Translating knowledge gaps to practice: a knowledge, attitude and practice study based on 5A approach to improve smoking cessation training for healthcare providers in Karachi, Pakistan

Abstract

Objectives

The 5A approach stands for chronological steps namely: Ask, Advise, Assess, Assist and Arrange that are recommended to healthcare providers (HCPs) for an effective smoking cessation counselling (SCC). This approach despite its effectiveness has not been evaluated in Pakistani healthcare settings. Understanding existing gaps in knowledge, attitudes and practices among HCPs is essential to inform targeted training interventions. Therefore, this study aimed to assess the knowledge, attitude and practice of HCPs regarding SCC, and to explore perceived barriers, facilitators and training needs related to its implementation.

Design

A cross-sectional study.

Setting

One public and one private tertiary care teaching hospital in Karachi, Pakistan.

Participants

364 HCPs participated who were medical or dental practitioners working for at least past 6 months in Karachi, Pakistan.

Outcome measures

Knowledge, attitudes and practice scores related to SCC were the primary outcomes, while perceived barriers and facilitators and training needs were secondary outcomes.

Results

The overall mean knowledge score was 16.8±4.8, indicating moderate knowledge levels. The mean attitude score was 5.7±1.58, reflecting generally favourable attitudes towards SCC, while the mean practice score was comparatively lower at 5.2±1.6, suggesting suboptimal implementation in routine practice. The most frequently reported barrier to delivering SCC was lack of patient willingness (64%), followed by limited access to cessation resources (52%). Improving patient health outcomes was the strongest motivator (64%), followed by a perceived professional responsibility (59%). Most HCPs reported that support from a multidisciplinary team (71.1%) and access to patient education materials (54.1%) would enhance SCC. Regarding training needs, behavioural counselling was the most requested (62.9%), followed by motivational interviewing (MI) (49.2%) and use of 5A approach (40.7%).

Conclusion

This study highlights important gaps between knowledge, attitudes and practices related to SCC among HCPs in tertiary care settings. While attitudes towards smoking cessation are generally positive, limited practical application of the 5A approach persists, largely due to patient-related, structural and resource-based barriers. The findings underscore a need for structured, context-specific training programmes emphasising behavioural counselling, MI and practical application of the 5A framework. Strengthening institutional support, improving access to cessation resources and integrating smoking cessation training into routine clinical education may enhance the capacity of HCPs to translate knowledge into effective practice, ultimately improving patient quit outcomes.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• The 5A framework use in a knowledge, attitude and practice design is a novel approach that has not previously been used to evaluate smoking cessation practices among Pakistani healthcare providers (HCPs).

• Knowledge assessment questions were designed to objectively evaluate participants’ understanding, offering greater rigour compared with subjective measures used in prior studies.

• The sample was restricted to two tertiary care hospitals in Karachi, which limits the generalisability of findings to broader healthcare settings and populations.

• The cross-sectional design and purposive sampling technique preclude causal inference and may introduce selection bias.

• Attitude and practice components relied on self-report, which may have introduced social desirability bias and led to overestimation of favourable behaviours among HCPs.

Introduction

Tobacco use remains one of the leading preventable causes of death worldwide, contributing to 2.2 million new lung cancer cases and 1.8 million lung cancer deaths annually.¹ Beyond this cancer, tobacco is the largest modifiable risk factor across multiple non-communicable diseases, driving a disproportionate burden of morbidity and mortality globally. In 2022, approximately 1.25 billion adults consumed tobacco worldwide,² and Pakistan ranked seventh globally with an estimated 25.4 million tobacco users.³ The prevalence of tobacco use in Pakistan stands at 13.4%, with higher rates in urban areas (16.3%) compared with rural (11.7%).⁴

In response to the global tobacco epidemic, the WHO developed the Framework Convention on Tobacco Control (FCTC), a landmark international treaty signed by Pakistan.⁵ Article 14 of the FCTC explicitly mandates that healthcare providers (HCPs) to offer treatment for tobacco dependence and ensure access to low-cost cessation interventions.⁶ Despite this commitment, implementation of the FCTC has been slow, particularly in low- and middle-income countries (LMICs). Strengthening its implementation in these settings has been projected to reduce the tobacco use burden by one-third by 2026.⁷

