Introduction

Worldwide, smoking and second-hand smoking (SHS) exposure prematurely kills more than seven million people annually. This is mostly attributed to cigarette and other combusted tobacco product use, according to the 2014 US Surgeon General Report [1]. Around 70% of premature deaths due to tobacco occur in developing countries, such as Egypt [2,3,4]. In 2003, in an effort to curb the tobacco epidemic, the World Health Organization (WHO) negotiated a treaty, the Framework Convention on Tobacco Control (FCTC), that focuses on supply and demand reduction strategies to combat the tobacco use epidemic [5]. Egypt signed the FCTC in 2003 and ratified it in 2005 [6]. In 2008, to scale up the implementation of tobacco control policy detailed in the articles of the WHO FCTC, the WHO introduced a comprehensive set of evidence-based measures called “MPOWER”: Monitor tobacco use and prevention policies (Article 20); Protect people from tobacco smoke (Article 8); Offer help to quit tobacco use (Article 14); Warn about the dangers of tobacco (Articles 11 and 12); Enforce bans on tobacco advertising, promotion and sponsorship (Article 13); and Raise taxes on tobacco (Article 6) [7,8,9].

Although Egypt has passed several laws and regulations to control the epidemic, such as banning smoking in the workplace and in public transportation [10], prohibiting tobacco advertising, promotion, and sponsorship [10], and requiring cigarette package pictorial warnings [11], it is one of the highest consumers of tobacco in the Middle East and North Africa region [12]. Previous reports indicated that the overall number of smokers in Egypt have grown on an average of 8% annually, with a declining age of initiation [11,12,13]. The extent to which these laws and regulations are being enforced is questionable [10, 14, 15]. The Global Youth Tobacco Survey (GYTS) offers a standardized monitoring tool to evaluate the progress of the global tobacco control efforts by monitoring youth tobacco smoking rates around the world, in addition to other indicators pertaining to various tobacco control measures and policies. The aim of the GYTS is to enhance the capacity of countries to monitor tobacco use among youth and to guide the implementation and evaluation of tobacco prevention and control programs, and policies [9].

In this study, we analyze GYTS data to evaluate the impact of demand-related articles of the WHO FCTC in addition to Article 16 concerning sales to minors as a supply reduction strategy. This study utilized indicators from the GYTS, grouped across the MPOWER measures, to evaluate changes, if any, before and after Egypt’s ratification to the WHO FCTC between 2001 and 2014. To date, there are no published evaluations pertaining to the tobacco control efforts in Egypt using the GYTS. We hypothesized that government implementation of the MPOWER strategies would result in less youth initiation and decrease in current smoking rates (i.e., past 30-day smoking) in Egypt.

Methods

Study Setting and Data Collection

This study utilizes cross-sectional secondary data (n = 15,255) of the GYTS waves 2001 (n = 3792), 2005 (n = 4196), 2009 (n = 4796), and 2014 (n = 2471) conducted in Egypt. Each wave was representative of the youth enrolled in the public school system. The overall response rate ranged from 77% in 2005 to 97% in 2009 [16,17,18,19]. The GYTS uses a common methodology for all waves. Public Egyptian schools (private schools were excluded) that had ≥ 40 students from 7th to 11th grade were eligible for sampling. The GYTS uses a two-stage cluster sample design (first for school level and then for classroom level). The probability of schools being selected is proportional to the enrollment size (i.e., large schools were more likely to be selected than small schools) [9]. Schools that did not agree to participate were not replaced. Prior to the assessment, the school director obtained a waiver of parental written consent (“passive consent”) by sending a letter to the parents explaining the purpose and contents of the GYTS and advised on opting out procedures. All students present on the date of the survey in the selected classroom were eligible to participate and were informed that participation was voluntary. Trained assistants, without the presence of the teacher in the classroom, conducted data collection in schools. The GYTS sample design produces representative, independent, cross-sectional estimates for each site [9, 17, 20].

A weighting factor was applied to each student record to adjust for non-responses and variation in the probability of selection at the school, class, and student levels. The questionnaire was self-completed in 30–40 min anonymously by the students. The GYTS uses a standardized core questionnaire with possibility for additional country or region-specific questions (e.g., hookah/waterpipe use in Egypt). The questionnaire was translated to Arabic with back translation into English to check the accuracy of questions. The GYTS core questionnaire gathers data on smoking prevalence, knowledge and attitudes against smoking, role of media and advertising, access to cigarettes, smoking-related school curriculum, SHS, cessation of smoking. Permission to conduct the study was obtained from the Egyptian Ministry of Education.

