Stroke Victims Who Continue to Smoke
Introduction
Patients who continue to smoke cigarettes after stroke face an increased risk of recurrent stroke and poor cardiovascular outcomes.1–3 The US Food and Drug Administration has approved 2 effective smoking cessation pharmacotherapies in addition to nicotine replacement therapy; bupropion was approved in 1997 and varenicline in 2006.4 Although the American Heart Association/American Stroke Association recommends such smoking cessation interventions after stroke,5,6 it is unknown whether the prevalence of active smoking in stroke survivors has decreased during the 2 decades since these drugs were approved. An improved understanding of tobacco smoking among stroke survivors may inform strategies to optimize secondary prevention efforts. Therefore, we evaluated trends in active tobacco cigarette smoking among stroke survivors in the United States using 2 complementary, nationally representative health surveys spanning 1999 to 2018.
Methods
Design
We performed trends analyses using cross-sectional data from 2 US health surveys. First, we assessed trends in active tobacco smoking using data from the National Health and Nutrition Examination Survey (NHANES).7 NHANES is an in-person health survey and examination conducted in 2-year cycles; we used data from 9 consecutive 2-year cycles spanning 1999 to 2016. The NHANES data were chosen for their large temporal span and the availability of biochemical markers of active smoking for many respondents. NHANES data are collected by the Centers for Disease Control and Prevention National Center for Health Statistics using a multistage, probability cluster survey method to generate nationally representative statistics for the US population. Participants undergo standardized home interview followed by physical examination and biological specimen collection in mobile examination centers. Data collection for NHANES was approved by the National Center for Health Statistics Research Ethics Review Board, and written informed consent was obtained from participants. Second, because the absolute number of stroke survivors included in each survey cycle is relatively low in NHANES, we also assessed trends in active tobacco smoking using data from the Centers for Disease Control and Prevention Behavioral Risk Factor Surveillance System (BRFSS).8 Given its telephone-based nature, it has a larger survey sample than NHANES. The BRFSS is an annual survey that collects information regarding chronic health conditions and health-related behaviors. We used data from 2011 to 2018, which represents all years of available data that can validly be used for trends analysis. In both surveys, each survey cycle is a cross-sectional sample; participants are not followed longitudinally. The data that support the findings of this study are publicly available at https://www.cdc.gov/nchs/nhanes and https://www.cdc.gov/brfss/annual_data/annual_data.htm. Analytic methods will be made available upon reasonable request from the corresponding author. The Weill Cornell Institutional Review Board certified these analyses as exempt from review.
Population
In the NHANES analysis, we included adult participants ≥20 years of age, as they were asked questions pertinent to this analysis. The BRFSS survey included only adult respondents ≥18 years of age. For both, we excluded participants who did not answer survey questions regarding stroke and smoking history (Figure 1). We assessed trends in active smoking among those with and without a history of stroke. In both surveys, participants were asked whether they had a history of a stroke. For example, in NHANES, participants were asked, "Has a doctor or other health professional ever told you that you had a stroke?" Self-report of stroke is validated in the general population and has been used in prior large-scale epidemiological studies of stroke survivors.9–11
- Download figure
- Download PowerPoint
Figure 1. Participant flowchart. We excluded participants under the age of 20 y in the National Health and Nutrition Examination Survey (NHANES) analysis and participants with missing stroke and smoking data in both analyses. BRFSS indicates Behavioral Risk Factor Surveillance System.
