Barriers to and facilitators of smoking cessation in pregnancy and following childbirth: literature review and qualitative study

Pregnant and postpartum women’s perspectives

Greater insight is required into how barriers to and facilitators of smoking cessation in pregnancy and after birth are perceived and experienced by women. To date, systematic reviews of qualitative studies involving pregnant women have had a broad focus on perceptions and experiences; barriers and facilitators may be identified but are not the primary concern. 47,48 For example, the systematic review by Flemming et al. 49 demonstrated how smoking in pregnancy was shaped by the contexts of women’s lives, including the embeddedness of smoking in their lives and the importance of it in the couple’s relationship. Although this review focused on studies of women’s experiences of smoking in pregnancy, it did not explicitly examine the barriers to and facilitators of smoking cessation faced by women while they are pregnant, and it also excluded papers that examined women’s experiences post partum.

Significant others’ perspectives

The smoking habits of women’s significant others (SOs) (i.e. partners, family and friends) are recognised as important barriers to sustained quitting,27,31,50 but these remain relatively under-researched in relation to smoking in pregnancy. Partners’ smoking status and attitudes to smoking cessation are identified as potentially important influences in a pregnant woman’s attempt to quit;30 for example, partners who try to quit at the same time as the pregnant woman can be seen as more supportive,51 whereas a partner’s failed quit attempt may reduce the chances of the woman succeeding. 52 Men may be less likely than their pregnant partners to receive advice to stop smoking from HPs and may be exposed to less pressure from friends and family to quit. 53 Further understanding of how SOs perceive the barriers to and facilitators of smoking cessation in pregnancy is needed, particularly from qualitative studies in which participants describe these issues in their own words.

Health-care professionals’ perspectives

The problems that women face in accessing effective support to stop smoking may involve structural, organisational and cultural barriers in the NHS or other services, rather than a reluctance or lack of knowledge among pregnant women. Recent studies in Scotland and England highlight a number of limitations to smokers being identified, referred and then supported to stop smoking. 54,55 For example, perceptions that HPs offered inconsistent advice on, and support for, quitting in pregnancy has emerged as a potential barrier. HPs themselves may fail to facilitate smoking cessation through, for example, midwives’ reluctance to ask about smoking, a lack of training in smoking cessation, organisational factors and HPs’ own smoking behaviour. 56 Fundamentally linked to the role of the HPs is the context in which services are delivered, the configuration and quality of which varies greatly throughout the country. To our knowledge, no review has been conducted of qualitative studies reporting HPs’ perceptions of the barriers and facilitators when addressing smoking in pregnancy. Given the pivotal role of many HPs involved in the care of women throughout their pregnancy, a better understanding is needed of the impact of the diversity of professional roles and practice, and how different organisational practices may both facilitate and create barriers to the delivery of cessation support that meets pregnant women’s preferences and needs.

Considering each of these three perspectives – pregnant and postpartum women, their SOs and HPs – the evidence suggests that a deeper understanding of what helps and hinders pregnant smokers to quit and remain abstinent post partum is needed; more specifically, how these barriers to and facilitators of smoking cessation are perceived and experienced by women and their families, and how they are accounted for by HPs in their approach to facilitating smoking cessation in pregnancy.

This study, therefore, takes into account the household and family context of smoking in pregnancy and compares and contrasts the views of women and their SOs with those of professionals involved in maternity and smoking cessation services. This is the first study in the UK to do this. By taking this more holistic approach, the study aims to both inform the delivery of existing interventions and serve as the foundation for the development of improved interventions to support pregnant women to quit.

Aims of the three reviews

1. To explore the barriers to and facilitators of smoking cessation experienced by women during pregnancy and post partum.

2. To explore the barriers to and facilitators of smoking cessation experienced by women’s SOs during pregnancy and post partum.

3. To explore HPs’ perceptions and experiences of the barriers to and facilitators of providing support for smoking cessation during pregnancy and post partum.

Approach to searching and data sources

For each review, we undertook comprehensive pre-planned searches that aimed to find all available studies. Search terms were developed in conjunction with an information specialist. A draft search strategy was developed for each review in the Cumulative Index to Nursing and Allied Health Literature (CINAHL), and these were then adapted to run in MEDLINE, PsycINFO, Social Sciences Citation Index, the Economic and Social Research Council database and Google Scholar Advanced (Google^TM, Mountain View, CA, USA), for both published and unpublished studies. In addition, a PubMed ‘ahead of print’ citation search identified papers yet to be indexed. We also used publication alerts to inform us of any papers published during the process of the review after the formal searches had been completed to ensure that the most up-to-date studies were included in the review. The reference lists of full-text papers were checked and consultation was undertaken with experts known to the project team, to identify papers not found through electronic searching. Details of the search strategies for each review are available in Appendix 1. Databases were searched from 1990 onwards for each review. As the reviews were undertaken sequentially, the end dates of searching varied as follows.

• Review 1: May 2013 (from January 2012). For the period to December 2011, searches for studies relating to smoking during pregnancy and post partum had been undertaken for an earlier review. 49 The recently published (September 2015) thematic synthesis of qualitative data on postpartum smoking relapse is therefore not included in this review. 84

• Review 2: January 2014.

• Review 3: January 2015.

Inclusion criteria

Published and unpublished studies were eligible for inclusion if they:

• were reported in English

• used qualitative research methods

• were conducted in a high-income country in which the stage of the cigarette smoking epidemic matched that reached in the UK (i.e. a strong association between social disadvantage and cigarette smoking).

Additional inclusion criteria relevant for each review were:

• review 1 – studies investigating the barriers to and facilitators of smoking cessation that women experience during pregnancy and post partum

• review 2 – studies investigating the views of women’s SOs of the barriers to and facilitators of smoking and smoking cessation in pregnancy and after childbirth

• review 3 – studies investigating HPs’ experiences of the barriers to and facilitators of supporting smoking cessation during pregnancy and post partum.

Data extraction and quality appraisal

Relevant data were extracted from papers in each review (aim, type and number of participants, methodology used, methods of data collection, analysis and results). The data were extracted and checked by two reviewers. The papers were appraised for quality by two reviewers using an established checklist,85 with disagreements in scoring resolved by consensus. There was no a priori quality threshold for excluding papers; assessment was undertaken to ensure transparency in the process.

Synthesis methodology

All three reviews were undertaken using meta-ethnography. 86 Meta-ethnography is an interpretative approach to research synthesis that enables conceptual translation between different types of qualitative research. This approach consists of four iterative stages (Table 1). In each review, the following processes occurred: in phase 1, two reviewers read all papers in depth. Phase 2 involved the line-by-line coding of data (participant accounts and authors’ interpretations) in each paper relating to the focus of the review using ATLAS.ti 2010 software (ATLAS.ti Scientific Software Development GmbH, Berlin, Germany).

The codes were then compared and grouped by the reviewers into broad areas of similarity through reciprocal translation analysis (phase 3) to generate a reduced set of codes (translations) about barriers and facilitators relevant to each review. Phase 4 focused on these translations; the reviewers examined and compared them to identify the ‘lines of argument’ emerging from the review.

Reporting

Reviews were reported in accordance with the Enhancing Transparency in Reporting the Synthesis of Qualitative Research (ENTREQ) statement,87 as recommended by the Equator Network for reporting syntheses of qualitative research (www.equator-network.org/reporting-guidelines/entreq/).

Aims of the three qualitative studies

1. To explore the barriers to and facilitators of smoking cessation experienced by women during pregnancy and post partum, and elicit views on interventions to support cessation.

2. To explore the barriers to and facilitators of smoking cessation experienced by women’s SOs during pregnancy and post partum, and elicit views on interventions to support cessation.

