INTRODUCTION
Tobacco use is a major risk factor for various non-communicable diseases, including cardiovascular diseases and cancer. It is responsible for about 8 million deaths globally each year, with the majority of these deaths occurring in low- and middle-income countries1. Although global smoking rates have decreased since 1990, smoking prevalence remains high in many countries. If past trends continue, it is projected that by 2050, smoking will cause 12.2 million deaths, leading to a significant health burden in the coming decades2.
The Philippines has long been known for having one of the most robust tobacco industries in Asia, wielding significant influence on public health policies3,4. According to the 2021 Global Adult Tobacco Survey (GATS) Philippines, approximately 18.5% of Filipino adults are current tobacco users, representing 14.4 million individuals5. Additionally, approximately 112000 Filipinos lose their lives each year due to diseases linked to tobacco use6.
In response to the global tobacco epidemic, the World Health Organization Framework Convention on Tobacco Control (WHO FCTC) was established to mitigate the harmful effects of tobacco consumption and exposure. The FCTC was ratified in the Philippines in 20054. To operationalize the FCTC, the WHO introduced the MPOWER strategy, which provides a practical framework for reducing tobacco use through six key measures: Monitoring tobacco use; Protecting people from tobacco smoke; Offering help to quit; Warning about the dangers of tobacco; Enforcing bans on tobacco advertising; and Raising tobacco taxes7.
A crucial aspect of tobacco control initiatives is smokers’ attempts to stop smoking8. However, research specifically examining the predictors of quit attempts among Filipino smokers is limited, if not absent. Addressing this gap can provide insights that may be useful for understanding smoking cessation behaviors in this population. Hence, the current research aims to examine the sociodemographic and smoking-related factors associated with quit attempts among Filipino daily smokers aged ≥15 years by using national survey data.
METHODS
Data source
This study implemented a secondary data analysis of the 2021 Global Adult Tobacco Survey (GATS) of the Philippines, a nationally representative household survey aimed at monitoring tobacco-related indicators such as tobacco use, exposure, and quitting attempts, among others. The 2021 GATS Philippines adopted the GATS global standardized methodology and was implemented by the Philippine Statistics Authority (PSA) in coordination with the Philippines’ Department of Health (DOH). The full methodology of the 2021 GATS Philippines is available on the second chapter of the 2021 GATS final report5. The data file for the 2021 GATS Philippines is publicly posted at the PSA website9. This study was assessed by the University of the Philippines Manila Research Ethics Board on 19 August 2024 and confirmed that the study appears to be similar to studies that are exempted from ethics review.
Participants
A total of 20971 households were sampled in the 2021 GATS Philippines via a multi-stage, geographically clustered sampling design. The survey collected data on persons aged ≥15 years from sample households. For each household, one individual aged ≥15 years was randomly chosen to respond to the survey questions. There were 18466 completed interviews. The current study covers current daily smokers. These are the individuals who reported they smoked daily through the question: ‘Do you currently smoke tobacco on a daily basis, less than daily, or not at all?’.
Variables
Supplementary file Table 1 lists the variables in this study, the questions that correspond to them (verbatim from the 2021 GATS), as well as the variable coding for the present research. The outcome variable of this research was an attempt to quit smoking in the past 12 months (categorical). The question ‘During the past 12 months, have you tried to stop smoking?’ was used to assess a quit attempt with ‘yes’ or ‘no’ as response categories. Predictor variables included demographic information, i.e. age, sex, education level, and employment status. Smoking-related variables were also included, such as smoking intensity (cigarettes/day), age at smoking initiation, minutes to smoking on waking up, smoking rules at home, awareness of the health harms of tobacco, and frequency of daily exposure to anti-smoking messages. Visit to health facilities was also included in the list of independent variables.