HCPs are uniquely positioned to drive smoking cessation at the individual level. Brief, simple cessation advice delivered by an HCP has been shown to increase the likelihood of successful quitting, even after extended follow-up periods, with more intensive counselling yielding incrementally higher quit rates.⁴ Clinician recommendations to quit tobacco not only enhance cessation but also produce even greater outcomes when combined with pharmacotherapy. Importantly, these interventions are considered cost-effective in the context of LMICs, making HCP-delivered smoking cessation counselling (SCC) a high-value public health strategy.^{5 6}

However, the landscape of nicotine use is rapidly evolving. The proliferation of novel nicotine products including e-cigarettes, vaping devices, heated tobacco products and nicotine pouches has outpaced the ability of many HCPs to accurately assess and counsel patients using these products.⁸ While these products are often marketed as safer alternatives to conventional cigarettes, their long-term health consequences remain uncertain and their growing use, presents a new layer of complexity for SCC.⁹ HCPs must therefore update their knowledge and assessment skills to remain effective in this changing environment.

A structured, evidence-based approach is essential to guide HCPs in this context. The 5A framework: Ask, Advise, Assess, Assist and Arrange, recommended by the Centers for Disease Control and Prevention (CDC), provides a systematic approach to SCC across healthcare settings.¹⁰ HCPs are advised to ask patients about their smoking status at every visit, advise all tobacco users to quit, assess their willingness and level of nicotine dependence, assist them through counselling or pharmacotherapy and arrange follow-up to reinforce quit attempts.¹¹ The 5A approach has demonstrated effectiveness in improving smoking cessation rates across diverse healthcare settings.¹² Notably, this framework has not been used in local studies within Pakistan, representing a significant gap.

Despite the scale of Pakistan’s tobacco burden, the evidence base for HCP-delivered SCC in the country remains limited. A national tobacco cessation guideline has been launched by Pakistan Chest Society¹³ followed by a toolkit for tobacco control in Pakistan by Tobacco Control Cell of the Government of Pakistan¹⁴; however, its uptake and integration into clinical practice among HCPs has not been explored. A prior study examining HCP knowledge in this domain was exclusively tailored to physicians managing tuberculosis patients, limiting its generalisability to broader clinical settings.¹⁵ Furthermore, previous studies have relied on subjective assessment methods and have not accounted for emerging nicotine products or the 5A framework, leaving critical gaps in our understanding of HCP preparedness for SCC in Pakistan.

Given these gaps, there is a pressing need for a comprehensive, objective assessment of HCP knowledge, attitudes and practices regarding SCC in Pakistan. Understanding these gaps through a behavioural science lens, specifically the COM-B model, which frames behaviour change in terms of Capability, Opportunity and Motivation can provide actionable directions for intervention design.¹⁶ The findings of this study will therefore contribute to the existing literature and inform the development of evidence-based training programmes for HCPs. The primary aim of this study was to assess the knowledge, attitudes and practices of HCPs using the 5A approach, and to explore barriers, facilitators and training needs in this context.

Methods

Study design, site and duration

This was a cross-sectional study, conducted at one private and one public tertiary care teaching hospital in Karachi, which is the capital city of the Sindh province of Pakistan. We selected these sites based on convenience sampling. The study spanned from November 2024 to November 2025.

Study participants

We included the doctors from specialties with a high likelihood of managing tobacco-related diseases. These comprised medical and allied disciplines: family medicine, internal medicine, pulmonology, cardiology, oncology, otorhinolaryngology, gastroenterology, neurology, endocrinology, nephrology and psychiatry as well as surgical and allied fields, such as cardiothoracic surgery, vascular surgery, urology, oral surgery, plastic and reconstructive surgery, orthopaedic surgery maxillofacial surgery and general surgery. Specialties such as obstetrics and neonatology were excluded, as smoking cessation interventions in these areas are typically indirect and their inclusion could introduce heterogeneity into the analysis. We used purposive sampling for participant recruitment.

Sample size calculation

A sample size of 364 was calculated to estimate the population proportion with a 95% confidence level and a specified absolute precision of 5%, assuming a population proportion of 61%¹⁷ using WHO sample size calculation software.

