System and method for the identification and subsequent alerting of high-risk critically ill patients
The system and method described herein represent a totally new paradigm for assessing risk in critically ill patients by identifying physiologic patterns associated with a poor outcome. It then uses this information to detect these patterns on future patients, sending alerts to clinicians caring for patients with one or more of them. Moreover, the system and method can be extended to outcomes other than mortality, and can be adapted for use in a specific hospital system's EMR. The “code footprint” for embedding the patterns that indicate a trigger is small and thus favorably suited for embedding into an existing clinical decision support system.
1. A method for identifying and managing high-risk patients comprising the steps of:
providing a clinical setting for a first patient wherein the setting includes a plurality of physiological monitors;
providing electronic devices to a caregiver in the clinical setting;
providing a first database that stores historical physiological measurements of a plurality of former patient physiological measurements and former patient outcomes of those former patients; and for each historical physiological measurement, dividing up the historical physiological measurements into ranges to form bins, calculating a distribution of 3-6 bins that results in a maximum variability in a historical mortality outcome rate, and assigning a different letter to each bin;
defining first database trigger measurements based on the historical physiological measurements, wherein those trigger measurements represent a set of historical physiological measurements that have been identified as indicating a high risk for a negative clinical outcome;
providing a second database that receives and stores a plurality of measurements from the plurality of the first patient physiological monitors wherein the first patient physiological measurements occur at regular time segments; and
wherein a bin letter is assigned for each time segment for each first patient physiological measurement depending on which bin the first patient physiological measurement is in;
saving the bin letters assigned for each first patient physiological measurement and for each historical physiological measurement, and collecting the letters for a predetermined time period which is comprised of a plurality of time segments;
forming a word with the plurality of bin letters associated with each time period, and identifying a trigger word that corresponds to a trigger measurement and that indicates an increase in a chance of a negative clinical outcome;
wherein the historical physiological measurements and first patient physiological measurements are recorded and saved for each of a predetermined time segment of 5 to 30 minutes, and for each predetermined time segment, the historical physiological measurement and the first patient physiological measurements are median values of all data points for a physiological measure for a single individual during the predetermined time segment;
providing a processor and comparing by the processor the first patient physiological measurement words to the trigger words to identify a high-risk patient based on whether the first patient set of physiological measurements correspond to the first database trigger measurements; and
sending by the processor an alert to the caregiver if the first patient is at a high risk for a negative outcome based on the comparing step.
2. A method for identifying and managing high-risk patients as described in claim 1 ,
wherein the historical physiological measurements and first patient physiological measurements are selected from a group consisting of heart rate, respiratory rate, mean arterial pressure, saturated oxygen, temperature, glucose level, platelet count and white blood cell count.
3. A method for identifying and managing high-risk patients as described in claim 1 ,
wherein the alert sent to the caregiver comprises visual indicia sent to a user interface on the caregiver electronic device with respect to the first patient.
4. A method for identifying and managing high-risk patients as described in claim 1 ,
wherein the alert sent to the caregiver comprises a visual and audible sound sent to the caregiver electronic device with respect to the first patient.
5. A method for identifying and managing high-risk patients as described in claim 1 ,
wherein the word has four letters which represent historical and first patient physiological measurements, one for each segment, for four time segments.
6. A system for identifying and managing high-risk patients comprising:
a clinical setting for a first patient wherein the setting includes a plurality of physiological monitors;
electronic devices for a caregiver in the clinical setting;
a first database that stores historical physiological measurements in the clinical setting of a plurality of former patient physiological measurements and former patient outcomes of those former patients; and for each historical physiological measurement, dividing up the historical physiological measurements into 3-6 ranges to form bins, wherein a distribution of bins results in a maximum variability in a historical mortality outcome rate, and assigning a different letter to each bin;
in the first database, a set of trigger measurements based on the historical physiological measurements, wherein those trigger measurements represent a set of historical physiological measurements that have been identified as indicating a high risk for a negative clinical outcome;
a second database that receives and stores a plurality of measurements from the plurality of the first patient physiological monitors wherein the first patient physiological measurements occur at regular time segments;
wherein in the second database a bin letter is assigned for each time segment for each first patient physiological measurement depending on which bin the first patient physiological measurement is in,
saving the bin letters assigned for each first patient physiological measurement and for each historical physiological measurement, and collecting the letters for a predetermined time period which is comprised of a plurality of time segments;
forming a word with the plurality of bin letters associated with each time period, and identifying a trigger word that corresponds to a trigger measurement and that indicates an increase in a chance of a negative patient outcome;
wherein the historical physiological measurements and first patient physiological measurements are recorded and saved for each of a predetermined time segment of 5 to 30 minutes, and for each predetermined time segment, the historical physiological measurement and the first patient physiological measurements are median values of all data points for a physiological measure for a single individual during the predetermined time segment;
a processor for comparing the first patient physiological measurement words to the trigger words to identify a high-risk patient based on whether the first patient set of physiological measurements correspond to the first database trigger measurements; and
an alert sent by the processor to the caregiver if the first patient is at a high risk for a negative outcome based on the comparing step.
7. A system for identifying and managing high-risk patients as set forth in claim 6 ,
wherein the alert sent to the caregiver comprises visual indicia sent to a user interface on the caregiver electronic device with respect to the first patient.