Measures

Table 1 presents an outline of the WHO FCTC articles, the corresponding MPOWER measures designed to support implementation of those articles, and the corresponding GYTS indicators that the authors extracted to evaluate these measures. Briefly, we utilized the appropriate measures for each FCTC article that were present in at least three of the four GYTS waves to enable the assessment of existing trends. For each FCTC article and its related MPOWER strategy, we assessed several indicators ranging from two indicators for “offering help to quit smoking” to six indicators for “monitoring tobacco use and prevention policies” (Table 1). All the questions either had the same wording and response categories in all waves, or response categories were collapsed in a similar fashion to yield comparable responses. The latter approach was used for five indicators in total. For the indicator of source of last cigarette purchased, responses were grouped into two response categories across all survey years depicting whether the respondent has (a) bought them on their own or (b) got them some other way irrespective of the source.

Table 1 WHO FCTC Articles, their supporting MPOWER measures, and primary GYTS indicators used in the study
Full size table

Other than the GYTS measures, we assessed cigarette affordability rather than absolute tax increase for the 4 years where the GYTS data were collected to better understand real-time affordability of cigarettes. The cigarette affordability measure was used to account for the affordability of a national (relatively cheaper) and more popular brand, Cleopatra, as well as another measure for the affordability of the most popular international (relatively more expensive) brand, Marlboro. The index was adapted from the CDC’s indicator guideline reference for the Global Adult Tobacco Survey [21]. We first obtained the prices of the two brands in local currency (Egyptian pound) and calculated the price in USD dollars where we adjusted for the exchange rates for different years. We also obtained the GDP per capita rates for Egypt for each of the 4 years corresponding to the GYTS waves from the World Bank portal [22], and calculated the indices accordingly. The index of cigarette affordability was the average price of 100 packs of manufactured cigarettes as a percentage of GDP per capita. The higher the percentage the lower the affordability for a given cigarette brand.

Data Analyses

For this study, GYTS data from waves of 2001, 2005, 2009, and 2014 were merged (n = 15,255) and analyzed in 2017. Changes of the prevalence of each indicator between each of the waves were the outcome of the analyses. In order to reflect the complex sample design of the GYTS survey, data were analyzed using the—survey—module of Stata, Version 12.0 (Statacorp, College Station, TX, USA). Weighted proportions were obtained as estimates of prevalence for each indicator used in our study (Table 1). The applied weights represent the total population of youth attending school in Egypt as per the information provided in the CDC’s manual for analysis of the GYTS survey. Weighted percentages and confidence intervals were used for descriptive analyses. The independent associations between the included indicators, sociodemographics (age and sex), and the survey year were assessed. To evaluate differences between waves among the indicators included in our study, we conducted logistic regressions, reporting adjusted odds ratios (aOR) for dichotomous outcomes and multinomial logistic regressions for categorical outcomes, reporting adjusted relative risk ratios (aRRR). This study used a significance level of α = 0.05. We also calculated an affordability index for a local cigarette brand and an international cigarette brand corresponding to the survey wave years, and we plotted the results while fitting polynomial trend lines that showed the best fit. We also presented the polynomial trend lines for ever cigarette use, and past 30-day cigarette use to contrast the findings of cigarette affordability and cigarette prevalence estimates (Fig. 1). Table 2 presents the descriptive results of the indicators included in our analysis by GYTS wave and Table 3 presents the results of the logistic regression analyses.

Fig. 1
figure 1

Cigarette affordability indices and youth cigarette smoking prevalence estimates in Egypt (2001–2014). The index of cigarette affordability is the average price of 100 packs of manufactured cigarettes as a percentage of gross domestic product (GDP) per capita. The higher the percentage, the lower the affordability for a given brand of cigarette (i.e., the trend of actual affordability is the reciprocal of the plotted trend line)

Full size image
Table 2 Weighted frequencies of selected demographics and tobacco measures from the Global Youth Tobacco Survey (GYTS) 2001–2014 (n = 15,255)
Full size table
Table 3 Weighted logistic regression to compare MPOWER tobacco measures by year for the Global Youth Tobacco Survey (GYTS), grouped by MPOWER strategies (n = 15,255)
Full size table

Results

Across the four GYTS waves, a total of 15, 255 students participated in the study of whom 8687 (52.1%) were males. The majority were in the 13-year-old age category (26.9%). Table 2 presents the characteristics and MPOWER indicators assessed in our sample for each wave.