3. To explore the perceptions and experiences of HPs with a role in supporting women to stop smoking regarding the barriers to and facilitators of providing support during pregnancy and post partum, elicit HPs’ views on the barriers to and facilitators of cessation experienced by women during pregnancy and post partum, and views on interventions to support cessation.

Inclusion and exclusion criteria

Eligibility criteria for participants in each group are listed in Table 2.

Sampling strategy

Purposive sampling was used to achieve maximum diversity within the recruited sample of pregnant women, their SOs and HPs. A sampling frame was used to obtain, as far as possible, a sample of pregnant women recruited for interview that took into account maternal age, that is 25% of the sample were to be aged < 25 years including two pregnant smokers aged < 20 years, and deprivation (i.e. 75% of sample from postcodes with the lowest Indices of Multiple Deprivation), and included both continuing smokers and quitters at around 20 weeks’ gestation. The professional sample was selected to reflect the perspectives of a range of professional groups on smoking cessation in pregnancy, and the various intervention options and opportunities available.

At the outset, the intention was to interview the following number of participants over the two study sites, as indicated in Figure 3: up to 40 pregnant women who were self-reported smokers at maternity booking, 20 of whom had stopped smoking by 20 weeks’ gestation and 20 of whom had not; 10 of these women a second time at up to 6 weeks after the birth of their babies; around 40–50 of these women’s SOs; and 40 HPs with a role in supporting pregnant women.

FIGURE 3.

Proposed participants for qualitative study at two study sites.

These targets were achieved, albeit by recruiting over a longer time period than planned, for pregnant women (9 months vs. the planned 5 months), postpartum women (7 months vs. the planned 4 months) and HPs (10 months vs. the planned 4 months). Interviews with HPs commenced earlier than planned (October 2013 rather than June 2014) because of delays in obtaining research and development (R&D) approvals to recruit pregnant women. In addition, for pragmatic reasons (i.e. when there was a significant number of relevant staff in similar roles, e.g. smoking cessation advisors and midwives) HPs (n = 26) participated in a focus group or a paired interview rather than an individual interview. Difficulties in recruiting women’s SOs, however, meant that the original target for this group of participants was unable to be reached despite an extended recruitment period (10 months vs. the planned 4 months). Further details of the changes made to recruitment approaches are described in Recruitment of participants.

A total of 121 individuals recruited from NHS study areas A and B participated (Table 3). A description of the characteristics of each group of participants is presented at the beginning of each of the qualitative findings chapters for pregnant and postpartum women (see Chapter 6), SOs (see Chapter 8) and HPs (see Chapter 10).

Recruitment of participants

To encourage participation in the study, pregnant and postpartum women and SOs were offered a £15 Love2shop voucher as an incentive and as a way of thanking them for taking part in the study. No incentives were offered to HPs to participate in the study. SOs were also interviewed around the same time as the pregnant women rather than a few months later, as originally planned, following feedback from potential participants at the start of the study. Some interviews, however, still took place after the findings from the systematic review element of the study became available, thus allowing discussion of any new data emerging from the review to be incorporated into the interviews.

Interviews and focus groups

All interviews and focus groups were conducted by female researchers from the University of Stirling and University of York. At area A, two researchers (AF and JM) conducted the interviews with women and SOs. Nine of these interviews had two researchers present for the purposes of shadowing prior to maternity leave for one researcher (n = 4), observation of qualitative fieldwork by a visiting researcher from the University of Ireland, Galway (n = 3), and safety concerns about visiting some participants’ homes (n = 2). In all cases the second researcher did not participate in the interview process. JM completed approximately 10 interviews with women and partners and one focus group with HPs (smoking cessation advisors) before going on maternity leave in December 2013. Three project team members (LB, LS and JM) conducted the interviews and focus groups with HPs. At area B, one researcher (DM) conducted all of the individual and group interviews.

Development of topic guides

Prior to commencing interviews and focus groups with the three groups of participants, topic guides were prepared to support semistructured discussions. The guides (see Appendix 2) were informed by project team members’ previous and ongoing work relating to smoking in pregnancy, the theoretical framework used in the study and a mapping of current and new interventions in this area.

In the early stages of the study, project team members identified key themes (informed by the TPB) for inclusion in topic guides for each of the participant groups: smoking behaviour and quit attempts, beliefs and feelings about smoking in pregnancy, perceived barriers to and facilitators of smoking cessation in pregnancy, perceptions concerning the SSS, and views on available and in-development interventions for smoking cessation in pregnancy. These key themes were used as a basis for developing interview questions. The HPs’ topic guide was the first to be developed. This was initially produced by one member of the project team (FN) and reviewed by other members. Draft topic guides for interviews with pregnant women and SOs were then produced by another member of the project team (JM), taking account of relevant themes but also the style of the HPs topic guide. The draft topic guides were then reviewed by the wider project team for content, comprehensibility and ‘flow’ of questioning. A fourth topic guide was produced for interviews with postpartum women using a similar process of identifying relevant themes prior to refinement and modification.

Coding and analysis

Three separate groups within the project team analysed the data from the three different participant groups: pregnant and postpartum women (AF, JM and KA), SOs (JM, DE and KA) and HPs (FN, SH and LS). The analysis was guided by Braun and Clarke’s phases of thematic analysis88 and set within the interpretivist paradigm. This paradigm considers participant accounts elicited during research interactions as representing one of many possible ‘truths’ and recognises that interpretations of these interactions are influenced by the researchers’ knowledge, beliefs and values. For the analysis of the HPs focus group data, we did not analyse interactions or dynamics between participants (e.g. participants looking at each other, laughing at another participant’s comments, nodding), as these data were combined with individual interview data in which there were no interparticipant interactions. NVivo (version 10, QSR International, Warrington, UK) was used to facilitate the coding and analysis.

Coding frameworks were developed for each qualitative study separately using mind-mapping techniques. These frameworks were informed by the SEF. 74 The following process was repeated for each of the three qualitative studies: after familiarisation with the transcripts (i.e. reading and rereading), researchers independently microcoded several transcripts and all identified codes were categorised and inserted into one of the three SEF levels being examined, individual, interpersonal or organisational; after this researchers agreed on which codes to include in the draft framework, several further transcripts were coded and any new concepts (codes) that emerged were added.

The adequacy of the coding frameworks and consistency of coding between researchers was assessed using several methods. When dual coding was employed (i.e. 10 instances of dual coding for pregnant women’s transcripts and five for HPs transcripts), NVivo’s ‘coding stripes’ and/or coding consistency assessment query function was used to identify coding with potentially poor overlap (e.g. kappa coefficient < 0.7) that should be investigated. Although the coding consistency was acceptable for most codes, some codes had poor reliability having been applied inconsistently by the different researchers. These codes were discussed and resolved by making a number of minor changes to the coding framework, including the merging of some codes and changing of code labels and descriptions to maximise coherence and validity. For the SO transcripts, two researchers coded separate transcripts and then discussed the adequacy of the coding framework, including the addition or removal of certain codes.

Owing to similarities in the topics discussed and shared environments, the coding frameworks for interviews with postpartum women and SOs drew on the coding framework developed for pregnant women. A read-through of a sample of interviews from each of the study sites allowed further development in relation to key additions and omissions. The postpartum coding framework was particularly focused on any changes in perceptions, behaviour and support as the pregnancy progressed and once the baby was born. The SOs framework focused on participant’s relationships with the pregnant woman with reference to smoking and smoking cessation and perceptions of available support for cessation.

