International associations recommend monitoring, screening, and preventing tobacco smoking among children and adolescents (1). Chronic exposure to tobacco is associated with bacterial proliferation, reduced periodontal immunity, and head and neck cancer (2, 3). Electronic cigarettes also deliver nicotine similar to cigarette smoking (4) and were suggested to cause cancer by affecting cell proliferation, periodontal cell migration, cytotoxicity, oxidative stress, apoptosis, and DNA maturation (5, 6). In addition, smoking affects oral and pharyngeal tissues causing irritation, stimulation of microbial growth, reduced periodontal tissue immunity, enhanced alveolar bone resorption, and delayed healing (3, 7). However, research is needed to identify differences between the use of regular and electronic cigarettes regarding determinants and effects.

In order to understand cigarette use, causes and control, multiple complex behavioral interventions need to be addressed. The Capability, Opportunity and Motivation for Behavior change (COM-B) framework (8) offers a structured approach for planning evidence-based interventions that could explain behavioral changes associated with cigarette use and control (8–12). The COM-B framework suggests that people engage in a behavior if they have the Capability (physical and psychological), the Opportunity, when the surrounding environment is physically and socially convenient and the Motivation to adopt this behavior. Within this framework, we conceptualized that the capability factors that may be associated with smoking during the study period include anxiety (13–15) and stresses caused by the COVID-19 pandemic (16) which are assumed to increase the risk of cigarettes' use (17, 18). Also, health problems such as COVID-19 infection may reduce the ability to smoke. Smoking is assumed to decrease in communities where there are legislative controls such as reducing access to cigarettes which decrease the opportunities to engage in smoking (18). The global variation in success of global strategies to control tobacco (18) emphasize the need for global evidence to inform tobacco control programs. Multi-country studies can shed light on person- and country-level factors associated with health problems. At the same time, assessing individual factors such as attitudes toward smoking can guide the development of tobacco cessation counseling activities.

The aim of this study was to assess the capability, opportunity and motivation factors associated with using regular and electronic cigarettes among adolescents and young adults (AYA) in several countries. The hypothesis of the study was that use of cigarettes was associated with capability, opportunity or motivation.

Complete data were available for 6,989 participants from 25 countries (Supplementary Table 1). Table 1 shows that most participants were 18–23 years old (74.4%), females (57.3%), with a university/college degree (56.8%). Most participants' parents had university degrees (42.6% of mothers and 53.3% of fathers). Also, 529 (7.6%) participants were infected with COVID-19, 712 (10.2%) currently used regular cigarettes and 395 (5.7%) currently used electronic cigarettes.

Also, 2,964 (42.4%) of the participants had normal anxiety and 504 (7.2%) had severe anxiety. The median scores for attitude toward smoking scale = 3 out of 5, for banning promotion on television = 2 out of 2, and for smoke free zones = 3 out of 4. The median affordability score was 3.8%.

Table 2 shows that daily use was reported by 588 (82.6%) of regular cigarettes users and 213 (53.9%) electronic cigarettes users. Also, 363 (51.0%) and 174 (44.1%) participants reported the same frequency of using regular and electronic cigarettes as before the pandemic, 201 (28.2%) and 123 (31.1%) reported less frequent use during the pandemic; and 300 (42.1%) and 177 (44.8%) participants reported increased use of regular and electronic cigarettes at night.

Table 3 shows the factors associated with the current use of regular and electronic cigarettes in 25 countries. Participants who were 18–23-years-old and 15–17-years-old had significantly higher odds of using regular (AOR = 5.70, P < 0.001 and AOR = 2.28, P =0.005) and electronic cigarettes (AOR = 2.88, P < 0.001 and AOR = 2.25, P =0.002) than respondents who were 11–14-years-old. Male respondents had significantly higher odds of using regular (AOR = 5.96, P < 0.001) and electronic (AOR = 2.31, P < 0.001) cigarettes than females. Those who got infected with COVID-19 reported significantly higher odds of using electronic cigarettes (AOR = 1.81, P = 0.02). Having normal or mild levels of general anxiety were associated with significantly lower odds of using regular (AOR = 0.34 and 0.52, P < 0.001) and electronic (AOR = 0.28 and 0.45, P < 0.001) cigarettes. A higher score of negative attitudes toward smoking was associated with significantly lower odds of regular and electronic cigarettes use (AOR = 0.75 and 0.78, P < 0.001). A higher affordability score was associated with significantly lower odds of using electronic cigarettes (AOR = 0.90, P = 0.004). The fixed effect variables explained 23.9% of the random variation among countries in using regular cigarettes and 32.1% in using electronic cigarettes.

Variance estimate of unconditional model for regular cigarettes = 0.67 and for electronic cigarettes = 1.217, full model variance estimate = 0.51 and 0.826, respectively. Regarding goodness of fit, the deviance decreased from 40818.4 in the unconditional to 37424.7 in the conditional model where using regular cigarettes was the dependent variable and from 44525.3 to 42732.0 where using electronic cigarettes was the dependent variable with significant differences in both cases indicating improvement in model fit after adding the fixed effect variables.

The study showed that most current users of regular and electronic cigarettes reported either the same or lower level of use as before the pandemic. Older participants and males had higher odds of using regular and electronic cigarettes. Those with normal or mild anxiety and with negative attitude toward smoking had lower odds of reporting smoking. Participants with a history of COVID-19 infection and those from countries where electronic cigarettes were more affordable had higher odds of using electronic cigarettes. The components of COM-B were, thus, associated with the use of regular and electronic cigarettes.