Monitor Tobacco Use and Prevention Policies

The prevalence of ever smoking cigarettes showed an increase from 12.6% in 2001 to 22.9% in 2014 (Table 2), where youth had increased odds of ever smoking cigarettes in 2014 compared to 2001 (aOR = 2.27; 95% CI = 1.45–3.55, p < 0.001; Table 3). However, past 30-day cigarette smoking initially more than doubled in 2009 (9.6%) compared to 2001 (4.1%) and later declined in 2014 (5.1%). Overall, there was no significant increase in the past 30-day smoking rates among youth in Egypt in 2014 compared to 2001. Figure 1 presents the trend line for ever smoking and past 30-day smoking across the GYTS waves. Similarly, waterpipe smoking almost doubled in 2009 (11.2%) in comparison to 2001 (6.0%) and later declined in 2014 (6.3%), with no statistically significant difference between 2001 and 2014. Moreover, among ever smokers, initiating cigarette smoking was significantly associated with being from the age group of 10–11 years in 2014 compared to 2001 (aRRR = 3.17, 95% CI 1.41–7.12, p < 0.05), and initiating waterpipe smoking was significantly associated with being from the age group 8–9 years in 2014 compared to 2001 (aRRR = 5.13; 95% CI 1.1–23.8, p < 0.05) with 14 years or older as the reference age category.

Protect People from Tobacco Smoke

Concerning parental smoking behavior, there was no significant overall change between 2001 and 2009, and there were no data available for 2014. Overall, almost half of the youth reported that at least one of their parents was a current cigarette smoker (range 42 to 48%). Reported exposure to SHS at home was significantly higher in 2014, compared to 2001 (aOR = 1.51, 95% CI 1.13–2.03, p < 0.05). Youth had increased odds of exposure to SHS in public areas in 2014 compared to 2001 (aOR = 1.85, 95% CI 1.34–2.56, p < 0.001). Finally, youth in 2005 were significantly less likely to support smoke-free policies in comparison to youth in 2001 (aOR = 0.54, 95% CI o.34–0.85, p < 0.05), and there were no data available for this measure as well in 2014.

Offer Help to Quit Tobacco Use

Among the total youth who smoked from 2001 to 2014, more than half (61%) reported having made a quit attempt in the past year and 85% reported receiving help to quit. However, there were no significant differences in the likelihood of youth making quit attempts or receiving help to quit between 2001 and 2014.

Warnings about the Dangers of Tobacco

Youth reported significantly more exposure to school programs that educate them about the acute harms (e.g., teeth discoloration) of tobacco in 2014 compared to 2001 (aOR = 1.85, 95% CI 1.26–2.72, p < 0.05). However, there was no significant difference between 2014 and 2001 regarding school programs that educate students about the long-term dangers of smoking, although awareness of danger was significantly associated with both 2005 (aOR = 1.53; 95% CI 1.20–1.95; p < 0.05) and 2009 (aOR = 1.84; 95% CI 1.37–2.48; p < 0.001). Finally, youth had decreased odds of exposure to public anti-smoking campaigns in 2014 compared to 2001 (aOR = 0.56, 95% CI 0.36–0.87, p < 0.05).

Enforce Bans on Tobacco Advertising, Promotion, and Sponsorship

There was no significant decrease in self-reported billboard ad exposure from 2001 to 2009, but there was a significant decline in self-reported exposure magazine ads (aOR = 0.64; 95% CI 0.46–0.90, p < 0.05) (Table 3). However, 70% of youth still reported exposure to billboard ads in 2009 (Fig. 2). These indicators were not assessed in 2014; they were removed from the survey given that these types of advertising was already prohibited by the Egyptian government in 2002 and the survey needed to be shortened (personal communication with WHO EMRO office). Nevertheless, indicators assessing other prohibited industry promotional tactic were still included in 2014; there was a decrease in students reporting owning a personal item with a tobacco company logo between 2001 (25.5%) and 2014 (8%) (aOR = 0.43; 95% CI 0.28–0.67, p < 0.001), and being offered a free tobacco product by a sales person from a tobacco company decreased from 26% in 2001 to 7% in 2014 (aOR = 0.25; 95% CI 0.15–0.41, p < 0.001). Figure 2 depicts the four aforementioned indicators. These observed declines in the above-mentioned indicators were consistent, as depicted in Table 3. Moreover, being exposed to smoking in movies, although it significantly declined through 2014 in comparison to 2001 (Table 3), it remained quite high as it ranged from 86% in 2001 to 75% in 2014.

