What Motivates Smoking Cessation? A Cross-Sectional Study in a Lower-Middle-Income Country

Our study will allow for a better understanding of the cultural specic motivating factors and strategies that most contributed to successful quit attempts. Based on these results, evidence based smoking cessation interventions can be developed tailored to the socioeconomic demographic of our country and region, including smoking cessation clinics and public outreach and media campaigns highlighting key elements of successfully quitting smoking.


Introduction
Smoking, with an average of 7 million deaths per year, is currently the leading cause preventable death in the world (1) , and causes a signi cant burden of oral and other cancers (2) . Literature pertaining to smoking cessation has shown that around two thirds of cigarette smokers are interested in quitting, while more than 50 percent report making a quit attempt in the past year (3) . However, fewer than one third of smokers who tried to quit used proven cessation methods, with only one in 10 smokers being able to quit successfully (3) . A UK based study showed that one third of quitting attempts were not preplanned and around half of those were made without the use of any support and thus were less likely to be successful (4) . Documented and validated support-based methods, and thus in extension a plan beforehand, contribute towards the success of any quit attempt (4) .
To facilitate those with the intention to quit smoking, it is imperative to identify factors motivating successful cessation in former smokers and use these to support others' quit attempts (5) . Cessation-aid interventions that are designed according to speci c motivations to quit smoking are likely to increase chances of successful cessation (6) . Factors motivating smoking cessation range from internal/individual factors (such as a smokers emotional state and willpower) and external factors (such as advice on why and how to quit from health professionals, environmental smoking restrictions, and expectations about the bene ts of quitting) (7) . The importance of internal/individual factors must not be undermined, as they have been shown to affect the e cacy of smoking cessation programs (7,8) .
While there is extensive literature exploring factors motivating smoking cessation amongst populations in developed countries (9) , such research is scarce from lower-middle-income countries (LMICs) such as Pakistan. Around 19.1% of Pakistan's adult population are tobacco users, with the majority being smokers (10) . However, a much lower percentage (24.7%) of smokers in Pakistan make attempts to quit smoking, as compared to other countries (40-50%) (11) . In addition, the success rates of quit attempts are also lower for smokers in Pakistan (2.6%), as compared to those reported by international literature (11) . These statistics highlight ineffectiveness or absence of adequate motivators of smoking cessation and interventions designed to motivate and support successful cessation attempts in Pakistan (12) . Thus, this study aims to identify factors motivating successful smoking cessation attempts in Pakistan, so that these may be incorporated towards the development of smoking cessation interventions that are targeted to the population of the country.

Study Setting and Population
This cross-sectional survey was carried out in Karachi, Pakistan, after approval from the institutional review board at the Aga Khan University Hospital (AKUH). The target population for this survey was adult former smokers, who were de ned as adult (≥ 18 years) individuals who had smoked at least 100 cigarettes in their lifetime but who had successfully quit smoking at the time of survey (13) . A quit attempt was de ned as deliberately stopping smoking for > 1 week, while successful quitting was de ned as having deliberately stopped smoking for > 1 month (13) . A quit attempt was categorized as unsuccessful if any smoking relapse (≥ 1 cigarette smoked) took place after a quit attempt.

Survey Characteristics
Data was collected by means of a questionnaire that was available in both English and Urdu, the national language of Pakistan. In the absence of a prior questionnaire suitable for our population, a comprehensive questionnaire was developed using elements from various sources (9,14,15) in close association with the faculty with expertise in tobacco cessation research at the Section of Pulmonary and Critical Care Medicine at AKUH. In addition, both the English and Urdu versions of the survey underwent pilot testing and subsequent modi cation for any ambiguous questions. The nal survey contained the following ve sections:

Usefulness of Public Health Interventions in Aiding Smoking Cessation
The survey was preceded by a consent form (available in both English and Urdu) explaining the nature and scope of the survey. In addition, preliminary screening questions based on current smoking status ensured that current smokers or those who had quit for < 1 month were not allowed to proceed with answering the survey.