One of the strengths of this study is the evidence generated on regular and electronic cigarettes use among AYA in several countries. Previous studies on using regular and electronic cigarettes during the pandemic had minor focus on AYA (30–32) or were restricted to only one country (16, 33–36) making it difficult to compare countries with different tobacco control policies. The use of the COM-B model as a conceptual framework for the study allowed the identification of individual and country-level factors that may be associated with smoking.

Despite the unique perspective offered by the present study, there were some limitations. First, there may be a social desirability bias in reporting smoking with potential for underestimating the level of smoking regular and electronic cigarettes. Second, the survey did not distinguish between the use of nicotine-containing and non-nicotine containing electronic cigarettes and this needs to be addressed in future studies. Third, because of COVID-19, data were collected electronically. The convenience sample may, therefore, not be representative of the profile of AYA in the countries included in the study as participants' recruitment was skewed toward those with access to the internet. Fourth, the study is cross-sectional and cannot prove causality; it can only suggest associations. Fifth, the attitude score was based on five items and this may possibly explain the relatively low level of internal consistency indicated by the Cronbach alpha (37) which might have also been reduced by the presence of more than dimension in the items assessing general attitude toward smoking and attitude toward the association of smoking and COVID-19 infection. Future studies assessing attitudes toward smoking may need to include a greater number of items to increase consistency. Despite the study limitations, there were important highlights from the study.

First, we found that a history of COVID-19 infection was associated with higher odds of using electronic cigarettes but not using regular cigarettes. This agrees with Gaiha et al. (16) who reported a significant association between smoking electronic cigarettes and history of COVID-19 infection. A recent meta-analysis of patients hospitalized for COVID-19 found no significant association between COVID-19 infection and smoking (38). However, this disagrees with some population-based studies (39). One possible explanation may be that AYA with a history of COVD-19 infection may opt to use electronic than regular cigarettes with the understanding that it would have less effect on their health. There are few studies that highlighted the interaction between smoking and COVID-19 (17). However, the association between electronic cigarette smoking and COVID-19 infection especially in AYA needs to be explored further. Electronic cigarettes smoking is associated with elevated chemokine CXCL8, extracellular matrix proteins and markers of mitochondrial dysfunction which decreases cellular viability and integrity. It also interferes with cellular mechanisms resulting in increased oxidative stress, inflammation, infections and airway remodeling in the lungs of device users (3, 38, 40–43). What is not known is whether this damage is worse for electronic cigarette smokers than it is for regular cigarette smokers; or if the damage is worse in AYA who smoke electronic cigarettes than it is for older smokers. These knowledge gaps need to be explored further. The study finding suggests that AYA who smoke electronic cigarettes should be counseled on the risk of contracting infection when sharing vaping devices (44).

Second, participants with severe anxiety had higher odds of using regular and electronic cigarettes. This agrees with studies reporting higher frequency of smoking to cope with stresses caused by the COVID-19 pandemic and the associated precautionary measures (45–47). These studies, however, mainly focused on adults and elderlies. We highlight that AYA equally need attention for psychological stress management even if they are less likely to have COVID-19 related deaths. Support provided to AYA should focus on helping them cope and maximize their resilience against surrounding stressors.

Third, less affordability of electronic cigarettes was associated with lower odds of smoking. Affordability is relevant for AYA because they may be less financially independent than adults (48). The rapid increase in electronic cigarette smoking among adolescence was associated with its affordability and availability (49–51). Further taxation of cigarettes during stressful events may be suggested to help reduce the risk of AYA using cigarettes to cope with stresses. However, it is important to note that capability and motivation factors may have stronger association with using electronic cigarettes than opportunity factors based on the present findings.

Fourth, banning tobacco promotion on television seemed to have a strong but non-significant negative association with using electronic cigarettes in the present study. This may be attributed to the impact of role modeling (52) on AYA.

The findings have implication for controlling tobacco smoking in AYA. At the individual level, educating AYA to manage stress without resorting to smoking, and promoting a negative attitude toward smoking may help control smoking. At a country level, it is important to ensure greater financial obstacles so that AYA are less able to afford tobacco products and to reduce the promotion of tobacco through advertisements in addition to exploring other policies. Adopting the MPOWER policies can help control smoking during the pandemic. The MPOWER package includes monitoring tobacco use and prevention policies (M), protecting populations from tobacco (P), offering help to quit tobacco (O), warning about the risks associated with tobacco (W), enforcing bans on tobacco promotion (E) and raising taxes on tobacco products (R). The package was created by the World Health Organization to help governments address the tobacco pandemic based on the framework convention for tobacco control (18). Enforcing the components of MPOWER may help reshape the surrounding environment, thus reducing the opportunities to engage in smoking as a behavior. The evidence generated in the present study supports the adoption of the MPOWER strategies to control smoking associated with stressful events. Further studies are needed to assess the impact of other MPOWER strategies on smoking, and smoking prevalence in AYA compared to other age groups.

The higher odds for AYA to smoke in association with anxiety highlight the need for action to reduce the capability, opportunity and motivation of AYA to smoke during stressful events. The adoption of the MPOWER strategies may provide the framework for governments to address the structural determinants of smoking thereby supporting global tobacco control efforts.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Research Ethics Committee of the Faculty of Dentistry, King Abdulaziz University, Jeddah Saudi Arabia. Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin.

HS, WA, MQ, and ME: conceptualization. MQ and ME: formal analysis. HS, MQ, NAA, JA, NQ, SQ, AH, NM, RO, NA-K, RJ, AB, MS, MF, OA, NG, RA, NAM, HE, MA, BS, RM, NA, SS, SA, and ME: investigation. HS, MQ, and ME: methodology. HS, WA, and ME: project administration and writing—original draft. HS, WA, NA-K, MF, and ME: writing—review & editing. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.