Method and apparatus for improving patient compliance with prescriptions

• US 6,587,829 B1
• Filed: 07/30/1998
• Issued: 07/01/2003
• Est. Priority Date: 07/31/1997
• Status: Expired due to Fees

• Current Assignee: Schering Corporation (Merck & Co., Inc.)
• Original Assignee: Schering Corporation (Merck & Co., Inc.)
• Inventors: Hanson, Gregory K., Camarda, Leonard T., Schectman, Stephen B., Wilbur, Curtis S.
• Primary Examiner(s): Smith, Jeffrey A.
• Assistant Examiner(s): Thompson, Jr., Forest O.
• Application Number:
  US09/126,369
• Time in Patent Office: 1,797 Days
• Field of Search: 705/3, 705/2, 707/10, 600/300, 436/111, 379/106.02, 345/327, 340/311.1, 128/905
• US Class Current: 705/3
• CPC Class Codes: B82Y 10/00 Nanotechnology for information processing, storage or transmission, e.g. quantum computing or single electron logic
• CPC Class Codes: G06N 3/123 DNA computing
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G16H 10/60 for patient-specific data, e.g. for electronic patient records
• CPC Class Codes: G16H 20/10 relating to drugs or medications, e.g. for ensuring correct administration to patients
• CPC Class Codes: G16H 40/67 for remote operation
• CPC Class Codes: G16H 50/30 for calculating health indices; for individual health risk assessment

Abstract

A method and apparatus for improving patient compliance with prescriptions utilizes computer terminals to convert prescription information into electronic form as records for each patient. Each such patient record includes information about the patient from the prescription and the prescription itself. In addition, demographic information about the patient is obtained from other commercial databases and added to the patient record. Then a regression analysis is run on the patient records using the various data elements versus the compliance of the patient with the prescription to determine the relative importance of each variable in the prediction. The analysis is used to segregate the patients into demographic clusters and to associate champion intervention messages with each cluster. When a patient'"'"'s prescription is entered in the system thereafter, the model is used to associate the patient with a cluster and to direct the champion message to that patient. Further, a regression analysis may be run on the patient data to create a model of likelihood of prescription compliance in general by the patient. The result is a probability equation that allows a score to be assigned to each record. Based on this score; the patients most likely to fail to comply with their prescription are sent the champion interventions. As new challenge interventions messages are created additional information is gathered and the regression analysis is re-run. If the challenge message is more successful, it is substituted as the new champion message. The results of these interventions are recorded and appended to the related patient record.

Claims

1. A method of improving patient compliance with prescriptions, comprising the steps of:

a) collecting prescription data in electronic form on a population of patients, said prescription data including at least an identification of the patient and the prescription;

b) forming electronic patient records in a database, said patient records including said prescription data for each patient;

c) providing intervention data in electronic form which at least generally indicates prescription compliance response in the form of the response of various patients to various predetermined interventions, said interventions being intended to get patients to comply with their prescription;

d) appending said intervention data to said patient records; and

e) automatically creating an intervention model using the data in said patient records, said intervention model segregating the patient records into clusters based on data in said patient records and associating each cluster with a champion intervention which is the predetermined intervention which is most effective in causing patients in that cluster to behave so as to create a general indication of compliance, said intervention model being created at least in part by regression analysis of the data in the patient records.

2. The method according to claim 1 further including the steps of:

obtaining demographic information on patients;

appending said demographic information to the related patient records; and

wherein said creating step involves segregating the patient records into clusters based on said demographic information.

3. The method according to claim 2 further including the step of intervening with patients in a cluster by sending the champion intervention for that cluster.

4. The method according to claim 1 further including the steps of

providing compliance information for at least some of the patients in the form of data indicating compliance of patients with prescriptions without an intervention;

appending said compliance information to related patient records;

creating a compliance model based on patient data for predicting which patients will comply with a prescription, said compliance model being created by regression analysis of the compliance information with regard to the other data in the patient records; and

assigning a score to each patient record based on the model, said score being a prediction of the relative likelihood that a patient will comply with his prescription.

5. The method according to claim 4 further including the step of:

j) intervening with patients in a cluster by sending the champion intervention for that cluster to those patients of the cluster with scores indicating a probability of lack of compliance above a certain level.

