Barriers experienced by nurses providing smoking cessation support to disadvantaged, young women during and after pregnancy

Abstract In Europe, smoking during and after pregnancy is still highly prevalent among socioeconomically disadvantaged women. Nurses caring for these women can play a key role in smoking cessation, but encounter many problems when providing support. This research aims to identify barriers in providing smoking cessation support, experienced by nurses working within a Dutch preventive care programme for disadvantaged young women (VoorZorg), and to understand the underlying reasons of these barriers. Sixteen semi‐structured interviews with nurses were performed. All interviews were recorded, transcribed and analysed deductively and inductively. We found that the VoorZorg programme provided nurses with training, resources and time to deliver smoking cessation support. Yet, nurses experienced important barriers, such as unmotivated clients and support methods that do not fit clients’ needs. Underlying reasons are competing care demands, unsatisfactory training for cessation support, lack of self‐efficacy in attending their clients, and conflicts with own professional attitudes. The results emphasise that nurses’ ability to provide smoking cessation support could be improved by proper training in interventions that fit their clients’ needs, and by time schedules and task definitions that help them to prioritise smoking cessation support over other matters.

the Dutch language and (e) face at least one additional risk factor (e.g. lack of social network, financial problems, psychosocial symptoms, drug/alcohol use and domestic violence). Certified specialised nurses support these women during home visits in a period of 2.5 years on multiple domains, such as health promotion and personal development (Mejdoubi et al., 2014). VoorZorg is offered in addition to standard maternal care in the Netherlands and includes regular check-ups by midwives or obstetricians (Mejdoubi et al., 2013). Clients may also receive help/support from other organisations, including youth care and mental healthcare, on issues such as finances, social work, domestic violence and child abuse.
Currently, the provision of smoking cessation support is standard unconditional care within the full VoorZorg programme. It includes inquiring about clients' smoking behaviour and exposure to passive cigarette smoking during every home visit. Nurses are instructed and trained to offer an intervention, named "V-MIS" to clients that smoke. Duration of the intervention depends on the pace with which the client proceeds through V-MIS. V-MIS, also used by Dutch midwives and gynaecologists, is a minimal intervention strategy based on the Integrated Change Model and transtheoretical model, adapted to pregnant women (de Vries, Bakker, Mullen, & van Breukelen, 2006;see Box;Bakker, Mullen, de Vries, & van Breukelen, 2003).
A previous study showed that at the onset of the VoorZorg programme (16-28 weeks of gestation), 43% of the clients smoked cigarettes. This was reduced to 33% at 32 weeks of gestation. Two months after birth, 48% of VoorZorg clients smoked cigarettes (Mejdoubi et al., 2014). Thus, although the prevalence of clients that smoked during pregnancy reduced, this reduction was modest and reverted after birth.
Our aims were to assess the barriers that nurses experience in providing smoking cessation support to socioeconomically disadvantaged young women during and after pregnancy, and to understand the underlying reasons of these barriers.

| ME THODS
In this observational study, we conducted semi-structured qualitative interviews among nurses who provided smoking cessation What is known about this topic • Various factors influence healthcare professionals' provision of smoking cessation support (e.g. training, experience and time available).
• Healthcare professionals find it hard to provide smoking cessation support to socioeconomically disadvantaged women.
What this paper adds • Barriers (e.g. clients' lack of motivation to quit, nurses' other priorities, inadequate training, support methods and lack of self-efficacy) persist even within a preventive care programme specifically designed for attending to socioeconomically disadvantaged young women during and after pregnancy.
• The results emphasise the need to train nurses in smoking cessation support methods that fit clients' needs and their poor motivation for smoking cessation. Also, the results underline that pregnancy care programmes should enable nurses to provide smoking cessation support next to the many other priorities. support to disadvantaged young women during and after pregnancy within the VoorZorg context. All participants gave their informed consent verbally, which was audio-recorded. The Medical Ethics Review Committee of Amsterdam UMC confirmed that the Medical Research Involving Human Subjects Act does not apply to this study, and thus no official approval of the committee was required.

