There are many situations where a health professions educator may be interested in tracking inaction. These can be broken down as follows:

  1. The educator wants to track instances where a learner appropriately ignored a distractor.
  2. The educator wants to track instances where the learner ignored a significant event or data (ie inappropriately ignoring).
  3. The educator has a checklist and wants to track whether the learner did or did not do a specific task in an accepted protocol (did the learner check the airway - yes or no?)

How do we describe these inactions in XAPI? A few questions to guide our discussion.

What might reports on inaction look like?

Let's say a simulation is rigged with a significant event to which the learner should react. For example, the mannekin patient becomes unresponsive during the simulation. The learner should call a code. But our learner, Pat Wrongway, does not call a code. Other learners, Jamie Rightway and A Firmative, do call a code. What might those reports look like? You could report only the action performed.

learners  who performed call a code     67%

We feel that this method of reporting is insufficient. You need to actively record that something should have happened but didn't.

Or you could have reports showing the percent of learners who performed "called a code." you could have reports showing the percent of learners who ignored event unresponsive patient.

learners who performed call a code     67%

learners who ignored event unresponsive patient    33%

In both cases, the person or application generating the reports from XAPI messages must know that performing called a code is the right thing to do and ignoring event unresponsive patient is the wrong this to do in this simulation. Another way to do this would be to use something like "failed to perform" to track the inaction. Then the report would look like this.

learners who performed call a code   67%

learners who failed to perform call a code   33%

The group felt this was a preferred option.

There may be other examples where there is a distractor. Let's say the patient has tachycardia, but given other clinical findings, ignoring the tachycardia is appropriate.

learners who ignored tachycardia  60%

learners who addressed tachycardia   40%

This would require adding the verb addressed and potentially linking to the action that reflects how it was addressed. There may also be interest in tracking what an action/inaction was in response to in the context or result of a statement. Also tracking the consequences of specific inaction as detailed in Scott Beck's note.

Two axes

In the reports described above, there are two different axes, two different types of inaction being tracked:

Appropriateness of inaction

In addition to the axes regarding the type of inaction, there is also the matter of whether or not the inaction is appropriate. Does the appropriateness of the inaction (or of any action?) need to be tracked in the XAPI statement? This could be beneficial for reports, as shown below.

learners who ignored tachycardia - appropriate  60%

learners who failed to ignore tachycardia - inappropriate   40%

In this model, you could also analyze the extent to which learners take appropriate actions.

Questions for consideration

  1. is it appropriate to differentiate inactions in response to events from inactions with regards to protocol?
  2. Does appropriateness need to be tracked?