Measurements
The outcome of interest was active smoking. The primary outcome was self-reported active smoking—a variable available in both NHANES and BRFSS. Participants in NHANES and BRFSS were asked whether they were currently smoking at the time of survey. Secondarily, in NHANES, we defined active smoking using serum cotinine measures in participants who had these measures available. Cotinine measures were available for participants who underwent in-person examination; participants with hemophilia or chemotherapy in the past 4 weeks were excluded. The laboratory methods are described in the NHANES manual.12 Cotinine is a nicotine metabolite with a long elimination half-life; elevated levels reflect active tobacco use.13 To categorize active smoking, we used validated cotinine thresholds that reflect active smoking rather than secondhand exposure and account for differences in smoking habits and nicotine metabolism.14 The thresholds were non-Hispanic white >4.85 ng/mL, non-Hispanic black >5.92 ng/mL, Mexican-American >0.84 ng/mL, and other >3.08 ng/mL.14 The sensitivity and specificity of these thresholds is >95%.14
We tabulated demographics and comorbidities to characterize individuals with and without prior stroke in the 2 survey populations. Demographic variables were age, sex, race/ethnicity, and health insurance status, categorized as private/commercial, Medicare, Medicaid, other, or uninsured. We tabulated the following smoking-related comorbidities: hypertension, heart disease, cancer, and pulmonary disease. In NHANES, we defined hypertension as a self-reported diagnosis, average measured blood pressure >140/90 mm Hg, or use of antihypertensive medications. In BRFSS, we defined hypertension as a self-reported diagnosis or use of antihypertensive medications. In NHANES and BRFSS, we defined heart disease as a history of myocardial infarction, angina, or coronary heart disease. Cancer and pulmonary diseases (asthma and chronic obstructive pulmonary disease) were defined using self-reported diagnoses for both surveys.
Statistical Analyses
We merged data for the 9 NHANES biannual cycles into one 18-year pooled dataset and the 8 BRFSS annual surveys into a separate 8-year pooled dataset. NHANES and BRFSS are population-based surveys that use nonrandom sampling design to ensure representation of certain population subgroups and assign sample weights to respondents to account for nonresponse and stratification, among other complexities in survey design. Our analyses used survey-specific SAS functions to account for survey design, survey strata, and sampling weights. We calculated the annual prevalence of active smoking in individuals with and without prior stroke. Then, we used logistic regression models with robust SEs to examine the association between time and the odds of active smoking.15,16 These logistic regression models evaluated whether the odds of active smoking decreased over time among those with and without prior stroke. In a sensitivity analysis, we evaluated trends in active smoking while additionally adjusting logistic regression models for age, sex, and race to account for possible changes in the demographics of stroke in the United States.17 Of note, age in BRFSS is provided in 5-year bands (eg, 25–29 years), so analyses in BRFSS treated age as a categorical variable with the youngest group serving as the reference. To test whether trends differed between those with and without prior stroke, we performed a test of interaction (stroke×time), with a threshold of P=0.10 for interaction significance on the multiplicative scale. The test of interaction provides formal testing of heterogeneity in the trend of active smoking by whether the respondent had a prior stroke. Apart from the test of interaction, the threshold of statistical significance was set at α=0.05. Statistical analyses were performed using SAS software, version 9.4 (SAS Institute, Inc, Cary, NC).
Results
The NHANES analysis spanned 1999 to 2016 and included 49 375 survey participants. The BRFSS analysis spanned 2011 to 2018 and included 3 621 741 participants. The BRFSS participants were older than NHANES participants but otherwise had similar characteristics, including when stratified by stroke (Table 1). In both surveys, individuals with prior stroke were older, were more often male, and generally had more medical diseases (Table 1).
