An article published in the June 4 issue of JAMA reports on the discordance between the clinical predictions and patient expectations of life expectancy and survival of heart failure patients. Larry A. Allen, M.D., M.H.S. (Duke Clinical Research Institute, Durham, N.C.) and colleagues find that younger patients and those with more severe disease significantly overestimate their remaining life span.
Every year in the United States, heart failure directly leads to 55,000 deaths and indirectly leads to 230,000 deaths. Patients with symptomatic heart failure have a median life expectancy of less than 5 years, even with the advances in care that exist today. About 90% of patients with the most advance disease die within one year. However, a patient’s prognosis is dependent on several factors, and researchers have developed several models to predict survival times for heart failure patients.
The issue on which Allen and colleagues focus is whether or not patients understand their prognosis after a heart failure diagnosis. The authors note that, “Patient perception of prognosis is important because it fundamentally influences medical decision making regarding medications, devices, transplantation, and end-of-life care.”
To investigate the topic, the researchers conducted a study to compare personal life expectancy predictions of 122 heart failure patients (who were not confined to a bed) to the predictions estimated by statistical models. The patients were about 62 years old and 47% African American. About 42% of them were classified as New York Heart Association [NYHA] class III or IV – an indication of more severe heart failure. The patients were asked what they predicted their life expectancy to be, and the researchers compared the responses with estimates computed using the Seattle Heart Failure Model (SHFM).
Patients, on average, predicted their life expectancy to be 13.0 years while the model predicted an average of 10.0 years. The researchers found that most patients (63%) overestimated their life expectancy compared to what was predicted by the SHFM. Using the ratio of patient-predicted to model-predicted life expectancy (life expectancy ratio or LER) the authors found that the median overestimation of predicted future survival was 40% – an LER of 1.4. Patients who had a higher LER did not necessarily have improved survival, and 29% of patients died within a median follow-up period of 3.1 years. Patients who were most likely to overestimate their life expectancy were younger, less depressed, and had more severe disease.
Regarding the finding that there was little relationship between patient-predicted and model-predicted life expectancy, the authors conclude:
“The exact reasons for this incongruity are unknown but they may reflect hope or may result from inadequate communication between clinicians and their patients about prognosis. Because differences in expectations about prognosis could affect decision making regarding advanced therapies and end-of-life planning, further research into both the extent and the underlying causes of these differences is warranted. Whether interventions designed to improve communication of prognostic information between clinicians and patients would improve the process of care in heart failure should be tested in appropriately designed clinical trials.”
An accompanying editorial, written by Clyde W. Yancy, M.D. (Baylor University Medical Center, Dallas), discusses concern for the accuracy of clinical prediction models like the SHFM.
“Currently, there is insufficient precision in the prognostication of heart failure, and decision making at the end of life is perhaps the most personalized of all decision making in medicine. Although well-intended and carefully constructed tools and awareness of the natural history of disease are helpful, it is the primacy of the patient-physician interface that must prevail. Until these questions are fully addressed, it is best to avoid adopting an imprecise method, instead continuing to embrace the individualized decision-making process guided by physician judgment that incorporates all patient care considerations,” he concludes.
Discordance Between Patient-Predicted and Model-Predicted Life Expectancy Among Ambulatory Patients With Heart Failure
Larry A. Allen; Jonathan E. Yager; Michele Jonsson Funk; Wayne C. Levy; James A. Tulsky; Margaret T. Bowers; Gwen C. Dodson; Christopher M. O’Connor; G. Michael Felker
JAMA (2008). 299:2533-2542.
Written by: Peter M Crosta