Abstract P204: Dyspnea as Predictor of All-cause Mortality: Reduction in Risk Over Time in a Prospective Cohort Study
Background: It has been suggested that risk in longitudinal studies may drop with time, but clear-cut documentation has been sparse. The purpose of this investigation was to determine risks at several time points in a recently completed/published longitudinal study to further evaluate this hypothesis.
Methods: A population-based sample of 11,533 Bangladeshsis aged 18 to 75 was recruited and followed for 11-12 years and all-cause mortality was evaluated in those with and without baseline dyspnea. Dyspnea, the exposure, was ascertained by trained physicians by questionnaire. Kaplan-Meier survival curves and Cox Proportional Hazard models were used to determine differences between groups at 3, 6, and 11-12 years.
Results: The Kaplan-Meier curves revealed a clear increase in mortality between dyspnea versus no dyspnea groups at the 3, 6, and 11.2 year time periods. The logrank test was significant with a p value less than 0.01 for all three time periods. At 3 years (3-yrs) the unadjusted hazard ratio (HR) was 3.43 (95% C.I.; 2.36 - 5.00). The 3-yrs HR adjusted for age, sex, arsenic well water concentration, education, blood pressure, BMI, and smoking was 2.58 (1.77 - 3.76)-for increased mortality in those with dyspnea versus no dyspnea. At 6-yrs, the crude and adj. HRs were 2.98 (2.28 - 3.88) and 2.23 (1.71 - 2.92), respectively. At 11.2 years, the crude and adj. HRs were 2.73 (2.27 - 3.28) and 2.10 (1.74 - 2.52) in those with dyspnea (relative to no dyspnea), respectively.
Conclusions: Risk appears to drop over time in longitudinal studies of all-cause mortality, consistent with greater loss of those with the exposure relative to the unexposed. This phenomena might be expected. Therefore, longer longitudinal studies may actually underestimate and probably do not overestimate exposure risk.
Author Disclosures: G.R. Pesola: None. V.M. Chinchilli: None. Y. Chen: None. M. Argos: None. F. Parvez: None. A. Ahmed: None. T. Islam: None. L. Tong: None. R. Hasan: None. A.I. Neugut: None. R.G. Barr: None. H. Ahsan: B. Research Grant; Significant; P42ES010349, RO1CA107431.
- © 2017 by American Heart Association, Inc.