| NHANES | BRFSS | |||
|---|---|---|---|---|
| 1999–2016 | 2011–2018 | |||
| n=49 375 | n=3 621 741 | |||
| No Stroke | Stroke | No Stroke | Stroke | |
| Respondents, n | 47 454 | 1921 | 3 470 350 | 151 391 |
| Age,† y (SE) | 46.2 (0.18) | 64.8 (0.46) | 55–59 | 65–69 |
| Women | 51.9% (51.5–52.3) | 56.5% (53.7–59.4) | 51.4% (51.3–51.5) | 52.3% (51.7–52.8) |
| Race‡ | ||||
| White | 68.5% (66.3–70.7) | 71.1% (67.8–74.3) | 64.4% (64.3–64.5) | 67.2% (66.6–67.8) |
| Black | 11.2% (12.4–14.9) | 14.9% (12.7–17.1) | 11.4% (11.3–11.5) | 15.9% (15.5–16.4) |
| Hispanic | 13.7% (12.1–15.3) | 7.6% (5.9–9.3) | 16.4% (16.3–16.5) | 10.4% 10.0–10.9) |
| Other | 6.6% (6.0–7.2) | 6.5% (4.9–8.1) | 7.8% (7.7–7.9) | 6.4% (6.1–6.7) |
| Insurance§‖ | ||||
| Private | 64.5% (63.4–65.6) | 47.2% (44.0–50.4) | 64.8% (64.5–65.1) | 30.9% (29.6–32.2) |
| Medicare | 8.0% (7.7–8.4) | 33.3% (30.6–36.0) | 19.7% (19.5–19.9) | 48.5% (47.1–49.8) |
| Medicaid | 4.4% (4.0–4.8) | 6.3% (4.8–7.9) | 8.8% (8.7–9.0) | 11.6% (10.7–12.5) |
| Other | 4.6% (4.3–4.9) | 4.5% (3.2–5.9) | 6.3% (6.2–6.4) | 8.4% (7.6–9.1) |
| Uninsured | 18.5% (17.6–19.3) | 8.6% (7.0–10.2) | 0.4% (0.3–0.4) | 0.4% (0.4–0.9) |
| Hypertension‖ | 34.6% (33.8–35.4) | 77.4% (74.8–80.0) | 31.0% (30.9–31.1) | 71.8% (71.1–72.5) |
| Heart disease | 5.1% (4.8–5.4) | 32.1% (29.2–35.1) | 5.7% (5.7–5.7) | 37.3% (36.8–37.9) |
| Pulmonary disease‖ | 17.4% (16.8–17.9) | 29.5% (26.7–32.3) | 18.8% (18.7–18.9) | 36.6% (36.0–37.1) |
| Cancer | 8.9% (8.6–9.3) | 20.8% (18.2–23.3) | 6.3% (6.3–6.3) | 16.0% (15.6–16.4) |
| Active tobacco cigarette use | 22.1% (21.4–22.9) | 24.3% (21.6–27.1) | 17.2% (17.2–17.3) | 23.4% (22.9–23.9) |
| Active smoking, by cotinine¶ | 27.7% (26.8–28.6) | 29.4% (26.2–32.7) | … | … |
NHANES
Based on the NHANES analysis, the prevalence of stroke among US adults during 1999 to 2016 was 2.8% (95% CI, 2.6–2.9). Overall, the average prevalence of active smoking was 24.3% (95% CI, 21.6–27.1) among stroke survivors and 22.1% (95% CI, 21.4–22.9) among those without prior stroke. The odds of active smoking decreased by 5% per 2-year cycle in individuals without prior stroke (odds ratio, 0.95 [95% CI, 0.93–0.96]), whereas there was no decrease for individuals with prior stroke (odds ratio, 1.00 [95% CI, 0.93–1.07]; P for interaction, 0.13; Table 2; Figure 2). Using serum cotinine measures, the overall prevalence of active smoking prevalence was 27.7% (95% CI, 26.8–28.7); self-report of active smoking underestimated biochemically ascertained active smoking by ≈5% in those with and without prior stroke (Table 1). Trends in biochemically ascertained active smoking for individuals with and without stroke were similar to the primary analysis (Table 2). When adjusting for age, sex, and race, results were consistent (Table 3).
| No Stroke | Stroke | |
|---|---|---|
| Active tobacco cigarette smoking | ||
| NHANES, 1999–2016 (n=49 375) | 0.95 (0.93–0.96) | 1.00 (0.93–1.07) |
| P for interaction, 0.13 | ||
| BRFSS, 2011–2018 (n=3 621 741) | 0.96 (0.96–0.96) | 0.99 (0.98–1.004) |
| P for interaction, <0.001 | ||
| Active smoking, biochemically assessed† | ||
| NHANES, 1999–2016 (n=44 125) | 0.96 (0.94–0.98) | 0.99 (0.92–1.06) |
| P for interaction, 0.37 | ||
| No Stroke | Stroke | |
|---|---|---|
| Active tobacco cigarette smoking | ||
| NHANES, 1999–2016 (n=49 375) | 0.96 (0.94–0.97) | 1.01 (0.93–1.09) |
| P for interaction, 0.19 | ||
| BRFSS, 2011–2018 (n=3 621 741) | 0.96 (0.96–0.96) | 0.99 (0.98–1.01) |
| P for interaction, 0.0001 | ||
| Active smoking, biochemically assessed† | ||
| NHANES, 1999–2016 (n=44 125) | 0.97 (0.95–0.99) | 1.00 (0.92–1.08) |
| P for interaction, 0.52 | ||
- Download figure
- Download PowerPoint
Figure 2. Trends in active tobacco cigarette smoking in stroke survivors in 2 nationally representative health surveys, 1999 to 2018. In contrast to individuals without stroke, active smoking has not decreased among stroke survivors between 1999 and 2018 in 2 nationally representative US health surveys. The black lines represent adults with prior stroke; the gray lines represent adults without prior stroke. BRFSS indicates Behavioral Risk Factor Surveillance System; and NHANES, National Health and Nutrition Examination Survey.
