Abstract 17576: Frequency, Timing and Co-Occurrence of Heart Failure Signs and Symptoms Among 50,625 Heart Failure Cases and Matched Controls in a Primary Care Population in the Months Preceding the Clinical Diagnosis of Heart Failure
The Framingham heart failure (HF) signs and symptoms are ubiquitous among primary care patients, leading to challenges in distinguishing a presentation attributable to HF versus other causes. A better understanding of the timing and co-presentation of HF symptoms could be useful in the earlier identification of high risk patients.
Methods: Validated text and data mining tools were developed and applied to longitudinal EHR progress notes to examine the temporal sequencing of HF symptoms before the clinical diagnosis of HF in a large primary care practice. Data were from 4644 incident HF cases and 45,981 group-matched controls.
Results: The prevalence of specific HF signs/symptoms varied widely between cases (<1%-65%) and controls (<1%-28%). DOE and ankle edema were the most common early findings, emerging up to 4 years before a HF diagnosis. However, these same findings were also the most common in controls, albeit with half the prevalence. Documentation of symptoms increased in cases in relation to proximal time to HF diagnosis and also in an equivalent time period sampled from controls, although, again, to a much greater degree among cases. Ankle edema and DOE most commonly co-occurred in both cases and controls. In the 18 months preceding the HF diagnosis date, prevalence of these two symptoms quadrupled in cases compared to doubling in controls. The FIGURE graphically demonstrates the changing frequency of co-occurrence of symptoms among cases (reduced and preserved EF) and controls in the 18 months prior to clinical diagnosis.
Conclusion: Signs and symptoms of HF are prevalent months to years prior to diagnosis, but are nonspecific and are also present in non-HF patients. A rapid increase in HF symptom co-occurrence is more prominent in cases relative to controls. Further refinement in the identification and timing of the co-occurrence of these and other HF signs/symptoms may allow for the earlier identification and improved treatment of HF in a primary care population.
- © 2012 by American Heart Association, Inc.