After all transcripts had been coded, the researchers read and reread the coded content, and summarised the key findings emerging from each code into higher-level categories. In line with ‘axial coding’,89 the researchers explored how the categories related to each other, leading to the emergence of preliminary themes. These themes were then refined through exploration and discussion of how well the preliminary themes mapped back onto the coded and raw data. Throughout this process, categories and themes were integrated into one of the three levels of the SEF being used to structure the codes, although interactions between categories and themes across these SEF levels were also examined. For all groups of participants, particular attention was paid throughout to similarities and differences in findings between research sites. In addition, similarities and differences in findings were examined between (1) women who had quit smoking and those who had continued to smoke through pregnancy, (2) partners and others with relationships with the pregnant women, and also SOs’ smoking status, and (3) the different groups of HPs, for example midwives, SSS advisors, etc. Once clear themes had been identified, these were named and described, accompanied by supporting quotations, as part of the process of writing up the findings. Supporting quotations were selected that generally expressed dominant views and demonstrated significant issues but also that reflected ‘deviant’ or ‘negative’ views. 90 As recommended when undertaking qualitative analysis,88 the process of analysis was recursive and often involved multiple iterations, particularly when identifying and refining themes from codes and categories.

Individual

An examination of women’s descriptions of their lives showed that many, in both research sites, lived in disadvantaged circumstances and faced a range of additional challenges associated with social disadvantage. Women described living in large households, not living with their current partner, being carers of children or parents in ill health, preparing to be single mothers, and being in financial insecurity with regard to welfare benefits and unstable jobs. Many described ongoing pregnancy and general health problems, including chronic conditions (depression, back pain, arthritis, asthma, diabetes). Smoking, a habit that was well established pre pregnancy, was described as a key part of their routines, as a break from work or family life, and as a way of coping with daily stress and boredom:

It’s just really hard, plus this wee one stresses you out sometimes and it’s like the only time I get like a 5-minute break sort of thing . . . I kind of look forward to that break.

Area A, pregnant woman 9, smoker

Most women said that they wanted to stop smoking in their pregnancy or remain abstinent after the birth. However, some, particularly in area B, said that they were satisfied with cutting down or continuing to smoke during the pregnancy. Risk perceptions and self-efficacy beliefs were key themes associated with smoking and smoking cessation.

Interpersonal

Organisational

Individual

Women reporting as non-smokers generally described finding it easy to stop smoking and remain abstinent. Some stated that protecting their baby from passive smoking had become a motivating factor for them after delivery. Others described how being able to purchase items for the baby using money saved from stopping smoking helped to keep them motivated. Several of these women said that they had not considered smoking because their lives were busy or stressful looking after a new baby. Others acknowledged that they had experienced a desire to smoke on occasion, usually at social events, but they had managed to resist:

I don’t have time [to smoke]. After the cleaning and looking after [baby] . . . I haven’t even thought about having a cigarette.

Area B, postpartum woman 3, non-smoker

When you are with other people that do [smoke], like at the wedding, everybody kept disappearing out for a cigarette and you are sitting there like that . . . and I think ‘why would you want to when you could think about what you could buy?’. Sometimes just going looking for something to focus on to buy him can be enough for me to think ‘well I can get him that if I don’t have them’.

Area A, postpartum woman 3, non-smoker

For the women who said that they had smoked throughout their pregnancy, refuting the risks of maternal smoking was a recurring theme. For example, one drew on her own children’s health as evidence of exaggeration of the risks. Another said that she found reassurance by receiving positive news from growth scans, relaxing her attitude towards smoking in pregnancy:

There is no proof I’d say because that’s me had two and they are both healthy, they have both been good weights, none of them were early and they’ve never had any problems.

Area A, postpartum woman 1, smoker

The motivation to quit was more at the start, and then as things progressed, it’s hard to explain, it was kind of like nothing had gone wrong; everything is fine, so what’s the point in torturing myself . . . it wasn’t ‘this baby is coming, I’ve got to stop’, it was kind of like ‘things are OK, I can get away with . . .’.

Area A, postpartum woman 2, smoker

All interviewees at follow-up were concerned about the effects of passive smoking on the baby and the baby’s exposure to toxins. Those who reported smoking described trying to minimise any risks by smoking outside the home, putting coats over their clothes or changing clothes, spraying toiletries, washing their hands and faces and using antibacterial gel:

We make sure that the living room door and the kitchen door are shut, but you can still you know put him in his cot, make sure he is safe and then we go out.

Area B, postpartum woman 5, smoker

I know when you are pregnant it goes into your blood seemingly, but when the baby is here, it’s a whole different . . . it seems worse to me . . . it seems to linger more, like on clothes and things like that . . . I wear a dressing gown when I am outside . . . wash my hands, use antibacterial . . . getting rid of as many toxins as we can.

Area A, postpartum woman 2, smoker

Interpersonal

The majority of interviewees described trying to influence some elements of the interpersonal environment. When visitors came to their homes it was usual for smokers to smoke outside. There appeared to be an implicit understanding that this is what people would do; several women mentioned that they had not needed to ask. Most women reported that when they visited friends or family members who were smokers these people usually went outside to smoke, either because this was where they usually smoked or because they had modified their behaviour because of the baby’s presence. In addition, a number of women said that should they ever be visiting people who continued to smoke inside with the baby present, they would leave:

Everyone smokes outside. That’s the rule . . . nobody’s ever asked . . . it’s just automatic that they go straight out.

Area B, postpartum woman 4, non-smoker

I say that to [partner] if we ever go to my mother’s and any of them are smoking we will just leave . . . We go to [visit sister-in-law] . . . she normally smokes in her kitchen, but she’s been going out the back since we’ve had the baby.

Area A, postpartum woman 3, smoker

The influence of partners came to the fore at follow-up. All of the women reporting as non-smokers said that their partners were also non-smokers, either because they had stopped smoking during the pregnancy or because they were former or never-smokers. The women whose partners had stopped smoking during pregnancy acknowledged that this was their main facilitator of success:

I would have struggled to stop [smoking if partner hadn’t stopped] because obviously I spend every minute of my life with him when he is not at work.

Area B, postpartum woman 3, non-smoker

[Partner] has helped me . . . when he quit as well I thought ‘oh well I definitely couldn’t go back to it because if I went back how much of a failure would I be when he smoked a lot more than me and he has really struggled to do it’.

Area A, postpartum woman 3, non-smoker

All but one of the women who said that they were smoking at postpartum interview had partners who also smoked. Their influence on the woman’s smoking was also shown:

I know for a fact [partner] will be a smoker for life . . . him still smoking is going to make it harder for me to quit one day.

Area A, postpartum woman 4, smoker

Organisational

Interviewees recalled little smoking-related discussion with HPs in the postpartum period. Some described being asked their smoking status by a midwife or a health visitor but then offered little support or encouragement to stop smoking if they said that they were smokers. Two of the women reporting as non-smokers said that they would have appreciated support after delivery to help them remain abstinent:

They sort of asked me if I was still smoking, that’s all they said to me. I think to be fair the midwives . . . they are busy.

Area B, postpartum woman 5, smoker

Maybe there should be a little note saying that they stopped smoking . . . like the health visitor never asked me if I needed patches or never asked me if I’d tried the chewing gum or like even just for someone to say, ‘listen if you need help, if you feel like that come to me’. . . it’s an ongoing thing, it’s always going to be there.

Area A, postpartum woman 5, non-smoker

Of the women who had accessed stop smoking support, those who said that they were smoking at follow-up generally reported more negative perceptions than those who stopped smoking during pregnancy. Three of those who reported smoking had engaged with a cessation advisor during pregnancy. After initial contact and meetings with an advisor, these women described subsequently avoiding the advisor’s telephone calls because they lacked motivation to stop smoking and, in one instance, were concerned over CO screening after relapsing. Two women said that, in hindsight, it would have been beneficial for them to maintain contact through face-to-face meetings, considered more helpful than telephone calls:

I do believe that if I had kept seeing her it might have given me that encouragement.