BRFSS
Based on the BRFSS survey, the prevalence of stroke among US adults during 2011 to 2018 was 3.1% (95% CI, 3.0–3.1). During this period, the overall prevalence of active smoking was 23.4% (95% CI, 22.9–23.9) among stroke survivors and 17.2% (95% CI, 17.2–17.3) among those without prior stroke. The odds of active smoking decreased 4% per year (odds ratio, 0.96 [95% CI, 0.96–0.96]) among individuals without prior stroke, whereas there was no decrease among those with prior stroke (odds ratio, 0.99 [95% CI, 0.98–1.004]; P for interaction, <0.001; Table 2; Figure 2). When adjusting for age, sex, and race, results were again consistent (Table 3).
Discussion
In this analysis of 2 complementary, nationally representative US health surveys, we found that the prevalence of active smoking among stroke survivors has not decreased in parallel with the general population.
The significant secular decrease in cigarette smoking in the United States is well-established18,19 and also seen in our data. Our study adds novel findings indicating that the prevalence of active tobacco smoking among stroke survivors has not decreased in the United States over the last 2 decades. While formal heterogeneity testing did not reveal a statistically significant interaction by stroke in NHANES, the difference in trends between those with and without prior stroke is supported by the consistent findings across analyses and survey datasets and the statistically significant interaction observed in the larger BRFSS analysis. Additionally, although biochemical ascertainment of active smoking revealed higher overall rates of active smoking in both those with and without prior stroke, trends in biochemically defined active smoking in NHANES recapitulated patterns seen in self-reported active smoking. Results were also consistent in sensitivity analyses that accounted for changes in the demographics of stroke.
Population-level antismoking interventions, such as tobacco pricing, smoke-free laws, and media campaigns, are, in part, credited for decreases in the prevalence of smoking in the general population.18 Differential reach or impact of such interventions, in addition to differences in smoking cessation, may account for the divergent trends in smoking between those with and without prior stroke. However, further study is needed to identify key determinants of unchanging smoking prevalence rates in stroke survivors. In addition to optimizing delivery of existing antismoking interventions to stroke survivors, efforts to identify treatment strategies specifically for patients with stroke may be needed.
The key strengths of our analysis pertain to the nationally representative nature of NHANES and BRFSS, the availability of cotinine measures in NHANES, and the long time period examined. Our results should be considered in light of the following limitations. First, we lacked data on prestroke smoking, and, therefore, we cannot directly compare smoking rates before and after stroke. Second, health surveys reach those who are able to participate; our study does not account for patients with severely disabling stroke who are institutionalized. The public health impact of our findings relates to the importance of secondary prevention for community-dwelling stroke survivors at risk of recurrent stroke, which is frequently more disabling. Third, we cannot determine whether smoking behavior differs by stroke type because survey data do not specify whether the prior stroke was ischemic or hemorrhagic. Fourth, self-report of stroke and smoking may cause misclassification. The consistent results in the analysis of biochemically ascertained active smoking, in addition to the low suspicion for differential misclassification, mitigate the risk of misclassification bias accounting for our findings. Fifth, although electronic cigarette use may be relevant for stroke risk,20 we were unable to account for trends in electronic cigarette use due to the limited availability of data. Last, there may have been insufficient power for the tests of interaction in NHANES analyses. The consistent results across models, and the presence of a statistically significant interaction in the larger BRFSS analysis, support our conclusions.
