Tobacco smoking is associated with a high global mortality rate since it is known to cause cancers and lung and heart diseases. To control and reduce annual mortality attributed to smoking, it is essential to design applicable smoke cessation programs based on realistic tobacco exposure risk assessment. In this regard, understanding the smoking habits of the smoker is crucial. Using self-report smoking habit surveys is a common approach in measuring basic variables of smoking habits. However, smoking topography measurement devices have recently become available for investigating smoking habit variables accurately. In this study, we conducted a self-report survey to investigate a group of Korean smokers’ smoking habit variables such as the number of smoked cigarettes per day, puff counts, and total smoking time. The survey also included items from the Fagerström Test for Nicotine Dependence (FTND). The results were compared with the corresponding variables from machine-determined data to investigate their correlation and reliability. Results indicate that Korean smokers have a reliable understanding of the average number of cigarettes they smoke daily (ρ = 0.517, Cronbach’s α = 0.754) and the time to first cigarette (TTFC) after waking up (ρ = −0.587, Cronbach’s α = 0.623), as fundamental items of the FTND score. Nevertheless, these smokers significantly under-reported the puff number and total smoking time, which can cause significant underestimation in the tobacco exposure risk assessment. Consequently, we suggest the application of self-report surveys that are based on the amount of daily smoked cigarettes (e.g. FTND) for clinical or risk assessment purposes. Using smoking topography measurement devices is recommended overusing self-report surveys in measuring smoking habit variables such as puff count and smoking time more accurately.
According to the World Health Organization (WHO), consuming tobacco is one of the most important global public health threats, causing more than seven million deaths annually [
There are several options to measure an individual’s smoking behavior, including self-report surveys and using smoking topography instruments [
Nevertheless, it is essential to investigate the reliability of the smoking habits survey by comparing it to the machine-determined smoking topography. In 2008, Shahab et al. [
Despite the importance of this matter, there is a limited number of studies on the evaluation of smoking behaviors of Korean smokers [
Recruitment of subjects was conducted as described by Yang et al. [
The raw results of the survey were compared with the associated ones measured using the portable CReSS Pocket device. The estimation of smoked cigarettes per day in the survey was done using
To investigate the smoking volume through the self-report smoking habit survey, a rank order scaling question is used to extract details on the smoke inhalation depth. The scale was ranged from 1 (no inhalation) to 10 (deep inhalation down to the lung), and the bigger number may indicate higher inhalation of smoking volume. The survey also included the Fagerström Test for Nicotine Dependence (FTND) [
The smoking topography experiments were conducted as described by Yang et al. [
Information provided by the self-report smoking habit survey is based on the knowledge of the smoker about himself. In contrast, using topography measurement devices may affect how they smoke, especially during the first few smoking events [
Results of machine-determined smoking topography and self-reported smoking habits are presented in
The average FTND score of the participants is calculated to be 3.13, which indicates a low nicotine dependency. More than 70% of the participants were evaluated to have low or very low nicotine dependency and only 4% of the participants had an FTND score of more than 8. The TTFCFTND for more than 50% of the participants was 30 minutes after waking up. The average CPD determined by the smoking topography device (CPDDevice) is close to the CPDSurvey, which is calculated based on the results of the questionnaire items for the average number of cigarettes smoked during weekdays and weekends separately. The difference between the average CPDWeekdays and CPDWeekends is only one cigarette. The averages of PCDevice and TSTDevice were 20.4 and around 3 minutes, respectively. According to the results of the self-report smoking habit survey, the majority of the participants declared that they try to smoke their cigarette with less than 14 puffs and within 1–1.5 minutes. Eleven subjects failed to report their wake-up time and were therefore excluded from the TTFCDevice analysis. Under this circumstance, the average of TTFCDevice was approximately 79 minutes.
Results also yield that the TTFCFTND and TTFCDevice have a strong negative and statistically significant correlation (ρ = −0.587, p<0.001). The negative sign of the correlation coefficient is mainly because of the scoring method of FTND [
Correlations between PCSurvey and TSTSurvey with the corresponding machine-determined variables, puff count and total smoking time, are positive but are neither strong nor statistically significant. The correlation between PCSurvey and machine-determined puff count in our study is lower than the one reported by Shahab et al. [
The Cronbach’s α was calculated for variables measured by the self-report smoking habit survey against the corresponding ones measured by the CReSS Pocket device. Cronbach’s α for CPDSurvey and CPDDevice was calculated to be 0.623, indicating a relatively high reliability of the average number of daily smoked cigarettes investigated by the self-report smoking habit survey. Similarly, the Cronbach’s α for TTFCFTND and TTFCDevice was 0.754. Accordingly, it may be possible to rely on surveys such as FTND that are designed based on the self-reported daily smoked cigarettes for clinical and tobacco smoke risk assessment exposures [
The results of inter-rater reliability evaluation between smoking habit variables obtained using the survey with the corresponding ones measured by topography device and calculated using Cohen’s Kappa value were consistent with the results of Cronbach’s α. In particular, the Kappa value for agreement between TTFCFTND and TTFCDevice was calculated to be 0.298, which was statistically significant (p=0.001). For TTFCFTND and TTFCDevice, the Kappa value was 0.226 (p<0.001). This is in contrast with the Kappa values calculated for PCSurvey and TSTSurvey against PCDevice and TSTDevice were −0.041 (p=0.252) and 0.009 (p=0.791), respectively, indicating very weak and statistically insignificant inter-rater reliability.
