Methods of assessing risk based on medical data and uses thereof

• US 8,355,925 B2
• Filed: 10/21/2009
• Issued: 01/15/2013
• Est. Priority Date: 10/21/2008
• Status: Active Grant

• Current Assignee: Spacelabs Healthcare LLC (OSI Systems Incorporated)
• Original Assignee: PeraHealth, Inc.
• Inventors: Rothman, Michael J., Rothman, Steven I.
• Primary Examiner(s): Morgan, Robert
• Assistant Examiner(s): HUNTER, SEAN KRISTOPHER
• Application Number:
  US12/582,942
• Publication Number:
  US 20100100392A1
• Time in Patent Office: 1,182 Days
• Field of Search: 705/2, 705/3
• US Class Current: 705/2
• CPC Class Codes: G16H 15/00 ICT specially adapted for medical reports, e.g. generation or transmission thereof
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16H 50/30 for calculating health indices; for individual health risk assessment
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients
• CPC Class Codes: G16H 70/60 relating to pathologies

Abstract

Methods of assessing risk based on medical data are disclosed herein. In an embodiment, a method of assessing risk associated with medical data includes creating a dataset representing a plurality of patients, the dataset comprising (x,y) pairs for each patient, wherein x represents the medical data collected at a first time, and wherein y is an outcome measurement collected at a second time; binning the (x,y) pairs to form a plurality of binned data sets; computing an average value for x and an average value for y for each binned data set; determining a minimum average value of y based on all of the average values of y; subtracting the minimum average value of y from each average value of y to get a new average value of y for each binned data set; and deriving a function for assessing risk associated with the medical data.

Claims

1. A method of assessing risk associated with medical data comprising:

creating, by a computer, a dataset representing a plurality of patients, the dataset comprising (x,y) pairs for each patient, such that x represents the medical data collected at a first time, y is an outcome measurement collected at a second time, and the collected medical data is selected from one of a continuous variable or an ordinal score;

binning, by a computer, the (x,y) pairs to form a plurality of binned data sets;

computing, by a computer, an average value for x and an average value for y for each binned data set;

determining, by a computer, a minimum average value of y based on all of the average values of y;

subtracting, by a computer, the minimum average value of y from each average value of y to get a new average value of y for each binned data set; and

deriving, by a computer, a first function for assessing risk associated with the medical data, such that the first function is defined from a minimum average value of x to a maximum average value of x;

deriving, by a computer, a second function for assessing risk associated with values of x less than the minimum average value of x; and

deriving, by a computer, a third function for assessing risk associated with values of x and greater than the maximum average value of x,

wherein the derived functions are used in calculating the risk associated with the medical data.

2. The method of claim 1 wherein the medical data for each patient is collected at discharge of the patient from a facility of care.

3. The method of claim 1 wherein the outcome measurement represents a mortality of each of the patients.

4. The method of claim 1 wherein the outcome measurement for each patient is collected one-year post discharge.

5. The method of claim 1 wherein the functions define excess risk due to deviation from normative values.

6. The method of claim 1 wherein the continuous variables are selected from the group consisting of albumin/globulin (A/G) ratio, an alanine aminotransferase (ALT or SGPT) value, an aspartate aminotransferase (AST or SGOT) value, an albumin value, an alkaline phosphatase value, a blood urea nitrogen (BUN) value, a calcium value, a carbon dioxide (CO₂) value, a chloride value, a creatinine value, a globulin value, a glucose value, a potassium value, a sodium value, a total bilirubin value, a total protein value, a tropon value, a base excess value, a fraction of inspired oxygen (FiO₂) value, a bicarbonate (HCO₃) value, a partial pressure of carbon dioxide (PCO₂) value, a partial pressure of oxygen (PO₂) value, a pH value, a hematocrit percentage, a hemoglobin value, a white blood cell count, a heart rate value, a diastolic blood pressure value, a systolic blood pressure value, a respiration rate, a percentage of arterial hemoglobin in oxyhemoglobin configuration (pulse Ox) and a temperature value.

7. The method of claim 1 wherein the ordinal score is a Braden Scale score.

8. The method of claim 1 wherein the derived functions are used in calculating an overall risk inherent in a patient'"'"'s medical condition.

9. The method of claim 1 wherein the derived functions can be used in determining a risk inherent in a value of corresponding medical data from a patient.