| Participants and recruitment
The study population consisted of nurses. Inclusion criteria were that nurses should be employed as VoorZorg nurse in the Netherlands, and have at least 6 months experience in providing smoking cessation support to VoorZorg clients. Managers of Youth Health Care organisations executing the VoorZorg programme were informed about the study and asked for permission of their nurses to participate in this study. Permission was granted by nine managers. Four managers denied permission due to time constraints (N = 2), and due to insufficient experience (N = 2).
Subsequently, nurses were informed about the study during a conference and were later asked by email whether they wanted to participate in the study. A reminder email was sent to nurses that had not replied. Inclusion was based on consecutive sampling.
Sixteen out of 36 nurses meeting the inclusion criteria participated (participation rate 44%). They worked at eight different organisations scattered over six regions in the Netherlands. Most non-participants did not respond to the email invitation, the four non-participants that did respond indicated that they denied participation due to other priorities (N = 2) or due to limited experience in providing smoking cessation support (N = 2).

| Data collection
Data were collected by telephone interviews (conducted by MD) held between June and October 2017. One interview was face-toface and took place at the workplace of the nurse. All interviews were in Dutch. Interviews were one-off and lasted approximately 45 min. Since data were collected during work hours, organisations received a compensation of €25, per participating nurse.
Participants were informed that the research aimed to gain insight into smoking cessation support that participation was voluntary, and that data would be anonymised. Interviews were audio recorded and field notes were made. A semi-structured interview guide was used, based on the framework of Fleuren, Wiefferink, and Paulussen (2002) which describes that, applied to our research, provision of smoking cessation support (i.e. implementation) can be influenced by characteristics of (a) the context of VoorZorg, (b) providing smoking cessation support, and (c) nurses. Topics covered in the interview guide were the current practice of smoking cessation support and perceived facilitators and barriers of the current practice based on their full work experience. Afterwards, participants completed a short questionnaire to collect baseline characteristics.

| Data analysis
Raw interview data were transcribed verbatim, except the data of one participant due to a failed recording. Prior to coding the data, a coding tree was prepared based on the interview guide (i.e. framework developed by Fleuren et al. (2002), see above), literature (Flemming et al., 2016;Seeleman, Suurmond, & Stronks, 2009) and our research questions (see Appendix I). MD analysed all data in three rounds of coding, using MAXQDA 12. Data analysis was an iterative process.
First, data were coded deductively using the predefined coding tree (i.e. directed content analysis; Hsieh & Shannon, 2005). Data analysis was complemented with an inductive approach: by constant comparison within and between interview transcripts. Newly emerging codes were added to the coding tree (i.e. open coding; Boeije, 2009) and applied to pieces of text that could not be coded with codes from the initial coding tree. Previously coded interviews were reanalysed for presence of the newly added codes. Next, relationships between codes were (re)defined: generating core-and subthemes (i.e. axial and selective coding; Boeije, 2009).
MF randomly and independently coded three interview transcripts in the first round of coding. MF and MD agreed on the majority of the coding, and the remaining disagreements between authors were resolved in a consensus session. These disagreements entailed pieces of text coded by one author, but not the other and vice versa.
Box 1 Seven steps of V-MIS (Bakker et al., 2003) Step 1. Classifying pregnant women to their stage of change, the first step is to identify her (and her partners') smoke profile and motivation to quit smoking.
Step 2. Increasing the motivation of the pregnant woman to quit smoking, adjusted to women's stage of change.
Step 3. Discussing and removing barriers to smoking cessation and mobilising support in the immediate environment (directed at women in the preparation stage).
Step 4. Choosing a date with the pregnant woman (and her partner) for smoking cessation (preparation stage).
Step 5. Distributing a brochure and flyer of tailored advice to pregnant woman and discussing potential useful tools (executed to women at all stages of change).
Step 6. Following consultations after quit date, returning to the topic smoking cessation and providing after care (action stage).
Step 7. Preventing of relapse after delivery (executed to women at all stages of change).
Regular research meetings were held to foster reflexivity in the coding process. Authors agreed that saturation was reached, at the moment that no new themes emerged during the last two interviews.
After data analysis, a native English speaker translated citations from the transcripts to English.