Statistical analysis
Analyses were limited to the 2432 daily smokers with complete information, as the 178 cases with refusals or don’t know responses were removed from the final dataset. Hence, only cases with complete information were analyzed. Sampling weights provided in the GATS dataset were applied in the study to account for the survey design. A significant level was set at 0.05. SAS® Studio was used to analyze data. Descriptive statistics were performed to summarize data. Specifically, means and standard deviations were computed for numeric variables, while frequency distributions were performed on categorical variables. Survey-weighted t-test via PROC SURVEYMEANS was performed to identify differences between those who attempted and did not attempt to quit smoking in terms of numeric variables. The Rao-Scott chi-squared test via PROC SURVEYFREQ was used to identify the association of quit attempts with categorical variables. Logistic regression via PROC SURVEYLOGISTIC was used to identify factors significantly related to quit attempts while controlling for other variables. Sociodemographic characteristics and smoking-related behaviors were included simultaneously in the model as independent variables to examine their associations with the outcome.
RESULTS
Characteristics of respondents
Table 1 summarizes the profile of the respondents. The majority of daily smokers were males (88.65%). The mean age was about 42 years (SD=14.49). Nearly 7 out of 10 daily smokers finished elementary level or higher (69.86%), and the majority of them were employed (84.95%). Less than one-fifth reported visiting a doctor or health provider in the past 12 months (18.05%). Among the daily smokers, 39.93% reported trying to quit smoking in the past 12 months prior to the survey. The percentages of daily smokers who usually have their first smoke >30 minutes after waking up, and ≤30 minutes were nearly equal (49.14% and 50.86%, respectively). Nearly 3 out of 10 (28.99%) of daily smokers reported that their homes allow smoking. The majority (93.09%) also believed that tobacco smoking can cause serious illness. The average number of manufactured cigarettes consumed daily was almost nine (mean=8.80; SD=7.42), while the average age of initiating daily smoking was 21 years (SD=6.52).
Table 1
Descriptive profile of responders, based on the 2021 GATS Philippines, a cross-sectional survey on adult tobacco daily smokers (N=2432)
Table 2 presents the results of chi-squared tests to identify which of the qualitative variables are associated with smoking quit attempts in the reference period. Table 3 outlines the results from the t-tests. Chi-squared tests indicate that smoking rules at home was the sole variable significantly associated with previous quit attempts [χ2 (1, N=2432)=9.66, p=002] with a higher proportion of individuals who attempted to quit coming from households where smoking was not allowed or where no rules were in place (79.56%) compared to those where smoking was allowed (20.44%). For the t-tests, there was a significant difference in the mean ages of initiating daily smoking between the two groups [t(1052)= -2.46, p=0.046]. Also, the mean number of daily exposures to anti-smoking messages was also significantly different, [t(1052)= -2.00, p<0.014].
Table 2
Weighted percentages for quit attempt status based on the 2021 GATS Philippines, a cross-sectional survey on adult tobacco daily smokers (N=2432)
| Variables | Attempt to quit smoking Unweighted n, (weighted column %) | p* | |
|---|---|---|---|
| Did not attempt | Attempted | ||
| Total, n | 1461 | 971 | |
| Sex | 0.463 | ||
| Male | 1305 (90.61) | 851 (92.14) | |
| Female | 156 (9.39) | 120 (7.86) | |
| Education level | 0.885 | ||
| Elementary level or lower | 471 (30.34) | 262 (30.93) | |
| Higher than elementary level | 990 (69.66) | 709 (69.07) | |
| Employment | 0.623 | ||
| Employed | 1252 (87.86) | 814 (86.63) | |
| Not employed | 209 (12.14) | 157 (13.37) | |
| Visited a doctor/healthcare provider | 0.060 | ||
| Yes | 210 (16.65) | 229 (23.97) | |
| No | 1251 (83.35) | 742 (76.03) | |
| Minutes to smoking after waking up | 0.815 | ||
| ≤30 | 768 (46.63) | 469 (47.74) | |
| >30 | 693 (53.37) | 502 (52.26) | |
