Capability

In the area of capability, most studies reported that clinicians lacked knowledge, skills and training to deliver smoking cessation advice. The most cited barrier was ‘Lack of knowledge’, cited in 56% of studies. The second most cited was a ‘Need for additional training’, cited in 39% of studies. Indeed, the finding that these attitudes were significantly less likely to provide cessation advice has repeatedly been shown in previous literature (Forman, Harris, Lorencatto, McEwen, & Duaso, Reference Forman, Harris, Lorencatto, McEwen and Duaso2017; Preechawong, Vathesathogkit, & Suwanratsamee, Reference Preechawong, Vathesathogkit and Suwanratsamee2011; Siddiqi, Dogar, & Siddiqi, Reference Siddiqi, Dogar and Siddiqi2013). These findings highlight a clear need for enhancing clinicians’ self-efficacy to deliver such advice. Over a quarter of the publications reviewed acknowledge the problem of a ‘lack of skills’ on the part of clinical practitioners to effectively promote smoking cessation interventions. This failing in communicative techniques is also evidenced by prior research on the subject, most notably in a 2013 Cochrane review by Carson et al., which shows that physicians who are trained in cessation delivery are more likely to use these skills and that their patients are more likely to quit smoking successfully (Berland, Whyte, & Maxwell, Reference Berland, Whyte and Maxwell1995; Carson et al., Reference Carson, Verbiest, Crone, Brinn, Esterman and Assendelft2012).

Interestingly, two separate publications in this review, acknowledged the negative impact that hospital facilities themselves can have on intervention effectiveness, encapsulated by the ‘absence of smoke-free hospital campus’ in the inpatient setting. This concept is supported by previous research regarding patients’ smoking in a smoke-free hospital and this particular finding highlights the necessity of making the inpatient setting a tobacco-free zone (Rigotti et al., Reference Rigotti, Arnsten, McKool, Wood-Reid, Pasternak and Singer2000). Overall, these findings not only match the capability dimension of COM-B, they also provide important evidence that these are subsumed within the policy context of the Behaviour Change Wheel, and that interventions to address these barriers will be rooted within an overall hospital or health services policy.

Opportunity

Most studies identified that a lack of time, support or resources were barriers to smoking cessation interventions. The most cited barrier was ‘Lack of time’, cited in 78% of studies, which was also the most cited barrier overall. The second most cited was a ‘Lack of support’, cited in 44% of studies. This perceived absence of adequate intervention support from colleagues hospital administration or from primary care physicians, could be seen as a policy issue, to be addressed. Perceived lack of professional support contributing to an inability to deliver smoking cessation interventions has been identified in previous research and eight different papers here support this view (Lala, Csikar, Douglas, & Muarry, Reference Lala, Csikar, Douglas and Muarry2017). The issue of time constraints being a leading obstacle to smoking cessation interventions has been well explored in previous literature and is supported by our findings here (Brotons et al., Reference Brotons, Bjorkelund, Bulc, Ciurana, Godycki-Cwirko and Jurgova2005; Twardella & Brenner, Reference Twardella and Brenner2005). However, this review specifically highlights healthcare workers’ difficulty with not possessing a clear hospital-directed mandate to intervene in smoking cessation. A heavy or overwhelming workload has been previously described by similar research (Carson et al., Reference Carson, Verbiest, Crone, Brinn, Esterman and Assendelft2012). The managerial issue of sparse resources for smoking cessation interventions was identified in over a third of the publications reviewed. This widespread recognition was expected, given the number of previous publications that have referenced the issue (Earnshaw et al., Reference Earnshaw, Richter, Sorensen, Hoerger, Hicks and Engelgau2002; Kanodra et al., Reference Kanodra, Pope, Halbert, Silvestri, Rice and Tanner2016; Sarna et al., Reference Sarna, Bialous, Wells, Kotlerman, Wewers and Froelicher2009).

