Download:
-
Pharmacologic treatment is among the most effective tobacco cessation interventions, significantly enhancing the likelihood of quitting (1–2). However, medication non-adherence remains a substantial barrier, reducing smoking cessation rates in both clinical trials and real-world settings (3). In China, three types of licensed medications are available: nicotine replacement therapy (NRT), varenicline, and bupropion. Despite this, the relationship between treatment adherence to smoking cessation medications and treatment outcomes among Chinese smokers is largely unexplored. This study utilized data from the nationwide, multicenter China National Tobacco Cessation Cohort Study (CNTCCS) to assess real-world adherence to smoking cessation medications. The results revealed that only 38.2% of smokers adhered to medication treatment at the 12-week follow-up. Those with good adherence were significantly more likely to quit smoking [adjusted odds ratio (OR)=1.90, 95% confidence interval (CI): 1.46, 2.47, P<0.001]. Enhancing adherence is a crucial strategy to promote successful smoking cessation and reduce smoking prevalence.
The CNTCCS was conducted from December 2017 to January 2022 and is recognized as the first and largest cohort study on tobacco cessation in China (4). Potential smoking cessation centers were identified from a list of 366 clinics, compiled from a previous national survey on smoking cessation services. Inclusion criteria for the centers required the capability to provide smoking cessation medications and representation from all seven geographical regions of China. Ultimately, 27 centers were selected to participate (
Supplementary Table S1 ). Recruitment was facilitated through posters displayed in each center. Eligibility criteria for participants included: 1) current smokers willing to quit, 2) aged 18–85 years, and 3) expired air carbon monoxide (CO) levels of at least 10 ppm. Exclusion criteria included severe cardiovascular or cerebrovascular diseases, neuropsychiatric disorders, significant liver or kidney impairment, severe psychiatric illnesses, pregnancy or breastfeeding, and allergies to cessation medications. All participants voluntarily joined the CNTCCS and provided informed consent.All study physicians received standardized training. Following preliminary screenings, a baseline survey was conducted to gather information for each recruited participant. The collected data included demographic information (gender, age, ethnicity, education, marital status, and income), health-related characteristics (self-reported health status and comorbid diseases), tobacco-related characteristics [number of cigarettes smoked per day, duration of smoking, and results from the Fagerstrom Test for Nicotine Dependence (FTND)] (2), and alcohol use. The Visual Analogue Scale was used to evaluate participants’ self-awareness of the importance of quitting, self-confidence in quitting, and readiness to quit smoking, with scores of 1–3, 4–6, and 7–10 defined as weak, medium, and strong levels, respectively (5). Study physicians prescribed treatment regimens based on the China Clinical Guideline for Tobacco Cessation (2015 version) (2), considering current scientific evidence and the patient’s preferences, especially for those with contraindications to medications due to specific comorbidities. All participants were advised to undergo 12 weeks of treatment and required to attend face-to-face visits at the study site at 1, 2, 4, 6, 9, 12, and 24 weeks after the treatment initiation. The follow-up questionnaire included self-reported information on smoking status and medication use. All data from each center were manually checked for completeness and accuracy by the research staff. An online Case Report Form (CRF) system, developed by the expert advisory panel, was used for data collection (
https://jieyan.einmatrix.com ).The CNTCCS complied with all relevant ethical regulations and received approval from the Institutional Review Boards at all study centers. The clinical trial registration number is ChiCTR1800016919.
Adherence to smoking cessation treatment was defined as taking more than 80% of prescribed medication from baseline to the 12-week follow-up. Participants who did not meet this criterion were considered non-adherent (5–6). The primary outcome was adherence at 12 weeks. Secondary outcomes included adherence at 1, 2, 4, 6, and 9 weeks; CO-validated 7-day point abstinence rate at 1, 2, 4, 6, 9, and 12 weeks; and factors associated with adherence at 12 weeks. Statistical analyses were conducted using SPSS (version 19.0; SPSS, Inc., Armonk, NY, USA.). Categorical variables were presented as numbers (percentages), and the chi-square test was used for comparisons. Logistic regression analysis measured the relationship between potential influencing factors and adherence at 12 weeks, as well as between adherence and successful smoking cessation at 12 weeks, with ORs and 95% CI. A significance level of 0.05 (two-tailed) was used to define statistical significance.
