Apparatus and method for categorizing health care utilization

• US 5,486,999 A
• Filed: 04/20/1994
• Issued: 01/23/1996
• Est. Priority Date: 04/20/1994
• Status: Expired due to Term

• Current Assignee: Andrew H. Mebane
• Original Assignee: Andrew H. Mebane
• Inventors: Mebane, Andrew H.
• Primary Examiner(s): McElheny, Jr., Donald E.
• Application Number:
  US08/232,512
• Time in Patent Office: 643 Days
• Field of Search: 364/401, 364/413.01, 364/413.02, 364/413.03, 395/21, 395/23, 395/924
• 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 40/20 for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms
• 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: Y02A 90/10 Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
• CPC Class Codes: Y10S 128/925 Neural network
• CPC Class Codes: Y10S 706/924 Medical

Abstract

An apparatus and method for categorizing health care utilization provides an efficient aid in identifying patients who are seeking inappropriate care. The invention involves a computer system having a neural network responsive to several input variables to categorize the utilization characteristics of the patient. The input variables define selected characteristics of a patient. In one embodiment, a screening process identifies patients who are at high risk to an immediate threat to their health and eliminates those least likely to be seeking inappropriate care.

Claims

1. A system to aid in identifying patients potentially seeking inappropriate health care, said system comprising:

a plurality of patient characteristics for a plurality of patients, said patient characteristics representing medical usage and psychological information about said patients; and

a neural network responsive to input variables generated from said plurality of patient characteristics, said neural network configured to categorize the plurality of patients according to utilization characteristics of said patients based on said input variables, one category of utilization characteristics identifying patients who are potentially seeking inappropriate health care.

2. A method for categorizing health care utilization of a plurality of patients, said method comprising the steps of:

defining a set of input variables having input values which define selected characteristics of each of the plurality of patients;

forming an input vector for a neural network based upon said set of input variables for each of the plurality of patients; and

analyzing said input vector with said neural network to categorize the health care utilization of at least some of the plurality of patients into at least one of a selected plurality of health care utilization categories.

3. The method of claim 2, wherein said at least one of said selected plurality of health care utilization categories identifies patients who are seeking inappropriate health care.

4. The method of claim 2, wherein at least one of said input variables corresponds to certain aspects of the psychological condition of one of said patients.

5. The method of claim 2, wherein at least one of said input variables corresponds to the diagnoses categories of one of said patients.

6. The method of claim 2, wherein said categories are selected from the group consisting of:

patient driven care, illness driven care, a mix of patient driven care and illness driven care, and possible patient driven care.

7. The method of claim 6, wherein said input variables form an input vector to a neural network trained to categorize patients according to health care utilization based upon said input variables.

8. The method of claim 7, wherein at least one of set of pre-screening variables comprises a first variable indicative of the existence of at least one alcohol or drug addiction disorder.

9. A method for training a neural network to categorize health care utilization of a plurality of patients, said method comprising the steps of:

(a) defining a set of input variables having input values which define selected characteristics of each of a plurality of patients, said plurality of patients forming a training set;

(b) defining at least two categories of health care utilization for said patients;

(c) selecting which of said at least two categories applies to each of said plurality of patients;

(d) providing said input variables for each patient and the selected category for each patient, to the neural network, the selected category for each patient category forming a target output value;

(e) analyzing the input variables with the neural network to obtain a first output value associated with said set of input variables;

(f) comparing said first output value with said target output value;

(g) adjusting weights in the neural network if said first output value does not equal said target output value to partially compensate for the difference between the target output value and the first output value; and

(h) repeating steps (d) through (g) for each of said plurality of patients in the training set until said first output value is substantially the same as said target output value or no more improvement is attained.