It could be argued that a ‘lack of support’, ‘lack of resources’ and ‘absence of mandate to intervene’ are barriers which are similar enough to be grouped as one heading, such as a ‘lack of structural support’, but given how frequently each barrier was specifically referenced throughout the studies in the review, it was decided to cite each barrier individually, to provide maximum clarity for healthcare professionals. Similarly, we decided that the barrier of a ‘heavy or overwhelming workload’ was conceptually different from a ‘lack of time’, as the three papers, which referenced this barrier, cited it as a separate issue from the large workload. This allows us to highlight both of these issues as unique factors which need to be ameliorated to provide the best possible care in the inpatient setting.

Motivation

In the COM-B model category of motivation, the most cited barrier was a clinician-expressed belief in a ‘Lack of patient motivation’, cited in 44% of studies. This had been identified by various publications as a substantial factor to poor provision of smoking cessation interventions and so, it is unsurprising that it was reported in the papers reviewed (Cabana et al., Reference Cabana, Rand, Powe, Wu, Wilson and Abboud1999; McLeod, Somasundaram, Howden-Chapman, & Dowell, Reference McLeod, Somasundaram, Howden-Chapman and Dowell2000; Mohiuddin et al., Reference Mohiuddin, Mooss, Hunter, Grollmes and Hilleman2007; Sarna et al., Reference Sarna, Bialous, Wells, Kotlerman, Wewers and Froelicher2009). This perception stands in sharp contrast to patient reports and behaviour; in numerous studies, a majority of patients (typically around 70%) want to quit and most are willing to make a quit attempt. Consistent with this, cessation interventions that use proactive recruitment at the population level typically enrol a large percentage of eligible smokers (Fu et al., Reference Fu, Van Ryn, Sherman, Burgess, Noorbaloochi and Clothier2014; Meyer et al., Reference Meyer, Ulbricht, Baumeister, Schumann, Ruge and Bischof2008; Tzelepis et al., Reference Tzelepis, Paul, Wiggers, Walsh, Knight and Duncan2011). The second most cited barrier was a ‘Lack of confidence’, cited in 38% of studies. This lack of clinical confidence in addressing patient's smoking practices is noteworthy as previous research had suggested that healthcare workers were largely confident and willing to raise the topic of smoking cessation with patients (Sarna, Wewers, Brown, Lillington, & Brecht, Reference Sarna, Wewers, Brown, Lillington and Brecht2001; Sarna et al., Reference Sarna, Bialous, Wells, Kotlerman, Wewers and Froelicher2009; Willaing & Ladelund, Reference Willaing and Ladelund2004). Our review indicates that a large portion of healthcare workers still view themselves as lacking confidence in addressing this issue. Personal discomfort on the part of clinicians when broaching the subject of tobacco cessation interventions has been previously identified (Pipe et al., Reference Pipe, Sorensen and Reid2009). One study identified the barrier of smoking being seen as ‘a coping mechanism for patients’ by healthcare workers, which has been similarly described in previous literature (Acquavita, Talks, & Fiser, Reference Acquavita, Talks and Fiser2017). Prior research has demonstrated that providing training to physicians and other healthcare workers substantially boosts their efficacy, which then translates into improved cessation intervention delivery (Carson et al., Reference Carson, Verbiest, Crone, Brinn, Esterman and Assendelft2012). It is possible that such attitudes could be addressed at an earlier stage, with students being trained to provide cessation counselling prior to qualification (Kumar et al., Reference Kumar, Ward, Mellon, Gunning, Stynes and Hickey2017). This attitude was seen as an altruistic act on the part of the healthcare professional and is interesting to note as it strays from the generally accepted view that all tobacco use, by inpatients, should be ended. These deficits could be ameliorated through implementation of targeted education and training programmes for healthcare providers.

