Methods and systems for assessing latent traits using probabilistic scoring

• US 8,834,174 B2
• Filed: 02/24/2012
• Issued: 09/16/2014
• Est. Priority Date: 02/24/2011
• Status: Active Grant

• Current Assignee: Patient Tools Incorporated
• Original Assignee: Patient Tools Incorporated
• Inventors: Malik, Alan David
• Primary Examiner(s): Utama, Robert J
• Assistant Examiner(s): Zaman, Sadaruz
• Application Number:
  US13/405,089
• Publication Number:
  US 20120270199A1
• Time in Patent Office: 935 Days
• Field of Search: 434/322, 434/323, 434350-353, 434/365
• US Class Current: 434/322
• CPC Class Codes: A61B 5/0002 Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network A61B5/07 takes precedence; transmission systems for measured values G08C; transmission H04B; transmission of digital information H04L; wireless communication networks H04W
• CPC Class Codes: A61B 5/16 Devices for psychotechnics using teaching or educational appliances G09B1/00 - G09B7/00; Testing reaction times ; Devices for evaluating the psychological state
• CPC Class Codes: A61B 5/165 Evaluating the state of mind, e.g. depression, anxiety
• CPC Class Codes: A61B 5/7475 User input or interface means, e.g. keyboard, pointing device, joystick
• CPC Class Codes: G16H 10/20 for electronic clinical trials or questionnaires
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

A method and system for assessing a latent trait such as a psychiatric disorder in a test subject. The method includes receiving a test subject'"'"'s responses to test items that are administered to the test subject to elicit the responses from the test subject. An initial first sub-region probability of the test subject lying within a first sub-region of a first latent trait is determined from the test subject'"'"'s response to the initial first test item. A subsequent first sub-region probability of the test subject lying within the first sub-region of the first latent trait is then determined using the test subject'"'"'s response to the subsequent first test item to ascertain a conditional response probability, and using the initial first sub-region probability as a prior first sub-region probability. The method and system can be used to more accurately and/or more rapidly assess one or more latent traits in a test subject.

Claims

1. A method for assessing a latent trait in a test subject, comprising the steps of:

providing a graphical user interface for allowing the input of a test subject'"'"'s first test item responses to a plurality of first test items that are administered to the test subject to elicit the first test item responses from the test subject;

receiving the test subject'"'"'s responses to at least an initial first test item and a subsequent first test item;

determining, using one or more processors, an initial first sub-region probability of the test subject lying within a first sub-region of a first latent trait from the test subject'"'"'s response to the initial first test item; and

determining, using one or more processors, a subsequent first sub-region probability of the test subject lying within the first sub-region of the first latent trait using;

the test subject'"'"'s response to the subsequent first test item to ascertain a conditional response probability, and

the initial first sub-region probability as a prior first sub-region probability.

2. The method recited in claim 1, wherein the conditional response probability is ascertained from one or more validated evidence data sets.

3. The method recited in claim 1, further comprising the step of determining, using the processor, a further first sub-region probability of the test subject lying within the first sub-region of the first latent trait using:

the test subject'"'"'s response to a further first test item to ascertain a further conditional response probability; and

the subsequent first sub-region probability as a prior sub-region probability.

4. The method recited claim 1, further comprising the steps of determining, using a processor, an initial second sub-region probability of the test subject lying in a second sub-region of the first latent trait from the test subject'"'"'s first test item response to the initial first test item and determining a subsequent second sub-region probability of the test subject lying within the second sub-region of the first latent trait.

5. The method recited in claim 4, further comprising the step of determining, using the processor, a differential probability value between the probability of the first and second sub-regions of the first latent trait after determining the subsequent sub-region probability of the test subject lying within the second sub-region and determining a subsequent sub-region probability of the test subject lying within the first sub-region.

6. The method recited in claim 5, further comprising the step of administering the plurality of first test items to the test subject through the graphical user interface.

7. The method recited in claim 6, wherein the first test items are administered to the test subject sequentially.

8. The method recited in claim 7, further comprising the step of terminating the administering of the first test items to the test subject when the differential probability value reaches a terminus differential probability value that meets or exceeds a threshold differential probability value.

9. The method recited in claim 5, further comprising the step of determining a severity score for the first latent trait in the test subject from the differential probability value.

10. The method recited in claim 9, wherein the severity score for the first latent trait is determined by mapping the differential probability value to a severity score validated evidence data set.

11. The method recited in claim 1, wherein the first latent trait is a psychiatric disorder.

12. The method recited in claim 11, wherein the first latent trait is selected from the group consisting of depression, dysthymic disorder, bipolar disorder, generalized anxiety disorder, obsessive-compulsive disorder, post-traumatic stress disorder, a psychotic disorder, a personality disorder, a sleeping disorder, an eating disorder, a developmental disorder or a substance dependency.

