Critique 215: Long-term effects of smoking and moderate drinking on the quality and longevity of life of elderly women — 9 May 2018

Nelson HD, Lui L, Ensrud K, Cummings SR, Cauley JA, Hillier TA. Associations of Smoking, Moderate Alcohol Use, and Function: A 20-Year Cohort Study of Older Women.  Gerontology & Geriatric Medicine 2018;4:1–9.  Pre-publication.

Authors’ Abstract

Objective: The objective of this study is to determine whether the health effects of smoking and moderate alcohol use persist with aging.

Method: Smoking status, alcohol use, and measures of function and health were obtained from 9,704 women aged ≥65 years at baseline and over 10- and 20-year follow-up periods. Adjusted multiple linear and logistic regression and Cox proportional hazard models estimated associations.

Results: Current versus never smokers had worse walking speed, self-reported health, difficulty with instrumental activities of daily living (IADLs), and depression at 10 years and higher death rates at 10 and 20 years. Moderate versus never drinkers had better grip strength, walking speed, self-reported health, and less difficulty with IADLs and were less likely to live in nursing homes at 10 years and die at 10 and 20 years.

Discussion: Among aging women over 20 years, smoking is associated with worse physical function, including death, while moderate alcohol use is associated with better outcomes.

Forum Comments

Data from most studies among middle-aged women (and men) have shown that cigarette smoking is associated with many adverse health outcomes, including greater risk of total mortality. On the other hand, most studies among middle-aged adults show better indices of quality of life and a lower risk of the “diseases of ageing,” especially cardiovascular disease, to be associated with moderate intake of alcohol. There have been fewer studies among the elderly, and guidelines often suggest that the elderly should avoid alcohol or drink less.  However, given that most types of physical disability, diseases, and death occur among the elderly, it is especially important that the effects of lifestyle factors on these outcomes are known for older subjects.

The results of the present study are especially interesting, as they deal with women being followed for up to 20 years from age 65 and older to evaluate quality of life and mortality associated with cigarette smoking and with moderate alcohol consumption. The authors’ conclusion: “In a cohort of older women followed over 20 years, smoking is associated with worse physical function and health, including death, while moderate alcohol use is associated with better outcomes. Alcohol use may serve as a marker of function and health in societies where it is common and related to social activity, and quitting alcohol may signal a decline in health among older women.”

Specific comments from Forum members: Forum members considered this to be a well-done prospective cohort study.  It was based on 9,704 women, at least 65 years of age at baseline, with a mean of 71.7 years (range 65-99), who were living independently in one of four areas in the United States.  Their smoking habits at baseline and repeated assessments of alcohol were recorded and related to morbidity and mortality over 20 years or more.  By the 20-year follow up, a total of almost 70% of the women had died.  Outcomes included grip strength, walking speed, prevalence of threshold scores for self-reported health, difficulty with three or more instrumental activities of daily living (IADLs), depression, prevalence of falls, living in a nursing home, and death after 10- or 20-year follow up.

Reviewer Finkel summarized this paper well: “Not new — smoking bad; drinking moderately good — but nicely, clearly done. I like the simple, concrete measures.  These data will help dispel the silly widespread anxiety about we mature enjoyers of life taking a dram now and then.”  Forum member Svilaas added: “I agree with Finkel’s short and concise comments.  This is a well-performed study with a simple and reliable conclusion.”  Reviewer Goldfinger noted that while the paper gives strong evidence against smoking, he felt that it was overly complicated.  “Further, the authors did not point out that smokers who also consume alcohol moderately have considerably less CVD than do smokers who do not drink (Friedman & Kimball).”

Forum member Pajak noted: “The relation between smoking and alcohol consumption at age over 65 years with physical functioning at very old age is interesting. Little was known on it.  To me it points more on the importance of a relation with ageing itself than on causing specific diseases as an explanation of shorter life span in smokers.  However, there are some limitations in the interpretation of the results.

  1.  The goal of the analysis and the paper was not an exact goal of the study. This could have an impact on the design of the analysis, number of persons in the studied subgroups, statistical power, etc.
  2. The article did not contribute to clarification of the problem of reverse causation. Moderate alcohol consumption could be related to better health both at baseline and at later stage. In fact there is an indirect conformation in the results of the baseline observation. This effect could even occur in heavy drinkers (14 drinks per week or more) who were, unfortunately, excluded from the analysis.
  3. Lack of control over confounding by psychosocial factors (SES) is a clear weakness of the analysis as it could explain the protective effect of moderate alcohol consumption, at least partially.”

Reviewer Stockley wrote: “I just came across a related paper by Daskalopoulou et al: a systematic review and meta-analysis of longitudinal studies which aimed to synthesise the associations of smoking and alcohol consumption with healthy ageing (HA). HA was defined by the authors of each separate study that was a part of that meta-analysis, and included terms such as ‘healthy ageing, successful ageing, active ageing, and healthy survival.’  Studies reporting HA as multiple outcomes or based solely on self-report were excluded. That study found results similar to those of the present paper by Nelson et al, i.e., poorer outcomes from smoking but better outcomes from light or moderate drinking.

“The results shown in the present paper by Nelson et al are complementary to those shown in a series of studies by Simons et al in the Dubbo Study of Australian Elderly, where it was also shown that after over 20 years of follow-up, men and women taking any alcohol survived 12 months longer than their peers taking no alcohol. For smokers, men lost 41 months of survival, women 25 months; with hypertension, men lost 20 months, women lost 17 months; with diabetes, men lost 20 months, women lost 30 months; with atrial fibrillation, men lost 20 months, women lost 25 months.  Simons (2014) concluded that ‘What are the public health implications of our findings, being derived from a population of ‘survivors’ in their senior years? In the presence of low to moderate intake of alcohol, senior citizens can be encouraged to continue this behaviour.’