| Participant characteristics
Nurses in our sample were on average 47 years old (range 31-62) and had 7 years of experience as VoorZorg nurse (range 0.5-11).
Nurses worked an average of 20 hr per week as VoorZorg nurse.
At the time of this study, nurses took care of 143 clients in total, of which 51 (35.7%) were current smokers.

| Themes
Data analysis resulted in three themes of nurses' perceptions and experiences in providing smoking cessation support to socioeconomically disadvantaged young women during and after pregnancy: (a) the VoorZorg context, (b) barriers in providing smoking cessation support, and (c) underlying reasons.

| Theme 1 -The VoorZorg context
Nurses considered providing smoking cessation support their responsibility. While the planned home visits are typically rich in topics that need to be discussed, nurses experienced no time constrains to provide smoking cessation support: In addition, nurses mentioned being equipped with and supported by materials of the VoorZorg programme to provide smoking cessation support and being trained to use these materials. Lastly, nurses emphasised that the way of providing smoking cessation support (i.e. motivational interviewing, empowering techniques) is used throughout the VoorZorg programme, making it compatible with the overall programme.

| Theme 2 -Barriers in providing smoking cessation support
Nurses immediately came across barriers following the implementation of V-MIS (i.e. step 1 identifying the smoke profile/ stage of change of a client). Since client support starts early after enrolment in the programme, nurses explained that they could not rely on their trusting relationship with clients in discussing sensitive topics such as smoking cessation, at that point.
Clients are generally reluctant to discuss their smoking habits at that stage: They really don't want you to come with a thing about smoking.
( Participant 3) Nurses also noticed that they tended to have the least motivated smoking clients with high levels of addiction, because more motivated clients already quit smoking before the onset of the VoorZorg programme: If they [clients] are [motivated to quit smoking], than they have often quit before we come into the picture, because they are able to flip a switch themselves and then we [nurses] do not have to do anything. It seems like either they quit, or they cannot quit and it will not happen.

(Participant 5)
With this unmotivated group of smoking clients, nurses found it very hard to enter the next steps of V-MIS, as they perceived that their clients should at least be somewhat motivated to continue providing smoking cessation support.
Most nurses indicated that they get stuck in step 2 of V-MIS, which entails increasing clients' intrinsic motivation to quit smoking, adjusted to clients' stage of change. Nurses believed that clients have other urgent matters on their minds and smoking often serves as a coping mechanism to deal with those matters: The motivation of the girls to quit is just not there. I try to empower them and to look at possible advan- In step 3, nurses and clients are also supposed to make a plan of how to overcome the clients' barriers. This is in itself an obstacle to clients, nurses explained that the support becomes theoretical rather than practical, resulting in a loss of interest among the clients: But it is also a barrier to nurses: nurses described once again that many things happen in the lives of the clients, causing plans to fail.
This results in disappointment among clients, which nurses wanted to protect them from: When making a plan, you have to think about when it might go wrong. But often so many things already go wrong in their lives, that the plans often already fail.
And that sounds incredibly negative, but that is often the reality. Whereby it [smoking cessation] is often such a disappointment, and that is what you want to protect the girls from. (…) That is a barrier: at a certain point you have to go through the various steps, but 9 out of 10 times, it is a disappointment.
(Participant 3) The last task in step 3 is to mobilise social support in clients' immediate environment. According to nurses, this mobilisation is impeded since most of the relatives and friends in the direct social environment of clients smoke themselves. This also makes it harder for clients to achieve (prolonged) smoking cessation, and for nurses to provide support:

Contextual barriers
A frequently reported underlying reason for barriers to provide smoking cessation support within VoorZorg, was that other (more urgent) issues are prioritised. Clients often wanted to discuss other topics that are more important to them than smoking. It seems that nurses want to comply with these needs, but also that nurses themselves are convinced that other matters often have more priority than providing smoking cessation support: They stressed that the way they provide support (e.g. their tone) is of great importance as it could easily provoke resistance among their clients: That is the way you make it [smoking cessation] discussable, that is important. That is what clients tell us, for example the midwife tells them (…) "you have to quit smoking because it is bad". So that way of saying that to them (…) provokes resistance.

