Redefining and Improving Hypothyroidism Care Using Real-World Data - PROJECT SUMMARY/ABSTRACT Hypothyroidism is a metabolic disorder that affects nearly 10% of the US population and directly contributes to over $400 million in healthcare costs annually. The primary treatment is to replace low thyroid hormone levels with levothyroxine (LT4) and normalize serum thyroid stimulating hormone (TSH). The TSH level provides clinicians with a measure of disease control over the last 2 months. If the TSH level is above or below the normal range, the clinician may adjust the LT4 dose and retest TSH. However, observational evidence suggests that the test-and-adjust treatment approach is inadequate for many patients, as one-quarter to one- half of patients experience uncontrolled TSH levels. Uncontrolled TSH levels cause systemic symptoms and increase the risk of adverse cardiovascular (CV) outcomes. However, our understanding of the impact of uncontrolled TSH levels on health outcomes is based on cross-sectional data and does not account for the cumulative exposure of uncontrolled TSH levels over time. In Aim 1 of this proposal, using real-world clinical data, the PI (Dr. Matthew Ettleson) will determine the association between cumulative TSH control and CV outcomes in patients with LT4-treated hypothyroidism using longitudinal analysis methods. Given the high frequency of uncontrolled TSH levels, strategies are also needed to improve the long-term management of hypothyroidism. In other chronic diseases, machine learning (ML) approaches and clinical decision support have been leveraged to individualize treatment and improve outcomes. In Aim 2, the PI will develop an ML model to identify which patients are at high risk of uncontrolled TSH levels within the next year of LT4 treatment. In Aim 3, the PI will seek to improve clinical practice by designing and implementing a clinical decision support system that identifies high risk patients, informs clinicians of monitoring gaps, and provides best practice recommendations at the point of care. The results of these studies will be used to inform an R01 proposal that will examine the effectiveness of risk-based monitoring approaches and clinical decision support to improve TSH control in LT4-treated patients. The purpose of this career development award is to support the PI in his long-term career goal of becoming a nationally recognized, independent health services and outcomes researcher with expertise in improving thyroid disease management. To meet his research and career goals, the PI will acquire skills in the following training areas that complement the research aims: 1) longitudinal analysis, 2) predictive modeling, 3) clinical informatics and implementation science, and 4) complete the Master of Science in Public Health Sciences for Clinical Professionals degree. The PI will work closely with his mentorship team, led by Dr. Neda Laiteerapong and Dr. Antonio Bianco, to meet the proposed research and training goals. The project will take place at the University of Chicago, a world-renowned institution with high-impact research programs in thyroid disease and health services and outcomes research.