Data collection tool

A questionnaire in English language was developed using literature,^{18 19} and based on COM-B model¹⁶ as shown in figure 1. Content validity was ensured through expert evaluation by a tobacco cessation specialist pulmonologist, a public health professional, a family medicine consultant and psychiatry specialists. The questionnaire comprised seven sections: (1) demographic information (12 items); (2) general knowledge regarding SCC (14 items); (3) knowledge related to the 5A components (5 items); (4) attitudes towards SCC (5 items); (5) practices related to SCC (7 items); (6) perceived barriers and motivators (2 items) and (7) additional resource or training needs (2 items). The complete questionnaire is provided in online supplemental file 2. Reliability was assessed using Cronbach’s alpha, which was 0.8.

Figure 1. Theoretical framework integrating knowledge, attitude, practice and barriers/facilitators within the COM-B model for smoking cessation counselling. 5A, ask, advise, assess, assist and arrange; COM-B, Capability, Opportunity, and Motivation; FCTC, Framework Convention on Tobacco Control; NRT, Nicotine replacement therapy.

Data collection procedure

A manual, self-administered questionnaire was distributed to participants at their workstations after obtaining written informed consent. The questionnaire was administered by the research team at a time convenient for each participant, and completion took approximately 15–20 min. A total of 400 HCPs were invited for participation out of which fifteen did not give consent due to lack of time and high workload. A total of 385 participants filled the form out of which 14 forms were discarded due to missing information and seven were discarded due to ineligible specialty. Therefore, the remaining eligible samples were 364.

Data analysis

Statistical analysis was performed using STATA V.17. Normality was assessed using the Shapiro-Wilk test. Mean and SD are reported for continuous variables (such as age, working hours and years of experience) and frequencies along with percentages are reported for categorical variables (such as gender, setting, designation and specialty). The variable ‘designation’ was recoded into two categories: interns, medical officers and residents were classified as junior HCPs, whereas senior registrars and consultants/professors were classified as senior HCPs. An independent t-test and one-way analysis of variance were applied to compare means between two groups and among more than two groups, respectively. P value of <0.05 was considered significant. Primary outcome was knowledge, attitude and practice scores. Secondary outcomes were perceived barriers, facilitators and training needs.

The knowledge composite score was created by adding up the correct responses for the knowledge-based questions (minimum to maximum range: 5–28), the attitude score was created by adding the options which depicted positive attitude (minimum to maximum range: 2–9), similarly the practice score was created by adding up the options which depicted good practice (minimum to maximum range: 0– 7). Simple linear regression was applied to find the effect of predictor variables (including age, gender, city of graduation, setting, designation, years of experience, working hours (per week), specialty, prior SCC training, perceived adequacy of SCC training, smoking status) with primary outcome. The multivariable regression model was applied keeping the variables which had p value <0.20. Multi-collinearity was checked using variance inflation factor (VIF) and a VIF greater than >5 was considered significant. The variables with high collinearity were removed in the final model.

Sensitivity analysis

Human resources for health in Sindh province are distributed across 80% private sector and 20% public sector employment.²⁰ We applied these proportions as weights to calculate weighted knowledge, attitudes and practice scores to confirm the robustness of our findings, as presented in online supplemental table S4.

Ethical considerations

We obtained approval from the ethical review committee of the Aga Khan University (2024–10671-31967) and Jinnah Postgraduate Medical Centre (F.2-81/2024-GENL/154/JPMC). Written and informed consent was taken from each participant before interview; they were informed about the publication of the findings.

Patient and public involvement

Patients or public were not involved in this study.

Results

A total of 364 HCPs participated, with a mean age of 33±9.8 years with 55.2% males. Most of the HCPs i.e. 61.3% graduated from Karachi and were working in public sector 63.7%. 22.3% (n=81) consultants and 9.3% (n=34) registrars were grouped together as senior HCPs. The remaining 31.3% (n=114) postgraduate residents, 21.2% (n=77) medical officers and 15.9% (n=58) interns were categorised as junior HCPs (table 1). The total knowledge mean score overall was 16.8±4.8, with public participants scoring lower than the private participants, with p value <0.001. The overall mean attitude score was 5.7±1.58, with no significant differences between public and private HCPs (p value=0.008). The overall mean practice score was 5.2±1.6, which did not have significant difference in public vs private settings (p value >0.05) (table 1). For weighted analysis, see online supplemental table S4.