| Smoking rules at home | 0.002 | ||
| Smoking allowed | 464 (33.16) | 241 (20.44) | |
| Not allowed/no rules | 997 (66.85) | 730 (79.56) | |
| Tobacco smoking can cause serious illness | 0.276 | ||
| Yes | 1364 (94.41) | 900 (92.03) | |
| No | 97 (5.59) | 71 (7.97) | |
Table 3
Weighted means between categories of quit attempts based on the 2021 GATS, a cross-sectional survey on adult tobacco daily smokers (N=2432)
| Variables | Weighted mean | p* | |
|---|---|---|---|
| Did not attempt | Attempted | ||
| Smoking intensity (cigarettes/day) | 9.38 | 7.97 | 0.058 |
| Frequency of daily exposure to anti-smoking messages | 1.35 | 1.72 | 0.014 |
| Age of daily smoking initiation (years) | 20.54 | 21.91 | 0.046 |
| Age (years) | 42.13 | 40.40 | 0.095 |
Predictors of quit attempt
Table 4 presents the weighted logistic regression results, which outline the factors predictive of smokers’ attempts to quit smoking over the past 12 months. Particularly, factors associated with higher odds of attempting to quit smoking among daily smokers in the past 12 months were visiting a
Table 4
Weighted and adjusted logistic regression coefficients based on the 2021 GATS Philippines, a cross-sectional survey on adult tobacco daily smokers, after eliminating cases with refused and don’t know responses (N=2432)
| Variables | OR | SE | 95% CI | p* | |
|---|---|---|---|---|---|
| Lower | Upper | ||||
| Male (ref. female) | 1.55 | 0.30 | 0.85 | 2.82 | 0.149 |
| Tobacco smoking can cause serious illness (ref. no) | 0.60 | 0.35 | 0.30 | 1.19 | 0.145 |
| Visited a doctor or healthcare provider (ref. no) | 1.72 | 0.25 | 1.05 | 2.82 | 0.030 |
| Smoking not allowed/no rules at home (ref. allowed) | 1.58 | 0.11 | 1.04 | 2.38 | 0.031 |
| Within 30 minutes to smoking after waking up (ref. >30 minutes) | 1.12 | 0.20 | 0.77 | 1.64 | 0.551 |
| Employed (ref. not employed) | 0.84 | 0.23 | 0.54 | 1.32 | 0.451 |
| Education higher than elementary level (ref. elementary or lower) | 0.72 | 0.22 | 0.46 | 1.12 | 0.143 |
| Smoking intensity | 0.98 | 0.02 | 0.95 | 1.01 | 0.232 |
| Frequency of daily exposure to anti-smoking messages | 1.14 | 0.06 | 1.01 | 1.28 | 0.029 |
| Age of daily smoking initiation | 1.05 | 0.02 | 1.01 | 1.08 | 0.006 |
| Age | 0.99 | 0.01 | 0.97 | 1.00 | 0.030 |
doctor or healthcare provider (OR=1.72; 95% CI: 1.05–2.82; p=0.030) versus not visiting, smoking being not allowed at home or having no rules in smoking at home (OR=1.26; 95% CI: 1.04–2.38; p=0.031) versus smoking allowed at home, higher frequency of daily exposure to anti-smoking messages (OR=1.14; 95% CI: 1.01–1.28; p=0.029) and older age at smoking initiation (OR=1.05; 95% CI: 1.01–1.08; p=0.006). Meanwhile, older smokers were slightly less likely to make quit attempts compared to younger smokers (OR=0.99; 95% CI: 0.97–0.99; p=0.029).
DISCUSSION
The present research aims to identify factors associated with attempts to quit smoking among daily smokers in the Philippines. Identifying these variables is crucial because it provides evidence on the factors that facilitate or hinder smoking cessation. To the author’s best knowledge, the present study is the first nationally representative study in the Philippines to aim to identify factors associated with attempts to quit smoking.
Visiting a doctor or healthcare provider was identified as a significant predictor of quitting attempts. This finding aligns with similar studies in Asia that used GATS data10,11 which highlights the crucial role healthcare providers play in promoting smoking cessation11,12. Although the 2021 GATS Philippines questionnaire included a question on whether respondents were advised to quit smoking by healthcare providers they visited, this was not included in the current study due to the small sample size of respondents eligible to answer the question. Furthermore, this study does not capture the intent of daily smokers when they had healthcare visits, nor the specific advice given by healthcare providers they visited. Nonetheless, it is reasonable to suppose that healthcare facilities often offer support and guidance on smoking cessation, which can motivate smokers to quit. Additionally, visiting healthcare providers may reflect a smoker’s general health awareness and proactive behavior10.
