Sounds reasonable, until you look at the Silicon Valley experience. Silicon Valley grew like a weed precisely because employees could…
Readers Write: Thinking Differently About OR Block Time
Thinking Differently About OR Block Time
By Michael Burke
Michael Burke, MBA is founder and CEO of Copient Health of Atlanta, GA.
The operating room is the hospital’s largest source of earnings, as well as the largest hospital cost category. Most OR time is allocated in advance to surgeons in chunks of time called blocks. Surgeons schedule cases into their allocated block time, such as Tuesdays from 7 a.m. to 3 p.m.
Block time often goes unfilled due to poor allocation decisions, case volume that can vary meaningfully from week to week, and surgeons neglecting to release block time when traveling or otherwise unavailable to use it. Often, OR time that sits empty can be filled with elective cases that have an average contribution margin of $2,000 per OR hour. Instead, the fixed costs from unused OR hours add up with no revenue to offset them.
Identifying block time that would otherwise go unfilled, getting it released, then refilling the time is something hospitals have attempted to do for quite a while. The process has been largely manual and has missed a meaningful portion of the opportunity, as evidenced by block utilization statistics.
New tools use machine learning to predict block time that is likely to go unfilled, along with mechanisms for seeking the release of the identified time and requesting the time. Finding more time, getting it released earlier, and getting it into the hands of those who can use it are all excellent reasons for adopting such a solution. Hospitals can make real gains with this approach. The core of the strategy is that any block time that would otherwise go unfilled should be filled with positive contribution margin cases whenever possible.
Surgeons are hesitant to release block time allocated to them, even if they don’t have cases to fill it. In most compensation scenarios, a surgeon has a financial incentive to hold on to any OR time allocated to them in the event that a case might come along later. Even if they are an equity holder in an ASC and benefit from facility earnings shared as dividends, they are still subject to a form of the prisoner’s dilemma. This affects their decision-making and can bias them against releasing allocated block time for which they don’t have cases to fill. Although some portion of unused time is collected from surgeons by proactive nudge reminders and the ad hoc efforts of the scheduling team, diverging incentives unnecessarily limit the amount of time that can be recaptured and repurposed.
In many ways, the math behind the predictions is the easy part. The difficulty lies in aligning incentives and driving changes in behavior. The structure of your incentives and your willingness to push will have as much or more impact on the success of an OR optimization effort as the predictive software you select. Maybe we should also consider taking lessons from other industries dealing with similar scarce resource challenges.
What if we thought of a hospital as an airline and an OR block day as a flight? Travelers or travel agents (schedulers) book seats on the plane (cases in the OR). However, from its predictive analytics, the airline knows that some seats will go unfilled, even if booked to capacity. The OR block appears to be booked to capacity in much the same way since 100% of the block’s time is allocated to the block holder.
But we know the block holder won’t fill all the allocated time, just like the airline knows that without intervention, many more seats on the plane would go empty due to no-shows or missed connections. The airline uses predictive analytics to intentionally and confidently overbook the flight to account for this.
The hospital should consider a similar process because the block holder often won’t fill an entire block with cases. To be clear, you wouldn’t be overbooking, since the chunks of time into which you would book cases are empty and predicted to remain so. The math behind the predictions for an OR is different from that of an airline flight, but the analogy still applies. By adopting this strategy, hospitals could fill much more time in their OR blocks with a high degree of certainty that the block holder won’t need it. This approach bypasses the behavioral challenge of seeking permission from the block holder early enough for the unneeded time to be usable, resulting in more recaptured OR time and more contribution margin.
“The difficulty lies in aligning incentives and driving changes in behavior.” I think this is very true. I don’t see how your airline analogy helps though. It does not deal with incentives at all. Plus, if an airline’s predictions are wrong and too many passengers show up, you remedy it with some coupons. If your TKA cannot go through as planned because the OR was overbooked, your patient is not going to be satisfied with some coupons. And the surgeon who was going to do that surgery is going to throw a hissy fit. So, while interesting as an analogy intellectually, I don’t think it translates to any gains in real life.
The behavior change and incentives are the right problem to address, but different solutions are needed than analogies with industries that are structured fundamentally differently.
You’re absolutely correct that overbooking as done in airlines would not work in an OR. The analogy in the article isn’t perfect, because airlines are in fact overbooking seats that are already sold to other travelers, while empty OR block time doesn’t contain any scheduled patient cases.
Right now, hospitals generally don’t have access to the data they’d need to make these decisions about what OR time will go unused and can be safely reallocated. Without that data, any such effort would be ill-advised. But if they had access to data with a sufficiently high confidence interval (say 99%), it could create flexibility to either increase case volume or consolidate resources to address staffing issues.