Table 1. Demographic characteristics with knowledge, attitude and practice scores of healthcare providers at public and private tertiary care hospitals in Karachi (n=364).

Characteristics	N (%)	Knowledge		Attitude		Practice
		Mean±SD	P value	Mean±SD	P value	Mean±SD	P value
Overall score	364 (100)	16.87±4.88		5.70±1.58		5.28±1.64
Age			0.001		0.003		0.877
Mean±SD	33±9.80
<30	184 (50.5)	16.05±4.86		5.45±1.46		5.29±1.63
≥30	180 (49.5)	17.69±4.78		5.95±1.67		5.27±1.67
Gender			0.401		0.475		0.057
Male	201 (55.2)	16.67±4.70		5.75±1.64		5.10±1.69
Female	163 (44.8)	17.10±5.10		5.63±1.51		5.43±1.60
City of graduation			0.628		0.044		0.003
Karachi	223 (61.3)	16.96±5.06		5.57±1.62		5.08±1.65
Other	141 (38.7)	16.71±4.61		5.91±1.50		5.60±1.59
Setting			<0.001		0.009		0.321
Public	232 (63.7)	16.01±5.09		5.53±1.49		5.34±1.61
Private	132 (36.3)	18.36±4.11		5.98±1.71		5.17±1.71
Designation			0.003		0.121		0.933
Junior HCP	249 (68.4)	16.35±4.85		5.61±1.51		5.29±1.62
Senior HCP	115 (31.6)	17.97±4.79		5.89±1.72		5.27±1.70
Years of experience			<0.001		0.016		0.551
Mean±SD	7.65±8.60
<5 years	201 (55.2)	16.02±4.96		5.52±1.48		5.23±1.62
≥5 years	163 (44.8)	17.90±4.59		5.92±1.68		5.34±1.67
Working hours/week			0.467		0.466		0.484
Mean±SD	54.1±18.4
<48 hours	115 (31.6)	16.59±5.39		5.61±1.59		5.19±1.50
≥48 hours	249 (68.4)	16.99±4.64		5.74±1.58		5.32±1.71
Specialty			0.064		0.215	5.40±1.59	0.06
Medicine	239 (65.7)	16.52±5.03		5.62±1.54		5.06±1.72
Surgery	125 (34.3)	17.52±4.54		5.84±1.66
Prior smoking cessation counselling training			0.052		0.435		0.126
Yes	44 (12.1)	15.52±4.52		5.52±1.27		5.23±1.66
No	320 (87.9)	17.05±4.91		5.72±1.62		5.64±1.48
Adequacy of training for smoking cessation counselling*			0.0003		0.026		0.005
Strongly disagree	43 (11.8)	16.72±4.89		5.42±1.64		4.79±1.88
Disagree	91 (25.0)	18.53±5.01		6.16±1.52		4.99±1.66
Neutral	117 (32.1)	16.88±4.89		5.54±1.53		5.27±1.60
Agree	84 (23.1)	15.96±4.42		5.61±1.70		5.65±1.56
Strongly agree	29 (8.0)	14.41±4.23		5.55±1.30		5.86±1.27
Smoking status*			0.432		0.985		0.137
People who never smoked	298 (81.9)	16.92±5.07		5.69±1.57		5.30±1.64
People who smoked in the past	34 (9.3)	15.91±4.36		5.74±1.73		5.56±1.48
People who smoke currently	32 (8.8)	17.38±3.32		5.72±1.57		4.78±1.79

^*Bonferroni-adjusted pairwise comparisons showed significant differences in knowledge (agree vs disagree, p=0.004; strongly agree vs disagree, p=0.001), attitude (neutral vs disagree, p=0.045) and practice (agree vs strongly disagree, p=0.047).

HCP, healthcare provider.

