Computerized method and system for measuring and determining neonatal severity of illness and mortality risk

• US 6,193,654 B1
• Filed: 11/19/1998
• Issued: 02/27/2001
• Est. Priority Date: 11/20/1997
• Status: Expired due to Fees

• Current Assignee: BC Research Institute For Children'S And Women'S Health Center, BETH Israel Deaconess Medical Center (Caregroup Incorporated), Kaiser Foundation Health Plan Incorporated (Kaiser Permanente Inc.), Permanente Medical Group Incorporated (Kaiser Permanente Inc.)
• Original Assignee: BETH Israel Deaconess Medical Center (Caregroup Incorporated)
• Inventors: Richardson, Douglas K., Lee, Shoo, Escobar, Gabriel J.
• Primary Examiner(s): O'Connor, Cary
• Application Number:
  US09/195,892
• Time in Patent Office: 831 Days
• Field of Search: 600/300-301, 600/481-486, 600/500-504, 600/529-534, 600/538, 128/900-925, 128/897-898, 705/2-3, 705/5-9
• US Class Current: 600/300
• CPC Class Codes: A61B 5/02055 Simultaneously evaluating both cardiovascular condition and temperature
• CPC Class Codes: A61B 5/021 Measuring pressure in heart or blood vessels A61B5/0205 takes precedence
• CPC Class Codes: A61B 5/145 Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissuemeasuring of blood pressure or blood flow A61B5/02; non-radiation detecting or locating of foreign bodies in blood A61B5/06
• CPC Class Codes: A61B 5/20 for measuring urological functions restricted to the evaluation of the urinary system A61B5/4375 takes precedence
• CPC Class Codes: A61B 5/4094 Diagnosing or monitoring seizure diseases, e.g. epilepsy
• CPC Class Codes: A61B 5/7275 Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
• CPC Class Codes: G16H 40/63 for local operation
• CPC Class Codes: G16H 50/30 for calculating health indices; for individual health risk assessment
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients

Abstract

A computerized method and system for measuring and determining severity of illness of a neonatal ICU patient uses a computer and a software program to process measured parameter values from preselected physical conditions. Measurement ranges for each measured physical condition are divided into contiguous zones; the contiguous zones are given predetermined weighting factors using the software program. The software program, using user inputs, optimally selects a single value of each measured physical condition from several measurements. The single selected value of each parameter is then modified using the software program. In one embodiment, for achieving the modification, the software program provides a predetermined weighting factor depending on the parameter value selection. For each selected measured value, an applicable zone and its predetermined weighting factor is determined to generate a modified partial score representing each measured physical condition. Values of modified partial scores for all the measured physical conditions are summed by the computer using the software program to provide an illness-severity measure which can be compared with data held in a database for similar patient population. As described in one embodiment, the physical conditions preselected are: lowest mean blood pressure, lowest pH, lowest temperature, lowest oxygenation ratio, urine output, and the presence of multiple seizures. Three additional measurements of birth weight, smallness for gestational age and low Apgar score, after optimal selection and modification as provided by the program, are used to provide a mortality rate assessment for a neonatal patient being monitored.

Claims

1. A computerized method of making an on-line determination of illness-severity of a neonatal patient in a predetermined time span, by using a software program and optimal weighted measurement values of a predetermined ‘

n’

number of on-line parameters from the patient being monitored, said parameters relating to n measurable predetermined physical conditions, said method comprising the steps of;

(a) obtaining, in said predetermined time span from the neonatal patient being monitored, a plurality of measured values of each of said n physical conditions and producing, using the software program, a single optimal value from said plurality of measured values for each of n measured physical conditions wherein the n physical conditions include mean blood pressure and urine output;

(b) using said software program, obtaining from said single optimal value a modified weighted partial score, thus generating n modified weighted partial scores for n on-line parameters from the patient being monitored; and

(c) summing the n modified weighted partial scores to provide an indication of on-line illness-severity of the neonatal patient, which on-line severity indication can be displayed and compared with other known values, said summing including, if the neonatal patient is prematurely born, at least one additional parameter defined as a function of at least one of birth weight, smallness for gestational age and Apgar.

2. The method as in claim 1, said method including the steps of:

(d) defining a range of possible measurement values for each of said n parameters, said range including and extending from known possible minimum to known possible maximum values of measurements of each of said n parameters;

(e) dividing, by using user-inputs, each said range into a plurality of contiguous zones;

(f) attributing, by using user-inputs and using said program, numerical weighting factors to characterize each zone for each of said n parameters;

(g) determining using a computer, a zone in which an optimal value of a measured physical condition from step (a) lies, and identifying its corresponding weighting factor from step (d) above, wherein the step of obtaining a modified partial score comprises;

(h) automatically determining, using said software program, as to in which zone said single optimal value of each of said n parameters from step (a) lies, and using an associated weighting factor from step (g) to obtain a modified and weighted partial score for each measured optimal value of each of n measured parameters, thus generating n modified weighted partial scores.