The study also found that allowing smoking at home is associated with a lower likelihood of quit attempts. This is consistent with a study among Bangladesh adults using GATS data13. The present study is consistent with a review exploring the connection between smoke-free homes and smoking behaviors among adults, which revealed that residing in smoke-free environments is consistently related to a higher probability of quit attempts and prolonged restraint14. Banning smoking at home is associated with behavioral changes that reduce smoking. For adolescents and young adults, such policies represent a socialization with anti-smoking norms15.
Anti-smoking messages were argued to be designed to emphasize the dangers of engaging in smoking and urge smoking cessation16. Previous studies using GATS data in Asian countries indicate a positive correlation between contact with anti-smoking messages and quitting attempts17-19. This study supports these findings, showing that increased contact with anti-smoking information is associated with a greater probability of a quit attempt. A randomized trial explored why image warnings on cigarette packs enhance quit attempts and found that such warnings provoke quit attempts by eliciting negative reactions and maintaining a strong anti-smoking message in smokers’ minds20.
The study indicates that initiating smoking at older ages is associated with higher odds of quit attempts. This study treated age at smoking initiation as a numeric variable rather than categorizing it. There were previous studies that studied the age of smoking initiation, but age groups were used. Transforming continuous data into categories can result in a loss of power21. The result of this study on the influence of age at daily smoking commencement on quit attempts is in congruence with a research study on adult men and women from the US22. In a study among university students in India, it was found that starting tobacco use at an older age is associated with greater motivation to quit23. The trajectories of age at smoking initiation can also be predicted. In a cohort study covering 55 years of the life course in Great Britain, it was revealed that early initiation of smoking corresponds to delayed quitting, relapse, and being a persistent smoker24. These findings suggest that the age at smoking initiation has profound effects on subsequent smoking-related behavioral outcomes.
Finally, results indicate that increasing age is associated with lower odds of attempting to quit smoking, which suggests that younger individuals are more likely to initiate quit attempts. Comparing this finding with previous studies that also used GATS data, the present study’s result in terms of age is different from a study in Vietnam, where older smokers (aged ≥55 years) were significantly more likely to attempt quitting compared to younger smokers (aged ≤24 years)25. However, the present results are consistent with studies in India and China, which also found that younger age groups were more likely to report quit attempts26,27.
Limitations
This study offers valuable evidence, but its limitations must be acknowledged. First, the reliance on secondary data constrained the study to the variables and their measures from the standardized methodology of GATS. For instance, the study could not consider motivational factors identified as a key predictor of quit attempt in a systematic review28. Although multiple sociodemographic and smoking-related variables were included in the regression model, residual confounding from unmeasured or imperfectly measured factors may still influence the observed associations. Also, the current research does not cover electronic cigarettes. Second, the cross-sectional design limits the ability to analyze trends in quitting attempts and draw causal inferences. Third, given the sensitivity of smoking behavior, self-reported data from the respondents may introduce social desirability bias, and participants’ recollection of past behaviors could introduce recall bias. It is also important to note that cases with ‘Don’t know’ or ‘Refused’ answers were excluded. Although few and likely missing completely at random, their exclusion could introduce selection bias. In terms of external validity, it should be noted that the findings are about daily smokers. The limitations highlight the need for careful understanding of the findings and underscore the importance of more comprehensive research on smoking cessation attempts in the future.
CONCLUSIONS
This study identified the significant predictors of quit attempts among Filipino aged ≥15 years using the 2021 Philippines GATS data. These are visits to a doctor or healthcare provider, home smoking rules, age, frequency of daily exposure to anti-smoking messages, and age of initiating daily smoking, mostly smoking-related behaviors as well. Future studies are warranted to further examine these associations, particularly through research designs that allow causal inference, as the present study analyzes cross-sectional data.