Knowledge regarding core smoking cessation concepts was variable among HCPs. Only 47.5% correctly identified the formula for calculating pack-years. Knowledge of pharmacological cessation methods such as nicotine gum was correctly identified by 82.6%, similarly for nicotine skin patches 73.9% and bupropion was identified by 42.5% HCPs (online supplemental table S1). Regarding e-cigarettes, only 28.8% correctly identified them as a last-resort option. Routine provision of SCC to all smokers was endorsed by 60.4%. Tele-consultation as a cessation option was identified by 48.9%. Always performing the ‘Ask’ step was reported by 64.6% of participants, while 54.4% reported always providing advice to quit smoking. The ‘Assess’, ‘Assist’, and ‘Arrange’ steps were reported as always performed by 35.2%, 26.4% and 21.7% of participants, respectively. Referral to smoking cessation or rehabilitation services was reported by 40.4% of participants. A significant difference was observed in the frequency of using MI techniques between public and private-sector participants (p=0.005) (online supplemental table S1).

The most frequently identified resource to improve SCC effectiveness was support from a multidisciplinary team, reported by 71.1% of the HCPs, followed by access to patient educational materials reported by 54.1% (figure 2). 62.9% of HCPs reported behavioural counselling as a desired area of training, followed by MI at 49.2% (figure 3).

Figure 2. Resources and support needed to improve smoking cessation counselling informed by healthcare providers in Karachi (n=364) (multiple response question).

Figure 3. Desired areas for smoking cessation counselling training informed by healthcare providers in Karachi (n=364) (multiple response question). 5A, ask, advise, assess, assist and arrange.

The most reported barrier to providing SCC was patients’ lack of willingness (63.5%). This was followed by limited access to cessation resources such as educational materials and medications (52.2%) (see figure 4A). Regarding motivators for providing SCC, improving patient outcomes was most frequently cited (64%), followed by a perceived duty to reduce morbidity and mortality (58.5%) (see figure 4B). The association between barriers and facilitators was also analysed with knowledge, attitudes and practices, as presented in online supplemental table S3.

Figure 4. Reported barriers (A) and enablers (B) to providing smoking cessation counselling by healthcare providers (n=364) (multiple response question).

Univariate analysis showed that higher knowledge scores and attitudes were significantly associated with working in a private setting, greater years of experience (online supplemental table S2). For practice, significant associations were observed with graduating from a city other than Karachi and perceptions of training adequacy (online supplemental table S2).

Multivariate linear regression analysis showed that working in a private setting was significantly associated with higher knowledge scores compared with the public sector (table 2). Years of experience, working hours, type of HCP, specialty, prior SCC training, city of graduation and perceived adequacy of training were not significantly associated with knowledge scores (table 2). For attitude domain, working in a private setting and increasing years of experience were positively associated with higher attitude scores (table 2). No significant associations were observed for gender, city of graduation, specialty or working hours (table 2). Lastly, for practice domain, perceived adequacy of previous training was positively associated with practice scores. Being in a surgical specialty was negatively associated with practice scores compared with medicine (table 2). Gender, city of graduation, years of experience and prior SCC training were not significantly associated with practice scores (table 2).

Table 2. Multivariable linear regression for knowledge, attitude and practices of healthcare providers at Karachi for smoking cessation counselling as outcome respectively (n=364).

Variable	Knowledge	Attitude	Practice
	Adjusted β (95% CI)	Adjusted β (95% CI)	Adjusted β (95% CI)
Gender Male (vs Female)	—	—	0.23 (-0.15, 0.62)
Setting Private (vs Public)	2.06 (0.96, 3.35)	0.51 (0.14, 0.89)	—
City of graduation Other (vs Karachi)	—	0.21 (-0.16, 0.58)	0.30 (-0.09, 0.69)
Type of HCP Senior (vs Junior)	0.74 (-0.81, 2.29)	−0.27 (-0.75, 0.21)	—
Years of experience	0.03 (-0.05, 0.11)	0.03 (0.01, 0.05)	0.02 (-0.01, 0.04)
Working hours/week	0.01 (-0.01, 0.05)	—	—
Specialty Surgery (vs Medicine)	−0.81 (-2.09, 0.47)	−0.12 (-0.52, 0.30)	−0.41 (-0.79, −0.03)
Prior SCC training Yes (vs No)	−0.83 (-2.46, 0.79)	—	0.14 (-0.40, 0.67)
Medical training was adequate
Strongly agree	−1.39 (-3.73, 0.94)	0.27 (-0.46, 1.01)	1.01 (0.23, 1.79)
Agree	−0.42 (-2.19, 1.34)	0.23 (-0.35, 0.80)	0.94 (0.35, 1.53)
Neutral	0.38 (-1.29, 2.03)	0.28 (-0.32, 0.79)	0.57 (0.01, 1.14)
Disagree	1.92 (0.19, 3.6)	0.80 (0.23, 1.37)	0.21 (-0.36, 0.81)
Strongly disagree	Ref	Ref	Ref

