Swarming for Improvement
By Dean Bliss
As we study the processes we use to do our work, we see opportunities that we may not have even considered before we learned about the Toyota Production System and Lean. At St. Luke's Hospital in Cedar Rapids, Iowa, we saw such an opportunity in the daily early-morning blood specimen collection.
Typically, the collection had been done by sending one phlebotomist, to each unit of the hospital. Before leaving the lab, the phlebotomist would get the bar-coded labels from the printer and place them on the collection tray. The specimens would then be collected and placed on the tray. When the collection for that trip to the units was complete, the phlebotomist would return to the lab, bringing the batch of specimens to be analyzed. Another batch of labels would be collected, and the process would be repeated until specimens for all units were collected.
The key thing that prompted us to look for ways to improve this process was the comments from the physicians, who are the customers of this process. Many of them would “round” on their patients starting at 7 AM, and when they did, they wanted to have the lab results on the charts by that time. In many cases, that wasn't happening, which would cause the physician to either have to call in later to get the results or to revisit the hospital.
So the team began to look at ways to get the specimens collected, analyzed, and resulted before 7 AM. We visited some other labs that have been using Lean methods and talked to the phlebotomy teams to see what they were doing and how they were doing it. We also looked at their physical layouts, their computer systems, and their specimen delivery systems.
As we looked at alternatives, one of the things we did that made a big difference was to demonstrate the difference between one piece flow and batching. Those of us that do Lean for a living know that sometimes it's tough to convince people that one piece flow is better. So we tested several different methods on different mornings and kept metrics on how the collection methods worked. The measures, which we were able to draw from our computer system, demonstrated that the “swarming” technique, combined with one piece flow, gave us the best results.
The technique that we settled on, and that we are continually reviewing, is to send a team of phlebotomists to each floor, rather than just one person. That's the “swarm”. The group has a leader who monitors the group and allocates the work, one collection at a time. Each team member is given one set of labels (which reduces the chance of mixing up specimens), does the collection, and returns to the leader for another set of labels. The leader sends the completed specimen to the lab in the pneumatic tube system so processing can begin. If a phlebotomist has difficulty with a collection, a team member can help before the team moves to the next unit – that reduces the number of missed collections. When all specimens have been collected, the team moves as a group to the next unit.
There were many details to work out to make this an effective collection method, including adjusting work schedules and gaining agreements on the sequence of units. There were also many, many small tests of change to refine the method. But once in place, it has had a very positive impact. It is now rare for a test result to not be on a chart before 7 AM. And the time from collection of a specimen to receipt in the lab has been reduced from a high of 28 minutes to an average of less than 10 minutes. Receipt to result time now happens within 23 minutes 97 percent of the time.
The “swarming” method has been an effective improvement at St. Luke's and has resulted in improvement not only in the lab, but on the inpatient units as well. And with continuous improvement, the method continues to get better. It's a great example of a team coming up with a wild idea, testing it to see if it works, bringing the team members on board, and making it happen.