smoking less than 1 cigarette/d assoc with inc mortality

Another article came out suggesting that even very light smoking is associated with increased mortality (see smoking occ and inc mortality jama2020 in dropbox or doi:10.1001/jamanetworkopen.2020.6436)

 

Details:

-- prospective cohort study of the Tobacco Use Supplements to the Current Population Survey (TUS-CPS), a National Cancer Institute/NIH sponsored national survey for tobacco use administered by the US Census Bureau, cosponsored in the past by the CDC and the FDA with data from 1992-2011; this database was linked to the National Death Index [of note, E cigarettes were not included in this analysis]

-- 505,500 participants aged 18-103

-- demographics (varied by smoking group):

    -- non-smokers: 30% over 55 years old/other age groups divided evenly, 58% female, 72% white/12% black/10% Hispanic, 14% less than high school/29% high school/27% some college/30% college

    -- daily smokers: evenly divided between 25 to >55 years old, 47% female, 82% white/10% black/4% Hispanic, 20% less than high school/43% high school/27% some college/11% college, started smoking age 17

    -- non-daily but previous daily smokers: even age distribution, 54% female, 70% white/12% black/6% Hispanic, 15% less than high school/34% high school/31% some college/21% college, started smoking age 18

    -- lifelong non-daily smokers most 25 to 34-year-old/48% female, 63% white/17% black/14% Hispanic, 17% less than high school/33% high school/30% some college/20% college, start smoking age 18

    -- former smokers: most >55yo, 49% female, 84% white/7% black/5% Hispanic, 14% less than high school/32% high school/27% some college/27% college, start smoking age 17

-- overall, lifelong non-daily smokers were younger than daily smokers and non-daily but previous daily smokers; lifelong non-daily smokers were more likely to be from racial/ethnic minority groups; more current non-daily smokers had a college education than current daily smokers

-- never smokers: 299,200 (59%)

-- daily current smokers: 78,870 (16%), median 600 per month (20 cigarettes per day)

-- non-daily smokers but previous daily smokers: 7605 (2%), median 75 cigarettes per month

-- lifelong non-daily smokers: 6292 (1%), median 40 cigarettes per month

 

-- 47,000 deaths occurred

Results:

-- all-cause mortality, with never smokers as reference:

    -- daily smokers, HR 2.32 (2.25-2.38) [ie 2.32 times the rate of never smokers]

    -- non-daily but previous smokers, HR 1.93 (1.80-2.07)

    -- lifelong non-daily, HR 1.82 (1.65-2.01)

-- cause-specific mortality, all with never smokers as the reference. HRs were respectively for current smoker, non-daily but prior daily, and lifelong non-daily:

    -- all cancers: HRs 3.14 (2.98-3.32), 2.46 (2.14-2.83), 1.76 (1.41-2.21)

    -- smoking-related cancers: HRs 4.88 (4.57-5.22), 3.65 (3.11-4.27), 2.16 (1.61-2.86)

    -- lung cancer: HRs 13.96 (12.50-15.59), 10.22 (8.28-12.60), 5.64 (3.89-8.18)

    -- cardiovascular disease: HRs 1.92 (1.82-2.20), 1.60 (1.40-1.83), 1.69 (1.40-2.04)

    -- cerebrovascular disease: HRs 1.51 (1.33-1.70), 1.57 (1.16-2.12), 1.45 (0.93-2.26)

    -- respiratory disease (COPD, influenza, and pneumonia): HRs 5.98 (5.42-6.58), 5.66 (4.60-6.94), 3.46 (2.38-5.02)

    -- other cause: HRs 1.75 (1.67-1.84), 1.48 (1.30-1.68), 1.82 (1.57-2.11)

-- for those quitting smoking:

    -- for daily smokers: all-cause mortality risk decreased progressively from HR 2.18 (2.03 2.35) if  quit <2 years before to 1.18 (1.15-1.22) if >10 years

    -- for former non-daily smokers: not enough people quitting <2 years to be significant, HR 1.57 (1.29 - 1.90) for 2 to 5 years, decreasing to 1.09 (1.04-1.15) for those >10 years [of note the numbers of people quitting >10 years was about 10 times those for the other categories

--  those who smoked just 6 to 10 cigarettes per month had almost twice as high all-cause mortality rates as nonsmokers, and not statistically significantly lower than those smoking > 60 cigarettes per month

   

Commentary:

-- overall findings:

    -- there was a gradient of all-cause mortality related to the number cigarettes smoked per day, though it hovered around twice the incidence of non-smokers, with not huge differences between those smoking about one cigarette a day vs those smoking a pack a day

    -- for all cancers, there was a steeper gradient, but still a residual increase in those who smoked only about one cigarette a day

    -- for smoking-related cancers there was a more profound gradient, though those smoking only about one cigarette a day had twice the incidence

    -- for cardiovascular disease, there is not a lot of variability between current smokers and those smoking only about one cigarette a day: all were increased over nonsmokers

        -- another study: a systematic review/meta-analysis found that smoking even one cigarette a day had about one half the attributable risk for cardiovascular disease as smoking one pack a day: see http://gmodestmedblogs.blogspot.com/2018/01/smoking-just-one-cigarette-one-too-many.html

    -- for respiratory diseases there was some gradient, though there was still more than a threefold incidence in those who were lifetime daily smokers

    -- and for "other causes of death" (presumably non-smoking related) there still remained a 50 to 80% increased risk. This may reflect the fact that smokers, even casual ones, have other unhealthy lifestyle issues explaining their significant increase in death (eg, they smoke when they drink, and smokers do drink more alcohol than nonsmokers). or, that our assumption is inaccurate that we have correctly identified that these causes of death are indeed not associated with smoking

-- this article is consistent with others published. for example a large NIH study (which seems to have a different database than the current one) found that consistently smoking <1 cigarette per day was associated with significantly increased mortality, as found in the above study: see http://gmodestmedblogs.blogspot.com/2016/12/light-smoking-and-mortality.html

                                          

limitations of study:

-- they had limited information on people, and they were able to adjust their models only for sex, race/ethnicity, and education

-- there are likely very different characteristics and confounders in the different groups: including diet, exercise, comorbidities, a wide array of psychosocial issues, etc. It is highly likely that those smoking more also had less healthy habits, such as diet/exercise/alcohol/etc

-- and, those who decreased their cigarette smoking or stopped may have had more underlying medical problems perhaps related to cigarette smoking that might have increased their mortality outcomes. Conversely, these patients might have “seen the light” and instigated an array of healthy behaviors as well as decreasing or stopping smoking, and this more global change may have been beneficial, moreso than just the smoking component. We do not have this information

 

So, despite the limitations of these studies, the point here is that smoking even occasional, nondaily cigarettes seems to be associated with significant increases in important clinical outcomes. And, though we as clinicians should be happy when patients are able to decrease their cigarette use, and appropriately provide a lot of encouragement and support, it is important that we and the patients realize the urgency of complete cessation....

geoff

 

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

 

to get access to all of the blogs:

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

 

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list