HCP, healthcare provider; SCC, smoking cessation counselling.

Discussion

This cross-sectional study at one public and one private tertiary care hospital in Karachi assessed HCPs knowledge, attitudes and practices regarding SCC, alongside barriers, motivators and training needs. The overall mean knowledge, attitude and practice scores were 16.8±4.8, 5.7±1.58 and 5.2±1.6, respectively. HCPs were more confident and active in the early steps of 5A approach (ask and advice) but engagement declined for more demanding steps (assess, assist and arrange). The most common barrier was patient unwillingness to quit (64%), followed by limited access to cessation resources (52%). Improving patient outcomes was the leading motivator (64%), followed by a sense of professional duty (59%). Only 12.1% of HCPs had received any prior SCC training. The most requested training topics were behavioural counselling and MI, and the most needed resources were patient educational materials and multidisciplinary team support.

The majority of HCPs in our study expressed interest in additional SCC training, consistent with findings from the Netherlands where 87.1% sought further training, with preferences for group workshops (71%), webinars (55.9%) and web-based content (35.5%).²¹ To optimise training design, the COM-B model, used in a Chinese smoking cessation context, offers a useful framework: addressing Capability (knowledge of smoking, withdrawal and cessation techniques), Opportunity (access to cessation services and supportive environments) and Motivation (personal beliefs about smoking harms and confidence in counselling).²² This framework is particularly applicable to the Pakistani context, where structured, multi-component training programmes remain scarce. Knowledge gaps reflect limited Capability, while restricted access to resources and referral systems indicates inadequate Opportunity, and declining engagement in advanced 5A steps suggests gaps in sustained Motivation. Additionally, as novel nicotine products gain prevalence, ongoing professional development through workshops, webinars and integration of vaping-related content into clinical curricula is essential to keep HCPs confident and current in addressing emerging tobacco-related challenges.²³ We found that use of MI was also limited, which is consistent with Champassak et al’s observation of limited MI skills in routine practice.²⁴ Importantly, nearly half of HCPs in our study (49.2%) expressed interest in learning MI, underscoring an opportunity for training.

Knowledge of the 5A approach was modest in our study, consistent with findings from Saudi Arabia where only 47.7% of HCPs reported familiarity with its components.²⁵ In terms of practice, HCPs performed better on the initial steps (‘Ask’ and ‘Advise’) but engagement declined for the more advanced components of ‘Assess’, ‘Assist’ and ‘Arrange’, a pattern similarly reported in a US study.²⁶ Use of computer-assisted decision support has shown promise in improving uptake of the latter 5A components.²⁷ Inclusion of mobile applications as a sixth ‘A’ to the framework has also been proposed.²⁸ A reason for this gap could be a lack of strong policies, as 36% of HCPs in our study reported the absence of policies or guidelines related to SCC. While, several countries such as the UK,²⁹ Saudi Arabia³⁰ and Kuwait³¹ mandate SCC to HCPs via established policies, in addition the U.S. Public Health Service Clinical Practice Guideline emphasises use of the ‘5 A’s’ approach.³² Encouragingly, 40% of HCPs in our study expressed willingness to undergo formal 5A training, indicating a receptivity of the HCPs for further training.

