Techniques for estimating charges of delivering healthcare services that take complicating factors into account

• US 6,061,657 A
• Filed: 02/18/1998
• Issued: 05/09/2000
• Est. Priority Date: 02/18/1998
• Status: Expired due to Fees

• Current Assignee: Crossflo Systems, Inc. (Patriot Scientific Corporation)
• Original Assignee: Iameter, Inc.
• Inventors: Whiting-O'Keefe, Quinn
• Primary Examiner(s): Trammell, James P.
• Assistant Examiner(s): Nguyen, Cuong H.
• Application Number:
  US09/025,321
• Time in Patent Office: 811 Days
• Field of Search: 705/10, 705/1, 705/8, 705/2-4, 705/34, 706/924, 706/923, 702/179
• US Class Current: 705/2
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 30/04 Billing or invoicing
• CPC Class Codes: G06Q 99/00 Subject matter not provided for in other groups of this subclass
• CPC Class Codes: G16H 10/60 for patient-specific data, e.g. for electronic patient records
• CPC Class Codes: G16H 40/20 for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms

Abstract

The expected charges for treating a patient are estimated by the use of linear regression techniques wherein variables and coefficients of estimate models are built from historic patient data of all episode types (inpatient, office visits, etc.) and which include secondary and collateral illnesses that greatly affect the cost of treating a patient for the primary illness. In addition to evaluating the primary illness, the evaluation of collateral illness is also made possible since the processing techniques are not simply averaging data for the primary illness as has been done in the past. This allows physicians to understand the magnitude and types of resource usage that are appropriate for particular illnesses. The present invention allows better management of many aspects of the healthcare system by significantly reducing a component of the difference between the estimate and actual charges for treating a particular illness of a patient that is due to the general state of health of the patient. This then allows the remaining difference to be attributed to differences in the way that health service providers treat each illness of a patient. Estimates are made for the expected cost of providing specific episodes of care, to treat each illness in its entirety by use of both in-patient and out-patient services, and to provide the specific procedures used for the appropriate management of each illness of each patient.

Claims

1. A method of managing delivery of services by healthcare providers to medical patients, comprising:

(A) creating a table of regression coefficients from individual records of encounters of a population of patients with healthcare providers, by a method including;

accumulating and storing data from the encounter records in a mass storage system of a computer, data of individual ones of the encounter records including at least (1) an identity of a single patient, (2) charges of the healthcare providers for the encounter, and (3) at least one diagnosis made or procedure performed,

grouping said encounter records, from information provided therein, into a plurality of summary records for individual ones of the population of patients and one of a plurality of primary illnesses,

establishing an estimate model of a total amount of charges for the encounters within a summary record as a function of a plurality of model variables and regression coefficients taken or derivable from the data within said summary records,

solving, separately for individual ones of the primary illnesses, the estimate model for the regression coefficients that optimizes fits of said estimate model with the data within said summary records, and

storing said regression coefficients in a table within the computer mass storage system, and

(B) estimating the charges for treating an illness of at least one patient, by a method including;

grouping, within the computer, data of the encounter records of said at least one patient, from information provided therein, into at least one summary record for one of the plurality of primary illnesses,

reading the regression coefficients from said stored table for the primary illness of said summary record,

solving the estimate model for estimated charges by use of the read regression coefficients, and

(C) utilizing the estimated charges to manage the delivery of health services by the healthcare providers.

2. The method according to claim 1, wherein accumulating and storing records of individual encounters includes accumulating and storing records of hospital and outpatient encounters for individual patients, and wherein grouping the encounter records includes grouping records of hospital and outpatient encounters into common ones of the summary records.

3. The method according to claim 2, wherein the outpatient encounters for which data is accumulated, stored and grouped include office visits, day encounters and therapeutic services.

4. The method according to claim 1, wherein data of individual ones of the encounter records include data of patient conditions that are collateral to the primary illnesses, and wherein one or more of the model variables and/or regression coefficients of the estimate model is taken or derived from data of the collateral conditions.

5. The method according to claim 4, wherein data of collateral conditions includes data of illnesses other than the primary illness.

6. The method according to claim 1, wherein grouping said encounter records includes determining a primary illness for individual ones of the summary records.

7. The method according to claim 6, wherein determining a primary illness includes reviewing summary record data for an individual patient for data of encounters occurring prior to those for which data are included in the summary record.

8. The method according to claim 1, wherein establishing the estimate model includes establishing more than one specific estimate model with an estimate of charges of one specific estimate model being used as a model variable of a second estimate model, and wherein solving the estimate model includes solving the specific estimate models in sequence.

9. The method according to claim 8, wherein said one specific estimate model is chosen from multiple alternative specific estimate models that use a different set of model variables and/or regression coefficients from each other, and wherein solving the specific estimate model includes solving one of the multiple estimate models having the greater number of variables and/or regression coefficients for which encounter record data is available.

10. The method according to claim 1, wherein a length of time for which data of encounters is included in one of the summary records is specified for the primary illness of the summary record.

11. The method according to claim 10, wherein said length of time is an indefinite duration for chronic ones of the primary illnesses.

12. The method according to claim 10, wherein said length of time is a specified finite duration for acute ones of the primary illnesses.