Nevertheless, two smokers who declared that they smoke more than 30 cigarettes per day in the self-report survey have over-reported. The Wilcoxon signed-rank test results indicated that there was no statistically significant difference between the median of CPDSurvey and CPDDevice measurements (z= −1.002, p=0.316, r=0.100). In contrast, there was a statistically significant difference between the puff counts (z = −7.639, p < 0.001, r = 0.764) and total smoking time (z= −7.928, p<0.001, r=0.793), with a relatively large effect size.
Regarding puff count, the Wilcoxon signed-rank test result shows that 85% of the smokers underestimate the puff number per cigarette. This finding is consistent with the results of Pulcu [
Since about 10% of the subjects failed to provide their wake-up time, we could not calculate the TTFCDevice for them. Accordingly, because the
This study should be considered with several limitations that are primarily due to limited funds and resources. The number of participants is relatively small compared to the nationwide population of South Korea and the majority of the participants were not heavy smokers. Due to the social limitations during the COVID19 pandemic, it was challenging to execute the investigation to involve more subjects. Moreover, the number of female smokers who volunteered to participate in this study was relatively lower than male smokers owing to cultural challenges. Also, considering the location of the venue where we conducted our study, we could not gather many subjects from provinces of South Korea that are of considerable distance from Seoul. Finally, since 11 subjects decided not to report their wake-up time, there were challenges in analyzing the TTFCDevice for the whole study community.
Nevertheless, we believe that these limitations may not have a significant effect on the results, which are consistent with those obtained by several prior studies such as Shahab et al. [
In this study, we have examined different smoking habit variables using a self-report survey and smoking topography measurement device for a group of Korean smokers. We then compared the results and investigated their association and reliability with the results obtained from the smoking topography measurement device. Our results demonstrate that Korean smokers have a relatively realistic understanding of the average number of cigarettes they smoke daily and the time to their first cigarette of the day. Nonetheless, they under-report their puff numbers and total smoking time, which are key elements in measuring exposure to tobacco smoke. Underestimating these key variables can significantly lead to underestimation of risks of exposure to tobacco smoke. Accordingly, it is recommended not to rely on self-report surveys for smoking topography measurements. Instead, using surveys that are based on measuring the number of cigarettes smoked daily such as FTND, is more reliable for clinical applications, risk assessment procedures, and in designing applicable smoking cessation programs. Accordingly, we recommend that the national organizations running nationwide health investigation programs such as the Korea National Health and Nutrition Examination Survey (KNHANES) to include both FTND and topography measurements for active smokers in their survey programs.
This work was supported by the Research Program funded by the Korea Disease Control and Prevention Agency (fund code 2019-E6704-00)
The authors declare no conflicts of interest.
JY: Conceptualization, Methodology, Validation, Formal Analysis, Investigation, Resources, Writing - Review & Editing; SH: Methodology, Formal Analysis, Investigation, Visualization, Writing - Original draft Preparation; CL: Methodology, Formal Analysis, Investigation, Data Curation; WH: Formal Analysis, Investigation, Visualization; YS: Formal Analysis, Investigation; YL: Conceptualization, Supervision, Writing-Reviewing and Editing.
Smoked Cigarettes per Day (Measured using CReSS Pocket device)
Smoked Cigarettes per Day During Weekends (Measured using FTND)
Smoked Cigarettes per Day (Measured using self-report survey)
Smoked Cigarettes per Day During Weekdays (Measured using self-report survey)
Smoked Cigarettes per Day During Weekends (Measured using self-report survey)
Framework Convention on Tobacco Control
Fagerström Test for Nicotine Dependence
Inter-Puff Interval
Korea National Health and Nutrition Examination Survey
Puff Counts (Measured using CReSS Pocket device)
Puff Counts (Measured using self-report survey)
Puff Time
Sustainable Development Goal
Total Smoking Time (Measured using CReSS Pocket device)
Total Smoking Time (Measured using self-report survey)
Total Daily Smoking Volume (Measured using CReSS Pocket device)
Time to First Cigarette (Measured using FTND)
Time to First Cigarette (Measured using CReSS Pocket device)
World Health Organization
Comparison of results measured using the CReSS Pocket device and the self-report smoking habit survey for (a) number of cigarettes smoked per day; (b) puff counts; (c) total smoking time.
