Method of continuous prediction of patient severity of illness, mortality, and length of stay

#The present application relates to a system or method for predicting outcome variables such as length of stay, mortality, need for specific therapy, and severity of illness of a patient in a health care environment. It finds particular application in improving and optimizing the prediction a patient's length of stay or probability of morality within a health care environment and will described with particular reference thereto. However, it will be appreciated that it is also applicable to intensive care units, post-anesthesia care unit, general care ward, rehabilitation unit, operating room, and the like.

#Presently, patient recovery/mortality prediction systems, such as the SAPS (Simplified Acute Physiology) system, are typically run at a patient's time of admission. The patient's length of stay or probability of mortality is estimated by weighting a combination of several admission criteria including, admitting vital signals, prior chronic disease, therapeutic and diagnostic investigations, procedures (planned or unplanned surgeries), demographics (age and gender), and the like. A problem exists if the prediction routine is run after the first day of admission, the same routine and relative weighting of the various criteria are used as would have been used as if the patient had just been admitting on the later date.

#The present inventor has observed that when predicting a patient's length of stay or probability of mortality after the patient has been under intensive care for several days, the prediction is less accurate. In clinical practice, the clinical ramifications of a failure to improve after receiving treatment(s) or a particular criteria, such as an abnormal value occurring several days into a hospital admission can have different significance relative to the condition on the first day of admission. For example, an abnormal serum bicarbonate level at the first day of admission may be due to a physiologic insult the patient suffered prior to coming to the hospital. However, if a patient were to develop severe abnormalities to a serum bicarbonate level several days into admission after intensive therapeutic and monitoring care, it may be surmised that the patient is in a critical state that could not be alleviated after several days of intensive care.

#The invention may take form in various components and arrangements of components, and in various steps and arrangements of steps. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention.

#FIG. 1 is a diagrammatic illustration of a patient recovery/mortality prediction system in accordance with the present application.

#FIGS. 2-4 are diagrammatic illustrations of the operation of the patient information server in accordance with the present application.

#With reference to FIG. 1, a patient (not shown) is monitored by various medical monitoring devices or sensors 10 that measure physiological parameters of the patient directly or in a lab and generate physiological data indicative thereof. These medical monitoring devices 10 may include an electrocardiographic (ECG) instrument with ECG electrodes, IV fluid pumps, blood pressure sensors, SpO2 sensors, pulse sensors, thermometers, respiratory sensors, exhaled gas sensors, and the like. Other medical monitoring devices 10 can be associated with a patient, and not all of the above-mentioned medical monitoring devices 10 have to be associated with a patient at any given time. It should be appreciated that while only two medical monitoring devices 10 are illustrated, more medical monitoring devices are contemplated. As used herein, medical monitoring devices signifies data sources indicating patient health, or the like. Electronics for receiving signals from the medical monitoring device 10 and for optionally performing signal processing on such signals are embodied in the illustrated embodiment as a multi-functional patient monitor device (PMD) 12, or may be embodied partly or wholly as on-board electronics disposed with one or more of the medical monitoring devices 10 or so forth. It should also be appreciated that the medical monitoring devices 10 and the PMD 12 could also be embodied into a single device. The PMD 12, for example, may be a monitor or monitoring system that travels with the patient, such as the transmitter of an ambulatory patient worn monitoring system, or the like.

#In one embodiment, the medical monitoring devices 10 transmit the generated physiological data via a body coupled network, Bluetooth, wired or wireless network, or the like to a controller 14 of the PMD 12. The PMD 12 serves as a gathering point for the physiological data measured by the medical monitoring devices 10, and provides temporary storage for the data in a memory 16. The collected physiological data is concurrently transmitted to a controller 14 in the PMD 12 which then transmits the physiological data through a hospital network (not shown) to a patient information server 18 where the physiological data is displayed and stored. The controller 14 of the PMD 12 also controls a display 20 to display the measured physiological data received from each of the medical monitoring devices 10 in the corresponding PMD display 20. The PMD 12 also includes an input device 22 that allows the user, such as a system administrator, to view, manipulate, and/or interact with the data displayed on the display 18. The input device 20 can be a separate component or integrated into the display 18 such as with a touch screen monitor. The controller 14 may include a processor or computer, software, or the like.

#A laboratory server 24 transmits laboratory data through the hospital network to the patient information server 18. The laboratory data may include white blood cell count data, serum potassium data, serum sodium data, serum bicarbonate data, bilirubin level data, and the like. Other laboratory data can be associated with a patient, and not all of the above-mentioned laboratory data have to be associated with a patient at any given time. As used herein, laboratory data signifies data sources indicating patient health, or the like.