Conclusions
In 2 nationally representative surveys, ≈1 of 5 stroke survivors were active smokers. Unlike among the general population, active cigarette smoking has not decreased among stroke survivors over the last 2 decades. Further emphasis on effective smoking cessation as part of secondary prevention after stroke may be helpful.
Sources of Funding
Dr Parikh is supported by the New York State Department of Health, Empire Clinical Research Investigator Program and by the Florence Gould Foundation.
Disclosures
Dr Merkler received funding from the American Heart Association and Leon Levy Foundation and has received personal compensation for medicolegal consulting on stroke. Dr Murthy received funding from the National Institutes of Health (NIH)/National Institute of Neurological Disorders and Stroke. Dr Iadecola has received personal fees from Broadview Ventures. Dr Navi serves as a Data and Safety Monitoring Board member for the Patient-Centered Outcomes Research Institute-funded TRAVERSE trial (Testosterone Replacement Therapy for Assessment of Long-Term Vascular Events and Efficacy Response in Hypogonadal Men Study) and has received personal compensation for medicolegal consulting on stroke. Dr Kamel serves as a co-principal investigator for the NIH-funded ARCADIA trial (Atrial Cardiopathy and Antithrombotic Drugs in Prevention After Cryptogenic Stroke), which receives in-kind study drug from the BMS-Pfizer Alliance and in-kind study assays from Roche Diagnostics, serves as Deputy Editor for JAMA Neurology, serves as a steering committee member of Medtronic Stroke AF trial (uncompensated), serves on an end point adjudication committee for a trial of empagliflozin for Boehringer-Ingelheim, and has served on an advisory board for Roivant Sciences related to Factor XI inhibition. The other authors report no conflicts.
Footnotes
References
- 1.
Epstein KA, Viscoli CM, Spence JD, Young LH, Inzucchi SE, Gorman M, ; IRIS Trial Investigators. Smoking cessation and outcome after ischemic stroke or TIA. Neurology . 2017; 89:1723–1729. doi: 10.1212/WNL.0000000000004524Google Scholar - 2.
Chen J, Li S, Zheng K, Wang H, Xie Y, Xu P, . Impact of smoking status on stroke recurrence. J Am Heart Assoc . 2019; 8:e011696. doi: 10.1161/JAHA.118.011696LinkGoogle Scholar - 3.
Zhang Q, Wang Y, Song H, Hou C, Cao Q, Dong K, . Clopidogrel and ischemic stroke outcomes by smoking status: smoker's paradox? J Neurol Sci . 2017; 373:41–44. doi: 10.1016/j.jns.2016.12.025CrossrefMedlineGoogle Scholar - 4.
Anthenelli RM, Benowitz NL, West R, St Aubin L, McRae T, Lawrence D, . Neuropsychiatric safety and efficacy of varenicline, bupropion, and nicotine patch in smokers with and without psychiatric disorders (EAGLES): a double-blind, randomised, placebo-controlled clinical trial. Lancet . 2016; 387:2507–2520. doi: 10.1016/S0140-6736(16)30272-0CrossrefMedlineGoogle Scholar - 5.
Kernan WN, Ovbiagele B, Black HR, Bravata DM, Chimowitz MI, Ezekowitz MD, ; American Heart Association Stroke Council, Council on Cardiovascular and Stroke Nursing, Council on Clinical Cardiology, and Council on Peripheral Vascular Disease. Guidelines for the prevention of stroke in patients with stroke and transient ischemic attack: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke . 2014; 45:2160–2236. doi: 10.1161/STR.0000000000000024LinkGoogle Scholar - 6.
Powers WJ, Rabinstein AA, Ackerson T, Adeoye OM, Bambakidis NC, Becker K, . Guidelines for the early management of patients with acute ischemic stroke: 2019 update to the 2018 guidelines for the early management of acute ischemic stroke: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke . 2019; 50:e344–e418. doi: 10.1161/STR.0000000000000211LinkGoogle Scholar - 7. National Center for Health Statistics. National Health and Nutrition Examination Survey (NHANES): Questionnaires, datasets, and related documentation.https://wwwn.cdc.gov/nchs/nhanes/. Accessed November 1, 2019Google Scholar
- 8. Centers for Disease Control and Prevention (CDC). Behavioral Risk Factor Surveillance System Survey Data . Atlanta, Georgia: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, 2016-2017. https://www.cdc.gov/brfss/index.html. Accessed November 1, 2019Google Scholar
- 9.
