Article Text
PDF
Abstract
Background Public attitude is a political driver in successful implementation of tobacco control policies. We assessed support for a range of tobacco control policies among smokers in Pakistan.
Methods We conducted a household survey among adult smokers in 10 cities of Pakistan, using a two-stage random sampling strategy to select households and Kish grid method to select one smoker per household. Attitudes were measured using a five-point ordinal scale on four policy statements: a complete ban on tobacco sale within 10 years; raising the legal age to buy tobacco from 18 to 21; increasing tobacco taxes to fund healthcare and a ban on smoking in cars with minors.
Results 6014 participants were interviewed between September 2019 and March 2020. Most participants demonstrated strong support for all policy statements: a ban on smoking in cars with minors (86.5%); a complete ban on tobacco sale within 10 years (82.1%); raising the legal age to buy tobacco (77.9%) and increasing tobacco taxes (68.1%). Smokers’ support for tobacco control policies increased with age but decreased with higher educational attainment and heaviness of smoking.
Conclusions There is strong support among smokers in Pakistan to strengthen tobacco control. Given this, policy-makers should strongly consider strengthening existing national policies on tobacco control.
- low/middle income country
- public policy
- public opinion
- surveillance and monitoring
- health services
Statistics from Altmetric.com
Background
For governments, public acceptability of measures to change health-related behaviours is important.1 Without evidence of public support, governments hesitate to impose such measures, fearing consumer backlash, accusations of being a nanny state and the cost of enforcement and subsequently losing authority.2 Such fears are frequently amplified by the industry for example, in case of tobacco control.3 4 Therefore, independent evidence of public support becomes even more important for policy change, as politicians often tend to underestimate the level of public support for public health measures.5
There is now strong evidence, although mostly from high-income countries (HICs), that the public supports policy measures to reduce consumption of tobacco.1 Emerging evidence from low-income and middle-income countries (LMICs) also suggests high level of public support for smoke-free workplaces, hospitality venues, schools and hospitals6–8 and for increasing tobacco taxes.9 For tobacco control policies, support from non-smokers as well as smokers has catalysed successful policy implementation in many HICs.10–13 A possible explanation for support among smokers particularly, comes from the theory of self control. That is, smokers willing to stop smoking may support smoking restrictions because they think they can help them to quit.14 For example, support for smoke-free laws among smokers is associated with their previous quit attempts and predicts future quitting.14 Similarly, tobacco tax increase is likely to have support among smokers, if the additional revenue is spent on providing cessation services.15 16 In contrast, evidence for smokers’ attitudes towards measures that protect young people from tobacco exposure17 and some of the proposed end-game strategies such as future tobacco ban18 19 is lacking. Only a few studies have measured support for a range of policies in the same cohort of smokers.20 21 Pakistan is a low resource, high-tobacco burden country. According to the 2014 Global Adult Tobacco Survey, 19.1% (23.9 million) adults (31.8% males, 5.8% females) were current tobacco users; of these, 12.4% smoked tobacco (22.2% males and 2.1% females).22 Tobacco use was unequally distributed across socioeconomic status; the smoking prevalence was more than twice in those with education up to primary school or below than those up to high school and above.23 In addition to 160 000 tobacco-related deaths every year, Pakistan suffers an annual/yearly economic loss of 198 billion rupees (US$1.3 billion) due to tobacco-related mortality and morbidity.24 In recent years, Pakistan has taken some positive steps towards tobacco control, for example, introducing bans on tobacco advertisement, promotion and sponsorship as well as bans on display of products at points-of-sale, yet many other evidence-based tobacco control policies such as tobacco taxes have not been implemented in a comprehensive manner.22 We investigated the level of support among smokers for a range of policy measures, with an aim to inform and strengthen tobacco control policies in Pakistan.
Methods
A cross-sectional household survey (Studying Tobacco users of Pakistan, STOP) was conducted in the 10 most populous cities of Pakistan representing 20% of the total population and 16% of all smokers in Pakistan. Participants were regular smokers aged 15 years or above.
