No one logged in. Log in

Managing emergency and elective patient flow through operating theatres

Alex Knight - Wednesday, November 30, 2016

Another rock and a hard place: the emergency-elective split

It is always interesting when people send questions to the website and there was one recently from New Zealand that I felt would be of interest to many people around the world. The topic raised was the ‘emergency-elective split’ in a hospital’s operating theatre handling both emergency and elective or planned patients through the same suite of theatres.

In this environment many hospitals allocate daily theatre capacity for the planned care activities of the specialist and general surgery practices based on some form of historical demand pattern, and they reserve some capacity on each day for the emergency demand. Such an approach is trying to meet two needs: operating theatre capacity is often at a premium and associated with high operating expenses. In order to successfully exploit this key resource one needs not just the theatre to be available but also the surgeon, an anaesthetist and a considerable number of support staff and amount of equipment to plan and carry out the operation, and then further specialist staff and equipment to manage the immediate recovery of the patient. It is no wonder there is considerable pressure to ensure a schedule is created where no time is wasted.

However, on the other hand, knowing the emergency demand for theatre capacity on any one day is extremely difficult to predict and yet failure to have enough capacity could be catastrophic for emergency patient care. Although the average emergency demand can easily be determined from a historical perspective, it is the wide variation on any one day that makes it so difficult to predict the actual demand requirements on each and every day: what happened yesterday is not a good predictor of the actual requirements for today.

If one errs on the conservative side and ensures there is always enough capacity for one of the more extreme demands from the emergency stream – such as when there is a string of car crashes or a long emergency procedure – then one has to accept that on many days this set of physical and staff resources with very high operating expense will stand idle. As a result, in today’s environment where operating expenses are rising faster than revenues, it is not uncommon for an approach to be implemented where the overall schedule allows for a more typical emergency demand and when there is a surge in demand then some of the planned care stream of activities are cancelled at short notice. However, this is also unacceptable for the patient who may have been waiting weeks or even months for their operation. It is often the case that this results in a constant, ongoing debate about how to set the capacity for emergencies and escalation strategies for handling the inevitable surges in demand. This constant search for a tolerable compromise often results in increasing frustration and can deteriorate into less than harmonious relationships across the emergency and planned care streams.

A summary of this situation is outlined below.

In some countries this dilemma has been raging for so long that a decision was made to split the resources into two and create one site just for emergency care and another just for planned care. This model of a ‘hot’ and a ‘cold’ site seems an attractive solution but it does not really deal with the underlying dilemma because, in essence, it is just swopping the risk of wasting operating expenses with the addition of capital expenses and a reduction in the overall return on investment for both the emergency and planned care facilities.

In my experience it is not long before questions are asked as to the long-term financial viability of running two sites that actually use many similar central support services. In its most grotesque form and combined with the ongoing drive to reduce operating expenses this deteriorates into operating theatre schedules being driven by centrally-shared resource availability. Frustration can then exceed even previous levels before the split was undertaken. This last approach violates one of the key principles Dr Goldratt derived from his Theory of Constraints: it can make sense to segment a market but that does not mean one should segment the resources serving the markets.

Examining some of the underpinning assumptions

The first assumption I would like to explore is that reducing operating expenses on a resource will increase the financial performance of the system as a whole. Clearly this is true if the savings in operating expenses has no effect on the throughput of the system or the timeliness of the provision of care. If there is a backlog of planned care then clearly the timeliness of the provision of care is already being compromised. If the hospital is being paid for the number of patients being treated then efforts to reduce operating expenses in this part of the overall system may be wrongly placed relative to the financial benefits of treating more patients. However, this route is often considered to be impractical as it is assumed that any increase in throughput will carry not only an increase in operating expenses through the theatre-related steps in the patient pathway, but also across all other steps. In Pride and Joy I challenge this assumption by demonstrating there is a simple, patient-centred and clinically led way of simultaneously creating a major reduction in length of stay without simply asking staff to work harder. This is achieved by addressing, in a systematic and focused manner, the answer to the two questions:

Of all the patients I could treat next which one should I treat next?

And, more importantly,

Which resource/task combination is most often causing the most delay across the most patients?

This is explained in detail in chapter six.

If this approach is successfully implemented first and significant free capacity is created in the bed base of the hospital then the opportunity is open to significantly increase throughput through the theatres, even accepting a corresponding increase in operating expenses around the theatres. All this is possible with little or no increase in the operating expenses through the rest of the patient’s journey of care. The impact on overall profitability of increased throughput will far outweigh any additional increase in theatre operating expense; although these benefits will only continue until the backlog is eliminated. At this point the hospital will be able to offer instant access, and a different strategy to exploit this new capability will then be required.

This is explained in detail in chapter thirteen and the financial implications are explained at the end of chapter ten.

‘Yes, but…’ Some of you may be saying that as the hospital is still being paid on what in some countries is called a ‘block’ contract or where volumes are capped, there will be no extra revenue and only extra totally variable costs associated with the increase in patients treated. This demonstrates that such attempts to limit overall expenditure create unintended outcomes and actually block hospitals from achieving the original objective of the block contracts and capped volumes, which was to improve productivity! Improvements in productivity are achieved when more patients are treated through the same resource envelope, as well as when the same number of patients are being treated through a smaller resource envelope. These examples of ‘block’ contracts and ‘capping’ are just two of the perverse local optimisation policies that we must break if we are going to improve the productivity of the whole system. I will deal with this in a later blog.

I suspect that even if a hospital did implement the approaches above and did dramatically increase their profitability with relatively little increase in operating expenses, and even went on to exploit the market opportunity through offering instant access, there would still be some frustration with the remaining, apparently inefficient use of these theatre-based resources. The second assumption that I would like to explore is one around the size of the protection that is required relative to the total volume of patients treated. When evaluating the size of these apparent inefficiencies it is vital to understand the difference between what is called ‘protective’ capacity and ‘excess’ capacity.

The protective capacity in this instance is the capacity that is required to ensure that no emergency need is unfulfilled due to a lack of theatre capacity while avoiding cancellation of a planned patient.

As described earlier, it is the combination of the variation in emergency and planned demand that determines the size of the protective capacity needed. If it is so difficult for us to affect the variation in demand of the emergency stream then we should scrutinize our assumptions around the planned stream of care. One can see that the variation in demand for planned care is considerably less than the variation in demand for emergency care and one could conclude that it is not possible to either significantly increase throughput of the planned patients through the same resources or treat the same number of planned care patients through fewer resources.

In chapter thirteen and fifteen I describe the significant opportunity that exists with the scheduling of the planned care stream using the buffer management techniques described to do exactly this. An added benefit in the situation would have improved the overall ratio of the amount of protective capacity required to meet both the planned and elective demand. This approach challenges the assumption that reducing variation is the main opportunity to improve throughput and replaces this with the approach that increased throughput can be achieved without addressing the variation in demand but rather by aggregating the variation in operating times within any one operating list.

I hope the two examples I have given demonstrate that when faced with such a seemingly intractable dilemma at an important step in the overall system, a breakthrough solution is more likely to be possible if one either examines how changes in other parts of the system can remove this seemingly impossible obstacle or how a solution may be found which does not actually mean choosing either one or the other of the current choices.

Post has no comments.
Post a Comment

Captcha Image

Trackback Link
Post has no trackbacks.