A total of 2,943 participants with reliable data were recruited in the CNTCCS. Among them, 1,852 participants were prescribed smoking cessation medication. Of these, 742 participants were lost to follow-up, resulting in 1,120 participants included in this analysis. Baseline characteristics are summarized in
Supplementary Table S2 . Overall, 94.5% of the participants were male, with an average age of 52.9±13.6 years. Additionally, 90.5% were married, 44.9% had a college education or higher, and 33.1% had a monthly income exceeding 6,000 Chinese Yuan (CNY). Regarding smoking behavior, 60.4% reported smoking more than 20 cigarettes per day, and 75.3% had smoked for at least 20 years. For nicotine dependence, 41.4% showed moderate dependence (FTND=3–6), and 25.6% showed severe dependence (FTND≥7). Of all participants analyzed, 40.4%, 51.5%, and 8.1% were treated with varenicline, bupropion, and NRT, respectively.The adherence rate gradually decreased from 83.5% at the 1-week follow-up to 38.2% at the 12-week follow-up. This trend was observed across all three types of medications: in the varenicline group, adherence dropped from 85.5% at 1 week to 33.6% at 12 weeks; in the bupropion group, from 83.8% at 1 week to 45.1% at 12 weeks; and in the NRT group, from 72.1% at 1 week to 17.6% at 12 weeks (Figure 1A). Multivariable logistic regression analysis indicated that participants with higher education levels were significantly more likely to adhere to smoking cessation treatment (middle and high school: OR=2.73, 95% CI: 1.58, 4.72, P<0.001; college and higher: OR=3.47, 95% CI: 1.98, 6.09, P<0.001) compared to those with primary school education or less (Table 1).
Figure 1.Participants’ adherence rate and abstinence rate at different time points of smoking cessation treatment in 27 centers, China, 2017–2022. (A) Adherence rate, overall and stratified by medication used, at different time points. (B) Abstinence rates, overall and stratified by adherence group, at different time points.
Abbreviation: NRT=nicotine replacement therapy.Characteristics N Adherent participants, N (%) Adjusted OR* (95% CI) P value Gender Women 62 27 (43.5) 1.00 (Ref.) Men 1,058 401 (37.9) 0.76 (0.45–1.31) 0.330 Age, years <40 248 89 (35.9) 1.00 (Ref.) 40–49 196 74 (37.8) 1.18 (0.78–1.80) 0.431 50–59 289 119 (41.2) 1.20 (0.81–1.77) 0.376 60 and above 387 146 (37.7) 1.07 (0.72–1.57) 0.751 Ethnicity Others 61 17 (27.9) 1.00 (Ref.) Han 1,059 411 (38.8) 1.55 (0.96–2.82) 0.148 Marriage Single 83 32 (38.6) 1.00 (Ref.) Married 1,014 391 (38.6) 1.01 (0.60–1.69) 0.970 Separated/divorced/widowed 23 5 (21.7) 0.45 (0.15–1.42) 0.174 Education Primary school or below 95 19 (20.0) 1.00 (Ref.) Middle and high school 522 202 (38.7) 2.73 (1.58–4.72) <0.001 College and higher 503 207 (41.2) 3.47 (1.98–6.09) <0.001 Monthly income, CNY <1,000 50 17 (34.0) 1.00 (Ref.) 1,000–2,999 215 84 (39.1) 1.19 (0.61–2.34) 0.605 3,000–5,999 485 211 (43.5) 1.42 (0.75–2.70) 0.281 6,000–9,999 244 69 (28.3) 0.74 (0.37–1.47) 0.389 >10,000 126 47 (37.3) 1.07 (0.51–3.49) 0.855 Self-reported health status Poor 220 83 (37.7) 1.00 (Ref.) Average 529 196 (37.1) 1.04 (0.65–1.65) 0.875 Good 371 149 (40.2) 1.03 (0.64–1.53) 0.907 Cardiovascular diseases at baseline No 999 385 (38.5) 1.00 (Ref.) Yes 121 43 (35.5) 0.99 (0.64–1.53) 0.957 Cancer at baseline No 1,067 409 (38.3) 1.00 (Ref.) Yes 53 19 (35.8) 1.11 (0.59–2.10) 0.740 Respiratory diseases at baseline No 770 295 (38.4) 1.00 (Ref.) Yes 350 133 (37.9) 0.93 (0.67–1.28) 0.650 Depression at baseline No 1,179 416 (38.6) 1.00 (Ref.) Yes 41 12 (29.3) 0.67 (0.18–2.46) 0.546 Anxiety at baseline No 1,180 415 (38.4) 1.00 (Ref.) Yes 40 13 (32.5) 1.28 (0.78–1.30) 0.966 Alcohol use No 460 180 (39.1) 1.00 (Ref.) Yes 660 248 (37.6) 1.01 (0.78–1.30) 0.966 Cigarettes