Smoking habits survey variables and questions.
Variable | Question type | Question | Choices for answers |
---|---|---|---|
CPDWeekdays | Text | How many cigarettes do you usually smoke per day during weekdays (Monday–Friday)? | |
CPDWeekends | Text | How many cigarettes do you usually smoke per day during weekends (Saturday–Sunday)? | |
PCSurvey | Multiple Choice | How many times do you usually puff when smoking one cigarette? | (1) 5–10 times; (2) 11–14 times; |
TSTSurvey | Multiple Choice | How long does it take for you to smoke one cigarette? | (1) < 30s; (2) 30s – < 1 min; |
Smoke Inhalation Depth | Rank Order Scaling | On a scale of 1 to 10, how deep do you inhale the smoke? |
1: No inhalation (keeping in mouth) ~ |
The answer has to be a positive integer number from 1 to 10.
Results of self-reported smoking habits and machine-determined smoking topography.
Measurement | Variable | Group or unit | n | Value (%n or mean±SD) | Rang (Min-max) |
---|---|---|---|---|---|
Gender | Male | 86 | 86% | ||
Female | 14 | 14% | |||
Age (years) | 20–39 | 42 | 42% | ||
40–59 | 40 | 40% | |||
>60 | 18 | 18% | |||
Residence area | Seoul | 63 | 63% | ||
Incheon | 8 | 8% | |||
Gyeonggi Province | 29 | 29% | |||
Machine-determined smoking topography |
CPDDevice | cig/day | 100 | 10.8±4.49 | 3–21 |
PCDevice | puff/cig | 100 | 20.4±8.39 | 1–44 | |
TSTDevice | s/cig | 100 | 179.4±69.6 | 30–385 | |
TSVDevice | ml/day | 100 | 14791±8238 | 2259–42620 | |
TTFCDevice | min | 89 | 79.3±84.0 | 1–420 | |
Self-report smoking habit survey | FTND Score | - | 100 | 3.13±2.37 | 0–9 |
Nicotine dependence evaluation | Very low (FTND Score=0–2) | 44 | 44% | ||
Low (FTND Score=3–4) | 27 | 27% | |||
Medium (FTND Score=5) | 11 | 11% | |||
High (FTND Score=6–7) | 14 | 14% | |||
Very High (FTND Score=8–10) | 4 | 4% | |||
TTFCFTND | ≤5 min | 22 | 22% | ||
6–30 min | 21 | 21% | |||
31–60 min | 27 | 27% | |||
>60 min | 30 | 30% | |||
CPDFTND | ≤10 | 50 | 50% | ||
11–20 | 42 | 42% | |||
20–30 | 6 | 6% | |||
>30 | 2 | 2% | |||
CPDWeekdays | cig/day | 100 | 10.3±6.76 | 0–40 | |
CPDWeekends | cig/day | 100 | 11.4±7.07 | 1–40 | |
CPDSurvey | cig/day | 100 | 10.6±6.56 | 1.43–38.6 | |
PCSurvey | 5–10 times | 36 | 36% | ||
11–14 times | 40 | 40% | |||
15–17 times | 20 | 20% | |||
18–20 times | 3 | 3% | |||
>20 times | 1 | 1% | |||
TSTSurvey | <30s | 0 | 0% | ||
30s–<1 min | 19 | 19% | |||
1 min–<1.5 min | 31 | 31% | |||
1.5 min–<2 min | 30 | 30% | |||
≥2 min | 20 | 20% | |||
Smoke inhalation depth | - | 100 | 6.89±1.48 | 3–10 |
For each participant, the 95% confidence interval of the data stored by the topography measurement device, except CPDDevice and TTFCDevice, is considered.
Results of Spearman’s correlation coefficient (ρ) for variables measured by survey and smoking topography device.
Measurement | Machine-Determined Smoking Topography | |||||
---|---|---|---|---|---|---|
| ||||||
CPDDevice | PCDevice | TSTDevice | TTFCDevice | TSVDevice | ||
0.571 |
−0.178 |
−0.090 | −0.276 |
0.312 | ||
0.029 | 0.048 | 0.079 | −0.071 | 0.007 | ||
0.065 | −0.047 | 0.061 | −0.050 | −0.048 | ||
0.465 |
−0.119 | −0.074 | −0.587 |
0.295 | ||
0.062 | 0.061 | 0.022 | 0.015 | 0.186 | ||
0.509 |
−0.130 | −0.081 | −0.477 |
0.322 |
p < 0.1;
p < 0.05;
p < 0.01;
p < 0.001.