Sampling Technique
In order to achieve a representative sample for this study, data collection was conducted on the premises of ve tertiary care hospitals in Karachi, including AKUH. Data collectors approached patients' attendants for participation in the survey. After initially introducing the study and obtaining consent from the individual, the data collectors screened potential participants according to the inclusion criteria and exclusion criteria. Once informed consent was taken and if the individual were suitable for inclusion, the data collectors verbally administered the survey in English or Urdu, according to the participant's preference.

Statistical Analysis
Statistical analysis was performed using IBM SPSS version 23. Continuous data was presented using mean and standard deviation/ median (interquartile range), and compared using independent sample ttests/Mann Whitney tests, as appropriate. Categorical data was presented using frequencies and percentages, and compared using chi-squared tests/Fischer's Exact tests. Multivariable logistic regression, adjusted for age, gender and monthly family income, was performed with number of quit attempts as the dependent variable (dichotomized as single attempt/successful on rst attempt and multiple attempts/one or more unsuccessful attempts before a successful attempt). A p-value <0.05 was considered statistically signi cant for all analyses.

Results
A total of 330 former smokers were included, with the majority male (92.7%) and aged between 18-30 years (43%) and 31-45 year (27.9%). Monthly family income was < Rs. 25,000 in 49.7% of respondents and > Rs. 75,000 in 18.2%. The mean age at which respondents at started smoking was 18.05 years, while the mean age at successful quitting was 31.37 years. Around half of the respondents reported having successfully quit smoking in their rst attempt (50.3%), while 17.9% reported > 6 quit attempts. Most respondents reported smoking < 10 cigarettes a day (68.2%) at the time they began their successful quit attempt (Table 1). The majority of respondents reported that they had abruptly stopped smoking (quit "cold turkey"; 62.1%). However, only 36 (10.9%) of respondents reported using a cessation aid during their successful quit attempt. Nicotine replacement therapy was the most common cessation aid used (n = 27; 8.2%). Additionally, 3 (0.9%) respondents reported using mint gums, while only 2 (0.6%) reported using pharmacological cessation therapy and 1 (0.3%) reported having attended psychotherapy/ counselling sessions for smoking cessation. Respondents also reported avoiding social company that encouraged smoking (46.4%), as well as triggers that caused an urge to smoke (28.5%). The majority of respondents believed that they had quit smoking de nitively (83.9%), although the majority felt that giving up smoking was very di cult/di cult (63.9%). Respondents reported using a variety of ways to discipline or distract themselves when they felt the urge to smoke, as well as various positive reinforcement strategies to aid cessation ( Table 2). The most frequently reported reasons for quitting smoking were to improve or protect one's own health (74.5%), followed by promptings by one's family (43%), and to improve/protect the health of family members (14.8%). 38.8% of respondents reported suffering from a smoking-related health problem (38.8%). Common sources of awareness regarding the need to quit smoking included family/friends/colleagues (37.6%), doctors (24.8%) and social media/online platforms (20.6%). Certain social pressures to quit smoking, such as peer-pressure to quit smoking (31.2%) and social avoidance by non-smokers (22.7%), were also reported. Respondents also reported having felt the need to give up smoking to be content with themselves (33.3%) and having felt upset whenever they felt the urge to smoke (30.9%). The various factors that encouraged smoking cessation are shown in Table 3.  The majority of respondents felt that anti-smoking public health interventions were not helpful at all. Consumer warnings on cigarette packs (4.5%), increased prices/taxes on cigarettes (4.5%), and smokefree public recreational places (4.2%) were most commonly reported to be helpful to a great extent in motivating cessation. Similarly, increased prices/taxes on cigarettes (4.8%) and consumer warnings on cigarette packs (4.2%) were most frequently reported to be help to a great extent in resisting relapse (Table 4). On multivariable logistic regression adjusted for age, gender and monthly family income (Table 5)

