Batching and Queuing – a Canadian Healthcare Paradigm?
I was recently engaged to provide consulting services to a primary care clinic in a large Canadian city. This particular clinic was considered to be at full capacity and the needs of the greater community could not be adequately served. The centre, staffed by some amazing people, provides support to a wide range of people in need: parents and their children, urban outreach (homeless), public walk-in patients, mental health, seniors… an extremely diverse community with equally diverse needs.
The clinic had recently been granted additional capital and annual operational funding from the Ministry of Health for the purpose of expanding their “footprint”. Work was already underway in terms of facility redesign and associated purposing of space for this much needed expansion.
One of the major issues with the current space, as universally cited by the clinic’s front-line care-providers: Not enough exam rooms! This was considered to be the biggest bottleneck in being able to serve more patients.
The Conflict: The Ministry funding formula, when considering all factors, including the size of the community, dictated that the expanded facility would have less exam rooms than the existing facility today! This made no sense to staff that has lived every day with the pain of the exam room “constraint”.
Although out of scope for the “value stream mapping” work I was engaged to do, I gathered a random sample of exam room utilization over a one week period, with a goal of identifying what the “occupancy rate” was for the exam rooms. The results were… interesting… but I kept them to myself.
A few weeks later, during a formal presentation of the value stream mapping work completed by the project team and their resulting improvement recommendations for both existing as well as future facility operations, I had the opportunity to share with the management team the results of exam room utilization sample.
Accounting for sample size, the week’s data suggested that exam room occupancy at the clinic ranged from as low as 56% to as high as 72%... the exam rooms might be empty 28% - 44% of the time! Note: Actual exam room occupancy rates were never tracked by the clinic even though this was considered a problem!
I then asked the team how this contrasted with their shared concern that the clinic did not have enough exam rooms and the “error” in the Ministry’s formula that dictated a reduction in the number of exam rooms for the future facility. We then discussed the booking process for appointments...
The majority of patients attending the clinic are booked in advance to a physician although a walk-in clinic is supported for part of every day.
Two exam rooms are assigned to each physician when they are on-site, (the majority of physicians are “part-time” holding positions with other healthcare organizations in the city or private practices). An exam room is “owned” by that physician for that day and not available for other physicians or staff to use.
The factors determining exam room assignment are physician seniority at the clinic, the location of the physician’s office in the clinic, (physicians want exam rooms close to their office), and exam rooms are not all created equal… There are “preferred” exam rooms in terms of “better” layout, equipment, lighting, aesthetic appeal, etc.
Some exam rooms are designated for “nursing assessment only” due to inadequate space, equipment, etc.
We then went back to the previously reviewed current state value stream map and considered the time elements that had been captured, but now with a focus on the variation that was observed at each step of the patient/client experience;
Arrival times at clinic (early, late, rescheduled, no-shows)
Wait time to register
Wait time to go to exam room
Nursing exam time
Wait time to go to physician exam room
Physician exam time
Wait time to complete post-appointment follow up at registration (scheduling of next appointment, outside lab referrals, etc.)
What was easily observable in the patient experience was the high degree of variability at each step of the process. I suggested to the management team that this variation was a major factor contributing to exam room occupancy rates! Of course, this also affected overall clinic efficiency including staff utilization.
I suggested to the management team that there were two options available to improve exam room occupancy rates.
Option 1: Tackle the sources of variation that have been identified in the value stream mapping with a goal of reducing overall variation, (even though some of it might be viewed as “uncontrollable”).
Option 2: Redesign the process so that it is robust to this variation. (Granted, a third option would be to do a bit of both but, for simplicity, I left that out of the discussion.
I explained to the team that a “robust” design, (Option 2), could be achieved by changing the engagement model by simply eliminating the batching that is evident throughout the process.
Rule #1: Do NOT batch the physicians to specific exam rooms.
Rule #2: Do NOT batch patients to physicians.
This model would be based on the next available patient going to the next available physician and the physician using the next available exam room.
One of the physicians finally spoke up and said that this model could not work… that studies show that a permanent patient-physician relationship yields better health outcomes.
I responded that I respected the physician’s opinion and experience but since we were in an environment that promotes evidence-based decision-making, (and this was an extremely important and costly one), where can we get these studies and review the findings? My hypothesis, until the data demonstrated otherwise, would be that a permanent patient-physician relationship does not affect the outcome, (my spin on the null hypothesis).
One of the social workers spoke caringly about their clients, the majority urban outreach and homeless, who have both chronic physical and mental health challenges. “We are fortunate to have these patients trust one physician let alone expecting them to come in and see any number of physicians that might be available. This next available physician model simply would not work for them!”
Perhaps so, but does one process have to fit all? The clinic already supports a walk-in model that does offer the next available physician model, (although rooms are still batched to the walk-in due to proximity to the walk-in entrance).
Certainly the panel of patients for the clinic are not homogeneous and “one size does not fit all”. Having said that, is it possible that 70% or more of the patients attending the clinic now or in the future could be serviced by a “next available” model?
Why is batching and queuing so prevalent in healthcare? Did it evolve at a time when we thought we could afford to support this system and its inherent inefficiencies? Have we created paradigms and expectations for healthcare care-providers and the community alike that we can no longer afford but are unwilling to challenge? Does the current model allow and even hide waste and risk, (i.e. no just in time access to patient data via a national health records system or national pharmaceutical registry, people-dependent processes that ask each day, “Please doctor or nurse… don’t make a mistake!”).
Are we so focused on the “quality of experience” for the patient in the room that we forget the patients that we will never see due to insufficient capacity?
Establishing a new paradigm: What if we grew up in a system where “next available” WAS the norm? What would that system look like?
I try to play it safe in these types of discussions by either starting off or ending by saying, “I don’t have all the answers… Heck! I don’t have all the questions!” But… healthcare, question we must!