3. The method as in claim 1, where n is at least six, and where said at least six on-line parameters relating to said optimal values of said measurable predetermined physical conditions comprise:

(1) lowest mean blood pressure;

(2) lowest serum pH;

(3) lowest temperature;

(4) lowest pAO₂/FiO₂ ratio, i.e., the oxygenation ratio;

(5) urine output; and

(6) presence of multiple seizures.

4. The method of claim 3, wherein in step (d), the defined complete range of possible measurement values for each said parameter comprises:

(1) more than 30 for said known maximum value and less than 20 for said known minimum value for lowest mean blood pressure;

(2) more than 96°

F. for said known maximum value and less than 95°

F. for said known minimum value for lowest temperature;

(3) more than 2.5 for said known maximum and less than 0.3 for said known minimum value for oxygenation ratio;

(4) more than or equal to 7.20 for said known maximum and less than 7.10 for said known minimum value for serum pH;

(5) zero for said known minimum and more than one for said known maximum value for number of seizures; and

(6) more than 0.9 for said known maximum and less than 0.1 for said known minimum value of urine output measured in cc/kg/hr.

5. The method of claim 4, wherein said numerical weighting factors recited in step (f) above comprise values:

6. The method as in claim 5 including the step of holding in memory the listed numerical weighting factor values and their corresponding parameter values for said normal, moderate, and severe zones.

7. The method as in claim 6, wherein the step (g) of determining the zone and its weighting factor for each said optimal value of an on-line measured physical condition comprises:

comparing each said optimal value with its corresponding zonal values held in memory for normal zone, moderate zone, and severe zone to;

identify in which zone a measured optimal value lies; and

determine from the identified zone a corresponding weighting factor which is held in memory.

8. The method as in claim 7, wherein step (h) above includes the step of checking an applicable weighting factor for correctness using said single optimal value of each of said n parameters, and automatically obtaining a modified partial score for each said measured parameter using step (g).

9. The method as in claim 1, wherein the patient is a prematurely born infant and wherein said summing includes three additional parameters comprising birth weight, smallness for gestatinal age and Apgar, where said three additional parameters have numerical weighting factors as follows:

10. The method as in claim 9, including the step of obtaining a score comprised of a score from claim 1 and adding a method score from said three additional parameters by way of:

10, if the birth weight is in moderate zone +OR 17, if birth weight is in severe zone +

8, if measured smallness for gestational age is in moderate zone +

18, if measured “

Apgar”

at 5 minutes is in severe zone, to obtain a SNAPPE-II measurement.

11. The method, as in claim 10, wherein an on-line determination is made at the time the infant is admitted into a neonatal intensive care unit, including the method steps of automatically:

(a) making a second on-line determination of subsequent illness-severity, during a predetermined time lapse of 12 hours from admission; and

(b) making a comparison of the illness-severity at the time of admission with the said subsequent illness-severity to generate an indication of a measure of progress of the patient.

12. The method of making an on-line determination of a patient illness-severity as in claim 10 including the step of assessing a measure of mortality risk of the patient in terms of probability of death, by using a measured patient severity illness score and a predetermined mortality risk equation.

13. The method as in claim 12 wherein:

17. The system as in claim 14, wherein the monitored patient is a prematurely born infant, and further comprising means to generate a Perinatal Extended Score (SNAPPE-II), which is the sum of the illness severity measure in step (f) of claim 1 together with the sum of:

(i) a weighting factor from step (c) above based on a birth weight zone;

(ii) a weighting factor from step (c) above based on a measured smallness for gestational age; and

(iii) a weighting factor from step (c) above based on measured Apgar at 5 minutes, to generate a measure of mortality.

18. The system as in claim 17 including means to evaluate a probability of death as

14. A system for making a computerized on-line determination of an illness-severity-score of a monitored neonatal patient by using measurement values of at least six on-line parameters from the monitored patient and a software program, said parameters relating to six different measurable physical conditions from the monitored patient, said system comprising:

(a) programmed means for defining a complete range of possible measurement values for each said parameter, said complete range including an accepted range extending from known minimum to known maximum values in medical practice for each said parameter, the parameters including one parameter relating to lowest mean blood pressure and another parameter relating to urine output;

(b) programmed means for dividing each said complete range from (a) into at least three contiguous range-zones of normal, moderate, and severe, said three range-zones when viewed collectively forming each said complete range;

(c) programmed means to attribute a numerical weighting and multiplying factor to characterize each zone for each measurable physical condition;

(d) means for taking measurements of on-line physical condition values for said at least six on-line parameters from the patient being monitored and for choosing an optimal value from said measurements;

(e) programmed means for determining from (c) above as to which range-zone each optimal value of a measured physical condition value falls within, and for obtaining a modified partial weighted score for each corresponding parameter; and

(f) on-line adding means connected to receive and automatically sum modified partial weighted score values to generate an on-line indication of the illness-severity-score of the monitored patient, if the monitored patient is prematurely born, the means for taking measurements further including means to make measurement values of at least one of three more parameters relating to birth weight, smallness measure for gestational age, and Apgar, and said adding means adjusting the illness-severity score to account for at least one of birth weight, smallness for gestational age and Apgar.

