Effectiveness of a Personalized Health Profile on Specificity of Self-Management Goals

Overview

With a growing number of people living with chronic diseases, the need to empower people for self-management is rising. A key element in self-management is goal setting but the extent to which meaningful actionable goals (i.e., SMART) can be set without direction from the health care team is not known. Providing people with specific feedback on actionable health outcomes may stimulate the setting of specific goals. To this end, a health outcome profile was computer generated from the existing outcome measures, at first and last recorded visits, of each person enrolled in the Positive Brain Health Now (+BHN) cohort from 5 sites in Canada. This profile will be tested with BHN participants who agreed to enrolled in sub-studies. The main outcome will be the extent to which goals are SMART by using specific words and actionable verbs. A measurement framework and an initial lexical (i.e., collection of vocabularies) has been developed for the goal evaluation. Using text mining techniques (i.e., tokenizing and pos-tagging), the specific components of each goal will be extracted and compared to the lexical using regular expression algorithms. The result will provide information on the specificity of participants' defined goals.

Introduction: With a growing number of people living with chronic diseases, the need to empower people to self-manage their condition is rising. A key element in self-management is goal setting, however, it is not known to what extent people living with chronic conditions are capable of setting their own actionable health-related goals. One possible strategy to improve goal-setting skills is to provide people with specific information about their health profile to stimulate the setting of specific goals. This study contributes to the understanding of the specificity of patient-formulated self-management goals by testing the effectiveness of providing feedback using a personalized health outcome profile. Objective: To estimate among people living with HIV, to what extent providing feedback on their health outcomes, compared to no feedback, will affect number and specificity of person-defined self-management goals. Design: A blinded, stratified, randomized controlled trial design will be conducted targeting all participants in the BHN2 (Action for Positive Brain Health Now) cohort in Canada. Methods: A personalized health outcome profile has been produced for each person enrolled in the BHN1 cohort (Positive Brain Health Now) at cohort entry and at the last recorded visit. The same profile will be developed for the participants who will enroll in the BHN2 as they go through their visits. The profile covers information on brain health outcomes, health and quality of life ratings, and lifestyle factors. Participants from the BHN1 who agreed to enroll in the BHN2 as well as the new participants will be contacted. The intervention group will receive their personal profile along with instructions on goal-setting and tips to improve brain health; the control group will receive only the goal-setting instructions and tips. Over 800 members of the BHN cohort have sufficient data to generate the profile and will be recruited into the study. Persons will be instructed to enter their goals on a specialized web-based goal setting platform within 2 weeks of receiving their profile. At the close of the two-week period, the control group will receive their profile. Text mining techniques will be used to extract information from the person-defined goals and the specificity of the goal will be scored based on word matches to a developed goal-setting lexical. Expected Results: The expectation is that the intervention group will set more goals and have more words matching the developed lexical than the control group. The total number of words per person-goal will be calculated for each group and Poisson regression will be used to estimate the rate ratio and 95% confidence intervals and compare rate ratios between men and women using an interaction term. Conclusion: The effect of providing outcome feedback on setting effective self-management goals will be tested in this study. The results will provide a thorough understanding of the quality of person-defined goals using text mining. Moreover, this study provides evidence necessary for future studies using text mining as an inexpensive and timely way for evaluation of textual information.