The Sol Plaatje district is part of the Frances Baard District Municipality of the Northern Cape Province, South Africa. It is named after Solomon Tshekisho Plaatje and includes the diamond mining city of Kimberley (The Local Government Handbook: South Africa 2023). The study population consisted of HCPs (medical officers, professional, enrolled or assistant nurses) working in the 13 Sol Plaatje District Municipality clinics, as obtained from the human resource employment database. The clinics include Galeshewe Community Health Centre, Masakhane Clinic, Beaconsfield Clinic, Betty Gaetsewe Clinic, Florianville Clinic, Mapule Matsepane Clinic, City Clinic, Dr Windston Torres Clinic, Greenpoint Clinic, Ritchie Clinic, Platfontein Clinic, Phutanang and Madoyle Clinic.

Medical officers and professional, enrolled or assistant nurses working in one of the 13 public clinics in Sol Plaatje District Municipality were included in the study. Healthcare providers not working in primary healthcare clinics were excluded.

The study included everyone who satisfied the inclusion criteria; a power calculation to determine an adequate sample size was unnecessary. The intended sample for the study consisted of a maximum of 165 respondents consisting of 18 medical officers, 99 professional nurses, 13 enrolled nurses and 35 enrolled nursing assistants. Only 156 of these HCPs completed the questionnaires from 01 to 28 June 2019 (94.5% response rate).

A confidential, structured, closed-ended questionnaire was used for the research instrument. The questionnaire was designed based on information from prior studies assessing the KAPs of HCPs in smoke cessation intervention (Cohen, McGinnis & Salsberg 2007; Delucchi, Tajima & Guydish 2009; Fiore et al. 2008; Van Zyl-Smit et al. 2013).

The questionnaire consisted of five sub-sections: The ‘knowledge’ section contains 19 questions designed to evaluate the clinician’s level of knowledge on smoking cessation intervention in the following areas: (1) smoking cessation methods; (2) benefits of smoking cessation interventions; (3) nicotine dependence and withdrawal symptoms; (4) effect of smoking on an unborn child; (5) medications recommended for treatment of tobacco dependence in South Africa. Most questions required a ‘true’, ‘false’ or ‘don’t know’ answer. Knowledge was categorised using Bloom’s cut-off point. This categorisation grades 80% and above as good knowledge, 60% – 79% as moderate and less than 60% as poor knowledge.

The ‘attitude’ section consisted of 12 items designed to assess the following: (1) the role of HCP in smoking cessation intervention; (2) the priority between health concerns and smoking cessation intervention; (3) the attitude of smokers to smoking cessation advice; (4) time constraints to provide smoking cessation advise during routine consultations; and (5) smoking status to be included as one of the ‘vital signs’ for patients attending primary care centres. Respondents were required to agree or disagree with the statements.

The ‘practise’ section was based on the South African tobacco smoking cessation clinical practice guideline, particularly the 5A’s strategy for providing smoking cessation, that is, ask, alert, assess, assist and arrange follow-up. The respondents were asked to state the frequency with which they performed the individual components of the 5A’s strategy in their day-to-day interactions with patients using the scales ‘Always’, ‘Sometimes’ and ‘Never’.

In assessing barriers in smoking cessation interventions, the primary investigator used a series of questions which required the respondents to rate each barrier as either ‘not a barrier’, ‘somewhat a barrier’ or ‘important barrier’.

The study was piloted at Robert Mangaliso Sobukwe Hospital Kimberley (RMSH). A medical officer, two professional nurses, two nurse assistants and five enrolled nurse assistants completed the questionnaires. However, data obtained from the pilot study were not included as changes were made to the datasheet. They included correction of grammatical errors and spelling out words in full rather than using abbreviations, which was confusing for some of the respondents.

The Department of Biostatistics, Faculty of Health Sciences, University of the Free State (UFS) assisted with data analysis using Statistical Analysis System (SAS) version 9 (SAS Institute Inc., Cary, North Carolina, U.S.). Frequencies and percentages were calculated for categorical data, and the presentation was performed using tables. Association between variables were assessed using Chi-squared or Fisher’s exact tests. A p-value of < 0.05 was set to be the level of significance.

The protocol for this study was approved by the Health Science Research Ethics Committee (HSREC), University of the Free State, with approval number (UFS-HSD2018/1351/2901). Permission for the study was also obtained from the Northern Cape Province ethics committee.

Number coding was used to ensure the confidentiality of the participant’s responses. No names or personal identifiers appeared on any research-related information or datasheet sent for statistical analysis. The researcher kept all paper-based records in a secure location, only accessible to those involved in the study. All information was managed confidentially.

The majority of respondents (n = 119), 76.3% had poor knowledge (< 60%), 21.2% (n = 33) had moderate knowledge (60% – 79%), while 2.6% (n = 4) had good knowledge (80% – 100%).