“I am curious whether Nelson et al found any relationships between smoking and alcohol consumption; that is, did the better outcomes from alcohol consumption attenuate the worse outcomes from smoking? To me this is quite important and would add significance to the current findings as well as previous findings that moderate alcohol consumption is one of four or five low-risk behaviours for longevity (Ford et al).”

Forum member de Gaetano noted: “It is an interesting paper. I’m wondering however why they compared ‘smoking’ tout court and ‘moderate drinking’: the former behavior without any specification (how many cigarettes/day, pack years, etc.).  Now it is well known (de Gaetano et al) that alcohol consumption and health outcomes are related by a J curve, while the relation with smoking is linear, but this was not discussed in the present paper.  I agree that data on the possible interaction between smoking and drinking would have been of interest.”

References from Forum critique

Daskalopoulou C, Stubbs B, Kralj C, Koukounari A, Prince M, Prina AM. Associations of smoking and alcohol consumption with healthy ageing: a systematic review and meta-analysis of longitudinal studies.  BMJ Open 2018;8:e019540. doi: 10.1136/bmjopen-2017-019540.  PMID: 29666127

De Gaetano G, Costanzo S. Alcohol and Health: Praise of the J Curves.  J Am Coll Cardiol 2017;70:923-925.

Ford ES, Zhao G, Tsai K, Li C. Low-risk lifestyle behaviors and all-cause mortality: Findings from the National Health and Nutrition Examination Survey III Mortality Study.  Am J Pub Health 2011;101:1922-1929.

Friedman LA, Kimball AW. Coronary heart disease mortality and alcohol consumption in Framingham.  J Epidemiol 1986;124:481-489.

Simons L, McCallum AJ, Friedlander Y, Ortiz M, Simons J. Moderate alcohol intake is associated with survival in the elderly: the Dubbo Study. Medical Journal of Australia 2000;173:1211–1224.

Simons, L. Alcohol intake and survival in Australian seniors: the Dubbo Study. Nutrition and Ageing 2014;2:85-90.

Forum Summary

While essentially all epidemiologic studies have shown adverse health effects of cigarette smoking and beneficial effects of moderate alcohol consumption, the present study is important as it follows a cohort of elderly women into very old age. It is a prospective cohort study of 9,704 women at least 65 years of age, who were living independently at baseline and were not heavy drinkers; they had a mean baseline age of 71.7 years (range 65-99) and underwent repeated assessments of lifestyle habits starting in 1986 to 1988.  They were followed longitudinally until death, loss to follow-up, or to April, 2016.  Outcomes included grip strength, walking speed, prevalence of threshold scores for self-reported health, difficulty with three or more instrumental activities of daily living (IADLs), depression, prevalence of falls, living in a nursing home, and death after 10- or 20-year follow up.  The main comparisons evaluated in this study were between women who were smokers and those who consumed alcohol moderately, with each analysis adjusted for the other exposure.

Forum members considered this to be a well-done study that provides some important data relating effects of smoking and moderate drinking on function and mortality in very elderly subjects (an age group where data are lacking). There were criticisms that the study does not report the effects of moderate drinking among smokers; previous studies have shown that the adverse effect of smoking is modified by moderate alcohol intake.  Further, given that the reasons why some women quit drinking during follow up are not known, the article does not contribute to clarification of the problem of reverse causation (e.g., healthier women may tend to continue drinking whereas sicker women stop alcohol consumption).  Also, the authors did not adjust for some strong confounders, especially those related to socio-economic status (SES).

Nevertheless, the results of this study make a real contribution to our knowledge of the effects of drinking on both quality of life and mortality among women who are 65 years or older. The follow up of subjects continued for a least 20 years, by which time almost 70% of the women had died.  Hence, the effects on mortality should be excellent estimates.  As expected, women who were smokers had greater risk for almost all adverse outcomes, while drinkers who continued to consume moderate alcohol into old age had the most favorable outcomes.  While the authors report that women who reported drinking at baseline but quit later on had mortality risks that were intermediate between non-drinkers and persistent drinkers, the reasons why some women quit drinking were not known.  Hence, it is appreciated that continued drinking could be only a marker of better health, and not the cause.

*                      *                      *                      *                      *                      *                   *

Comments on this critique by the International Scientific Forum on Alcohol Research were provided by the following members:

Giovanni de Gaetano, MD, PhD, Department of Epidemiology and Prevention, IRCCS Istituto Neurologico Mediterraneo NEUROMED, Pozzilli, Italy

R. Curtis Ellison, MD, Professor of Medicine, Section of Preventive Medicine & Epidemiology, Boston University School of Medicine, Boston, MA, USA

Harvey Finkel, MD, Hematology/Oncology, Retired (Formerly, Clinical Professor of Medicine, Boston University Medical Center, Boston, MA, USA)

Tedd Goldfinger, DO, FACC, Desert Cardiology of Tucson Heart Center, University of Arizona School of Medicine, Tucson, AZ, USA

Professor Andrzej Pająk, Epidemiology and Population Studies, Jagiellonian University Medical College, Kraków, Poland

Creina Stockley, PhD, MSc Clinical Pharmacology, MBA; Health and Regulatory Information Manager, Australian Wine Research Institute, Glen Osmond, South Australia, Australia

Arne Svilaas, MD, PhD, general practice and lipidology, Oslo University Hospital, Oslo, Norway