Fig. 2
figure 2

Exposure to tobacco advertising among youth across Global Youth Tobacco Survey (GYTS) waves in Egypt (2001–2014). In 2002, tobacco advertising, promotion, and sponsorship were completely prohibited (Article 6 of Law 85 for 2002). Questions regarding billboard ads and newspaper/magazine ads were not included in the 2014 wave

Full size image

Raise Taxes on Tobacco and Prohibiting Sales to Minors

Figure 1 shows the trends of change in the cigarette affordability index and its variability with the cigarette smoking prevalence estimates. Despite the substantial increase in prices between 2001 and 2014 for both, the local (approx. 1.25 to 7 Egyptian pounds) and international brands (approx. 4.5 to 17 Egyptian pounds), cigarette affordability did not considerably differ for the local brand between 2001 and 2014, with a slight increase for the international brand. Cigarette smoking prevalence estimates appear to have increased with relatively higher affordability for cigarettes in 2009 and decreased when affordability was relatively low in 2005 and 2014.

Among those who attempted to buy cigarettes, there were no significant differences in youth reporting being refused sales of cigarettes due to their age between 2001 and 2014. However, more youth reported that they bought cigarettes themselves in 2014 (85%) compared to youth in 2001 (60%), and these differences were statistically significant, where youth in 2014 had decreased odds of obtaining cigarettes some other way compared to youth in 2011 (aOR = 0.21, 95% CI 0.08–0.61, p < 0.05).

Discussion

We evaluated the MPOWER measure implementation in Egypt by assessing the GYTS data collected in four consecutive GYTS waves in 2001, 2004, 2009, and 2014. We hypothesized that effective implementation of the MPOWER strategies would likely prevent youth initiation and decrease current smoking rates (i.e., past 30-day smoking). We found that MPOWER strategies were not effectively implemented. Youth in 2014 were more likely to initiate smoking cigarettes and waterpipe at a younger age compared to 2001. Overall, there was no significant decline in past 30-day smoking rates among youth in Egypt in 2014 compared to 2001. Further, youth experimentation and past 30-day smoking rates fluctuated with affordability, with more adolescents initiating cigarette smoking, particularly at a younger age. While it is difficult to ascertain the reason behind this rapid fluctuation in smoking rates, it is plausible that cigarette smoking increased when cigarettes were more affordable (Fig. 1). However, since the experimentation and past 30-day use decline of cigarettes in 2014 compared to 2009 was not accompanied with similar decline in other indicators suggesting that youth might have been quitting, another plausible hypothesis is that there might have been an oversampling of smokers in the GYTS of 2009, or there might have been under-sampling of smokers in 2014, especially given that the sample size in the 2014 wave was considerably lower. Finally, the prohibition of tobacco advertising, promotion, and sponsorship in 2002 [23] was associated with a decrease of offering free tobacco product samples to youth; however, these prohibitions failed to be associated with a decrease in advertisement exposure (e.g., more than 60% of youth reported seeing ads in magazines and on billboards in 2009), and laws banning public smoking did not help decrease youth exposure to SHS in public places and at home.

The reported prevalence estimates recorded in the GYTS are comparable with those from the Survey of Young People in Egypt among the 15–17 years age group, where cigarette smoking rates were found to be 5.9% in 2001 [24], as compared to an average of 5% past 30-day smoking across four waves of the GYTS. Similar to our findings, other studies have demonstrated that cigarette smoking experimentation and current use (i.e., past 30-day use) among youth were correlated with the fluctuation of cigarette affordability. Youth are particularly sensitive to cigarette price changes [25, 26] and overall cigarette affordability [27]. Previous literature and our results suggest that policymakers in Egypt should account for the cigarette affordability indices, not just for absolute price changes, to guide the taxation scheme of cigarettes in the future.

Prior research demonstrates the declining age of smoking initiation in Egypt [12, 13, 28]. In support of this trend, our current findings show a significant increase in initiating both cigarettes and waterpipe among younger age groups [29•]. Currently, Egyptian tobacco control laws target cigarette smoking and do not often include other tobacco products, such as waterpipe. Waterpipe is perceived to be less harmful and could be attracting youth more than cigarettes [30,31,32]. Moreover, poly-tobacco is increasing among youth in many countries [29•, 33,34,35, 36••, 37•], including Egypt [36••, 38]. For example, GYTS findings demonstrated past 30-day dual cigarette and waterpipe use was 4.3% among youth in Egypt [36••]. Given the rapidly changing landscape of tobacco use [39], and that electronic cigarettes (e-cigarettes) are currently available in Egypt [40], future tobacco surveillance efforts need to consider assessing emerging tobacco and nicotine-containing products in more detail (e.g., waterpipe, e-cigarettes).