(Participant 1)
Nurses also experienced situations in which clients claimed that their midwife/gynaecologists permitted them to smoke a few cigarettes: She even says that the midwives and gynaecologist told her to smoke a few cigarettes otherwise she would get too stressed out. (Participant 13) Yet, nurses are convinced that other healthcare professionals would not permit smoking during pregnancy: That is not true, I know that for sure. Because every midwife says that you're not allowed to smoking during pregnancy. (Participant 4) Nevertheless, nurses seemed to realise that whether other healthcare professionals actually do or do not tell clients this, they have to refute this perception and that it is their word against that of a professional regarded as more of an expert in pregnancy matters.

Barriers in provision of smoking cessation support
Although not mentioned as a barrier by nurses themselves, gener- Consequently, nurses predominantly developed their own methods for providing smoking cessation support: We work with V-MIS.

| D ISCUSS I ON
This qualitative study among nurses showed that nurses experience important barriers in their roles, such as competing care demands, unsatisfactory training for cessation support, support methods that do not fit to clients' needs, lack of self-efficacy in attending their clients and conflicts with own professional attitudes. They also felt that clients were poorly motivated to stop smoking, due to multiple stressors and other challenges they had to face.

| Interpretation of findings
The VoorZorg programme provided a promising context for smoking cessation support to nurses by offering training, resources and time to deliver smoking cessation support. Barriers were found in the V-MIS intervention available to nurses, which did not meet clients' needs for smoking cessation support. Due to this lack of congruence, most nurses experience difficulties in implementing V-MIS. We found that most barriers were face in the first steps of V-MIS. This is in accordance to the findings of other studies that found that, when moving through the stages of V-MIS, more and more healthcare professionals did not make it to the next step (Segaar, Willemsen, Bolman, & De Vries, 2007;Oude Wesselink, Lingsma, Robben, & Mackenbach, 2015). In terms of the transtheoretical model of health behaviour change, clients of nurses in our study were at the (pre)contemplation stage, that is, not ready or unmotivated to change their smoking behaviour), as was observed in another study (Bull, 2007). V-MIS, appears to undervalue addressing the early stages, like most traditional health promotion programmes (Prochaska & Velicer, 1997). V-MIS only attends these stages in step 2, when healthcare professionals attempt to increase clients motivation, following steps of V-MIS are directed to preparation and action stages.
Nurses reported that relatives and friends of clients are a barrier early in the process of provision of smoking cessation support as they often smoke themselves. Other qualitative research showed that having a partner, close family or friends that smoke could impede smoking cessation of women that smoked during pregnancy as they provide temptation to the women (Koshy, Mackenzie, Tappin, & Bauld, 2010). The importance of social context is also underlined in the Theory of Planned Behaviour which states that subjective norms of relevant others, besides attitudes and perceived behavioural control, influences behavioural intention and consequent behaviour (Ajzen, 1991). We also found that nurses often prioritised other matters over smoking cessation. This finding seems in contrast to the finding of Flemming et al. (2016), who found that healthcare professionals provide smoking advice to their clients even without support of the latter, because they felt this was their professional obligation. This may differ from our study in that our nurses were concerned with the lives of their clients in general, and all problems they have to cope with (e.g. housing and financial problems, lack of social network, psychosocial symptoms, substance use and domestic violence). Rather than being exclusively concerned with matters directly linked to pregnancy and physical health. Moreover, nurses reported inadequate training in smoking cessation support. Training of healthcare professionals in providing smoking cessation has been found to increase their readiness to carry out tasks, such as discussing quit dates and providing smoking cessation materials (Carson et al., 2012). We also found that, as the relationship between nurses and their clients is delicate, an emphatic provision of smoking cessation support might risk affecting their relationship. Flemming et al. (2016), who found that the lack of such a relationship constitutes a barrier to providing cessation support, underlines the importance of a positive, supportive relationship. Lastly, a concern mentioned by nurses in our study was that their clients pleaded that other healthcare professionals allowed them to continue to smoke a few cigarettes. A systematic review on pregnant women's views also found that women claimed that their healthcare professional permitted cutting down as an option besides to cessation (Graham, Flemming, Fox, Heirs, & Sowden, 2014). However, it remains uncertain whether this practice is widespread in The Netherlands. The Dutch National Expert Centre Tobacco Control disseminates to professionals the message that they should support their clients in complete smoking abstinence, no