Most large, complex health systems rely on block schedules for managing the complexity of operating room utilization. The concept has a visceral appeal to it — give the surgeon or the service line a full-day or a half-day block, and leave the scheduling of procedures within the block to the block owner since they are presumably the best positioned to use each block most efficiently.
Yet, leadership teams of most health systems would likely agree that the three challenges they face are:
1. Lack of Actionable Visibility and Transparency: Accessing timely and accurate information about performance metrics easily and regularly is a challenge every senior executive faces. Existing "dashboards" give them large quantities of rear-view mirror data at best once a week but most often monthly, and the relevant portions of the data are seldom shared across surgeons uniformly in formats they can easily and quickly digest. For example:
a. How are surgeons performing on block utilization — across centers, across block types, and rooms?
b. Who are the best and worst performers on utilization, late-day starts, abandoned or released blocks? What can we (or they) do about improving performance?
c. Admins and surgeons typically look at different reports and varied performance data with different frequencies leading to a lack of common understanding of relative performance and standards and, more importantly, root causes.
2. Dealing with Daily Unplanned Events:
a. When surgeons release blocks just a few days/weeks before the surgery date (e.g., because the surgeon needs to go to a conference, teach, unplanned travel, etc.), the blocks tend to go underutilized because the process to release and ask for blocks is cumbersome, process heavy and takes unnecessary cycles.
b. Surgery day cancellations, delays in surgery, and the need for overtime often causes problems in staffing and leads to high costs.
3. Lack of Transparent Planning: Supply and demand is rarely "matched" based on sophisticated forecasting and predictions for blocks, staff or rooms but more on historical rules for how blocks have been allocated and redistributed. This typically involves an OR committee that meets periodically and decides on the plan with the facts they have available at that time. Matching the ever-increasing demand for surgical resources (blocks, staff, equipment, etc.) with the available capacity of those resources becomes particularly difficult.
The stakes are enormous — even a one point improvement in utilization is worth $100K per OR per year. For most large systems that have 70-80 ORs, the value represents millions of dollars and hundreds of extra procedures per year, all of which represent a significant impact on the access and cost of care provided to patients.
The macro trends would indicate a higher need for surgical services (ACA, aging of the population, more chronic illness, earlier detection of cancer, etc.) while simultaneously confronting a significant downward pressure on reimbursement levels. This would suggest a heightened urgency for being able to "do more with less," and the operating room is an excellent place to start given that it is one of the most significant revenue drivers of the hospital as well as a major contributor to inpatient bed utilization (along with the Emergency Department).
The Current OR Block Scheduling Process Is Subjective and Time-Consuming
The current processes used by many health systems are simply insufficient to capture the potential opportunity. Although the data for every surgery (date, time, type, surgeon, room, duration, etc.) is captured on a timely basis and exists within the EHR, accessing it to conduct insightful analytics is slow, difficult and labor intensive. As a result, a certain "inertia" sets in — blocks are assigned to surgeons and service lines and examined on a monthly (sometimes quarterly) basis by the OR committee which reviews trends in utilization performance by surgeon and service line in order to make informed decisions on the potential reallocation of blocks.
Perioperative business managers feel a significant amount of pressure from such a system since they try and meet the demands of surgeons who want more block time by identifying low-performing blocks that could potentially be reallocated. Between OR committee meetings, these managers conduct a series of "bilateral" negotiations, persuading some surgeons to give up a block in order to meet the legitimate request for additional block time from a new surgeon or a surgeon whose case load has been growing significantly over the past few months. All of this has to be done while keeping the other real-world constraints in mind; these could include equipment in a specific OR (e.g., robotics) or calendar constraints (e.g., the surgeon requesting a block can only operate on Tuesdays, and the candidate blocks that could be released are only on Thursdays and Fridays). Since the OR committee does not have a sufficiently robust fact base for allocating OR blocks to different surgeons, there is an inherent subjectivity to the allocation approach which reduces the perception of fairness in the minds of many surgeons. This, in turn, makes them naturally more resistant to giving up their allocated blocks even if it is the right thing to do from an overall effectiveness perspective.
Imagine a series of timely, accurate charts that are automatically — on a daily or weekly basis — "pushed" to the smartphones of every surgeon, their schedulers and the administrative personnel responsible for managing the utilization of the OR. In addition to providing key stats on utilization, first case on-time starts, cancellations, cases running long, etc., it will create a high level of awareness in the minds of each surgeon about their utilization performance both in absolute terms as well as relative to their peers. Surgeons are data driven, fact based and competitive; accurate, transparent, automated feedback will go a long way toward improving the utilization even if nothing else were to change.
Abandoned or Released Blocks Are a Significant Drain on the OR Capacity
As many as 10 percent of allocated blocks are either abandoned (unused without a replacement surgeon using the block) or released (the original block owner requests that the block be used by some other surgeon). By definition, abandoned blocks have a performance of zero utilization. An abandoned block is like an empty seat on a departing flight; once the flight takes off, that seat capacity is lost forever.
Released blocks represent a different type of problem since they tend to have utilization levels that are ~10 percentage points lower than the original owner of that block would have achieved. In some ways, it is the surgical equivalent of the reality that drivers tend to treat rental cars less carefully than their own cars.