Okura Y, Urban LH, Mahoney DW, Jacobsen SJ, Rodeheffer RJ . Agreement between self-report questionnaires and medical record data was substantial for diabetes, hypertension, myocardial infarction and stroke but not for heart failure. J Clin Epidemiol . 2004; 57:1096–1103. doi: 10.1016/j.jclinepi.2004.04.005CrossrefMedlineGoogle Scholar - 10.
Lin MP, Ovbiagele B, Markovic D, Towfighi A . "Life's simple 7" and long-term mortality after stroke. J Am Heart Assoc . 2015; 4:e001470LinkGoogle Scholar - 11.
Jamrozik E, Hyde Z, Alfonso H, Flicker L, Almeida O, Yeap B, . Validity of self-reported versus hospital-coded diagnosis of stroke: a cross-sectional and longitudinal study. Cerebrovasc Dis . 2014; 37:256–262. doi: 10.1159/000358583CrossrefMedlineGoogle Scholar - 12. Centers for Disease Control and Prevention (CDC). National Center for Health Statistics. National Health and Nutrition Examination Survey Examination and Laboratory Procedures.https://wwwn.cdc.gov/nchs/data/nhanes/2013-2014/manuals/2013_MEC_Laboratory_Procedures_Manual.pdf. Accessed: Novemrber 1, 2019Google Scholar
- 13. SRNT Subcomittee on Biochemical Verification.Biochemical verification of tobacco use and cessation. Nicotine Tob Res . 2002; 4:149–159Google Scholar
- 14.
Benowitz NL, Bernert JT, Caraballo RS, Holiday DB, Wang J . Optimal serum cotinine levels for distinguishing cigarette smokers and nonsmokers within different racial/ethnic groups in the United States between 1999 and 2004. Am J Epidemiol . 2009; 169:236–248. doi: 10.1093/aje/kwn301CrossrefMedlineGoogle Scholar - 15. Center for Behavioral Health Statistics and Quality. (2017). National Survey on Drug use and Health: An overview of trend testing methods and applications in NSDUH and other studies. Substance Abuse and Mental Health Services Administration, Rockville, MD.https://www.samhsa.gov/data/report/nsduh-overview-trend-testing-methods-and-applications-nsduh-and-other-studies. Accessed November 1, 2019Google Scholar
- 16.
Ingram D, Malec D, Makuc D, Kruszon-Moran D, Gindi R, Albert M, . National Center for Health Statistics guidelines for analysis of trends. National Center for Health Statistics. Vital Health Stat . 2018; 2:1–71.Google Scholar - 17.
Ramirez L, Kim-Tenser MA, Sanossian N, Cen S, Wen G, He S, . Trends in acute ischemic stroke hospitalizations in the United States. J Am Heart Assoc . 2016; 5:e003233LinkGoogle Scholar - 18.
Jamal A, King BA, Neff LJ, Whitmill J, Babb SD, Graffunder CM . Current cigarette smoking among adults - United States, 2005-2015. MMWR Morb Mortal Wkly Rep . 2016; 65:1205–1211. doi: 10.15585/mmwr.mm6544a2CrossrefMedlineGoogle Scholar - 19.
Benjamin EJ, Muntner P, Alonso A, Bittencourt MS, Callaway CW, Carson AP, ; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart disease and stroke statistics-2019 update: a Report From the American Heart Association. Circulation . 2019; 139:e56–e528. doi: 10.1161/CIR.0000000000000659LinkGoogle Scholar - 20.
Parekh T, Pemmasani S, Desai R . Risk of stroke with e-cigarette and combustible cigarette use in young adults. Am J Prev Med . 2020; 58:446–452. doi: 10.1016/j.amepre.2019.10.008Google Scholar
Source: https://www.ahajournals.org/doi/full/10.1161/STROKEAHA.120.029084
0 Response to "Stroke Victims Who Continue to Smoke"
Postar um comentário