Discussion
This study was conducted to explore factors associated with successful smoking cessation in former smokers in Pakistan, a lower-middle-income country (LMIC) in South Asia. Our study identi ed personal health, promptings from one's family, and one's family's health, as the most important motivating factors.
Social pressures to quit smoking included peer-pressure to quit and social avoidance by non-smokers.
Lastly, successful cessation on one's rst quit attempt was associated with being married, quitting cold turkey, having a negative self-image of oneself due to smoking, and having strong willpower to quit.
The commonest reasons for quitting smoking were to improve/protect own health (74.5%), family's promptings (43%), to improve/protect the health of family members (14.8%), and to save money (14.5%).
Respondents reported receiving awareness regarding the need to quit smoking most commonly from their family, friends, and colleagues (37.6%). Moreover, social pressures, such as peer-pressure to quit smoking (31.2%), social avoidance by non-smokers (22.7%), and non-smokers asserting rights to smokeless public spaces (9.1%), were also major deterrents. Studies from the United States, Poland and France have demonstrated similar results, with health concerns, discouragement of smoking at home, and the high cost of cigarettes being important deterrents (16)(17)(18) . In addition, social pressure, such as having a smokefree social network that pressurizes towards cessation, has also been found to be a strong motivator of cessation across different populations (17)(18)(19) . It is interesting that promptings by doctors were reported as being a reason for quitting by only 13% of respondents, and only one quarter (24.8%) of respondents received cessation-related awareness from their doctors. A study from the United Kingdom revealed that most patients were skeptical about doctors smoking cessation advice, which was often generic and of a preaching nature, and suggested that doctors practice a more personalized approach to cessation counseling (20) .
Around half (50.3%) of the respondents in our study reported quitting successfully on their rst attempt, while the remaining reported needing 2-5 attempts (31.8%) and > 6 attempts (17.9%). These ndings are in great contrast with what is usually suggested by smoking cessation programs. These vary from 8-14 attempts, as suggested by The American Cancer Society, the Australian Cancer Council, and the Centers for Disease Control (21)(22)(23) . However, there is some literature that aligns with our ndings, as it has been suggested that though the number of quit attempts may be quite high on average, between 40-52% may be successful on their rst serious attempt (24,25) .
On multivariable regression, successful cessation on rst attempt was associated with being married, quitting cold turkey, having a negative self-image on oneself because of being a smoker, telling oneself they have the willpower to resist the urge to smoke and quit de nitively, and consciously diverting one's thoughts to distract oneself from smoking. While the concept of willpower has been debated for a long time for its actual contribution to smoking cessation (26) , it has been demonstrated to be an important factor in Pakistan previously (11) . Moreover, personal willpower is an essential feature of the "5A's" model in "Treating Tobacco Use and Dependence" (27) , of which the rst three A's build towards willingness to quit and the last two A's facilitate those willing to quit to take the nal decision to quit. This concept of personal willpower being an important factor in single-attempt cessation is strengthened by how family's promptings as a major reason for cessation was negatively associated with single-attempt cessation in our study. This suggests how personal motivation that arises from within the individual is more likely to lead to successful cessation than when it arises externally. Additionally, quitting cold turkey has been recommended as more successful in smoking cessation, as compared to gradually tapering off cigarette use (28) . Interestingly in our study, use of a smoking cessation aid was negatively associated with quitting on the rst attempt, a nding corroborated by a survey by Manis et al. in Switzerland (29) . With regards to self-image, while having a negative self-image due to one's addiction may cause distress to the smoker (30) , it can also function as a powerful motivator to quit smoking as it negates the perceived bene ts of smoking (31) . Lastly, being with a spouse or partner who is a non-smoker, a former smoker, or who encourages and motivates quitting, is associated with a greater likelihood of success on cessation attempts (32)(33)(34) .
Self-distraction by consciously diverting one's thoughts to other matters (37.3%), trying to keep one's hands and ngers occupied (34.5%), and engaging in work (28.8%), were useful strategies reportedly used by respondents. Moreover, consciously diverting one's thoughts to other matters was signi cantly associated with single-attempt cessation on multivariable regression. These are encouraging ndings, as they are simple yet effective. More technological methods of distraction, such as mobile phone applications and games (35,36) , that have been piloted in the setting of developed countries may not be feasible for a resource-constrained like Pakistan. In addition, positive reinforcement strategies, such as expecting rewards (23.6%) and receiving rewards (19.1%) from others for resisting the urge to smoke, were also employed by respondents. Rewards and incentives, often monetary, are helpful in motivating smoking cessation, especially when individualized (37,38) .
Lastly, none of the public health interventions mentioned in our survey were perceived by respondents as particularly useful for helping smoking cessation or resisting relapse, with less than 5% of respondents rating any intervention as helpful to a great extent. This is indirect contrast with studies from developed countries, such as the United States (39,40) , and may be explained by several reasons. Firstly, interventions such as government or private sector mass media anti-smoking campaigns, anti-smoking advertisements, and health warnings preceding/during lms, may not effectively be effective amongst those of lower socioeconomic and less educated backgrounds. Secondly, although Pakistan subscribes to the MPOWER model of tobacco control outlined by the World Health Organization (41) , it is possible that these interventions are not practically implemented in an optimal manner. Thirdly, since our results highlight how former smokers predominantly attribute the success of their cessation to personal factors, such as willpower, self-discipline, and distraction strategies, they are perhaps unable or hesitant to acknowledge the potentially subconscious impact of external motivators. Nevertheless, further studies are required to determine the e cacy of such large-scale public health interventions in the setting of a LMIC like Pakistan, in terms of both improving cessation and cost-effectiveness.
Despite the major burden of tobacco consumption in the country, Pakistan lacks any major smoking cessation programs or clinics facilitating rehabilitation, which along with the low cost and easy availability of tobacco, can prove the di cult task of quitting even more challenging (11) . The results of our study provide a comprehensive and unique understanding of the factors that motivate smoking cessation in Pakistan. Strengths of our study include its generalizability, as shown by the varied distribution of socio-demographic characteristics, which was achieved by targeting ve different settings for data collection in the metropolis of Karachi. Further research must investigate patterns speci c to gender, age, socioeconomic status, education level, and other demographics. These would help develop evidence-based personalized programs for smoking cessation across the population.