In 1994, the Egyptian government introduced an environmental protection law that banned smoking in public places and all modes of public transportation [23]. However, this study suggests a significant overall increase in the proportion of Egyptian youth exposed to SHS in public places as well as at home (p < 0.05). Youth exposed to SHS are more likely to initiate smoking [20]. The GYTS does not include the type of combustible product that youth report being exposed to at their homes. Recent research suggests that SHS resulting from waterpipe smoking in homes is more hazardous than SHS resulting from cigarettes in the home because waterpipe smoking emits substantially higher levels of carbon monoxide and leads to at least twice the air particulate matter (2.5) levels of cigarette smoking [41••]. In future GYTS waves, it might be beneficial to ascertain the type of tobacco that youth are being exposed to at home to inform future public health campaigns and interventions. Further, while laws banning public smoking exist, more vigorous enforcement of this regulation would benefit public health [42,43,44]. Perhaps enforcement of a smoke-free work environment would be a readily impactful action because, among Egyptian adults, those that worked in a smoke-free environment have been found to be significantly more likely to report also living in a smoke-free home [45••]. Enforcing smoke-free work places could benefit youth by increasing the likelihood of having smoke-free homes and decreasing SHS exposure.

Findings from this study indicated that a high proportion of the youth surveyed in the GYTS recall being exposed to school programs that raise awareness of the acute effects of smoking on health, such as teeth discoloration, as well as long-term effects on smoking such as increased cancer risk. However, these youth also reported being significantly less often exposed to anti-smoking media campaigns in 2014 in comparison to 2001, which is a an essential element to de-normalize smoking [46,47,48]. This might also be a factor for why there were significantly less youth in 2009 who favored smoke-free policies, despite having the majority of youth, in general, in favor of smoke-free polices [49]. However, whether existing programs in Egyptian schools are effective warrants further research, [50,51,52]. Additionally, authors were not able to find existing literature supporting that evidence-based youth prevention or cessation program has been put in place in Egypt.

A recent review suggested that youth are more likely to recall and think about advertising that includes personal testimonials; a surprising narrative; and intense images (e.g., pictorial warnings), sound, and editing [48, 53]. Moreover, youth recall pictorial warning more often than text-only warnings, where it sparks greater negative emotional reaction compared to text-only warning [53]. In 2007, a presidential decree mandated all cigarette manufacturers to include pictorial warnings on cigarette packs in compliance with the FCTC recommendations (e.g., 50% or more of and not less than 30% of the principal display areas, and to be pictorial) [8]. To date, no studies have evaluated the effectiveness of these warnings within the Egyptian context. Finally, the effectiveness of health warnings on waterpipe has shown similar effects to that of cigarettes [54••, 55,56,57]; however, to date there are no comparable warnings that are mandated for waterpipe tobacco.

This study had some limitations: the data was self-reported which may result in under- or over-reporting; however, using survey weights likely corrected for this limitation [58]. There is considerable variability in the questionnaires used, despite supervision from the CDC and the WHO. For example, questions on noticing tobacco advertising on billboards and magazines/newspapers were not included in the 2014 survey. While there is sufficient reasoning for removing these questions, as they were outlawed more than a decade before the 2014 wave, our findings indicate that banning advertising on billboards was yet not enforced till 2009, which was the last wave where this indicator was assessed. Specific to Egypt, school enrollment rates are 83% in the 12–14 years age group and 69% in the 15–17 years age group, decreasing the representativeness of the GYTS results. Additionally, the sample size of the 2014 data was considerably lower than previous years. These data are repeated cross sections, which could, in principle, reflect impacts by time or selection confounders. E-cigarette use is not yet included in the GYTS for Egypt. Exploring its use in subsequent waves could be beneficial to fully understand the tobacco and nicotine use behavior among youth in Egypt.

Conclusion

In Egypt, between 2001 and 2014, no significant observable gains were made in the battle against the tobacco epidemic. Despite Egypt’s ratification of the WHO FCTC, the indicators assessed in this study did not seem to have had significant impact on tobacco product use prevalence among youth wherein the overall tobacco product use did not decline over the four GYTS waves. While Egypt has enacted tobacco control regulatory policy, this study demonstrates their further enforcement could have a large impact of public health. In conclusion, developing novel approaches that can overcome the barriers to implementing the MPOWER recommendations in Egypt could enable researchers and policy makers in Egypt to form a comprehensive tobacco control program. Continuous monitoring and evaluation of the tobacco control activities in Egypt is needed.