Area A, postpartum woman 4, smoker

She said ‘do you want to meet up again or do you want me to just phone you?’. And I think if I’d said to meet up again things might have . . . because you are face to face, whereas the phone calls were just ‘oh here we go’.

Area A, postpartum woman 2, smoker

Four of the women reporting as non-smokers had accessed the SSS. These women generally described positive experiences of the support, primarily because of the advisor–woman relationship but also because of NRT provision and CO screening. They felt that they could call on their advisor for support at any time, even after delivery, although nobody had done this:

She sent me a letter and said that if I am ever at that thinking point I’ve just to give her a call . . . she will help me through it whether that’s talking or offering some kind of help. I know that I could always call her.

Area A, postpartum woman 5, non-smoker

Smoking and smoking cessation

Analysis of the qualitative interviews indicated that many different factors within women’s lives interacted to influence their smoking behaviour either positively or negatively. A SEF was used to unpick these factors into different levels of women’s environments: individual, interpersonal and organisational.

Overall, women’s perceptions and experiences were mixed, demonstrating that individual factors can act as barriers to or facilitators of smoking cessation depending on the context of women’s lives. However, there appeared a predominance of barriers at the individual level. Challenges related to a high incidence of disadvantaged circumstances and lack of self-belief in the ability to stop smoking made smoking cessation difficult for some women. Although all women acknowledged some potential harm from maternal smoking, risk perceptions were varied, with some women expressing uncertainty on this issue.

Women’s beliefs about smoking in pregnancy were greatly influenced by close relationships with family and friends at the interpersonal level. The accounts of women suggested that risk messages from their social networks often prevailed over information delivered by HPs, which women reported was received some way into the pregnancy (usually at booking, at around 8–12 weeks’ gestation) and described as inconsistent. Other women’s stories relating to smoking in pregnancy were often used to refute the risks of smoking. Among some women, however, these could also be used as motivation to stop smoking. It appeared women looked for, and received, reassurance that their choices around smoking were acceptable, especially from family and friends, but also from HPs, midwives in particular.

Women’s accounts also suggested that at the interpersonal level partners could significantly influence their smoking behaviour. Women described a high occurrence of partner support – emotional and practical. However, although this was generally appreciated, a partner’s smoking behaviour appeared to have a greater influence in facilitating smoking cessation or making it difficult for women to stop.

Within the organisational level women described how workplace routines, demands and relationships could make it easier or more challenging for them to stop smoking. Negative attitudes, expressed or perceived, from all those around them could lead to a defiant attitude in some women with regard to their smoking.

For the 10 women interviewed post partum, there was little change in their perceptions of the factors identified at the different levels of the SEF. However, the importance of some themes was shown at follow-up. There appeared to be a difference between those who accessed and kept attending stop smoking support and those who did not. Among those who stopped attending, there was a view that if they had maintained face-to-face contact with their advisor, this might have helped them to quit. In addition, post partum, the difference between the perceptions of risk from maternal smoking and the perceptions of risk from passive smoking came to the fore. Although risk perceptions of smoking in pregnancy were mixed, in the postpartum period all women were extremely concerned about their baby’s exposure to passive smoke and had detailed plans in place for minimising this. At follow-up, women described how the influence of their partner’s smoking status was crucial to their success or otherwise in stopping smoking and remaining abstinent. In the postpartum period, women also described an apparent lack of smoking-related discussion, advice and support by HPs.

Improvements to existing services/interventions

Factors emerging from the health-care setting were mostly helpful to women. Engagement with SSSs was key to women accessing support, although some described being discouraged by a negative service image. In this regard an opt-out referral pathway best facilitated women’s engagement. Although perceptions of SSSs were mixed, they did contribute to many women being successful in stopping smoking, although the study recruitment procedures likely influenced this finding. Women often described a positive relationship with their cessation advisor. This was a key feature in the continuity of stop smoking support and providing women with detailed information and advice. Women especially liked being offered flexibility in service delivery, home appointments and one-to-one support. There were mixed perceptions surrounding the use of NRT in pregnancy but some described successfully quitting using patches and the inhalator. CO screening was a powerful motivational tool for many women.

Interpersonal

At the interpersonal level, barriers and facilitators were identified in SOs’ own smoking behaviour, attitudes to smoking in pregnancy, relationships with pregnant women and the ways in which they attempted to support the pregnant women’s quit attempts. The influence of others within women’s wider social network was also shown. Mention of pregnant women’s physical environment, with specific regard to the smoking behaviour of others, is incorporated into discussion of interviews with reference to both interviewees and members of wider social networks.

Organisational

At the organisational level, SOs identified barriers to and facilitators of smoking cessation for pregnant women in their contacts with statutory health-care services, encompassing initial discussion of smoking in pregnancy with HPs; the nature of the cessation support provided; CO monitoring; and the use of NRT.

Smoking

Smoking or former experience of smoking was common among SOs; however, smoking itself was largely perceived as undesirable. Attitudes to smoking in pregnancy were even more stringent, with most SOs showing disapproval and a preference for women to quit during pregnancy. This largely remained the case, even when interviewees had experienced other women smoking during pregnancy with no apparent ill effects for the baby. SOs also expressed concern about exposing babies and children to SHS.

According to participants’ accounts, smoking appeared to have an important role within relationships between SOs and pregnant women, although this was not always acknowledged by interviewees themselves. Stopping smoking, however, was often recognised as a fraught issue for relationships, with SOs admitting to avoiding placing emphasis on women to stop, for fear of provoking tension or conflict within the relationship or putting women under excessive pressure.

Smoking cessation

Significant others and members of women’s wider social networks were generally supportive of pregnant women’s attempts to stop smoking and were often willing to change their behaviour to assist quit attempts including, most commonly, avoiding smoking in front of the pregnant woman. Others, particularly partners, had attempted to stop smoking at the same time as the pregnant woman, having recognised the potential negative influence of others’ smoking, and some were willing to attend SSS appointments simultaneously.

Significant others felt it was appropriate for generic HPs (i.e. midwives and GPs) to raise smoking in pregnancy with women but recognised the benefits of specialist cessation support, particularly when provided on a one-to-one basis. Interviewees relayed mixed experiences of NRT use from both their and the pregnant woman’s perspectives but appeared to have few or no concerns about its use during pregnancy. CO monitoring was generally considered to be a helpful motivational tool, although accounts cautioned against inappropriate interpretation of readings.

Improvements to existing services/interventions

Most SOs appeared largely satisfied with current interventions available to pregnant women in relation to smoking cessation. However, a minority suggested that brief advice provided by generic HPs and the range of NRT on offer for pregnant women may not necessarily be sufficient to support quit attempts. One possible improvement to services expressed by a similarly small group was for SSSs to pay more attention to and support the role of SOs, or more specifically partners, in supporting pregnant women’s quit attempts.

Organisational

Interpersonal

Individual

Smoking

Health-care practitioners noted that alongside the psychological, physical and social environment of pregnant smokers, service and professional issues could have an important influence on smoking in pregnancy. Similar to the systematic review of HPs, a number of HPs in this study reported a lack of knowledge and confidence around raising and delivering information about smoking in pregnancy, and, particularly in opt-in areas, several felt that tackling smoking was not a routine part of their role. In both areas, smoking was frequently left for midwives to address. However, midwives, particularly in area B, often reported struggling to tackle smoking because of limited time, feeling undertrained and not having smoking as a priority in their work.