Conclusion
Major motivations for smoking cessation in a Pakistani population include to protect the health of oneself or family members, and due to promptings from family members. Self-discipline, personal willpower, distraction strategies, and positive reinforcement play an important role in a population where smoking cessation aids may be inaccessible to many. Moreover, peer-pressure to quit and social exclusion also motivate smokers towards quitting, as does the negative self-image one associates with themselves because of their addiction to smoking. Lastly, most public health interventions, such as mass media campaigns and anti-tobacco advertisements, were not perceived as being helpful for motivating cessation.

Declarations
Authors' Contributions RSM conceptualized and supervised the investigation, along with devising the methodology and analyzing the data. RSM was a major contributor in writing and editing the manuscript.
MUJ conceptualized the investigation, along with devising the methodology and analyzing the data. MUJ was a major contributor in writing and editing the manuscript.
MSK supervised the investigation, along with devising the methodology and analyzing the data. MSK was a major contributor in writing the manuscript.
NA collected the data by verbally administering the survey. NA also contributed to analyzing the data and writing the manuscript.
ZZF collected the data by verbally administering the survey. ZZF also contributed writing the manuscript.
MU collected the data by verbally administering the survey. MU also contributed to analyzing the data.
FS collected the data by verbally administering the survey and supervised the investigation.
JAK supervised the investigation and contributed to editing the manuscript.