Not viewing smoking cessation intervention as a priority has been previously reported (Rice, Hartmann-Boyce, & Stead, Reference Rice, Hartmann-Boyce and Stead2013) with over a quarter of publications in the present review acknowledging this barrier. This attitude requires examination at a training and educational level to ensure adequate provision is made to aid patients in quitting smoking. The potential barrier of insufficient healthcare worker motivation or interest, which was described by four different publications, also has precedence in previous research (Borrelli et al., Reference Borrelli, Hecht, Papandonatos, Emmons, Tatewosian and Abrams2001; Hall, Vogt, & Marteau, Reference Hall, Vogt and Marteau2005). This belief is heavily linked with the reported belief that healthcare workers require financial reward or career recognition to initiate smoking cessation interventions, which was also identified in four studies in our review, as a ‘lack of incentive’. This attitude runs contrary to the typically altruistic notion of healthcare workers, but it also has been cited in previous literature, especially in the outpatient setting (Brotons et al., Reference Brotons, Bjorkelund, Bulc, Ciurana, Godycki-Cwirko and Jurgova2005; Fu et al., Reference Fu, Van Ryn, Sherman, Burgess, Noorbaloochi and Clothier2014). Our review demonstrates that this perceived insubstantial incentive, for addressing smoking cessation, is equally present in the inpatient environment, albeit this may depend on the healthcare system in which professionals are employed.

Four publications reviewed cited, the scepticism healthcare workers may have regarding smoking cessation intervention effectiveness. This was seen as a tangible obstacle to delivery of interventions and is another attitudinal belief identified by our review, which is congruent with previous research on the subject (Tong, Strouse, Hall, Kovac, & Schroeder, Reference Tong, Strouse, Hall, Kovac and Schroeder2010). It is interesting to note that three of the reviewed studies detailed the negative impact a physician's own smoking history could have on the success of cessation intervention. While healthcare worker tobacco use has been reported before (Pipe, Sorensen, & Reid, Reference Pipe, Sorensen and Reid2009), it is interesting that this personal practice would impact on providing optimal care for patients. A previous research study, in Syria, showed that physicians who smoke were less likely to deliver cessation interventions, support non-smoking policies and believe that smoking was harmful (Asfar, Al-Ali, Ward, Vander Weg, & Maziak, Reference Asfar, Al-Ali, Ward, Vander Weg and Maziak2011). This suggests that to increase the effect this potential barrier has upon future interventions, healthcare workers should be mindful of not permitting their own personal tendencies and habits from negatively affecting patient care.

The one noted instance of ‘social pressure’ being seen as a barrier to a healthcare worker providing smoking cessation intervention is unusual. Although there is precedence for it in the literature, it may be specific to the demographic of male physicians in China, where these studies were carried out (Cheng, Ernster, & He, Reference Cheng, Ernster and He2000; Kohrman, Reference Kohrman2004). Future research should investigate whether this barrier is encountered in other regions.

Finally, three studies expressed the belief that healthcare workers ‘negative past intervention experience’ could impact the success of smoking cessation interventions. This viewpoint has been identified by previous research on the subject (Fu et al., Reference Fu, Van Ryn, Sherman, Burgess, Noorbaloochi and Clothier2014). The fact that such a small portion of our studies reviewed identified this (17%) suggests that it is perhaps not as pressing a failing as some of the more common barriers which are categorised under the banner of motivation. Nevertheless, to achieve the optimum outcomes for smoking cessation interventions, each of the 20 barriers identified by this review must be considered as potential areas for improvement, in the inpatient setting.

An interesting point to note is viewing these various studies in the context of self-efficacy to deliver cessation interventions. Several of the barriers identified in the study detail a failing in healthcare professional efficacy, such as ‘lack of confidence’, ‘lack of healthcare worker interest or motivation’ and ‘lack of incentive’. Negative past experiences likely reflect poor efficacy, due to awareness of one's lack of expertise and/or lack of success in getting patients to quit, which makes the physician reluctant to intervene in the future. For healthcare workers to provide the optimum care for patients, it is crucial that each identifies the personal efficacy areas, in which they can improve, to correctly deliver clear and effective intervention advice.