Objectives: My overall career goal is to use novel methods to develop and implement systems that support the effective management of serious, high-cost subspecialty conditions within and outside of VA. In this CDA I plan to use inflammatory bowel disease (IBD) as a model condition to accomplish the following specific aims: 1) to compare the accuracy and calibration of traditional regression vs. machine-learning models for predicting IBD exacerbations;2) to develop and use a microsimulation model to compare the clinical and economic impact of making patient decisions using current guidelines, a traditional regression-based model and a machine learning-based model;and, 3) to develop and pilot a personalized medical decision support tool for veterans with IBD. Research Plan: Veterans with "High Expense, Low Prevalence" (HELP) diseases such as rheumatoid arthritis, multiple sclerosis, and inflammatory bowel disease often have exacerbations that can result in preventable mortality or major morbidity. Although some of these patients require lifelong, expensive, and potentially harmful medications to prevent serious complications, many others are at lower risk and are better treated with less expensive and less harmful medications, or by using "as-needed" therapy as flares occur. In addition, it has been clearly demonstrated that physicians do not prescribe medications in an efficient manner, and in fact may over-treat, when given the freedom to make choices. Therefore, stratifying veterans with HELP diseases, using biomarker-driven tools, into those at higher vs. lower risk offers great promise to significantly improve both the quality and efficiency of veteran care, and to minimize harm associated with more aggressive therapies. Developing tools and decision support systems to guide clinicians in personalizing medical decision- making for veterans with HELP diseases has particular application for the VA, because having a physician at every facility that subspecializes in each HELP disease is not feasible. However, to implement this "targeted" or "tailored" prevention approach to risk stratifying individuals for disease exacerbation and treatment, a clinician must know both the individual's baseline risk of disease complications and the probability that the individual would benefit (or suffer harm) from therapy. Having risk stratification tools developed and validated within the veteran population is an important first step towards realizing efficient patient-centered care for HELP diseases through the VA. Towards this goal, this CDA proposes to develop "targeted-prevention" prediction tools and decision support systems to facilitate the delivery of timely and cost-effective therapy for HELP diseases and to compare it to the current "symptom-driven" model used by clinicians. The proposal focuses on IBD as a model condition for HELP diseases. Methods: My research plan involves a series of sequential studies.
In Aim 1 (year 1 and 2 of the award), I will compare the accuracy of regression models and machine learning approaches for predicting exacerbations of disease among veterans with IBD, conducting discrimination, calibration and re-classification analyses. In years 2-4, I will develop and use a microsimulation model to compare a biomarker-driven strategy based on the risk prediction model developed in Aim 1 to a symptom-driven disease management (usual care) strategy, as well as assessing a combination approach. The clinical and economic effects of the two strategies will then be compared (Aim 2). Finally, starting early in year 4, I will use the above work to develop and pilot test a personalized medical decision support tool (Aim 3). An IIR will also be submitted during year 3-4, to test the clinical intervention and the implementatio intervention, at multiple sites, in a Hybrid Type II implementation trial. !

Public Health Relevance

Veterans with "High Expense, Low Prevalence" (HELP) diseases such as rheumatoid arthritis, multiple sclerosis, and inflammatory bowel disease (IBD) often have exacerbations that can result in preventable mortality or major morbidity. Developing tools and decision support systems to guide clinicians in personalizing medical decision-making for veterans with HELP diseases has particular application for the VA, because having a physician at every facility that subspecializes in each HELP disease is not feasible. Having risk stratification tools developed and validated within the veteran population is an important first step towards realizing efficient patient-centered care for HELP diseases through the VA. Towards this goal, this CDA proposes to develop "targeted-prevention" prediction tools and decision support systems to facilitate the delivery of timely and cost-effective therapy for HELP diseases and to compare it to the current "symptom- driven" model used by clinicians. The proposal focuses on IBD as a model condition for HELP diseases.

Agency
National Institute of Health (NIH)
Type
Veterans Administration (IK2)
Project #
1IK2HX000775-01A2
Application #
8396278
Study Section
HSR&D Career Development Award (CDA0)
Project Start
Project End
Budget Start
Budget End
Support Year
1
Fiscal Year
2014
Total Cost
Indirect Cost
Name
Veterans Health Administration
Department
Type
DUNS #
City
Ann Arbor
State
MI
Country
United States
Zip Code
48105