Tulva AI for hospitals
Abstract
Do you feel like the emergency department of your hospital is more often crowded than not? Do you struggle with providing your patients with the followup care beds they require? You are not alone. In this article we discuss why crowding is not just a nuisance but a serious patient safety issue and tell how Tulva AI can help you and what it takes to get started.
Patient flow and crowding
Let us paint a picture of a simplified flow of bedoccupying in-hospital patients depicted in Figure 1. First, patients arrive in the ED at unexpected quantities and at unexpected intervals. Majority of the patients are discharged from the ED and rest are admitted to the wards. Out of the admitted patients a minority are directed to emergency surgery which later end up occupying a bed in a surgical ward. At the same time, elective surgery depends on the same inpatient beds the emergency department and emergency surgery depend upon. This is obviously a crude simplification and omits many of the potential and important pathways and components such as the ICU or downstream units such as intensive service housing but bear with us.
This is great when everything works as expected and the patient stream flows unobstructed. Unfortunately, more often than not, anomalies appear. These can be triggered by unexpectedly high number of arrivals at the ED, which becomes crowded and starts feeding patients to both emergency surgery and hospital wards. The results, as you might guess, is that the wards become crowded too, which causes an obstruction in all of it’s upstream dependencies. This exacerbates crowding in the ED and can result for example in cancellation of elective surgery.
Needless to say, the negative implications of this to the patients are evident. From strictly the ED’s point of view, crowding has been associated with increased mortality [1–8], increased operating cost [9] and delays in potentially life-saving treatments such as time to antibiotics [10] or thrombolysis [11]. At the same time, the cancellation of elective surgeries further exacerbates the already unsustainable growth of treatment waiting lists [12].
A simplified overview of patient flow of in-hospital patients, dependencies of different components of the care chain and implications of obstructed patient flow.
Saving lives by predictability
Now we have established that something has to be done. One obvious solution would be to ensure that there is always an excess amount of reserve capacity, so that the hospital never runs out of resources. The problem with this approach is that it is expensive. In Finland, the government has recently announced 500 million budget cuts in public health care [13] while in Europe populations has been projected to age up until year 2100 [14]. In this environment of financial scarcity and ever increasing demand for service, over resourcing is not a viable option.
This is where Tulva AI comes in. With predictions of the future service demand, a hospital can ensure that the supply for care matches the demand without the need for planned over-resourcing or unexpected periods of under-resourcing.
From theory to practice
We hope this conceptual overview was useful. If you want to hear more and move theory to practice, don’t hesitate to contact. Come and see the future with us!
References