Current issues of ACP Journal Club are published in Annals of Internal Medicine


Smoking increased and smoking cessation decreased the risk for mortality, cardiovascular mortality, and cancer among older men and women

ACP J Club. 1991 Sep-Oct;115:60. doi:10.7326/ACPJC-1991-115-2-060

Source Citation

LaCroix AZ, Lang J, Scherr P, et al. Smoking and mortality among older men and women in three communities. N Engl J Med. 1991 Jun 6;324:1619-25. [PubMed ID: 2030718]



To determine the association between smoking and 5-year mortality among men and women who were ≥ 65 years of age.


Cohort analytic study, with yearly follow-up for 5 years.


1 rural and 3 urban U.S. communities.


Men and women recruited for the Established Populations for Epidemiologic Studies of the Elderly, between 1981 and 1983, representing 80% to 84% of those eligible in each community. 7178 men and women ≥ 65 years old with no history of myocardial infarction, stroke, or cancer were selected. Rates of current smoking were 12.9% for women and 21.2% for men.

Assessment of risk factors

Information collected included demographic characteristics, medical history, health behaviors, and functional status. Smoking status, number of cigarettes smoked, and duration of smoking were determined by self report.

Main outcome measures

Death certificates were requested for all deaths, with information coded by 1 nosologist using the International Classification of Diseases, Clinical Modification (9th edition). Mortality rates were calculated as the number of deaths per person-years of follow-up, adjusted for age and community. Risks were calculated relative to nonsmokers. Deaths were classified as total deaths, deaths from cardiovascular causes, and death from cancer.

Main results

1442 deaths occurred from all causes, 729 from cardiovascular disease, and 316 from cancer. Relative risk (RR) for total mortality for current smokers was 2.1 (95% CI 1.7 to 2.7) for men and 1.8 (CI 1.4 to 2.4) for women. For former smokers, the RR was 1.5 (CI 1.2 to 1.9) for men and 1.1 (CI 0.8 to 1.5) for women. For current smokers aged 65 to 69 years, 70 to 74 years, and ≥ 75 years, the RRs for men for total mortality were 2.3 (CI 1.5 to 3.8), 3.4 (CI 2.1 to 5.5), and 1.3 (CI 1.0 to 1.7), respectively, and for women, 2.4 (CI 1.5 to 3.8), 1.8 (CI 1.2 to 2.7), and 1.2 (CI 0.8 to 2.7), respectively. All current smokers were at increased risk for death from coronary heart disease and other cardiovascular causes (RR ranging from 1.5 to 2.2), but former smokers were not, regardless of when they stopped. All current smokers and men who were former smokers had a higher risk for all cancers and smoking-related cancers. Total and cause-specific mortality were highest in those who had smoked ≥ 40 years.


Among men and women who were ≥ 65 years of age, smoking continuation was associated with higher 5-year rates of total mortality and mortality caused by cardiovascular disease and cancer. Smoking cessation was associated with a rapid decline in cardiovascular mortality.

Source of funding: National Institute on Aging.

Address for article reprint: Dr. A.Z. LaCroix, Center for Health Studies, Group Health Cooperative of Puget Sound, 1730 Minor Avenue, Suite 1600, Seattle, WA 98101-1448, USA.


This well-done and important study strongly supports the benefits of quitting smoking for people > 65 years of age. Previous smaller or case-control studies yielded conflicting results for both the effects of smoking on the mortality of those > 65 years of age and the potential benefits of quitting. This study is methodologically robust and clearly shows the strong association of smoking with cardiovascular and cancer mortality in older persons. Although smoking status was determined by self-report only, any underreporting of smoking status would have lessened the statistical effect of smoking on mortality and the benefits of quitting. Therefore, the findings are impressive.

We can now confidently tell our older patients that smoking is associated with approximately twice the rate of mortality for cardiovascular disease and an even higher risk for cancer. More important, stopping smoking will rapidly reduce the risk for cardiovascular mortality. The risk for cancer death remains high in men after quitting smoking, whereas it falls rapidly in women, probably because the men were heavier smokers for longer periods. Yet to be reported for this cohort are the relations between smoking and morbidity or functional status, a potentially even more important message for older persons who smoke.

Previous studies of smoking cessation have generally excluded older persons. Studies are needed to better understand the behavioral and clinical characteristics of these older persons who smoke and learn how to best help them quit.

One of the most exciting aspects of this report is that it will not be the last. The Established Populations for Epidemiologic Studies of the Elderly cohort should give us much valuable information about the health of older persons.

David L. Bronson, MD
University of VermontBurlington, Vermont, USA