from the group. The group culture at this hospital seemed different from the other groups…very silent during the session…not very many showed up..for their reports”. Project manager 1: “[The participants were] young. Did they feel uncomfortable speaking up?”

Question choice

Innovation

When asked about nature of the project case 3:

Project manager 2: “The goal was hazy…very broad—things that were common”.

Relationship building

Recipients and context

Describing the recipients (case 3):

Facilitator: “This group is super engaged. A bunch of XXX residents who may have been friends before, or became friends afterwards—they are all about the same age, early millenials mostly all on the same page. The group was on the same page. NAME is a competent, quiet leader … very practical, down to earth, that’s why NAME was asked to do this. The group is tight because they cover for one another and cross-cover. The mean age of this group was particularly young – sense of buy in to data, computers, digital natives. Had many suggestions …came up with at least ten potential projects. They were keen – lots of sharing of one another’s data in the sessions”.