The majority of the respondents answered correctly to the following knowledge items ‘Patients should only be asked about their smoking history if they have a smoking-related disease or illness’, ‘Smoking cessation is not a critical component of management of Chronic Obstructive Pulmonary Disease’, ‘It is not necessary to assess nicotine dependence prior to initiation of Nicotine Replacement Therapy’, ‘When advising patients to stop smoking, the advice should never be linked to the patient’s current health/illness’. In others, the majority answered either incorrectly or were not sure of the following knowledge items: ‘Patients who have their first cigarette within half an hour of waking are likely to be less dependent on nicotine than patients who have it much later in the day’, ‘Most of the withdrawal symptoms from smoking cessation disappear within 4 weeks of abstinence’. See Table 2 for other knowledge items.

Table 3 shows the respondents’ attitude scores. Most respondents strongly agree that smoking cessation counselling is an important part of their job.

Most of the respondents disagree with the items ‘Smoking is a personal decision which does not concern the health care worker’ (n = 117), 75%, ‘It is uncomfortable to counsel my smoking patients on quitting smoking’ (n = 120), 77%, and ‘Patients are not interested in receiving smoking cessation information’ (n = 64), 41%. However, many agree with the item, ‘Clinicians should discuss smoking cessation with patients even if it’s not the reason for the visit’ (n = 147) 94%. See Table 3 for respondents’ agreement with the other attitude items.

Table 4 depicts the respondents’ practice scores, with the majority of respondents sometimes asking patients about their smoking status. Still, many sometimes or never record the patients smoking history in their medical records.

Many respondents sometimes encourage the use of NRT (n = 68), 43.6% and always encourage patients not to smoke in the presence of infants and children (n = 115), 73.7%. See Table 4 for healthcare providers’ responses for other practice items.

Table 5 shows the respondents’ perceived barriers to smoke cessation counselling. The top three perceived barriers to smoking cessation counselling, in descending order, were ‘Lack of pharmaceutical medication for Nicotine Replacement Therapy’ (n = 109), 70%, ‘Lack of copies of smoking cessation guidelines in the facility’ (n = 100), 64.1%, and ‘Some health care providers are themselves smokers’ (n = 95), 60.9%.

Table 6 shows an analysis of the background characteristics of respondents with KAP. It indicates that the healthcare practitioners’ smoking status was significantly associated with the practice of assessing a smoking patient’s willingness to quit smoking (p = 0.033). When knowledge was compared with some variables, knowledge was significantly associated with profession (p ≤ 0.001) and year of practice (p = 0.008) but was not associated with smoking status (p = 0.479).

There was a significant association (p = 0.002 and p = 0.001) between the professional ranking and their responses to the items ‘My patients’ acute health problems take precedence over smoking counselling’ and ‘Smoking cessation counselling negatively affects my relationship with patients’, respectively. An association was found between the item ‘Clinicians should discuss smoking cessation with patients even if it’s not the reason for the visit’ and the smoking status of the HCPs (p = 0.008). ‘I do not have sufficient time to provide advice and counselling to all my patients who smoke during routine consultations’ was significantly associated with HCPs’ years of practice (p ≤ 0.001).

Asking the patients how many cigarettes they smoke a day was significantly associated with HCP’s smoking status (p ≤ 0.001) and profession (p = 0.049) but not their years of practice (p = 0.100). A significant association was found for items ‘recording the patient’s smoking history in their medical records’, ‘discussing the risks of smoking and benefits of quitting smoking’ with HCPs profession (p ≤ 0.001). Asking patients about previous attempts to quit smoking was significantly associated with the smoking status of the HCPs (p = 0.044). Assessing a smoking patient on willingness to quit smoking was also significantly associated with HCP’s smoking status (p = 0.033) and profession (p = 0.017).

Table 7 compares the association between KAP items.

Knowledge was not significantly associated with the item, ‘clinicians should discuss smoking cessation with patients even if it’s not the reason for the visit’. When attitude items were compared with knowledge, no significance was found except for the item ‘Smoking cessation counselling negatively affects my relationship with patients’ (p ≤ 0.001). Practice items, such as asking the patients how many cigarettes they smoke daily and previous attempts to quit smoking, were strongly associated with knowledge (p = 0.049 and 0.023), respectively.

This study included several HCPs with various levels of knowledge about smoking cessation guidelines and has identified several issues. Among the findings, most HCPs lacked knowledge of existing smoking cessation guidelines. This is similar to findings in a study by Mahoto, Mitonga and Oladimeji (2023), where most HCPs in Zambezi region, Namibia, were unaware of the national tobacco legislation. Batini et al. (2019) reported that HCPs are poorly equipped to support smokers willing to quit smoking. Banandur, Kumar and Gopalkrishna (2017) reported that in India, most of the persons responsible for compliance and authorised officers lacked appropriate awareness of the provisions in the Cigarette and Other Tobacco Products Act (COPTA) guidelines. The reasons for the lack of knowledge of the existing guidelines, as found by our study, may be because of the lack of priority on the subject; there appears to be no any form of continuing HCP education on smoking cessation in the study setting.

