Persistent cigarette smoking and other tobacco use after a tobacco-related cancer diagnosis

Abstract

Introduction: People who continue to smoke after a cancer diagnosis have an increased risk for recurrences or development of new malignancies. These risks may be even higher among tobacco-related cancer survivors (TRCS). We describe tobacco use behaviors among TRCS, other cancer survivors, and people without a history of cancer.

Methods: We used 2009 Behavioral Risk Factor Surveillance System data to describe demographic characteristics, smoking history, current smoking prevalence, and smokeless tobacco use among TRCS, other cancer survivors, and people without a history of cancer (cigarette smoking and smokeless tobacco use were calculated after adjusting for age, sex, race, and insurance status). Tobacco-related cancers were defined as lung/bronchial, pharyngeal, laryngeal, esophageal, stomach, pancreatic, kidney/renal, urinary bladder, cervical, and acute myeloid leukemia.

Results: A total of 20 % of all cancer survivors were TRCS. TRCS were primarily female (68 %) and white (78 %). Smoking prevalence was higher among TRCS (27 %) compared with other cancer survivors (16 %) and respondents without a history of cancer (18 %). Smokeless tobacco use was higher among respondents without a history of cancer (4 %) compared with TRCS (3 %) and other cancer survivors (3 %).

Conclusions: The self-reported smoking prevalence among TRCS is higher than among other cancer survivors and people without a history of cancer. Targeted smoking prevention and cessation interventions are needed for cancer survivors, especially those diagnosed with a tobacco-related cancer.

Implications for cancer survivors: We recommend all cancer survivors be made aware of the health risks associated with smoking after a cancer diagnosis, and smoking cessation services be offered to those who currently smoke. We provide the first population-based report on demographic characteristics and tobacco use behaviors among self-reported tobacco-related cancer survivors.

Fig. 1

Unadjusted smoking prevalence by sex among TRCS, other cancer survivors, and respondents without cancer [32]. BRFSS Behavioral Risk Factor Surveillance System. All data are based on self-report and are weighted. For cancer survivors reporting more than one cancer diagnosis, the cancer site reported was the most recently diagnosed cancer. TRCS include respondents who report a prior diagnosis of lung/bronchial, pharyngeal, esophageal, stomach, pancreatic, kidney/renal, urinary bladder, or cervical cancer. Tobacco-related cancers are defined according to a 2004 Surgeon General report [8]. To capture acute myelogenous leukemia survivors, all patients who reported a leukemia diagnosis were included in the tobacco-related cancer survivor population. Other cancer survivors include respondents who report a prior diagnosis of any cancer, not designated as tobacco related. Respondents without cancer include anyone who reports never being diagnosed with cancer. Current smoking defined as regular or intermittent smoking and having smoked >100 cigarettes ever. *p<0.05; p values calculated from Rao Scott chi-square test

Fig. 2

Unadjusted smoking prevalence by age among TRCS, other cancer survivors, and respondents without cancer [32]. BRFSS Behavioral Risk Factor Surveillance System. All data are based on self-report and are weighted. For cancer survivors reporting more than one cancer diagnosis, the cancer site reported was the most recently diagnosed cancer. TRCS include respondents who report a prior diagnosis of lung/bronchial, pharyngeal, esophageal, stomach, pancreatic, kidney/renal, urinary bladder, or cervical cancer. Tobacco-related cancers are defined according to a 2004 Surgeon General report [8]. To capture acute myelogenous leukemia survivors, all patients who reported a leukemia diagnosis were included in the tobacco-related cancer survivor population. Other cancer survivors include respondents who report a prior diagnosis of any cancer not designated as tobacco related. Respondents without cancer include anyone who reports never being diagnosed with cancer. Current smoking defined as regular or intermittent smoking and having smoked >100 cigarettes ever. *p<.05; p values calculated from Rao Scott chi-square test

Fig. 3

Adjusted smoking prevalence by time since diagnosis among TRCS and other cancer survivors [32]. BRFSS Behavioral Risk Factor Surveillance System. All data are based on self-report and are weighted. For cancer survivors reporting more than one cancer diagnosis, the cancer site reported was the most recently diagnosed cancer. TRCS include respondents who report a prior diagnosis of lung/bronchial, pharyngeal, esophageal, stomach, pancreatic, kidney/renal, urinary bladder, or cervical cancer. Tobacco-related cancers are defined according to a 2004 Surgeon General report [8]. To capture acute myelogenous leukemia survivors, all patients who reported a leukemia diagnosis were included in the tobacco-related cancer survivor population. Other cancer survivors include respondents who report a prior diagnosis of any cancer, not designated as tobacco-related. Current smoking defined as regular or intermittent smoking and having smoked >100 cigarettes ever. Smoking prevalence adjusted for race/ethnicity, sex, age, insurance coverage, and employment status

Fig. 4

Unadjusted smoking prevalence by race and ethnicity among TRCS, other cancer survivors, and respondents without cancer [32]. BRFSS Behavioral Risk Factor Surveillance System, NH Native Hawaiian, PI Pacific Islander, AI/AN American Indian/Alaska native. All data are based on self-report and are weighted. For cancer survivors reporting more than one cancer diagnosis, the cancer site reported was the most recently diagnosed cancer. TRCS include respondents who report a prior diagnosis of lung/bronchial, pharyngeal, esophageal, stomach, pancreatic, kidney/renal, urinary bladder, or cervical cancer. Tobacco-related cancers are defined according to a 2004 Surgeon General report [8]. To capture acute myelogenous leukemia survivors, all patients who reported a leukemia diagnosis were included in the tobacco-related cancer survivor population. Other cancer survivors include respondents who report a prior diagnosis of any cancer, not designated as tobacco-related. Respondents without cancer include anyone who reports never being diagnosed with cancer. Current smoking defined as regular or intermittent smoking and having smoked >100 cigarettes ever

Fig. 5

Adjusted smoking prevalence by site among TRCS [32]. BRFSS Behavioral Risk Factor Surveillance System. All data are based on self-report and are weighted. For cancer survivors reporting more than one cancer diagnosis, the cancer site reported was the most recently diagnosed cancer. TRCS include respondents who report a prior diagnosis of lung/bronchial, pharyngeal, esophageal, stomach, pancreatic, kidney/renal, urinary bladder, or cervical cancer. Tobacco-related cancers are defined according to a 2004 Surgeon General report [8]. To capture acute myelogenous leukemia survivors, all patients who reported a leukemia diagnosis were included in the tobacco-related cancer survivor population. Data for pharyngeal, esophageal, stomach, pancreatic, urinary, and leukemia cancer sites were suppressed because of sample size of <50 or half width of CI of >10. Current smoking defined as regular or intermittent smoking and having smoked >100 cigarettes ever. Smoking prevalence adjusted for race/ethnicity, sex, age, insurance coverage, and employment status. *p<.05; p values calculated from linear contrasts between cervical and other cancer sites