Between-patient comparisons for risk stratification of future heart failure decompensation

• US 8,768,718 B2
• Filed: 12/27/2006
• Issued: 07/01/2014
• Est. Priority Date: 12/27/2006
• Status: Active Grant

• Current Assignee: Cardiac Pacemakers Incorporated (Boston Scientific Corporation)
• Original Assignee: Cardiac Pacemakers Incorporated (Boston Scientific Corporation)
• Inventors: Cazares, Shelley, Peterson, Jon, Sachanandani, Haresh G., Stalsberg, Kevin J., Sweeney, Robert J., Siejko, Krzysztof Z., Carlson, Gerrard M.
• Primary Examiner(s): Sereboff, Neal
• Assistant Examiner(s): COUPE, ANITA YVONNE
• Application Number:
  US11/616,430
• Publication Number:
  US 20080162182A1
• Time in Patent Office: 2,743 Days
• Field of Search: 705 2- 3
• US Class Current: 705/2
• CPC Class Codes: A61B 5/0031 Implanted circuitry
• CPC Class Codes: A61N 1/37282 characterised by communication with experts in remote locations using a network
• CPC Class Codes: G16H 40/67 for remote operation
• 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 50/50 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

This document discusses, among other things, systems and methods for using between-patient comparisons for risk stratification of future heart failure decompensation. A method comprises receiving patient data of a current patient, the patient data collected by a patient monitoring device; determining a reference group related to the patient; determining a reference group dataset selected from the reference group, wherein the dataset includes patient data that is of a similar type received from the patient monitoring device; generating a model of the reference group dataset; and automatically comparing the received physiological data to a model to derive an index for the patient.

Claims

1. A method comprising:

receiving, at an implantable medical device, patient data of a current patient;

determining, at the implantable medical device, a reference group related to the current patient, the reference group including patients with at least one similar characteristic to the current patient;

detecting, at the implantable medical device, a triggering event, the triggering event comprising an interrogation of the implantable medical device;

updating, at the implantable medical device, the population of the reference group related to the patient in response to detecting the triggering event;

determining, at the implantable medical device, a reference group dataset selected from the updated reference group, wherein the dataset includes patient data that is of a similar type received from the implantable medical device;

generating, at the implantable medical device, a model of the reference group dataset, wherein the model is generated using a probability distribution function; and

deriving a numerical index by calculating an area under the probability distribution function and using the area as the numerical index that indicates an activity index of the current patient, and wherein calculating the area under the probability distribution function comprises;

identifying a plurality of points on the probability distribution function, the plurality of points having equivalent probability density and defining an interval or area of integration; and

calculating the area of the interval or area of integration; and

using the numerical index to determine at least one of;

a similarity of the patient to the reference group, how likely the patient is to decompensate within a timeframe relative to a reference group'"'"'s likelihood of heart failure decompensation in the timeframe, how likely the patient is to die in a timeframe relative to a reference group'"'"'s likelihood of dying in the timeframe, how likely the patient'"'"'s quality of life is to change a timeframe relative to a reference group'"'"'s likelihood of quality of life changing in the timeframe, or stratify the patient into one or more discrete groups associated with one or more risk levels.

2. The method of claim 1, wherein the reference group is selected from patients associated with a centralized patient management system.

3. The method of claim 1, wherein the reference group is selected from patients similar to the current patient using one or more criteria, wherein the criteria include age, gender, ethnicity, clinic, left ventricular ejection fraction (LVEF), New York Heart Association (NYHA) heart failure classification, results of a six-minute walk test (6MW), quality of life, heart failure etiology, body mass index (BMI), blood pressure, medication, co-morbidity, arrhythmia history, implant history, treatment compliance, health care system, circadian rhythm, geographic location, patients who have survived for a particular time frame, or patients who have not decompensated in a particular time frame.

4. The method of claim 3, wherein the triggering event comprises detecting a change of one or more of the criteria;

and wherein the updating the reference group comprises using the changed criteria.

5. The method of claim 1, wherein the reference group includes one or more patients that are considered to be in stable condition using one or more criteria, wherein the criteria include patients who did not die, decompensate, or experience a decrease in quality life during a particular time period.

6. The method of claim 1, wherein the risk levels are indicative of a likelihood of heart failure decompensation in a particular timeframe.

8. The method of claim 1, wherein the patient data includes one or more of an electrocardiogram, an activity level, a heart rate variability, a heart rate, a heart sound, a blood pressure, an impedance, a body weight, or an O₂ saturation.

9. The method of claim 1, wherein the triggering event includes one or more of an implantation of a new patient device, an introduction of a new patient device, an approval of a revised reference group, a periodic or recurring time interval, a change in a health state of one or more patients, a user command, or a request to access the reference group.

