It’s not “lean” to wait 150 days for an MRI
This news story, “Ottawa Hospital seeks fourth MRI machine,” from Ottawa describes an unfortunate situation where 90% of patients had been waiting 360 days for an MRI. So the hospital added capacity – by adding a third MRI machine (an expensive proposition). They also improved some administrative processes. Waiting times fell to about 150 days, still a long time to wait. What does the hospital want now?The hospital wants another expensive MRI machine, which will probably cost a couple of million dollars.
“…the Ottawa Hospital needs a fourth MRI machine to bring them down any further, says the head of diagnostic imaging at the Ottawa Hospital.”
It could be that they need another machine, but the news story doesn't report the utilization rate of the machine. Is it running anywhere near 75 or 80% capacity (the practical maximum before queuing theory says patient flow will suffer)? Have they improved their scheduling processes and looked at the whole value stream to reduce delays in patient flow caused by problems like missing orders, incorrect orders, and bottlenecks at the registration desk?
There are often cheaper ways to get more capacity. Have they reduced the turnover time between patients, using lean “quick changeover” methods?
We don't know. But considering the deficits at the hospital ($25M) and the deficits for Ontario ($25B), I hope they are considering every alternative short of buying a new machine.
I worked with a children's hospital U.S., where the team I was coaching from radiology was able to reduce outpatient MRI waiting times from 14 weeks to just 2 weeks. Sure, they worked a little overtime to help get the backlog down, but the main backlog and waiting time reductions came from process improvements and breaking a bottleneck with a very selective staffing increase (this is different than blindly throwing people at the problem). They got waiting times down without buying a new MRI. In fact, when they first started looking at the data, the MRI utilization had only been about 40% during the 12-hour working day (which really means the ultimate utilization was just 20%!). With improved value stream flow and better staffing patterns, they quickly got utilization up over 60% of the working hours. That was far cheaper than buying another MRI (one for which no space was available, anyway).
From my headline, the comment that bothered me from the story was this:
“We are as lean as we almost can be, but I really need that fourth MRI to get any better,” said Dr. Mark Schweitzer.
There's the problem with the word “lean.” In common, everyday usage, people use it and it has nothing to do with the Toyota Production System. Everyday usages of “lean” means “we don't have enough people” or “we don't have enough equipment.” Even if they have high utilization and they are truly as efficient as they can be with existing resources, making patients wait five months can't really be anybody's definition of Toyota “lean,” which would mean meeting customer needs (demand) at the highest quality, at the lowest cost.
One reason waiting times are down is an artificial constraint of demand for MRI:
Many doctors have been “self-censoring” by ordering CT scans and ultrasounds instead of joining the Ottawa Hospital MRI queue, said Schweitzer.
I wonder if the physicians were pressured by those controlling the budgets to not order MRI? Is that necessarily the best thing for the patients? It depends – is the MRI medically necessary? Is it value to the patient?
It's one of the complexities in healthcare… it's often argued that supply creates demand (or lack of supply constrains demand). It's like adding more lanes to a congested freeway… traffic delays don't get any worse because more people choose that road (as explained here).
Will adding a fourth MRI machine reduce waiting times? It might, but that's one expensive “Do” for the Plan-Do-Check-Act cycle! If I were advising them, I'd make sure every possible process improvement had been investigated first.