6. The method according to claim 4 further including the step of including the score in the related patient record.

7. A method of improving patient compliance with prescriptions, comprising the steps of:

a) collecting prescription data in electronic form on a population of patients, said prescription data including at least an identification of the patient and the prescription;

b) forming electronic patient records in a database, said patient records including said prescription data for each patient;

c) automatically providing an intervention model based on regression analysis which defines patient clusters and a predetermined champion intervention associated with each cluster;

d) applying patient data in the patient records to said intervention model;

e) providing a challenger intervention for at least one cluster;

f) sending champion interventions to some patients and challenger interventions to other patients of a cluster;

g) collecting compliance information related to the champion and challenger interventions;

h) comparing the compliance information related to the champion and challenger interventions; and

i) substituting the challenger intervention for the champion intervention when the challenger intervention is more effective.

8. The method according to claim 7 further including the steps of:

obtaining demographic information on patients, and

appending said demographic information to the related patient records; and

wherein said providing step involves providing an intervention model which defines demographic patient clusters.

9. The method according to claim 8 where in the step of comparing comprises the step of creating a new intervention model using the data in said patient records and compliance information, said new intervention model performing the steps of:

segregating the patient records into clusters; and

associating each cluster with a new champion intervention, from among said champion and challenger interventions, that is the predetermined intervention which is most effective in causing patients in that cluster to behave so as to create a general indication of compliance, said intervention model being created at least in part by regression analysis of the data in the patient records.

10. A method of improving patient compliance with prescriptions, comprising the steps of:

a) providing prescription data in electronic form on a population of patients, said prescription data including at least an identification of the patient and the prescription, said prescription data being in the form of electronic patient records in a database, said patient records including said prescription data for each patient;

b) automatically providing an intervention model based on regression analysis which defines patient clusters and a predetermined champion intervention associated with each cluster;

c) applying new patient data to said intervention model; and

d) sending champion interventions based on the cluster to which the intervention model assigns the patient.

11. The method according to claim 10 further including the steps of

obtaining demographic information on patients, and

appending said demographic information to the related patient records; and

wherein the regression analysis defines demographic clusters.

12. The method according to claim 10 further including the steps of:

providing compliance information for at least some of the patients in the form of data indicating compliance of patients with prescriptions without an intervention;

appending said compliance information to related patient records;

creating a compliance model based on patient data for predicting which patients will comply with a prescription, said compliance model being created by regression analysis of the compliance information with regard to the other data in the patient records; and

assigning a score to each patient record based on the model, said score being a prediction of the relative likelihood that a patient will comply with his prescription.

13. The method of claim 12 wherein the interventions are made only with those patients with scores indicating a probability of lack of compliance above a certain level.

14. The method according to claim 12 wherein the intervention occurs just after the step of applying the new patient data and at the same location.

15. The method according to claim 10 wherein the clusters are based on the condition of the patient as implied by the prescription.

16. The method according to claim 10 wherein the intervention occurs just after the step of applying the new patient data and at the same location.

17. The method of claim 16 wherein the step of applying new patient data is performed on a computer terminal at one of a physician'"'"'s office and a pharmacy.

18. A method of improving patient compliance with prescriptions, comprising the steps of:

a) collecting prescription data in electronic form on a population of patients, said prescription data including at least an identification of the patient and the prescription;

b) forming electronic patient records in a database, said patient records including said prescription data for each patient;

c) providing compliance information for at least some of the patients in the form of data indicating compliance of those patients with prescriptions without an intervention;

d) appending said compliance information to related patient records;

e) automatically creating a compliance model based on patient data for predicting which patients will comply with a prescription, said compliance model being created by regression analysis of the compliance information with regard to the other data in the patient records; and

f) assigning a score to each patient record based on the model, said score being a prediction of the relative likelihood that a patient will comply with his prescription.

19. The method according to claim 18 further including the step of converting the score into an odds ratio that indicates the probability that the patient will comply with his prescription.

20. The method according to claim 18 further including the step of generating an intervention designed to cause the patient to comply with his prescription and causing said intervention to be directed to a responsible party capable of effecting the patient'"'"'s compliance.