smoked per day 1–9 51 18 (35.3) 1.00 (Ref.) 10–19 392 160 (40.8) 1.14 (0.61–2.16) 0.690 20–29 502 193 (38.4) 0.99 (0.52–1.86) 0.963 30 and above 175 57 (32.6) 0.87 (0.44–1.74) 0.693 Smoking duration, year 1–9 36 13 (36.1) 1.00 (Ref.) 10–19 241 82 (34.0) 0.86 (0.40–1.84) 0.697 20–29 275 101 (36.7) 1.07 (0.50–2.27) 0.862 30 and above 568 232 (40.8) 1.44 (0.69–3.01) 0.336 FTND 0−3 369 140 (37.9) 1.00 (Ref.) 4–6 464 174 (37.5) 1.02 (0.75–1.37) 0.924 7 and above 287 114 (39.7) 1.13 (0.80–1.60) 0.498 Self-awareness of the importance of quitting Weak 146 57 (39.0) 1.00 (Ref.) Medium 325 112 (34.5) 1.16 (0.45–2.97) 0.759 Strong 645 256 (39.7) 1.35 (0.52–3.49) 0.537 Self-confidence in quitting Weak 154 60 (39.0) 1.00 (Ref.) Medium 407 138 (33.9) 1.07 (0.42–2.73) 0.896 Strong 555 227 (40.9) 1.37 (0.51–3.68) 0.529 Self-readiness in quitting Weak 163 66 (40.5) 1.00 (Ref.) Medium 335 115 (34.3) 0.58 (0.24–1.40) 0.225 Strong 618 244 (39.5) 0.55 (0.22–1.38) 0.550 Abbreviation: OR=odds ratio; CI=confidence interval; Ref.=reference; FTND=Fagerstrom Test for Nicotine Dependence; CNY=Chinese Yuan.
* Adjusted for gender, age, ethnicity, marital status, education, monthly income, self-reported health status, alcohol use, cigarettes smoked per day, smoking duration, FTND, self-awareness of the importance of quitting, self-confidence in quitting, and readiness to quit.Table 1. Factors associated with medication adherence to smoking cessation treatment at 12-week follow-up in 27 centers, China, 2017–2022.
The CO-validated 7-day point abstinence rates of participants who adhered to their medication regimen were consistently higher than those of non-adherent participants across all follow-up intervals: at 1 week, 29.7% vs. 20.8% (P=0.007); at 2 weeks, 45.4% vs. 23.2% (P<0.001); at 4 weeks, 50.4% vs. 33.3% (P<0.001); at 6 weeks, 64.8% vs. 37.3% (P<0.001); at 9 weeks, 62.8% vs. 50.9% (P<0.001); and at 12 weeks, 60.7% vs. 47.5% (P<0.001) (Figure 1B). After adjusting for variables such as gender, age, ethnicity, marital status, education, monthly income, self-reported health status, alcohol use, daily cigarette consumption, smoking duration, FTND score, perceived importance of quitting, self-confidence in quitting, readiness to quit, and the specific medication allocated, adherence to smoking cessation medication at 12 weeks was significantly associated with higher CO-validated 7-day point abstinence rates (OR=1.90, 95% CI: 1.46, 2.47, P<0.001). This significant association remained consistent across each specific medication (Table 2).
Group N Abstinent participants, N (%) adjusted OR* (95% CI) P value Overall Non-adherent 692 329 (47.5) 1.00 (Ref.) Adherent 428 260 (60.7) 1.90 (1.46–2.47) <0.001 Varenicline Non-adherent 300 165 (55.0) 1.00 (Ref.) Adherent 152 93 (61.2) 1.23 (1.06–1.55) 0.041 Bupropion Non-adherent 317 130 (41.0) 1.00 (Ref.) Adherent 260 158 (60.8) 2.45 (1.70–3.52) <0.001 NRT Non-adherent 75 34 (45.3) 1.00 (Ref.) Adherent 16 9 (56.3) 1.67 (1.22–1.97) 0.026 Abbreviation: OR=odds ratio; CI=confidence interval; Ref.=reference; NRT=nicotine replacement therapy.
* Adjusted for gender, age, ethnicity, marriage, education, monthly income, self-reported health status, alcohol use, cigarettes smoked per day, smoking duration, Fagerstrom Test for Nicotine Dependence, self-awareness of the importance of quitting, self-confidence in quitting, and self-readiness in quitting, allocated medication.Table 2. Association between participants’ medication adherence to smoking cessation treatment and successful quitting at 12-week follow-up in 27 centers, China, 2017–2022.
Additionally, there was little difference in adherence rates before and after the coronavirus disease 2019 (COVID-19) lockdown period, except for 12-week adherence to varenicline (Table 3).