22. The method of claim 1 wherein the second function and the third function are constant functions.

10. A method of determining an overall risk in a patient'"'"'s current condition comprising:

receiving, by a computer, a plurality of medical data about a patient at a plurality of time points over a period of time, such that the received medical data includes continuous variables;

converting, by a computer, each of the admitted patient'"'"'s medical data for each of the plurality of time points to values using a set of functions, wherein each of the functions defines risk associated with each of the medical data, and wherein the set of functions for each of the medical data is derived by;

creating, by a computer, a dataset representing a plurality of discharged patients, the dataset comprising (x,y) pairs for each discharged patient, such that x represents a single type of medical data collected at a first point in time and y is an outcome measurement collected at a second point in time;

binning, by a computer, the (x,y) pairs to form a plurality of binned data sets;

computing, by a computer, an average value for x and an average value for y for each binned data set;

determining, by a computer, a minimum average value of y based on all of the average values of y;

subtracting, by a computer, the minimum average value of y from each average value of y to get a new average value of y for each binned data set;

deriving, by a computer, a first function for assessing risk associated with the medical data, such that the first function is defined from a minimum average value of x to a maximum average value of x;

deriving, by a computer, a second function for assessing risk associated with values of x less than the minimum average value of x; and

deriving, by a computer, a third function for assessing risk associated with values of x and greater than the maximum average value of x;

combining, by a computer, for each of the plurality of time points, the values;

generating, by a computer, a Health Score from the combined values for each of the plurality of time points, each Health Score representing the patient'"'"'s health; and

continually displaying, by a computer, a graph of each of the Health Scores,

wherein the derived functions are used in calculating the overall risk in the patient'"'"'s current condition.

11. The method of claim 10 wherein the medical data for each discharged patient is collected at discharge of the patient from a facility of care.

12. The method of claim 10 wherein the outcome measurement represents a mortality of each of the discharged patients.

18. The method of claim 10 wherein the second function and the third function are constant functions.

19. The method of claim 10 wherein the received medical data further includes nursing assessments.

20. The method of claim 19 wherein each of the nursing assessments are converted, by a computer, to values using a set of functions, wherein each of the functions defines an excess risk associated with failing of a nursing assessment, and wherein each function is derived by:

creating, by a computer, a dataset representing a plurality of patients, the dataset comprising (x,y) pairs for each patient, wherein x represents a pass or a fail score of the nursing assessment collected at a first point in time, and wherein y is an outcome measurement collected at a second point in time;

binning, by a computer, the (x,y) pairs to form two binned data sets;

computing, by a computer, an average value for y for each binned data set;

determining, by a computer, a minimum average value of y based on all of the average values of y;

subtracting, by a computer, the minimum average value of y from each average value of y to get a new average value of y for each binned data set; and

calculating, by a computer, the function for defining excess risk associated with the failing of the nursing assessment.

21. The method of claim 20 wherein the function is a constant function.

13. A method of defining excess risk associated with medical data comprising:

creating, by a computer, a dataset representing a plurality of patients, the dataset comprising (x,y) pairs for each patient, such that x represents a pass or a fail score of a nursing assessment collected at a first point in time and y is an outcome measurement collected at a second point in time;

binning, by a computer, the (x,y) pairs to form two binned data sets;

computing, by a computer, an average value for y for each binned data set;

determining, by a computer, a minimum average value of y based on all of the average values of y;

subtracting, by a computer, the minimum average value of y from each average value of y to get a new average value of y for each binned data set; and

calculating, by a computer, a function for defining excess risk associated with the failing of the nursing assessment,

wherein the calculated function is used in determining the excess risk associated with the medical data.

14. The method of claim 13 wherein the nursing assessment is selected from a food assessment, a neurological assessment, a psychiatric assessment, a safety assessment, a skin assessment, a genitourinary assessment, a muscular-skeletal assessment, a respiratory assessment, a cardiac assessment, a peripheral vascular assessment, a gastrointestinal assessment and a pain assessment.

15. The method of claim 13 wherein the nursing assessment for each patient is collected at discharge of the patient from a facility of care.

16. The method of claim 13 wherein the outcome measurement represents a mortality of each of the patients.

17. The method of claim 13 wherein the outcome measurement for each patient is collected one-year post discharge.

23. A method of defining excess risk associated with medical data comprising:

creating, by a computer, a dataset representing a plurality of patients, the dataset comprising (x,y) pairs for each patient, such that x represents a categorical class determined at a first time and y is an outcome measurement collected at a second time;

binning, by a computer, the (x,y) pairs to form a plurality of binned data sets;

computing, by a computer, an average value for y for each binned data set;

subtracting, by a computer, the minimum average value of y from each average value of y to get a new average value of y for each binned data set; and

calculating, by a computer, a function for defining excess risk associated with the categorical class,

wherein the calculated function is used in determining the excess risk associated with the medical data.

24. The method of claim 23 wherein the categorical class is selected from the group consisting of sinus bradycardia, sinus rhythm, heart block, paced, atrial fibrilation, atrial flutter, sinus tachycardia and junctional rhythm.

25. The method of claim 23 wherein the function is a constant function.