The core reason for the loss of capacity due to abandoned or released blocks is the structural inefficiency of finding a taker for a newly released block. Schedulers and administrators are expected to fire off emails and place dozens of phone calls. Given all of the other tasks on their plate, they understandably take the path of least resistance by offering the block to another surgeon in the service line or to the first surgeon who appears to have a need for the block.
Instead, imagine an Uber-like experience where an efficient market automatically connects a person looking for a ride with a driver willing to make the trip. In this situation, a surgeon releases a block that they do not plan on using due to another commitment. Other surgeons (or their schedulers) would be notified of the block that is now available and could indicate their interest in obtaining the block. This enables the hospital to apply a fair set of rules that are consistent with their stated objective — e.g., egalitarian, performance-first or service-line-first. In the egalitarian approach (or first-come, first-served), the first claim on the newly released block is granted. In a performance-first model, the newly released block would be communicated only to the top decile of surgeons (measured by historic block utilization). If none of them requests the block for a predetermined period of time (e.g., 12 hours), the next quartile is notified and so on. In the service-line-first approach, if a block is released by a surgeon in the Neurosurgery service line, the block would first be offered to that service line for a predetermined period of time before rotating through the other service lines based on a predetermined sequence.
Over a period of time, this will create a vibrant marketplace for blocks that is likely to be more efficient than the current Request for Time (RFT) approach or the manual phone/email tree approach.
Allocations of Blocks to Surgeons and Service Lines Are Suboptimal
The initial allocation of blocks to surgeons tends to stick for a long time. Soft factors like seniority and tenure make it difficult to reduce the block allocation to some of the surgeons. The blocks are also often based on whatever is "easiest" to communicate and manage (e.g., full-day block on Mondays even though the case volume for that surgeon could have easily been achieved by two full-day and two half-day blocks on Mondays).
Instead, imagine an approach that could create smarter block schedules. It begins with a sophisticated approach to developing the prioritized demand for blocks. Rather than trying to forecast the exact number of each type of procedure that a surgeon is likely to perform on a given day, it is better to take a slightly longer time horizon (e.g., 13 weeks spanning the entire quarter) since many procedures are elective and may therefore slide forward or backward by a week or two to accommodate patient preferences. It is also helpful to abstract the slate of procedures into duration buckets (one-hour procedures, two-hour procedures, etc.). As an example, the forecast for Dr. X could be that she needs 41 hours of block time next quarter in order to complete 13 one-hour procedures, eight two-hour procedures and four three-hour procedures. Assuming a nine-hour block, this would be five blocks over the course of the next quarter (because it is preferable to round up). Having built the entire demand table for blocks, it becomes feasible to establish rules for prioritizing each individual request in order to determine the exact number of new blocks that need to be allocated.
Having built the demand table, the supply table can be developed in an analogous manner. First, the obvious sources of supply are blocks that are released voluntarily by surgeons who may be traveling or no longer intend to practice in the coming quarter. This is followed by identifying surgeons whose current allocation is larger than strictly necessary or the surgeons whose utilization performance warrants a reduction in their block allocation.
Both the demand as well as the supply side will continuously use machine learning algorithms in order to automatically prioritize the tables, thereby getting closer to making recommendations that are likely to be accepted by the OR committee. The intent is not to get to automatically generated block schedules. Instead, it is to generate schedules that are smart enough to enable the OR committee to produce an optimal block allocation with the least amount of manual effort possible.
Balancing the demand and supply of blocks through a fact-based methodology builds confidence among the surgeons about the fairness of the process even while enabling a more productive conversation at the OR committee.
In summary, by using data science and machine learning, hospitals can substantially improve the value that is realized from scarce OR assets. The three pieces of the solution reinforce each other — timely, smart metrics pushed automatically to surgeons and administrators raise awareness thereby improving performance, and smart block schedules provide a better starting point for assigning the right block to the right surgeon consistently and automatically. An Uber-like block exchange capability makes the "market" more efficient. Hence, even if the smart block schedule wasn't perfect, the ability to make small, one-off swaps could help bring it closer to perfection.
###
Mohan Giridharadas is an accomplished expert in lean methodologies. During his 18-year career at McKinsey & Company (where he was a senior partner/director for six years), he co-created the lean service operations practice and ran the North American lean manufacturing and service operations practices and the Asia-Pacific operations practice. He has helped numerous Fortune 500 companies drive operational efficiency with lean practices. As the founder and CEO of LeanTaaS (a lean and predictive analytics company), Mohan helped transform healthcare operations at more than a dozen leading organizations including Stanford Health Care, University of Colorado Health, UCSF, Wake Forest and more. Mohan holds a B.Tech from IIT Bombay, MS in Computer Science from Georgia Institute of Technology and an MBA from Stanford GSB. He is on the faculty of Continuing Education at Stanford University and UC Berkeley Haas School of Business. For more information on LeanTaaS, please visit http://www.leantaas.com.
The views, opinions and positions expressed within these guest posts are those of the author alone and do not represent those of Becker's Hospital Review/Becker's Healthcare. The accuracy, completeness and validity of any statements made within this article are not guaranteed. We accept no liability for any errors, omissions or representations. The copyright of this content belongs to the author and any liability with regards to infringement of intellectual property rights remains with them.