Granted Patent B2

US 12,224,068 · App. 16/961,266 · Granted Feb 11, 2025

Inventors: Raymond L. Ownby (Fort Lauderdale, FL); Amarilis Acevedo (Fort Lauderdale, FL); Drenna Waldrop-Valverde (Atlanta, GA)

Assignees: Nova Southeastern University; Emory University

G16H50/30G16H10/20G16H10/60

Abstract

Claims (31)

1. A method of automatically evaluating an assessment of a patient's health literacy in order to present therapeutic instructions which are understandable by the patient, comprising:

using an automated agent having a form of instructions carried out by at least one electronic processor, the automated agent acting to:

access health literacy data from members of a population of at least 100 individuals, wherein the health literacy data indicates that members of the population understand health-related information and health-related words;

access demographic data from the members of the population corresponding to age, race, and level of education;

collect skills data from the patient corresponding to a set of questions presented to the patient relating to skills needed for understanding therapeutic instructions;

collect personal data relating to the age, race, and level of education of the patient;

assign a health literacy level to the patient using a polytomous logistic regression with i) the health literacy of members of the population, ii) the demographic data of members of the population, iii) the skills data of the patient, and iv) the personal data of the patient,

wherein the health literacy level is one of a plurality of groups based upon the abilities, skills and knowledge (ASK) predictors of Tests of Functional Health Literacy in Adults (TOFHLA), wherein the plurality of groups include;

proficient includes individuals who can, at least, independently read text at a high school level;

intermediate includes individuals who can, at least, independently read text at an eighth-grade level;

basic includes individuals who can, at least, read sentences below the eighth-grade level; and

below basic includes individuals who can read words but cannot understand sentences; and

select a communication strategy for the patient corresponding to the assigned health literacy level of the patient, by a computer, to enable the patient to have a communication of therapeutic instructions when the patient has responded to the set of questions, wherein the communication of therapeutic instructions is based upon the assigned health literacy of the patient to include:

for a below basic health literacy level, use of at least one of graphics, audio narration, and repetition;

for a basic health literacy level, use of single step instructions with text in sentences below a reading competence of the eighth-grade level;

for an intermediate health literacy level, use of multi-step instructions; and

for a proficient health literacy level, use of multi-step instructions and additional content in written form.

2. The method of claim 1 , further comprising:

communicating to the patient, by the automated agent, the therapeutic instructions based on the health literacy level.

3. The method of claim 1 , further including performing a binary logistic regression upon results of the polytomous logistic regression.

4. The method of claim 3 , wherein the binary logistic regression is performed for individuals determined to be at one of two lowest levels assigned for health literacy level groups.

5. The method of claim 1 , wherein the polytomous logistic regression produces a predicted variable which is a natural log of a probability of any individual being a member of one of the groups.

6. The method of claim 1 , wherein the polytomous logistic regression uses a linear combination of an intercept constant and other variables, the variables being weighted so as to provide a composite that is a least-squares solution.

7. The method of claim 1 , wherein the polytomous logistic regression forms a maximum likelihood approach in which the automated agent iteratively derives a solution that is a closest approximation to an actual data while minimizing a second derivative of a function of a model corresponding to a likelihood of the patient being a member of one of the groups.

8. The method of claim 1 , wherein the assignment of a group includes the automated agent calculating log odds of any group membership for each individual as a linear function of performance on the health literacy using the measures of age, education, race, and gender.

9. The method of claim 8 , wherein the linear function of performance includes a native language of the patient in addition to age, education, race, and gender.

10. The method of claim 9 , wherein log odds group membership-β 0 +β 1 ·age+β 2 ·education+β 3 ·language+β 4 ·race+β 5 ·gender+β 6 ·health literacy test score, where β i refers to numerical values drawn from a statistical output of the polytomous logistic regression.

11. The method of claim 1 , wherein a probability of group membership of an individual is further derived by taking an antilog of a result of the polytomous logistic regression and converting odds into probabilities, with the greatest probability taken as the group membership of the individual.

12. The method of claim 1 , wherein the health literacy assigned to each member of the population is based upon one or more known health literacy assessment techniques.

13. The method of claim 12 , wherein the known techniques of assessing population health literacy data include at least one of Tests of Functional Health Literacy in Adults (TOFHLA) and FLIGHT/VIDAS (F/V).

14. The method of claim 1 , wherein the set of questions includes not more than 30 questions.

Assignments (3)

Continuity (2)

References Cited (27)