In terms of clinical paperwork, contrary to review findings, many participants in this study reported that prompts in clinical paperwork and systems positively facilitated the discussion of smoking. However, at a local level, shifting priorities in targeting risk factors or health behaviours in pregnancy appeared to affect the HPs’ perceptions of the importance of addressing smoking in pregnancy. Across all professional groups, smoking in pregnancy training opportunities appeared to be limited and, when available, were usually accessed only by self-motivated individuals. Although lack of training is a common theme in prior research, participants in this study reported that training issues were a potential cause of inconsistent messages provided by different HPs.

A central issue raised by HPs was the importance of the relationship developed with pregnant clients. As with the review, many HPs in this study were anxious about damaging relationships by discussing smoking and felt inhibited to discuss it as a result. This study extends the review findings by identifying how in some cases, protecting the client relationship appeared to be a factor in HPs downplaying the risks of prenatal smoking and failing to emphasise the importance of abstinence. In some cases, however, good rapport was described as enabling frank discussions about smoking behaviour and a minority believed that pregnant smokers expected and wanted to be told the risks. This reinforced the review’s conclusion that the professional’s relationship with the pregnant smoker can be both a barrier to and facilitator of addressing smoking.

Smoking cessation

New insights were generated about views of opt-out referral pathways combined with routine CO monitoring. Some midwives felt that opt-out referral pathways simplified referral and increased access to support, whereas others reported concerns about referring women who they felt were not at the ‘right stage’ to quit. Among SSS advisors, opt-out referral pathways were seen to increase client awareness about the SSSs, but some also discussed how the pathway clashed with their service philosophy of client choice. Other advisors expressed concerns that the information given to pregnant smokers around CO monitoring could lead to early dropout. These concerns about opt-out referral and client choice linked with the findings of the systematic review of HPs of the perceived importance of putting the woman at the centre of the decision to quit.

The relationship between HPs and SSS advisors was raised as a further important factor in smoking cessation work. In some cases, poor communication between antenatal HPs and SSS advisors was identified as contributing to deterioration in interservice relationships and mistrust. SSS advisors sometimes described frustrations with the reduction in quality of the referral information that they received from other HPs, although the HPs themselves sometimes felt that they lacked referral feedback from advisors. The quality of communication varied between sites, however, with some examples of effective relationships and efforts made to maintain good communication among both antenatal HPs and SSS advisors.

Aside from referral, HPs were very cautious about NRT use in pregnancy, despite near universal provision in the SSSs. This has been an enduring theme in prior research as identified by the systematic review of HPs. However, these findings identified additional barriers to NRT provision within SSSs, in which delays to provision were perceived as leading to disengagement among pregnant women.

Improvements to existing services/interventions

Several potential improvements were identified from this study’s findings. At an organisational level, improving existing paperwork to include more effective prompts and providing clear information for HPs regarding smoking could effectively elevate its perceived priority when it is inadequately addressed (contrary to local guidance). Several SSS advisors felt that having more designated time available for telephone and text contact with pregnant smokers between appointments and having greater opportunities for flexibility around appointment times and locations could all help to improve engagement. In addition, HPs and SSS advisors across both study sites felt that more staff members were needed for the specialist SSS pregnancy advisor role. Enabling staff members to be more fully involved in service development could potentially enable the service to make better use of staff expertise, increase staff buy-in and reduce the feeling among some staff that changes are imposed from elsewhere. Provision of NRT in a timely manner was also perceived to increase the chances of pregnant smokers quitting successfully.

At individual and interpersonal levels, improving communication pathways between HPs could lead to improvements in the referral process and greater trust between different professional groups; suggested approaches included having update newsletters, a presence at each other’s team meetings and increased guidance around local statistics, cessation rates and dealing with referrals. For HPs lacking confidence to conduct CO testing and/or deliver the opt-out pathway, providing mandatory training in this area could improve HPs’ confidence and equip them with the skills and knowledge to effectively raise smoking as a topic, provide very brief advice and implement the opt-out referral protocol. Improved communication together with the delivery of basic smoking cessation training for all HPs involved with pregnant smokers could then lead to greater consistency in health messages delivered to pregnant smokers.

Pregnant women

The pregnant women interviewed had mixed perceptions on the use of financial incentives to encourage smoking cessation in pregnancy. Some had very positive views, suggesting ways in which incentives could work for them or other pregnant women. This was often seen in the context of the expense of having a baby. A number of women reported that incentives would be particularly motivating for those with financial concerns; the focus here was on being able to afford to buy things for the baby:

Yeah it probably would [help people to stop smoking] because to be fair you know, when you fall pregnant the first, I think your main worry is ‘how are you going to financially cope?’. So you know I suppose if you are getting a wee bit more of an incentive when it comes to a voucher, yeah it probably would, it would help.

Area A, pregnant woman 18, non-smoker

Offering a financial incentive was also seen by some as a useful motivational tool, one that could help pregnant women by having something to look forward to or being rewarded for their efforts. It was also considered useful in providing women with an extra reason to try to stop smoking; this could help those struggling to find motivation or willpower to stop smoking:

It would have given me more reason, more, I think it would have given me more willpower to stop.

Area B, pregnant woman 7, non-smoker

Negative perceptions focused more on moral arguments against using financial incentives rather than a perception that they would not be helpful to women. Some women likened it to ‘bribery’ or ‘blackmail’. Others said that the baby’s health should be motivation enough for stopping smoking in pregnancy. Other women were conflicted by the idea of incentives, on the one hand thinking it was a good idea but on the other hand concerned that it would lead to women being motivated by financial rather than health reasons:

I think it’s handy, because I know, even before you become pregnant, it’s, you struggle, everything’s so expensive nowadays, . . . it’s like, it’s handy because it means like, obviously you’ve got that wee bit more money to spend on your baby – or on other things that you need – but then at the same time . . . I don’t know, I think it feels a bit rude, ‘do this, and we’ll give you money’, I think it’s like a bit greedy.

Area A, pregnant woman 5, smoker

I don’t know, I think if you want to stop you should stop for your baby not for money. But then I am saying that and I am absolutely skint so it sounds like a great idea.

Area B, pregnant woman 14, smoker

Other negative perceptions focused on the possibility of gaming the financial payments, on uncertainty about the idea of incentives to help with smoking cessation and on remaining abstinent once the incentives ceased:

Yeah it would [help people to stop smoking], because a lot of people would do anything for money won’t they . . . but as soon as it stopped then they would go back on to smoking.

Area B, pregnant woman 8, smoker

Significant others

Significant others had mixed opinions on the idea of using financial incentives to encourage smoking cessation in pregnant women. Some interviewees perceived this idea positively, thinking that it could be helpful to pregnant women by providing them with a reward for stopping smoking. Others felt that using financial incentives would mainly attract, and therefore be most effective among, women on lower incomes, with some reporting that receiving shopping vouchers would be very helpful to women preparing for the arrival of a new baby:

I think that is a bloody good idea because again you can use the vouchers to go and get baby clothes and actually see the rewards and have the rewards there for it. I think that is, I think that is probably the best one I’ve heard actually.

Area A, partner 48, smoker

It might be for some people yeah, I think some people it could possibly work if they needed the money. For those that don’t need the money they would probably sit down and go, I don’t know, no sort of thing.

Area B, partner 5, smoker

A substantial proportion of SOs interviewed viewed the prospect of using financial incentives negatively, with some believing that it amounted to bribery. Others simply did not approve of paying pregnant women to stop smoking:

It’s more like bribery isn’t it; I think it should be a personal choice to be honest.

Area B, partner 5, smoker

I don’t see the point in paying someone for a bad habit, it would be like paying someone to stop taking drugs, you wouldn’t want to do that would you?