Being in a surgical specialty was negatively associated with good SCC practice scores in our study, suggesting that surgical HCPs are less likely to engage in SCC compared with their medical counterparts. This finding resonated with a Finnish cohort study evaluating preoperative SCC, which found that preoperative cessation was initiated in merely 2.2% of current smokers reflecting weak smoking awareness within surgical settings. Despite smoking being a risk factor for postoperative complications, SCC remains a missed opportunity. This pattern reflects a discipline-specific gap, where surgical HCPs prioritise procedural outcomes, and SCC is not yet embedded into preoperative workflows. These findings collectively point to the need for structured perioperative cessation protocols that engage surgical teams as active participants in SCC.³³

Lack of referral options for smoking cessation specialists was reported by almost half of HCPs in our study, as it has been reported that Pakistan has one of the lowest psychiatrists to population ratios in the world,³⁴ making specialist referral almost inaccessible in routine practice. A practical response is the training of non-specialist providers including general practitioners to deliver cessation support services within structured referral networks.³⁵ Another significant barrier reported was limited access to medications, high relapse rates and patient unwillingness. A Chinese study demonstrated that subsidising pharmacotherapies significantly improves quit rates,³⁶ meanwhile the Taiwanese government introduced a subsidisation policy which substantially increased uptake of cessation treatments.³⁷ In the context of Pakistan, such steps need advocacy to the policy makers. As lack of institutional policies, incentives and organisational support were reported by also reported in our study. A systematic review highlighted that cessation-oriented institutional cultures and inter-professional collaboration can enhance provider engagement.³⁸ While, a study from Malaysian HCPs suggested that introducing provider incentives alongside active promotion of cessation services can meaningfully shift provider behaviour.³⁹ Considering the context of LMIC, it is noteworthy to mention that the provider incentivisation does not always require a large budget; non-financial recognition such as appointing tobacco champions, giving badges and publicly acknowledging them can reinforce positive behaviours.^{40 41}

Regarding motivators, improving patient outcomes and a sense of professional duty identified in our study are consistent with findings from quantitative and qualitative US study each,^{24 42} in which HCPs viewed SCC as integral to comorbidity management. However, guideline adherence was cited as a motivator by only 26% of our HCPs, contrasting with a Dutch study where guideline familiarity and inter-professional collaboration agreements were among the primary drivers of SCC engagement.⁴³ This suggests that while intrinsic motivation exists, external drivers such as guidelines and institutional reinforcement remain limited.

Among HCPs in our study, 8.2% of actively smoked, consistent with rates reported in comparable regional studies: 11% in Qatar,¹⁹ 10.8% in Saudi Arabia²⁵ and 8.2% in Malaysia.⁴⁴ In the Qatari study, people who currently smokers had significantly lower odds of good cessation practice compared with non-smokers; however, smoking status was not a significant predictor in either the Malaysian study or our own. Similarly, prior training in SCC was reported by only 12.1% of HCPs in our study lower than the 27.1% reported in the USA,²⁶ but comparable to rates in Malta (8.5%)⁴⁵ and Malaysia (14.1%).⁴⁴ The Qatari study identified training as a significant predictor of good cessation practice, whereas neither the Malaysian study nor our own findings demonstrated a significant association between training and practice. These differences could be due to a better sample size of Qatari study (n=7214) and the fact that the data collection was online in Malaysian study. However, this finding reflects variability in training quality, lack of standardisation or insufficient reinforcement of learnt skills in clinical settings.

In our study, HCPs also indicated that telemedicine and smartphone applications can be an option for SCC. Similarly, a recent study conducted in Türkiye reported that telemedicine interventions were highly effective, with a smoking cessation rate significantly higher than traditional in-person counselling (44% vs 16.9%, respectively).⁴⁶ These findings are relevant to the Pakistani context, where mobile health platforms and telemedicine could serve as practical adjuncts, particularly where dedicated cessation clinics are unavailable and infrastructure for telemedicine is already in place.⁴⁷

