Image interpretation vs. workflow interruption: Quantified, analyzed, strategized

Upon commissioning a business process improvement team to meticulously track interpretative uptime, a 13-radiologist group learned its members had been spending nearly as many hours navigating interruptions as reading images.

What’s more, they found the interruptions measurably decreased efficiency and increased reading time regardless of imaging modality.

Corresponding author Lamya Atweh, MD, and colleagues at Nationwide Children’s Hospital in Columbus, Ohio, share lessons learned from the exercise in a study published June 27 in Current Problems in Diagnostic Radiology [1].

For the study, the business observers sat behind or alongside the study subjects and logged their actions into one of three columns—study interpretation, active interruption (initiated by the radiologist) or passive interruption (initiated by someone else).

Tallying the results after 61 hours of observation conducted over a three-month period, Atweh and co-authors found image interpretation claimed not a lot more than half the rads’ combined clock (52%).

The rest—close to half the tracked hours—went to active interruptions (29%) or passive interruptions (18%).

More findings of note:

• Around 50% of non-interpretive time involved in-person conversations or consults. Of these, 16% involved phone calls, 67% of which were incoming.
• The longest uninterrupted span was 20 minutes.
• Most of the time (85%), an interruption came within the first three minutes of an interpretation. These interruptions lasted one minute or less 70% of the time.

In addition, the authors cite previous research showing the negative effects interruptions can have on interpretation accuracy.

As for modality-specific findings, interruptions expanded read times by one minute for X-rays, two minutes for ultrasounds, six minutes for CTs and 10 minutes for MRIs.