Area B, partner 2, non-smoker

Other interviewees reported reservations about using financial incentives to promote smoking cessation among pregnant women with a key concern relating to the extent of effectiveness. Some interviewees felt that offering a financial incentive would be insufficient to motivate and/or sustain smoking cessation, particularly once women were no longer receiving incentives, and one interviewee felt that willpower remained more important than the offer of financial incentives:

If somebody was offering you a few quid a month, and it helps you then all good, but I don’t see how a small financial reward could take away that craving. Because if somebody stops smoking what they gain in money from stopping smoking is far more than what they will be given. A bit sceptical but, for some, again if it works for a few people then that sort of thing.

Area A, partner 53, smoker

. . . well I suppose it would get, it would get more people to stop, but then again it’s like are they stopping to get something for it? Are they then going to get that and then go back to smoking? . . . it might not always work, but it might work sometimes I suppose.

Area B, partner 55, smoker

Those who had reservations about using financial incentives also reported concerns about the possibility of gaming the incentive, believing that some pregnant women would accept the incentive but continue to smoke:

I just think it might be, people might just be taking advantage of it saying, ‘Oh aye I’m stopping, I am stopping’, and with the money that they’re getting spending it on cigarettes . . . I think maybe some, some women might take it serious, but I don’t think every woman would.

Area A, partner 16, smoker

God can you imagine how many people will say that they smoke when they don’t, say they have stopped to get the money . . . Could you imagine though, 90%?

Area B, partner 9, smoker

In addition, SOs who reported the possibility of gaming the incentive among pregnant women, including most of those who perceived financial incentives to be useful to some, demonstrated limited confidence in CO monitoring:

That’s the only thing, like you could say to the women, ‘Right, do the CO thing in the morning, or in the afternoon’, but a’ she has to do is not have a cigarette for 2 and a half hours and it looks likes she’s stopped . . . have a shower before she goes, put clean clothes on and you would never know . . .

Area B, partner 16, smoker

Health-care professionals

Health-care professionals expressed mixed opinions on both the effectiveness of financial incentives in helping pregnant smokers to quit and how such incentives would be perceived by the general public. Some felt that financial incentives would be a positive way to increase engagement; others thought that they might help in the short term but not in the longer term, and still others felt that pregnant smokers should stop for their own benefit and not for financial incentives:

Well, about the financial benefit, if you are trying to target young people, yes the financial benefit will work, but once they have got a bit more money, they might just go back to it anyway after a while because they are like oh well I’ve got some money now, I am out with my friends and start smoking again. Because you haven’t got that real, I really, really want to stop smoking for me, not for the money, but for me.

Area A, SSS group 1

I just think they have got to want to do it for themselves and their baby and that money is maybe, it seems like a quick fix but it won’t necessarily work and it could be spent wrongly and the government is already short of money, the NHS is already short of money.

Area B, midwife 5

One GP expressed concerns that financial incentives could be counterproductive and could encourage people to relapse in order to get further incentives:

To pay people to stop? That just sounds horrendous, em [laughs] . . . you could, and I’m sure there’s, I’m sure it would work, whether or not that would have a lasting benefit, I would doubt because as soon as you’re not being paid anymore there’s no excuse and also it’s perhaps an incentive to start smoking again and get paid the next time, so I would worry about that.

Area A, GP 1

Alternative forms of incentive, such gym vouchers or relaxation classes, were seen as potentially helpful for increasing engagement among pregnant smokers. However, again, some expressed reservations, with particular concern about how these would be perceived by the public:

I don’t think it’s a great idea because I think you might get the, you know, the rest of the population’s back up.

Area A, midwife group 1

Although there were concerns about the perception of financial incentives, one HP also felt that uptake might not be that high and reflected on a conversation with a colleague in a neighbouring area, where incentives had not generated the interest expected:

I was speaking to the champion from [location] and she was saying we’ve been doing it. I said how do they afford that? I thought that women would snap your hand off, but she said no, the uptake actually was not very good. We know that a lot of smokers are in socially deprived circumstances, so you would think that a financial incentive – but I suppose it just proves the level of the addiction.

Area A, midwife group 1

Pregnant women

Some women did not have smartphones, had never downloaded smartphone applications and did not perceive these to be a helpful resource for stop smoking information or support. Other women, although aware of applications, were not interested in using them; either they preferred to get information from another source or they thought that the impetus for quitting had to come from the smoker rather than from an application:

I’m not really interested in all that kinda stuff, I’d prefer to talk to someone and get all my advice and all my help that way. Like, ‘cause as well as that, like the [applications] and stuff sometimes they don’t work and then it’s just, you have to download it all and, just hand me a bit of paper, it’s much easier . . . does that sound weird coming from a 23-year-old eh?

Area A, pregnant woman 5, smoker

On the other hand, some women had quite a positive attitude towards applications specifically for smoking cessation in pregnancy. Among women who were aware of, or had tried, applications for pregnancy or weight loss, there was discussion about the lack of a specific application for pregnant smokers or about smoking not being covered in the existing pregnancy applications:

. . . the only ones I have really downloaded are pregnancy ones so, but they don’t give you any advice. You know like the Mothercare one, I think it would be good if they did a you know are you a smoker part to their [application], I think that would be good because most women do download those ones. You get free money, vouchers and things. I am just doing that. So I think that would be a good way. I think they do a diet part to it . . . and things like that, but nothing for smokers.

Area B, pregnant woman 4, smoker

Attitudes towards general smoking applications that women either had used themselves or were aware of other people using were quite negative. The only type of smoking application that generated positive attitudes among the women was a money savings calculator; being able to visualise the money saved and watching the figure grow was quite powerful for some:

It says so far that’s £348.55, and 1049 fags is what I haven’t smoked . . . it tells you health, motivations, stuff like that. Yeah, how many different hours, yeah, look you see, coming up to 6 months, it tells you all the different things that are changing in your body over time. That’s quite good isn’t it?

Area B, pregnant woman 2, non-smoker

There were mixed views on the usefulness of the internet and websites. Some women were not aware of any online resources for smoking cessation in pregnancy. Others felt that having things online was not convenient and that social network websites were too ‘public’ for them; there was also the issue of credibility of information posted on websites. Conversely, a couple of woman described the internet as a useful and convenient source of information, and another described how online social network websites could be supportive. Some women also mentioned particular websites, the NHS and BabyCentre, that they had looked at for information:

I can look on the internet and stuff but then I don’t really, you know you don’t know what to trust and stuff on there so I prefer you know proper information . . .

Area B, pregnant woman 14, smoker

Significant others

Many SOs were either unaware of the existence of smoking cessation websites and applications for smartphones, or had never used them. A couple of respondents, however, described their actual experience of using smartphone applications for smoking cessation, one alongside their partner, and mentioned the interactivity, which they perceived as an expediter:

We both downloaded something about, you put in about how many cigarettes you smoke a day and how much they cost and the date and the sort of time that you gave up and then every day you can look and it tells you sort of how much you have saved, how much time you’ve saved. All that sort of stuff and then it’s got like milestones so you’ve not smoked 20 fags, you have not smoked a 100, you have saved 10 minutes of your life, and things that you can personalise and things that you can see that are relevant to yourself and it’s really good. And if you don’t look at it for a few days and you go back and you see how much money you have saved you know it’s a good incentive to do it.

Area B, partner 1, non-smoker

Some SOs believed that applications were a self-help approach that would not necessarily work for them but acknowledged that they could be useful for pregnant women. Others had more mixed views on both the usefulness and the effectiveness of applications. Lack of technical aptitude, however, was perceived by SOs as a barrier to their pregnant partners using applications or websites:

I think doing it more interactively is definitely a better way than charts and booklets and leaflets and stuff like that.

Area B, partner 3, smoker

I would say on a scale from 1 to 10 I’d say halfway, it’s really about the willpower you have got inside yourself really.