13. The method recited in claim 11, wherein the first latent trait is depression.

14. The method recited in claim 1, wherein the subsequent sub-region probabilities of all first latent trait sub-regions are determined and are normalized to one.

15. The method recited in claim 1, wherein the plurality of first test items comprise agreement test items that elicit the test subject to indicate a level of agreement or disagreement with a concept.

16. The method recited in claim 15, wherein the agreement test items elicit a level of agreement on a Liked scale.

17. The method recited in claim 1, wherein the test items comprise physiological test items that elicit the test subject to indicate the presence or absence of a physiological condition.

18. The method recited in claim 1, wherein the initial first sub-region probability is determined using an apriori probability of the test subject lying in the first sub-region of the first latent trait.

19. The method recited in claim 1, wherein the initial and subsequent first sub-region probabilities are determined using Bayesian inferences.

20. The method recited in claim 1, wherein the step of administering the plurality of first test items to the test subject comprises administering at least three first test items to the test subject.

21. The method recited in claim 1, wherein the step of administering the plurality of first test items to the test subject comprises administering at least four first test items to the test subject.

22. The method recited in claim 1, further comprising steps of:

receiving the test subject'"'"'s responses to a plurality of second test items that are administered to the test subject to elicit second test item responses from the test subject;

determining an initial first sub-region probability of the test subject lying within a first sub-region of a second latent trait from;

i) the test subject'"'"'s response to an initial test item from the plurality of second test items to ascertain a conditional sub-region probability for the second latent trait, and

ii) the subsequent first sub-region probability of the test subject lying within the first sub-region of the first latent trait as a prior second sub-region probability for the second latent trait.

23. The method recited in claim 22, wherein the second latent trait is a psychiatric disorder.

24. The method recited in claim 23, wherein the first latent trait is depression and the second latent trait is generalized anxiety disorder.

25. The method recited in claim 1, wherein the initial first sub-region probability of the test subject lying within the first sub-region of the first latent trait is determined before receiving the subsequent first test item from the test subject.

26. A method for assessing a latent trait in a test subject, comprising the steps of:

(a) administering, through a graphical user interface, a test item from a plurality of first test items to a test subject to elicit a response to the first test item from the test subject;

(b) determining, using one or more processors, sub-region probabilities of the test subject lying within each of a plurality of sub-regions of a first latent trait from the test subject'"'"'s response to the first test item;

(c) determining, using the one or more processor, a differential probability value between at least first and second adjacent sub-regions of the first latent trait; and

(d) repeating steps (a) to (c) until the differential probability value between the first and second sub-regions is a terminus differential probability value that meets or exceeds a threshold differential probability value.

27. The method recited in claim 26, further comprising the step of:

determining a severity score for the first latent trait in the test subject from the terminus differential probability value.

28. The method recited in claim 27, wherein the severity score for the first latent trait is determined by mapping the terminus differential probability value to a severity score validated evidence data set.

29. The method recited in claim 26, wherein the step of determining the sub-region probabilities of the test subject lying within each of the plurality of sub-regions of the first latent trait comprises determining the sub-region probabilities using:

i) the test subject'"'"'s response to the test item to determine a conditional sub-region probability; and

ii) a previously determined sub-region probability of the test subject lying within each of the plurality of sub-regions of the first latent trait as a prior sub-region probability.

30. The method recited in claim 26, wherein the sub-region probabilities are determined using Bayesian inferences.

31. The method recited in claim 26, wherein the first latent trait is a psychiatric disorder.

32. The method recited in claim 26, wherein the first latent trait is selected from the group consisting of depression, dysthymic disorder, bipolar disorder, generalized anxiety disorder, obsessive-compulsive disorder, post-traumatic stress disorder, a psychotic disorder, a personality disorder, a sleeping disorder, an eating disorder, a developmental disorder or a substance dependency.

33. The method recited in claim 26, wherein the first latent trait is depression.

34. The method recited in claim 26, wherein the sub-region probabilities of the test subject lying within each of the plurality of sub-regions of the first latent trait are normalized to one before determining a differential probability value.

35. The method recited in claim 26, wherein the plurality of first test items comprise agreement test items that elicit the test subject to indicate a level of agreement or disagreement with a concept.

36. The method recited in claim 35, wherein the agreement test items elicit a level of agreement on a Liked scale.

37. The method recited in claim 26, wherein the plurality of first test items comprise physiological test items that elicit the test subject to indicate the presence or absence of a physiological condition.