Medication n Adherence rate, % 1 week 2 weeks 4 weeks 6 weeks 9 weeks 12 weeks Overall 2017–2019 884 83.7 80.7 75.1 67.6 42.3 38.3 2020–2022 236 82.8 86.0 77.8 70.3 38.8 37.7 P value 0.721 0.061 0.410 0.452 0.357 0.858 Varenicline 2017–2019 360 85.7 86.0 80.8 69.7 43.2 36.7 2020–2022 92 84.9 89.4 81.4 65.9 35.2 21.7 P value 0.839 0.389 0.888 0.490 0.178 0.007 Bupropion 2017–2019 451 83.8 76.3 70.7 65.9 40.4 44.8 2020–2022 126 84.0 79.9 73.8 72.5 42.2 47.2 P value 0.933 0.919 0.767 0.113 0.802 0.638 NRT 2017–2019 73 74.2 78.8 79.0 70.7 42.9 19.2 2020–2022 18 63.6 71.4 62.0 56.9 50.0 11.1 P value 0.325 0.468 0.078 0.111 0.626 0.421 Abbreviation: NRT=nicotine replacement therapy. Table 3. Participants’ adherence rates at various time points during smoking cessation treatment, stratified by medication type and time period, across 27 centers in China from 2017 to 2022.
-
After 12 weeks of follow-up, only 38.2% of smokers adhered to the allocated smoking cessation medication. A higher adherence rate was observed among those with higher education levels. Furthermore, medication adherence significantly impacted smoking cessation outcomes. These findings suggest that more intensive management of adherence to smoking cessation medication is essential to enhance the effectiveness of pharmaceutical interventions in smoking treatment practices.
In our study, 16.5% of participants were non-adherent to smoking cessation medication at the 1-week follow-up. This aligns with previous literature indicating that 20% of smokers who receive prescriptions for smoking cessation medication never fill them (3). This is concerning because initiation is a fundamental phase, and medication use during the first few weeks predicts the effectiveness of a full-course smoking cessation treatment (6). The American Thoracic Society’s Clinical Practice Guideline recommends that health professionals counsel patients about the relative safety and efficacy of pharmacological therapies to promote adherence when treatment is initiated (7). Implementation is another critical phase of medication adherence. Our results indicated a decrease in adherence over time, with better adherence observed for participants using varenicline and bupropion at 12 weeks. This is consistent with previous studies in China. For NRT, the adherence rate is reported to be 16% (8), while for varenicline/bupropion, 48.5% of participants demonstrated good adherence (5). The reasons for these adherence disparities across different medications may be attributed to several factors. Varenicline is highly efficacious, and bupropion is relatively inexpensive, both of which may promote continued use. Conversely, NRT, as an over-the-counter medication, is often limited in availability in most Chinese hospitals. Despite the convenience and popularity of e-commerce, there remains a mistrust of online medication purchases (3), contributing to non-adherence to prescribed NRT. The decline in adherence to varenicline following COVID-19 may also be due to its unavailability in 2021. Given the significant correlation between adherence and abstinence rates, improving the sustainable availability of medications is crucial for enhancing adherence and achieving better smoking cessation outcomes.
Non-adherence can be both intentional and unintentional, influenced by a person’s motivations and capabilities. Intentional non-adherence primarily relates to perceptions of medication treatment, while unintentional non-adherence pertains to practical factors such as medication availability and affordability (9). Our study indicates that a higher education level is associated with better adherence, which is logical as education correlates with knowledge, beliefs, and socioeconomic status, thereby influencing adherence factors. Current interventions targeting increased medication adherence vary widely in content and characteristics (10). Further in-depth research is necessary to identify effective adherence interventions for Chinese smokers, particularly those with lower education levels.
To the best of our knowledge, CNTCCS is the first and largest cohort study on tobacco cessation in China, providing a comprehensive view of real-world adherence to smoking cessation medications.
However, this study has several limitations. First, adherence data were self-reported by participants, which may introduce bias. Second, the lost-contact rate was relatively high (nearly 40%), partially due to the COVID-19 pandemic. Third, there were relatively few participants using the NRT treatment regimen, potentially causing some bias in the results. Finally, although the 27 centers were selected to provide diverse inclusion across the country, they may not fully represent all smoking cessation centers in China.
In conclusion, although evidence-based smoking cessation interventions have been implemented, adherence to medication treatment among smokers in China remains low. Our results indicate that better adherence to medication is significantly associated with successful quitting. Therefore, it is necessary to develop effective strategies and policies to facilitate medication adherence, thereby enhancing the effectiveness of smoking cessation treatments.
-
No conflicts of interest.
-
We thank all participants and supporters of China National Tobacco Cessation Cohort Study.
HTML
| Citation: |
|