Area A, partner 52, smoker

For smokers in general, applications or websites were mostly perceived as a positive tool for cessation because of their accessibility and convenience, the availability of information and the ability to see others’ successes with applications on Facebook (Facebook, Inc., Menlo Park, CA, USA; www.facebook.com):

I think that’s a better approach now, because I mean even, even the older generation now, my dad’s generation, I mean he has got a smartphone, my mum has got an iPad and an iPhone [Apple Inc., Cupertino, CA, USA], she uses her sat nav [satellite navigation] and all that and I think everybody uses it and it’s just a lot more modern, like you say it’s a quicker way, people text you or to have [applications].

Area B, partner 3, smoker

Some SOs gave their thoughts on the content of smoking cessation applications and websites, and made suggestions for improvement, including the development of an officially recognised application and the inclusion of stronger messages about harms:

I don’t like the word ‘official’, but if they created an official application that had all that kind of stuff and it was well made, because all the ones you can currently get are quite cheap and a bit rubbish, I mean they do the job, but if you could get something like that I think you would, you know you would get people more involved because it was more personal to them.

Area B, partner 1, non-smoker

Health-care professionals

Generally, HPs had fairly positive views on the usefulness of utilising websites such as Facebook or having a presence on Twitter (Twitter, Inc., San Francisco, CA, USA; www.twitter.com). Some felt that these tools would be appealing and accessible to people who had busy schedules or who experienced other social or physical barriers to accessing support:

. . . that’s how a lot of people live their lives now don’t they you know and if you have a Facebook group or whatever for stopping smoking in pregnancy, you would probably get a lot of helpful support and advice from other people going through the same thing . . . and it breaks down barriers of time commitments and so on.

Area B, GP 1

Facebook and texting seems to be the way to communicate.

Area B, midwife 3

However, issues were raised about Facebook, including concerns about confidentiality. SSS advisors explained that this was one reason why their service was not currently using Facebook. Some felt instead that having an increased community presence generally would help improve visibility, accessibility and convenience:

A couple of years ago we did a, I suppose it was kind of like a pop-up it was like a stall at the local market here . . . and it was testing people you know the sort of lung function for COPD [chronic obstructive pulmonary disease] they were doing a sort of mini spirometry thing, and you know we were inundated and people were so interested and that sort of oh I didn’t realise.

Area A, SSS group 1

Significant others

Only about one-third of SOs commented on the use of text messaging to support cessation, and responses were mixed. Approximately half of this group felt that text messaging could be supportive of cessation; for some this was because of accessibility and convenience, whereas for others text messaging was seen to provide more opportunities for support that was less intrusive than telephone calls from smoking cessation advisors:

Well yeah, that is quite good, like an instant support, yes, yes, yeah, something like that yeah just somebody or yeah, on hand just to when you are feeling like that yes, yes that would be more useful I would say, yes.

Area B, partner 8, non-smoker

One partner cautioned that text messaging could also be counterproductive if it was not used appropriately:

. . . as long as you don’t get a text through the day saying I hope you are doing well not smoking because then you will read that and go oh bloody hell smoking, if you had forgotten about it it’s just going to bring it back into your head again.

Area B, partner 2, non-smoker

The remainder of those commenting on text messaging included some who were less sure of its usefulness because texts could easily be ignored, and one partner who felt that it could simply be irritating:

I don’t think it would appeal to me. I think, I get sort of texts sometimes of even from like . . . I will get a text and I will see it’s from . . . and I just won’t want to read it and I think I have one from a few days ago and it’s still not read. I am not bothered, some people might.

Area B, partner 1, non-smoker

Health-care professionals

Text messaging was raised as a good way of keeping in contact with pregnant women, as it has benefits over traditional forms of communication. In particular, it was reported that pregnant smokers and families do not necessarily have landline telephones and, similarly, might not necessarily have enough credit on their mobile phones to be able to pick up voicemail messages:

A lot of people don’t have credit so if you leave an answering message they can never access the answer machine message. So the texting is very much the way to communicate.

Area B, midwife 3

Very few of our families have landlines now they have mobiles. And texting is the way they communicate with one another so often if I have got a hard-to-reach family, I am texting them rather than ringing through a landline.

Area B, health visitor 2

Text messaging was also seen as preferable to telephone calls, for example for cases when pregnant smokers feel uncomfortable talking on the telephone or feel anxious about having to make quick decisions:

In the last couple of years certainly texting is increasingly a communication tool, particularly with this client group and that is how they prefer to communicate. And I think sometimes when you get a phone call out of the blue or it’s not pre-arranged or you are not expecting it, you have to make a decision then and there whereas a text allows you that thinking time so you don’t have to respond straight away.

Area B, SSS administrator 1

Pregnant women

There were mixed opinions about social network interventions. Views tended to be more positive when women felt that they had already experienced this type of support from their partners, family and friends. More negative views expressed related to women’s perceptions of this type of intervention increasing pressure on them to quit:

Yes, I think it’s a good idea. [. . .] Because you will have more support around you not just like once a week. And you’d have each other for support so.

Area A, pregnant woman 1, smoker

It’s more pressurised though isn’t it . . . Like all together if it’s a big group, I can understand the ones that smoke sat in a group, yeah but if there is like the family members and everything else . . . it would be just pressure, because they are eager for mum to go and she didn’t actually really want to stop just like that, then it’s going to mean more pressure which is going to stop her from wanting to physically to do it.

Area B, pregnant woman 15, smoker

Suggestions for the format of social network support to overcome this potential for increased pressure included encouragement from the woman’s network rather than ‘nagging’ to quit smoking and conveying that it is very difficult to quit an addiction:

That would be actually good because, as I say my partner’s been supportive at times but other times we’ve been arguing and he was pushing me, so maybe if he went to groups like that maybe it would be easier for him to understand; it’s like being a drug addict, or any other addict, you know what I mean?

Area A, pregnant woman 14, non-smoker

Social network support, however, was most often thought to be suitable for other people but not these particular pregnant women or their friends and family, either as a result of typical family interaction or simply because cessation was a much more internalised process for that individual:

And I think if that was my family they would be like well you are the one who is pregnant, you are the one who is having a child, why should we stop? So I don’t think that anyone that I know would go to a group thing, but it might be alright for some people.

Area B, pregnant woman 7, non-smoker

Significant others

Significant others were generally positive about involving partners, family and friends in a scheme to support pregnant women to stop smoking and most said that they would be willing to participate if invited to do so. The main motivation for participation was to support the pregnant woman but some recognised that it could also help them or others who took part. A key reason for the usefulness of such a scheme was the opportunity for the pregnant woman to quit smoking alongside others, which was seen to make the experience easier. This was acknowledged by those SOs who were already trying to support the pregnant woman in this way:

I think it’s, well to me it’s better because there’s a group o’ you, like you and your friends, and it’s, it’s not directly one on one but it’s more than going to sit in a room full o’ people that you don’t know and just one person talking tae everyone.

And what else is helpful about that kind of support?

Like it’s individual as well, so if you’ve . . . if you’re needing like more or less support then you can ask.

Area B, friend 44, smoker

The idea of those closest to the pregnant women taking part in a scheme to aid cessation was perceived to provide a type of support that could not be provided either by HPs and/or by people who had already quit smoking or were in the process of quitting. SOs emphasised the benefit to women of being supported by those with long-standing relationships with them, thus offering greater familiarity and trust. One partner liked the idea of involving family in supporting cessation, as he felt that it was important that all those living in a household take responsibility for the baby’s health rather than only the pregnant woman, thereby minimising the risk of isolating the pregnant woman with related negative consequences:

I think that’s really good, because you are going to go through it together, you are going to have each other for support and it is somebody that you know and trust rather than somebody in an office with a bit of paper.