Most HCPs knew that smoking cessation advice given by a healthcare professional increased the patient’s chances of quitting smoking. This was similar to findings by He et al. (2021) in a study from China, which reported that doctors’ advice was associated with a smoker’s intention and action to quit smoking. Most of the HCPs were knowledgeable that most smokers would not successfully quit smoking without assistance. This was supported by similar suggestions from other studies on smoking cessation (Batini et al. 2019; He et al. 2021; Tamirat 2021; Van Zyl-Smit et al. 2013). Thus, despite HCPs’ awareness of the positive effects of smoking cessation counselling, there seems to be a lack of motivation to practice it.

Many HCPs were knowledgeable that smoking increases the risk of developing tuberculosis and mortality from tuberculosis, as reported in other studies (Batini et al. 2019; Van Zyl-Smit et al. 2013). Almost all HCPs were knowledgeable that smoking during pregnancy is unsafe for the baby. In a study involving midwives in the public sector antenatal clinics, Murphy, Steyn and Matthews (2016) reported that all midwives agreed that smoking during pregnancy harms the unborn baby. Chamberlain et al. (2017), in their study, reported that psychosocial interventions to support pregnant women could increase the number of women who stop smoking in late pregnancy and decrease the number of babies born with low birth weight.

Despite the existence of pharmacological interventions for smoking cessation and possible drug–drug interaction with other medications the patient might be taking, the HCPs in this study were more knowledgeable of only one type of pharmacological intervention – NRT. They were not knowledgeable about the other pharmacological modalities such as antidepressants and nicotine receptor agonists (NRA). The HCPs knew about NRT probably because it is the most commonly known of the pharmacological interventions but was not readily available in our study setting. Similar to this finding, a systematic review by Ilesanmi, Agwai and Afolabi (2021) reported a low percentage of physicians who prescribed pharmaceutical interventions to help patients quit smoking. This was because of poor knowledge and the unavailability of NRT.

Most HCPs were knowledgeable that combining counselling and medication was more effective than either intervention alone, which was confirmed by other studies reviewed (Banandur et al. 2017; Ilesanmi et al. 2021; Van Zyl-Smit et al. 2013). The combination of pharmacological medications and counselling will assist in addressing the psychosocial influence that causes a person to smoke or prevent them from overcoming their smoking addictions.

When overall knowledge was categorised in this study, most of the HCPs were found to have poor knowledge about smoking cessation. This is in keeping with findings from other studies on smoking cessation (Ilesanmi et al. 2021; Mahoto et al. 2023). The lack of training of HCPs on smoking cessation methods and non-implementation of available guidelines are likely to be blamed for the poor knowledge in our study setting.

Generally, smoking often leads to several adverse health consequences and is considered a basic health education topic that HCPs should emphasise. This study found that most HCPs agreed that counselling on smoking was an important part of their job. This was supported by other studies (De Frel et al. 2022; Ilesanmi et al. 2021; Van Zyl-Smit et al. 2013). However, Mahoto et al. in their study in the Zambezi region, reported that although the majority of HCPs regard themselves as providers of preventive healthcare, they do not necessarily consider the provision of smoking cessation intervention as part of their duty (Mahoto et al. 2023).

Offering basic smoking cessation advice is likely to assist in further encouraging those contemplating quitting to take the bold step. Most HCPs in this study disagreed with the statement that it is not worth discussing the benefits of smoking cessation with patients as they already know they should quit. Carsten and Linley share a similar view, as they suggested that even with minimal advice, the likelihood of patients quitting smoking increases (Carstens & Linley 2020).

Healthcare providers were unsure if the patients’ acute health conditions should take precedence over smoking counselling. Their uncertainty regarding this statement is understandable; while the significance of smoking cessation is undisputable, it is crucial to consider the timing of this intervention when faced with acutely ill patients. This is in keeping with findings by Mahoto et al. (2023) who reported that addressing the patients’ immediate healthcare needs took precedence over smoking cessation counselling.

Half of the HCPs disagree that time constraints are an issue for them. This contrasts with reports from other studies where time constraints have been noticed to be a reason why HCPs skip smoking cessation counselling (De Frel et al. 2022; Mahoto et al. 2023). Despite time constraints in healthcare delivery, incorporating vital health promotion counselling remains crucial and can be achieved through effective time management.

The majority of HCPs agree that patients do not comply with the information given on smoking cessation, and this was supported by a report by Klink et al. (2011) in a study involving HCPs in China, where only 13.3% of the respondents believe their patients will comply. Although patients have to take responsibility for their health, designing healthcare interventions to enhance compliance with HCPs’ advice should be encouraged. Most HCPs also agree that smoking should be included as a ‘vital sign’ for patients attending primary care facilities in keeping with the guidelines (Van Zyl-Smit et al. 2013). This will assist in reminding HCPs to provide smoking cessation counselling during the consultation with patients.

The study benefitted from a high response rate from the required number of respondents, which enhanced the accuracy and data quality. However, as the study was conducted in only one municipality in the Northern Cape, the external validity of applying these findings to the entire Northern Cape and South Africa is limited. Furthermore, the use of self-reported interviews with HCPs may have introduced bias, as they might have over-reported their provision of smoking cessation services. Additionally, the study’s ability to differentiate between the expected knowledge level for each professional group was limited, as a doctor may be more likely to possess better smoking cessation guideline knowledge compared with a professional nurse.