Year of Publication



Public Health

Degree Name

Master of Public Health (M.P.H.)

Committee Chair

Dr. Steven Browning

Committee Member

Dr. Daniela Claudia Moga

Committee Member

Dr. Anna Kucharska-Newton

Committee Member

Dr. Roberto Cardarelli


Introduction: Heart failure constitutes a global pandemic affecting at least 26 million globally (an estimated 5.7 million of which are Americans), and its burden on our healthcare system is projected to increase.[1] As a result, it is imperative that diagnostics and treatment among that population be evidence based in order to contain costs and optimize patient outcomes. Over the last 20 years, thyroid function screening has been recommended for patients upon initial diagnosis of heart failure by the American College of Cardiology and other professional organizations. However, it remains unclear whether this recommendation is followed by practitioners and how much thyroid dysfunction screening uncovers. The aim of this capstone is to examine the association between thyroid dysfunction and heart failure and explore a practical example of thyroid function evaluation for patients newly-diagnosed with that disease.

Objectives: The primary objective of this study is to explore one medical center’s hospitalizations associated with incident heart failure and determine how often and in which patients thyroid function is screened (using TSH). A secondary objective was to determine the amount of thyroid dysfunction discovered through that screening.

Methods: Participants in the current study were sampled from inpatient electronic medical records for patients at the University of Kentucky Albert B. Chandler Hospital from 2007- 2017. Only first hospitalizations during which heart failure was documented were included in the study (N = 15900). Logistic regression was used to analyze the association between patient characteristics and the probability of receiving thyroid function screening during that hospital stay. Similar analysis was done to explore the association between patient characteristics and thyroid dysfunction among those screened.

Results: A minority (36.09%) of our patient population had their TSH tested during initial heart failure admission. In multivariable analysis, longer length of stay (OR: 1.33-2.18), year of encounter (OR: 1.03, 95% CI: 1.02-1.04), and female sex (OR: 1.29, 95% CI: 1.21-1.38) were significantly associated with increased odds of patients having their TSH tested. Among those tested, multivariate analysis revealed that female sex (OR: 1.34, 95% CI: 1.19-2.07) and longer length of stay (OR: 1.20 for 8-14 days; OR: 1.45 for 22+ days) were associated with increased odds of thyroid dysfunction (abnormal TSH). African American race [vs white], on the other hand, was associated with decreased odds of thyroid dysfunction (OR: 0.74, 95% CI: 0.59-0.93). Age 65+ was also associated with decreased odds of dysfunction (OR: 0.82, 95% CI: 0.67-0.998). Of all patients tested, 28% had thyroid dysfunction. The most common abnormality detected was high TSH with normal FT4 (11%).

Conclusion: This study suggested that, despite recommendations, only a minority of newly diagnosed heart failure patients who are hospitalized have their thyroid function screened. Furthermore, our data hint that thyroid function is not consistently tested on all heart failure patients but is tested more often in women and in patients with longer hospitalizations.

Among those screened, thyroid dysfunction appeared to be more common than that reported in the general population but was still a minority of patients. The most common abnormality detected was subclinical hypothyroidism. We hope that knowing how often thyroid screening is performed, who specifically is screened, and what testing uncovers may

lead to more targeted, cost-effective screening in those who would derive the most benefit from that practice.

Included in

Public Health Commons