Pakistan’s national tobacco-cessation guidelines recognise both NRT and sustained-release bupropion as first-line pharmacotherapies.¹³ While evidence suggests combination NRT may offer superior efficacy compared with bupropion alone, single-form NRT and bupropion demonstrate comparable effectiveness.⁴⁸ NRT is often preferred in Pakistani settings due to its flexible delivery options, suitability for patients with contraindications to bupropion and greater accessibility. Varenicline is not included as a treatment option in Pakistan due to its unavailability in the country.⁴⁹ Local evidence on smoking cessation pharmacotherapy remains sparse; a recent randomized controlled trial demonstrated that both behavioural change communication (BCC) and BCC combined with bupropion significantly improved smoking abstinence among newly diagnosed tuberculosis patients, yet the trial on adults without tuberculosis remain scarce.⁵⁰ Regarding e-cigarettes, regulatory measures in Pakistan are minimal, restricted primarily to restricting sales near schools⁵¹ and use in public places,⁵² while their popularity continues to rise. A systematic review reveals that when e-cigarettes are used for cessation, there should be structured and carefully guided approaches including behavioural relapse-prevention strategies, and use of NRTs alongside leading to tapering.⁵³ Therefore, HCPs need to be trained specially for this domain as in our study there were variability in responses related to use of e-cigarettes for SCC.

Our findings broadly align while adding important local texture to a growing international literature on HCP-delivered SCC. Compared with the Qatari study, our cohort showed higher pharmacotherapy awareness: correct identification of nicotine gum (82.6% vs 65.9%), bupropion (42.5% vs 27.8%) and varenicline (31.0% vs 25.3%). Within the globally identified cost-effective smoking cessation medications, cytisine is also recommended for inclusion in national guidelines.⁵⁴ However, it remains unlicensed in Pakistan,⁵⁵ which likely explains why only 7.1% of HCPs in our study correctly identified it as a cessation medication.

A systematic review found that e-referrals were the most common, with most patients directed towards quit-lines or local cessation services. In our study, 40.7% of HCPs reported referring patients to smoking rehabilitation or specialist clinics. This might be due to the fact that, in Pakistan, there is no active smoking cessation quit-line at a national level.⁵⁶ The same review highlighted that institution-level supports practices, with interventions such as staff training and performance feedback reports, stressing that systematic approaches to embedding cessation into routine care are important. Encouragingly, efforts to integrate SCC into clinical care have been piloted in Khyber Pakhtunkhwa, Pakistan.¹⁵ An implementation study spanning Pakistan, Bangladesh and Nepal revealed that brief behavioural cessation support could be successfully embedded within tuberculosis care programmes, with short 1–2 hour trainings increasing provider confidence by 17% (95% CI: 14 to 20%). System-level adaptations including reporting protocols and structured supervision facilitated both institutional scale-up and broader coverage.¹⁵ However, qualitative findings revealed that busy clinical environments often led to partial implementation, showing that one-off training is insufficient without sustained support and tailored delivery.¹⁵ These findings are particularly relevant to our study context: despite demonstrated feasibility, significant gaps in HCP knowledge, attitudes and practice around SCC persist across Pakistani healthcare settings, highlighting the need for structured, system-embedded and continuously reinforced cessation training programmes.¹⁵

This study has several strengths, including the use of the COM-B model to guide questionnaire development and the incorporation of the 5A approach, which has not been extensively used in previous research. In addition, the knowledge component was assessed objectively, in contrast to prior studies that have largely relied on self-reported measures. However, certain limitations should be acknowledged. The sample was drawn from only two tertiary care hospitals in Karachi, limiting the generalisability of the findings. Moreover, the cross-sectional design and purposive sampling technique further restrict causal inference and representativeness. Attitudes and practices assessment relied on self-reported data and may be subject to social desirability bias among the participants.

Conclusion

This study identifies important gaps between knowledge, attitudes and practices regarding SCC among healthcare professionals in tertiary care settings. Although attitudes towards smoking cessation were generally positive, the practical implementation of the 5A approach was limited, primarily due to patient-related, structural and resource constraints. These findings highlight the need for structured, context-specific training programmes with an emphasis on behavioural counselling, MI and the practical application of the 5A framework. Strengthening institutional support, improving access to cessation resources and integrating SCC training into routine clinical education may enhance healthcare professionals’ capacity to translate knowledge into practice and improve patient quit outcomes. Furthermore, although national guidelines and cessation toolkits exist, they need to be incorporated in the design of future training initiatives.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.