Abstract 18350: Impact of Amount vs Duration of Smoking on Incident Heart Failure and Mortality Among Community-dwelling Older Adults: Need to Reassess the Role of Pack-years as a Tool to Estimate Smoking
Background: Clinicians and public health professionals often use pack-years (PY) to quantify amount and duration of smoking. Thus, 40 PY of smoking may mean 1 pack of cigarette a day for 40 years or 4 packs a day for 10 years. However, it is unknown if the impact of smoking as measured by PY may vary by amount and duration.
Methods: Of the 3185 community-dwelling adults ≥65 years without prevalent heart failure (HF) in the Cardiovascular Health Study (CHS), 2556 were never-smokers. The 629 current smokers smoked a median of 1 (20 cigarettes) pack a day (PPD) for a median of 51 years resulting in a median of 42 PYs of smoking. Of 319 smoking ≥1 PPD, 125 smoked <51 years (higher-amount-“shorter”-duration; median 49 PY) and of 336 smoking ≥51 years, 142 smoked <1 PPD (lower-amount-longer-duration; median 27 PY). Age-sex-race-adjusted hazard ratios (aHR) and 95% CI for centrally-adjudicated incident HF and all-cause mortality during 13 (mean, 10) years of follow-up for the two smoker groups (vs. never smokers) were estimated in Cox regression models.
Results: Patients (N=2823) had a mean (±SD) age of 74 (±6) years, 70% were women, and 16% African American. Incident HF occurred in 21%, 24% and 28% of never smokers, higher-amount-“shorter”-duration (aHR, 1.32; 95% CI, 0.91-1.91) and lower-amount-longer-duration smokers (aHR, 1.60; 95% CI, 1.14-2.16), respectively (Figure; left panel). Respective rates for all-cause mortality were 39%, 48% (aHR, 1.34; 95% CI, 1.04-1.74) and 58% (aHR, 1.73; 95% CI, 1.39-2.17; Figure; right panel). .
Conclusions: Smokers who survived into older age and smoked larger amount for “shorter” duration appear to have numerically higher risk for incident HF and death than those who smoked smaller amount for longer duration, despite the fact that former group had lower pack-year estimates. These findings suggest that the use of pack-year to estimate smoking may underestimate the risk of prolonged smoking.
Author Disclosures: A.A. Ahmed: None. V. Bittner: None. C. Morgan: None. S. Zhang: None. G.C. Fonarow: Consultant/Advisory Board; Modest; Medtronic, Amgen, Johnson & Johnson, Bayer, Boston Scientific. Research Grant; Significant; NIH. Consultant/Advisory Board; Significant; Novartis. P. Deedwania: None. J. Butler: None. W. Wu: None. R.M. Allman: None. M. White: None. R.D. Fletcher: None. G. Howard: None. R.O. Bonow: None. W.S. Aronow: None. A. Ahmed: None.
- © 2015 by American Heart Association, Inc.