13. The method according to claim 12, wherein said specified finite duration of time is extended when data exists of a succession of related encounters that begin within the specified duration but extend beyond said specified duration.

14. The method according to claim 1, wherein grouping data of the encounter records into the summary records includes first grouping such data into episodes of one of the plurality of primary illnesses and then grouping the data of the episodes into the summary records of the same primary illnesses.

15. The method according to claim 1, wherein the data of the encounter records of said at least one individual patient are not within the data from encounter records used to create the table of regression coefficients.

16. The method according to claim 1, wherein creating the table of regression coefficients and estimating the charges for treating an illness of at least one patient each include determining, from the summary records data and estimated charges, charges for one of a plurality of specific procedures performed.

17. A method of estimating charges of healthcare providers to at least one patient for the purpose of providing advice on the efficiency of such providers, comprising:

accumulating and storing, in a mass storage system of a computer, data from records of encounters of said at least one patient with healthcare providers that includes at least (1) an identity of said at least one patient, (2) charges of the healthcare providers for the encounter, and (3) at least one diagnosis made or procedure performed,

grouping data of said encounter records, from information provided therein, into at least one summary record of said at least one patient for one of a plurality of primary illnesses,

solving an estimate model of a total amount of charges for the encounters within a summary record as a function of a plurality of model variables and regression coefficients taken or derivable from the data within said at least one summary record, using regression coefficients previously determined with the same estimate model to optimize a fit of said estimate model for a population of patients with data within a summary record corresponding to said at least one summary record, and

utilizing the estimated charges to advise on the efficiency of the healthcare providers in the delivery of health services.

18. A computer readable mass storage system storing the following:

a plurality of raw data files of individual records in multiple different formats of patient encounters with both in patient and out patient healthcare providers,

a common patient data file containing an indexed version of the raw data contained in the patient encounter data files,

a plurality of tables containing definitions including those of various illnesses,

a table of regression coefficients calculated from at least the indexed patient data file and the definition tables for various combinations of primary and collateral illnesses,

processing software including an estimate model that accesses the table of regression coefficients according to illnesses of specific patients to cause the computer to calculate from healthcare encounter data of such specific patients estimated charges to treat these patients, and

an output table containing data of the calculated estimated charges.

19. A computer readable mass storage system storing the following:

a plurality of data files of individual records of patient encounters with both in patient and out patient healthcare providers, said records including patient illness diagnoses,

a table of regression coefficients that have been calculated for various combinations of illnesses,

processing software including an estimate model that accesses the table of regression coefficients according to illnesses of specific patients to cause the computer to calculate from healthcare encounter data of such specific patients estimated charges to treat those patients, and

an output table containing data of the calculated estimated charges.

20. A method of estimating charges of healthcare providers to at least one patient, comprising:

accumulating and storing, in a mass storage system of a computer, data from records of encounters of said at least one patient with healthcare providers that includes at least (1) an identity of said at least one patient, (2) charges of the healthcare providers for the encounter, and (3) at least one diagnosis made or procedure performed,

grouping data of said estimate encounter records, from information provided therein, into at least one summary record of said at least one patient for one of a plurality of primary illnesses,

solving an estimate model of a total amount of charges for the encounters within a summary record as a function of a plurality of model variables and regression coefficients taken or derivable from the data within said at least one summary record, said regression coefficients having been previously determined with the same estimate model to optimize a fit of said estimate model for a population of patients with data within a summary record corresponding to said at least one summary record, and

utilizing the estimated charges as an indication of the efficiency of the healthcare providers in the delivery of health services to said at least one patient.

21. The method according to claim 20, wherein accumulating and storing records of individual encounters includes accumulating and storing records of hospital and outpatient encounters for individual patients, and wherein grouping the encounter records includes grouping records of hospital and outpatient encounters into common ones of the summary records.

22. The method according to claim 21, wherein the outpatient encounters for which data is accumulated, stored and grouped include office visits, day encounters and therapeutic services.

23. The method according to claim 20, wherein data of individual ones of the encounter records include data of patient conditions that are collateral to the primary illnesses, and wherein one or more of the model variables and/or regression coefficients of the estimate model is taken or derived from data of the collateral conditions.

24. The method according to claim 23, wherein data of collateral conditions includes data of illnesses other than the primary illness.

25. The method according to claim 20, wherein grouping said encounter records includes determining a primary illness for individual ones of the summary records.

26. The method according to claim 25, wherein determining a primary illness includes reviewing summary record data for an individual patient for data of encounters occurring prior to those for which data are included in the summary record.

27. The method according to claim 20, wherein a length of time for which data of encounters is included in one of the summary records is specified for the primary illness of the summary record.

28. The method according to claim 27, wherein said length of time is an indefinite duration for chronic ones of the primary illnesses.

29. The method according to claim 27, wherein said length of time is a specified finite duration for acute ones of the primary illnesses.

30. The method according to claim 29, wherein said specified finite duration of time is extended when data exists of a succession of related encounters that begin within the specified duration but extend beyond said specified duration.

31. The method according to claim 20, wherein grouping data of the encounter records into the summary records includes first grouping such data into episodes of one of the plurality of primary illnesses and then grouping the data of the episodes into the summary records of the same primary illnesses.