Smith, Lindsay Garnier2016-01-042016-01-042013-10https://hdl.handle.net/11299/175904University of Minnesota Ph.D. dissertation. October 2013. Major: Epidemiology. Advisor: Alvaro Alonso, MD, PhD. 1 computer file (PDF); xv, 152 pages.Background information on the epidemiology of atrial fibrillation (AF), including descriptive data and risk factors, pathophysiology, clinical aspects and outcomes, as well as three original manuscripts that together form the basis of this doctoral dissertation, are presented. The objectives of this dissertation were to assess temporal trends in the occurrence and prognosis of AF among acute myocardial infarction (MI) patients, to determine the usefulness of administrative data to identify incident AF, and to describe the impact of AF on healthcare utilization. AF in the setting of MI occurs frequently and is associated with increased mortality. Nonetheless, temporal trends in the occurrence of AF complicating MI and in the prognosis of these patients are not well described. In a population-based sample of 20,049 validated first incident nonfatal hospitalized MIs from the Atherosclerosis Risk in Communities (ARIC) Study, prevalence of AF in MI increased from 11% to 15% (adjusted odds ratio [OR] for prevalent AF: 1.11; 95% confidence interval [CI]: 1.04 - 1.19 per five-year increment) from 1987 through 2009. In patients with MI, AF was associated with increased 1-year mortality (adjusted OR 1.47, 95% CI 1.07-2.01) compared to those without AF. However, there was no evidence that the impact of AF on MI survival changed over time or differed over time by sex, race or MI classification. In the setting of MI, co-occurrence of AF should be considered a critical clinical event and treatment needs unique to this population should be explored further. Increasingly, epidemiologic studies use administrative data to identify AF. Capture of incident AF is not well documented. ARIC cohort participants without prevalent AF enrolled in fee-for-service Medicare, Parts A and B, for at least 12 continuous months between 1991 and 2009 were included. Of 10,134 eligible participants, 738 developed AF according to both ARIC and Centers for Medicare and Medicaid Services (CMS); an additional 93 and 288 incident cases were identified using only ARIC and CMS data, respectively. Incidence rates per 1,000 person-years were 10.8 (95% CI: 10.1-11.6) and 13.6 (95% CI: 12.8-14.4) in ARIC and CMS, respectively; agreement was 96%; the kappa statistic was 0.77 (95% CI: 0.75-0.80). Additional CMS events did not alter observed associations between risk factors and AF. Drawbacks of CMS are its inapplicability to those <65 years and inability to capture AF for those with Medicare Advantage. AF is associated with increased risk of hospitalizations. However, little is known about the impact of AF on non-inpatient healthcare utilization or about sex or race differences in AF-related utilization. ARIC cohort participants with incident AF (n=944) enrolled in fee-for-service Medicare, Parts A and B, for at least 12 continuous months between 1991 and 2009 were matched on age, sex, race and center to up to three participants without AF (n=2,761). The average annual days hospitalized were 13.1 (95% CI: 11.5-15.0) and 2.8 (95% CI: 2.5-3.1) for those with and without AF, respectively; the annual numbers of outpatient claims were 53.2 (95% CI: 50.4-56.1) and 23.0 (95% CI: 22.2-23.8) for those with and without AF, respectively. Most utilization in AF patients was attributable to non-AF conditions, particularly other- cardiovascular disease-related reasons. There was suggestive evidence that sex modified the association between AF and inpatient utilization, with AF related to greater utilization in women than men. The association between AF and healthcare utilization was similar in whites and blacks. In addition to rate or rhythm treatment, management of AF also should focus on the accompanying cardiovascular comorbidities. Overall, the results from this dissertation indicate that co-occurrence of AF in MI is a critical clinical event, that administrative data can be useful in AF epidemiologic research, and AF patients have substantial healthcare utilization, especially for other-cardiovascular disease-related reasons.enAtrial fibrillationconcordanceHealthcare utilizationSurveillanceEpidemiologyThesis or Dissertation