10. The method of claim 1, wherein the triggering event comprises a change in the patient'"'"'s health condition, and wherein the change in the patient'"'"'s health condition comprises at least one of:

an implantation of a new patient device, an introduction of a new patient device, or an introduction of a new drug.

7. The method of 1, wherein using the numerical index includes comparing the numerical index to one or more threshold values to categorize the patient into a risk level.

11. A system comprising:

an implantable patient device comprising;

a communication module adapted to receive patient data associated with a patient;

a reference group module adapted to;

select and store reference group data corresponding to the patient;

detect a triggering event, the triggering event comprising an interrogation of the patient device; and

update the population of the reference group corresponding to the patient in response to detecting the triggering event; and

an analysis module adapted to;

generate a model of the updated reference group data; and

compare the patient data to the model to calculate a numerical index for the patient by calculating an area under a probability distribution function, and using the area as the numerical index that indicates an activity index of the current patient, and wherein to calculate the area under the probability distribution function, the analysis module is adapted to;

identify a plurality of points on the probability distribution function, the plurality of points having equivalent probability density and defining an interval or area of integration; and

calculate the area of the interval or area of integration; and

use the numerical index to determine at least one of;

a similarity of the patient to the reference group, how likely the patient is to decompensate within a timeframe relative to a reference group'"'"'s likelihood of heart failure decompensation in the timeframe, how likely the patient is to die in a timeframe relative to a reference group'"'"'s likelihood of dying in the timeframe, how likely the patient'"'"'s quality of life is to change a timeframe relative to a reference group'"'"'s likelihood of quality of life changing in the timeframe, or stratify the patient into one or more discrete groups associated with one or more risk levels.

12. The system of claim 11, comprising a centralized patient management system.

13. The system of claim 11, wherein the patient data includes patient compliance data.

14. The system of claim 11, wherein the patient data includes physiological data.

15. The system of claim 11, wherein the triggering event comprises a change in the patient'"'"'s health condition, and wherein the change in the patient'"'"'s health condition comprises at least one of:

an implantation of a new patient device, an introduction of a new patient device, or an introduction of a new drug.

16. A non-transitory computer-readable medium, including instructions that, when performed by an implantable medical device, cause the implantable medical device to:

receive patient data of a current patient, the patient data collected by a patient monitoring device;

determine a reference group related to the current patient;

detect a triggering event, the triggering event comprising an interrogation of the patient monitoring device;

update the population of the reference group related to the current patient in response to detecting the triggering event;

determine a reference group dataset selected from the updated reference group, wherein the dataset includes patient data that is of a similar type received from the patient monitoring device;

generate a model of the updated reference group dataset, wherein the model is generated using a probability distribution function; and

derive a numerical index by calculating an area under the probability distribution function and using the area as the numerical index that indicates an activity index of the current patient, and wherein the instructions to calculate the area under the probability distribution function comprises instructions to;

identify a plurality of points on the probability distribution function, the plurality of points having equivalent probability density and defining an interval or area of integration; and

calculate the area of the interval or area of integration; and

use the numerical index to determine at least one of;

a similarity of the patient to the reference group, how likely the patient is to decompensate within a timeframe relative to a reference group'"'"'s likelihood of heart failure decompensation in the timeframe, how likely the patient is to die in a timeframe relative to a reference group'"'"'s likelihood of dying in the timeframe, how likely the patient'"'"'s quality of life is to change a timeframe relative to a reference group'"'"'s likelihood of quality of life changing in the timeframe, or stratify the patient into one or more discrete groups associated with one or more risk levels.

17. The non-transitory computer-readable medium of claim 16, wherein the reference group is selected from patients similar to the current patient using one or more criteria, wherein the criteria include age, gender, ethnicity, clinic, left ventricular ejection fraction (LVEF), New York Heart Association (NYHA) heart failure classification, results of a six-minute walk test (6MW), quality of life, heart failure etiology, body mass index (BMI), blood pressure, medication, co-morbidity, arrhythmia history, implant history, treatment compliance, health care system, circadian rhythm, geographic location, patients who have survived for a particular time frame, or patients who have not decompensated in a particular time frame.

18. The non-transitory computer-readable medium of claim 17, wherein the triggering event comprises detecting a change of one or more of the criteria;

and wherein the updating the reference group comprises using the changed criteria.

19. The non-transitory computer-readable medium of claim 16, wherein the triggering event includes one or more of an implantation of a new patient device, an introduction of a new patient device, an approval of a revised reference group, a periodic or recurring time interval, a change in a health state of one or more patients, a user command, or a request to access the reference group.

20. The non-transitory computer-readable medium of claim 16, wherein the triggering event comprises a change in the patient'"'"'s health condition, and wherein the change in the patient'"'"'s health condition comprises at least one of:

an implantation of a new patient device, an introduction of a new patient device, or an introduction of a new drug.