Using an average of 20 households per primary sampling unit (PSU), a sample size of 6313 households and 316 PSUs was estimated. This was based on assuming a smoking prevalence of 10% in urban areas and a design effect equal to 2.22 Furthermore, the population at risk and the average household size were assumed to be 64% and 6.2, respectively.25 The response rate and a margin of error were set to 95% and 0.055, respectively.
Households were identified using a two-stage random sampling. At the first stage, we used stratified random sampling to select Union Councils (PSUs) from the 10 cities proportional to their population size. We identified eligible households (having at least one regular smoker aged 15+) by screening every household (range 200–400) within each PSUs and from these, randomly selected 20 households in the second stage. One participant per household was identified using Kish grid method.26 All eligible households were offered written study information and selected participants consented prior to their recruitment.
Our field investigators (11 females and 36 males; age range from 20 to 40) received 3-day training to conduct the survey including in-field supervision. The training also included how to ask smoking-related questions from women in a culturally sensitive manner. The survey methods were also piloted in two cities. Data were collected by field investigators in real time using handheld digital tablets. Attitudes towards tobacco control policies were gauged by responses to four statements adapted from the German National household survey on smoking behaviour and cessation (DEutsche Befragung zum RAuchverhalten -DEBRA):20 (1) The sale of cigarettes and tobacco in Pakistan should be banned completely within the next 10 years; (2) The legal age of sale of cigarettes and tobacco in Pakistan should be raised from 18 to 21; (3) Tobacco industry sales should be taxed in order to use the money to address problems caused by tobacco (eg, health issues) and (4) When minor children are in a private car, smoking inside the car should be banned and subjected to punishment. For each statement, participants indicated their level of support by choosing one of the following five options: ‘strongly support’, ‘tend to support’, ‘no opinion either way’, ‘tend to oppose’ and ‘strongly oppose’. Demographic variables included participants’ age, sex and highest level of education attained. Smoking-related variables included the Heaviness of Smoking Index 27 and the strengths of urges to smoke.27
The level of support for each of the four policy statements was assessed by calculating the proportion of participants in each response category (strongly support, tend to support, no opinion, tend to oppose, strongly oppose). We further assessed whether participants’ responses (support/no opinion/oppose) were associated with their age, educational status or heaviness of smoking. All responses were weighted to account for the survey design. We used the χ2 test to test for associations, or the Fisher’s exact test where expected cell values were less than 5. P values<0.05 were considered statistically significant. All statistical analyses were conducted on STATA V.16.28
Results
The STOP survey was conducted between September 2019 and March 2020. We approached 97 345 households and 12 127 were found eligible. We randomly selected 7225 smokers (1 per eligible household) and out of these 6014 (83.3%) participated in the survey; rest were either unavailable (8.2%) or refused (8.6%). The respondents were predominantly male (98.5%). Response rates to the four policy statements were high, ranging between 98.7% (complete ban on tobacco in 10 years) and 97.5% (increase taxes to pay for healthcare). The proportion of participants who strongly support all four policy statements was high: a ban on smoking in cars with minors (86.5%); a complete ban on tobacco in the next 10 years (82.1%); increasing legal age of cigarette sales to 21 (77.9%) and increasing tobacco taxes (68.1%) (figure 1). For all but one policy statement (ie, increasing tobacco taxes) support was higher among participants aged >35 years. An inverse association between participants’ support and their educational status was observed across all policy statements—the inverse association was also observed with participants’ heaviness of smoking across three of the four policy statements (raise legal age of cigarettes, increase tobacco taxes, ban on smoking in cars with minors) (table 1).
STOP survey participants’ responses (weighted %) to the four policy statements. STOP, Studying Tobacco users of Pakistan.
Association between STOP survey participants’ responses to policy statements and their sociodemographic and smoking-related characteristics
Discussion
Our study suggests that most smokers in Pakistan are in strong support of more stringent tobacco control policies. Tested across a range of statements, this support was fairly consistent; being stronger for a ban on smoking in cars with minors and a complete ban on tobacco sales in the next 10 years, compared with increasing the legal age of cigarette sales to 21 and increasing tobacco taxes.