15. The system as in claim 14, wherein said programmed means of element (e) includes:

a comparator means for comparing each measured parameter relating to one measurable physical condition of the patient with an applicable normal, moderate, and severe zone group, to identify which zone a measured parameter falls in, and to determine a corresponding weighting factor taken from said programmed means from element (c) above.

16. The system as in claim 15 including a second comparator for comparing the generated illness-severity-score with standard illness-severity scores stored in memory to assess the degree of illness severity.

19. A method of making an on-line determination of an illness-severity of a neonatal patient being monitored by using measurement values of at least six on-line parameters from the patient in a predetermined time span, said parameters relating to six measurable predetermined physical conditions, said method comprising the steps of:

(a) obtaining, from the patient being monitored, a plurality of measurements in a predetermined time span for each said measurable physical condition and choosing, using a program, an optimal value from said plurality of measured values, thus generating six optimal values for six said predetermined physical conditions in said predetermined time span, at least one parameter relating to lowest mean blood pressure and another parameter relating to urine output;

b) defining a range of possible measurement values for each said parameter, said range including a known minimum and extending to known maximum values of said parameter measurements in medical practice;

(c) dividing each said range into at least three contiguous zones of normal, moderate, and severe;

(d) attributing, by using said program, numerical weighting factors to characterize said zones;

(e) determining a zone and its weighting factor for each said optimal value from step (a), as decided by in which zone an optimal value would fall;

(f) producing a modified partial score for an optimal value of each measured parameter, thus generating six modified partial score values; and

(g) automatically summing at least six said modified partial score values for the six measured parameters to obtain an on-line illness-severity measurement which can be displayed, said summing including adjusting the on-line illness-severity measurement as a function of premature birth of the patient.

20. The method as in claim 19, where said at least six on-line parameters relating to said optimal values of said measurable predetermined physical conditions comprise:

(1) lowest mean blood pressure;

(2) lowest serum pH;

(3) lowest temperature;

(4) lowest pAO₂/FiO₂ ratio, i.e., the oxygenation ratio;

(5) urine output; and

(6) presence of multiple seizures.

21. The method of claim 20, wherein in step (b), the defined complete range of possible measurement values for each said parameter comprises:

(1) more than 30 for said known maximum value and less than 20 for said known minimum value for lowest mean blood pressure;

(2) more than 96°

F. for said known maximum value and less than 95°

F. for said known minimum value for lowest temperature;

(3) more than 2.5 for said known maximum and less than 0.3 for said known minimum value for oxygenation ratio;

(4) more than or equal to 7.20 for said known maximum and less than 7.10 for said known minimum value for serum pH;

(5) zero for said known minimum and more than one for said known maximum value for number of seizures; and

(6) more than 0.9 for said known maximum and less than 0.1 for said known minimum value of urine output measured in cc/kg/hr.

22. The method of claim 21, wherein said numerical weighting factors recited in step (c) above comprise values:

23. The method as in claim 22 including the step of holding in memory the listed numerical weighting factor values and their corresponding parameter values for at least said normal, moderate, and severe zones.

24. The method as in claim 23, wherein the step (e) of determining the zone and its weighting factor for each said optimal value of an on-line measured physical condition comprises:

comparing each said optimal value with its corresponding zonal values held in memory for normal zone, moderate zone, and severe zone to;

identify in which zone a measured optimal value lies; and

determine from the identified zone a corresponding weighting factor which is held in memory.

25. The method as in claim 24, wherein step (e) of claim 19 above includes the step of automatically checking a zone of a measured optimal value of a parameter and its determined weighting factor to obtain a modified partial score for each said measured parameter, wherein further an applicable weighting factor is the modified partial score for an applicable measured optimal value of a parameter.

26. The method of making an on-line determination of an illness-severity-score of a monitored patient, as in claim 19 wherein the step of adjusting in the summing step (g) includes the step of using three additional optimized measured parameters.

27. The method as in claim 26, wherein said three additional optimized parameters comprise birth weight, smallness for gestational age and Apgar at 5 minutes, all relating to the infant, to determine a severity illness score termed SNAPPE-II.

28. The method, as in claim 27, wherein an on-line determination is made at the time the infant is admitted into a neonatal intensive care unit, including the method steps of automatically:

(a) making a second on-line determination of subsequent illness-severity, during a predetermined 12-hour time lapse from admission; and

(b) making a comparison of the illness-severity at the time of admission with the said subsequent illness-severity to generate an indication of a measure of progress of the patient.

29. The method of making an on-line determination of a patient illness-severity as in claim 28 including the step of assessing a measure of mortality risk of the patient in terms of probability of death, by using a measured patient severity illness score and a predetermined mortality risk equation.

30. The method as in claim 29 wherein: