Toward High-Value, Cost-Conscious Care – Supporting Future Doctors to Adopt a Role as Stewards of a Sustainable Healthcare System

Figures & data

Figure 1. Illustration of the Realist Evaluation approach in data collection and data analysis.

Table 1. An overview of HVCCC carriers and building blocks.

Carrier I – Continue to promote HVCCC awareness	1. Build a shared understanding of HVCCC 2. Increase knowledge about costs 3. Provide quantitative feedback 4. Promote reflective decision-making
Carrier II – Create an institutional structure that fosters HVCCC	1. Install HVCCC ambassadors 2. Promote open discussion about HVCCC
Carrier III – Continue the focus on projects for embedding HVCCC in practice	1. Facilitate motivated residents to develop HVCCC projects 2. Provide the principles of project-based working 3. Build a HVCCC platform
Carrier IV – Generate evidence	1. Identify the contexts and mechanisms that influence project outcomes 2. Identify quantifiable variables 3. Generate outcomescenarios to indicate potential HVCCC project

Table 2. Contexts and mechanisms that lead to a lack of awareness and deliberate decision-making.

Context	Mechanism	Outcome
People train to become doctors because they want to help people. During their training, residents are continuously reminded that they bear final responsibility for their patients’ care.	Externally imposed (financial) constraints in medical decision-making conflict with the physicians’ duties (and their professional autonomy) to deliver the care that their patients need.	Positioning HVCCC as a means to reduce costs will only lead to resistance amongst physicians (in training).
Physicians (in training) work in a setting of high working pressure and they have to meet production targets.	It is difficult to take a step back to reflect on existing processes of decision-making. Many decisions are based on routine: choices based on “what we usually do.”	Physicians (in training) do not sufficiently consider the medical necessity of tests and the consequences of unnecessary testing.
Medical residents can initially be overwhelmed at the start of their residency. “From one day to the next, you suddenly have two hundred patients that are added to your caseload.”	Medical residents’ minds are overloaded with getting through their caseload each day, leaving little room to consider: “what is actually stated in the guideline?” or “was this additional test really necessary?”
Health costs are not addressed as part of the medical curricula.	Physicians (in training) are largely unaware of the costs of the care that they prescribe on a daily basis. There is no sense of urgency to deliver HVCCC.	Doctors may unknowingly choosing expensive interventions over cheaper, just as effective alternatives.

The CMO configuration illustrates how specific contextual factors fire specific lines of reasoning (mechanisms) that guide behaviors of (some of the) medical residents. The outcome is an expression of the interactions between contexts and mechanisms.

Table 3. Contexts and mechanisms that lead to a lack of room for HVCCC in the information educational system. ¹

Context	Mechanism	Outcome
There is a hierarchical structure within hospitals. Even though hierarchy used to be much more apparent back in the day, it still exists today.	Attendings say that their door is always open to residents. However, residents may hold themselves back from approaching their attending for questions about current working methods, “why is this organized like this?”	Contextual factors including hierarchy, dependency and lack of supervisor-initiated HVCCC discussions fire mechanisms that explain why it is less likely that residents bring up HVCCC in conversations with their attendings. As a result, it is unlikely that residents learn how to improve HVCCC in daily practice or set up HVCCC projects.
Medical residents work in different organizations during their residency.	Residents are temporary employees and therefore often considered outsiders.
Residents are dependent; they rely on their supervisors’ assessment.	Some interviewed residents indicated that they feel that they need to be able to execute medical procedures, such as ordering lab tests, independently. Because of this paradigm, residents tend to ask questions to fellow residents instead of their supervisor, because they do not want to come across as someone who cannot figure these “basic medical procedures” out themselves. In addition, residents want their supervisor to like them. They do not want to be known as the resident who asks a million questions all the time.
There are several interaction moments between the residents and supervisor, such as grand-rounds discussion, multidisciplinary deliberation (MDO) and shift transfers, during which patient cases are discussed and residents can ask questions. HVCCC is currently not embedded in these interaction moments. Supervisors do not encourage HVCCC reflection on medical decision-making.	Because residents are inexperienced, they tend to order an excessive amount of tests to create certainty. Some residents do take the initiative to ask questions about HVCCC to their attendings. Other residents think, ‘Oh, but that is the attending’ and do not ask questions that could help them to enhance HVCCC.

Table 4. Contexts and mechanisms that lead to tensions between high work demands and HVCCC innovation projects.

Context	Mechanism	Outcome
The current working and training environment is densely packed, with fulltime residents working at least 50 hours a week. Residency training contains an increasing number of subjects, explained by the continuous advancement of medicine. Meanwhile, the duration of residency is shortened, which implies that more subjects need to be acquired in a shorter time frame.	HVCCC project have to compete with other educational commitments.	Motivated residents cannot (or can barely) find the time to act on their ideas to promote HVCCC. The residents that do work on a HVCCC project give up their spare time to run the project and/or implement projects mostly in outpatient clinic settings that have relatively less demanding work schedules.
Residents, and doctors in general, are “doers.” Most residents have no prior experience with project-based working.	Residents have a tendency to start a project without analyzing the underlying problem first. Several residents described that they never fully grasped the goal of the project or that actions were not well-aligned with the intended end-goals.	Energy may be wasted on a project that does not promote HVCCC in the end.
	The project process is generally characterized by a trial and error approach with residents reactively seeking solutions for encountered problems.	A trial and error approach leads to significant delay, often caused by negative responses to the project such as resistance from stakeholders.
Hospitals, which have grown considerably in size due to mergers, have complex organizational structures.	Residents are unaware of the different functions that are represented in the hospital.	Residents often overlook the importance of locating and involving stakeholders in order to get their project off the ground. Failure to include important stakeholders may trigger negative responses to the project due to resistance from stakeholders, risking delay or discontinuance of the HVCCC project.

Table A1. Example context-mechanism-outcome configuration.

Context	Mechanism	Outcome
Due to time constraints, lab testing is rarely discussed during shift transfers and other interactions between supervisors and medical residents.	Although supervisors can be approached for questions, residents hold themselves back from discussing the necessity of certain lab orders. Medical residents often think that they should be able to figure out the “basics” themselves and not discuss it during supervision rounds. “We should be able to order lab tests ourselves”	Lab testing can be more efficient. Too many tests are ordered that are not medically necessary.
Corresponding interview quotes (interviewee 2)
“Especially in regional teaching hospitals, you have a time issue (..) Sometimes you have only 5 minutes [for supervision] in the afternoon. Well, then you have to discuss the main things, [lab tests] you will have to figure out yourself”	“Actually, it’s a bit of a taboo to discuss the ‘basic stuff’, such as lab tests, with your supervisor (..) if you choose to discuss this, you discuss it with a fellow resident instead of the internist (..) It depends on how much effort you want to put into it yourself [as a resident] to discuss this with your supervisor. The opportunity is there (..) but I would hold myself back when supervision has already taken place and I forgot to ask something about lab testing. I would not call my supervisor only for that (..) in theory [you can ask supervisors anything]. Whether they will appreciate [all these questions], is another thing…	“We often see recurrent CRP test orders that have no clinical follow-up, [another example is] the ordering of standard combinations of lab tests that we repeat too often, when actually.. they are not really necessary. We order these tests out of habit, rather than having a real good reason for it”

Table A1 depicts a CMO configuration based on one interview transcript related to inefficient lab testing (the problem that this respondent aimed to address by means of the HVCCC project). The third row illustrates how interview quotes were summarized in the context-mechanism-outcome configurations, depicted in the second row.

Data sharing statement

Due to the nature of this research, participants of this study did not agree for their data to be shared publicly, so supporting data are not available.