A limited number of studies from other LMICs have also reported support for strict tobacco control measures mainly smoking bans in workplaces, hospitality venues, shopping centres and transport terminals.7 21 29 30. This support has come from the general public as well as smokers, and is consistent with observations in HICs.13 ,31 In our survey, the support was strongest for a ban on smoking in cars with minors, similar to that observed among smokers in New Zealand, Australia, the UK, Canada, Germany and the USA.10 13 20 We also observed overwhelming support for banning tobacco sales in the next 10 years—a tobacco endgame policy that has received support in Hong Kong18 and in the UK.19 The majority of smokers supported raising the legal age of tobacco sale to 21 years. Such policies have also received support among smokers in other HICs.17 20 The level of support for taxing tobacco industry and use the revenues to support healthcare and antitobacco control measures in our survey was also comparable to the findings in similar surveys in Germany,20 New Zealand,15 England,16 and Nigeria.9
Our study has several strengths. First, it investigated smokers’ opinions across a range of tobacco control policies in a South Asian country; it was also by far the largest study to be conducted on active smokers in Pakistan—for comparison, GATS surveyed 7831 individuals which included only 177 active smokers. Our study sample draws similarities with smokers surveyed in GATS (online supplemental table 1), which suggests representativeness. We were also able to achieve a good response rate (83.3%) and a limited amount of missing data. The study does, however, have some weaknesses. We recruited fewer women than men even after allowing for their low prevalence of smoking (1.5%) in the urban population. We did not have non-smokers or ex-smokers to compare smokers’ responses. Moreover, since smokers were interviewed, some of their responses may be subject to social desirability bias. This might explain a slightly higher level of support among less educated smokers than the ones more educated. Lastly, for the statement on a complete tobacco ban, participants were not specified any underlying legal mechanism.
Supplemental material
It is encouraging to see a high level of support among smokers for a range of tobacco control policies in Pakistan. The findings provide additional support for improving implementation of measures such as tobacco taxes, which can bring the country’s tobacco control policy in line with international standards. It also highlights support for expanding smoking bans to private vehicles with minors, increasing the minimum age for cigarette sales to 21 years and considering tobacco endgame strategies for the future. Our findings would be valuable to the strong anti-tobacco advocacy network in Pakistan supported by the Bloomberg Initiative in lobbying for political support and countering the industry’s protobacco narrative. While the industry continues policy interference by amplifying public reaction,32 our findings should encourage legislators in Pakistan to introduce bold tobacco control policies without a fear of any widespread dissent among smokers or a lack of compliance.
What this paper adds
What is already known on this subject
Smokers in high-income countries tend to support tobacco control policies such as smoke-free laws.
What important gaps in knowledge exist on this topic
The evidence indicating level of support for a range of tobacco control policies among tobacco users in low-income and middle-income countries is limited.
What this paper adds
Smokers in a low-middle income setting such as Pakistan strongly support strengthening and expanding the scope of tobacco control.
Ethics statements
Patient consent for publication
Ethics approval
The study was reviewed and approved by the Health Sciences Research Governance Committee (HSRGC), University of York, UK and by the National Bioethics Committee of Pakistan Health Research Council, Islamabad, Pakistan.
Acknowledgments
The authors wish to thank all the field investigators and participants for their time and contribution to the study. We wish to thank Mr Salman Sohail for coordinating the survey in the field and Mr Saeed Ansari for data management.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Twitter @Fsiddiqui532, @zoheb_dr
Contributors KS (principal investigator) along with AK, ZI, ZK and MB (coinvestigators) were involved in designing the original study and provided inputs throughout the project period. KS also wrote the introduction, methods and discussion section. FS analysed the data, drafted results and parts of methods sections and all tables and figures. FD was involved in the interpretation of the results and contributed to the write up. MK supervised the quantitative analysis and contributed to the manuscript. All the above listed authors reviewed and approved the final manuscript draft.
Funding The study was funded by European Union Horizon 2020, grant no 680 995.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.