Area A, partner 48, smoker

I think that would be a lot better than like I say singling the pregnant woman out and saying you have got to give up smoking because you are the pregnant one. Because I think that makes them, there is a lot of them sort of rebel against that and so you know it’s not just me, why should I have to do it myself.

Area B, partner 3, smoker

In addition, when discussing this approach, a mother said that she felt that group support was preferable to individual support, which she believed placed too much pressure on an individual:

I think the support groups are really, really good, I think they are probably better than the one to ones . . . the bad thing about the one to one with [name 2] is a lot of the time you feel like, you are not getting picked on but it’s like the spotlight is on you, whereas when you are in a group, you are all in the same situation and you relate to each other and things, so.

Area B, mother 20, smoker

Some SOs reported that a family or social network approach would mean that those closest to and most keen to support the pregnant woman would be given the necessary information to help them provide the correct support. A mother-in-law spoke of how such a scheme could help close others to avoid well-meaning but unhelpful advice, and instead provide consistent and constructive support.

A minority of SOs were not supportive of the idea of a family or social network approach or questioned its effectiveness. A key theme within these responses was the belief that no amount of support from others would help a pregnant woman stop smoking if she did not want to stop or did not have the willpower to stop:

It’s all about the wanting to do it, willpower and then being able to do it . . . If people aren’t fully committed to wanting to stop they aren’t going to stop. They have got to want to.

Area B, partner 9, smoker

During discussions about social or peer network interventions, there was disagreement among interviewees about the smoking status and intentions of those who may take part to support the woman. Often interviewees assumed that those taking part in such an intervention would be smokers themselves or those who have had some success in quitting before. This raised concerns about the expectations that would be placed on SOs to stop smoking. Although simultaneous quitting or desire to quit was thought by some to be very important in this situation, others felt that SOs may be willing to take part to support the pregnant woman only if it did not also require them to stop or consider stopping; one partner wondered how this type of intervention would work if he were to take part and continue to smoke. Some interviewees appeared to indicate that non-smokers and even ex-smokers would not be interested in taking part:

I wouldn’t have thought they would be that keen on it no, because they wouldn’t really see that as their issue I don’t think. So as supportive as they are, I don’t think that they can offer any insight. Only my dad smokes so nobody else smokes so nobody else knows what it is like so it’s quite hard for them to sit through something without being sort of critical of us being smokers so they wouldn’t have any understanding I don’t think.

Area A, partner 51, non-smoker using e-cigarette

I would quite happily give [name 1] support if it helped her stop smoking . . . but then again she might just say right you can support me by stopping as well and then I’d be like oh well, but no the theory behind it, I would quite happily support her in anything she does so I’d be interested.

Area A, partner 53, smoker

Health-care professionals

Related to family interventions, some HPs felt that increasing service capacity for peer support could be beneficial. One midwife illustrated this by describing a time when she had witnessed a pregnant smoker talking in a group about her knowledge and experiences to others:

. . . she was pregnant, she was about 18. And she said this is what happens, this is why your baby gets more [CO], I know when I stop I am going to crave, when my pregnancy ends I am going to crave it more so I am prepared for that and she said all this. And I thought this is just brilliant. I thought this is better isn’t it because she is living it you know. And the girls all sat and listened actually.

Area B, midwife 3

One commissioner suggested that, for young pregnant smokers in particular, visiting formal services is quite an unnatural thing to do and suggested that peer support would improve people’s motivation to attend. However, some HPs were concerned that peer support could potentially feel stigmatising and that this might deter pregnant smokers from attending:

Sometimes I think it would be good if it was like a group, but maybe that’s sort of stigmatising that as well like oh gosh I am going to be the one in the pregnant smokers group.

Area A, midwife group 2

Health-care professionals

Several HPs felt that increasing the capacity of SSSs to support families was important, with a number of advisors suggesting family therapy and providing more support for partners to stop. Some HPs emphasised that supporting mothers of pregnant smokers could also be particularly helpful, and one commissioner noted that although pregnant women’s mothers were not turned away from support, this was a less proactive part of the service than supporting partners to stop. One midwife summed it up by describing the potential ‘knock-on effect’ that could occur through offering more support to families:

I think families are definitely key but they are probably the hardest to reach you know, the pregnant woman has at least got some motivation that her family don’t always have . . . so maybe, maybe getting grandparents and fathers would have a knock-on effect.

Area A, midwife group 2

Although family interventions were seen as potentially positive, it was noted by some that they would probably be difficult to orchestrate, in terms of both family practicalities and service implications, as running a family intervention effectively would be time-consuming and resource heavy:

I mean I’ve seen mothers and daughters before together but it’s just again as you know if people can’t come at a certain time, getting them all in the one room for a start isn’t easy and then getting them all in one room and getting them all to agree to do that and stopping smoking is different matter. But yeah I think when it works it works really well.

Area A, SSS group 2

Some also reported that partners might not necessarily see quitting as their responsibility and as a consequence might not attend after the first one or two sessions:

I don’t think it would work, because I think that, when we talk about the people we’re dealing with, I often find that the mums will always buy into all the antenatal care, dads will come, there’s certain classes, they’ll come for labour, they’ll come to a token few, and that already shows you that they don’t see that part as being part o’ their responsibility, you know they see it as the women going to the classes, so I don’t know how you change the culture o’ that.

Area A, health visitor 1

One HP felt that more flexibility to visit families in their homes might help to alleviate some of these difficulties:

Maybe the service maybe needs to think more about going out to people’s homes all the time when families are around. Because as you say trying to get them when they are busy doing other things. You know if they are all home in the evening, Because we don’t actually really say to them oh we will come out to you, because you know we don’t have the facility to do that.

Area A, SSS group 2

Despite being seen as potentially beneficial, one HP noted that although families quitting together could motivate often each other, they could also cause a ‘domino effect’ (area A, SSS group 1) of relapses if somebody in the family intervention were to lapse:

They are just living the life that they live sort of thing, the sort of . . . and I went and there was the dog, there was the grandparents . . . home, but it was a really good experience, but they all fell like dominos, once one of them started.

Area A, SSS group 1

Pregnant women

Although e-cigarettes were not part of the pregnant women’s topic guide, some women talked about them, and their use was on the rise during the study period. Among those who raised the topic, e-cigarettes were viewed as a harm reduction tool that could potentially help women to stop smoking. There were comments relating to e-cigarette use both in and outside pregnancy. Some women described how family and friends had successfully stopped smoking using these devices. Others saw switching to e-cigarettes as not properly quitting smoking. Although there was interest in using e-cigarettes in pregnancy, generally women were uncertain about their safety. At the time of the study, midwives did not have any national guidance (either in England or in Scotland) regarding these devices. It was, therefore, not surprising that interviewees tended to follow their midwife’s advice that e-cigarettes should not be used in pregnancy:

[I tried] an e-cig[arette] which was really good . . . It’s good to have this you know the smoke coming out and going through all the motions, but I wasn’t allowed anything like that during pregnancy because they have not tested things like that properly yet.

Area A, pregnant woman 18, non-smoker

A few women had used e-cigarettes while pregnant. One woman had used a nicotine-free e-cigarette but had reverted to cigarettes after 2 days and was contemplating trying an e-cigarette with nicotine. Another woman had managed to quit for 2 weeks with an e-cigarette in a previous pregnancy. Previous experience, however, also resulted in some negative views on the effectiveness of e-cigarettes. A number of women considered them an expensive option:

I had one of them e-cig[arette] things you know the ones with the oil, and it lasted 3 weeks and then I got rid of it because it was rubbish to be fair. It was, you had to charge it all the time and then you had to buy the fluid and then it just ended up costing like the same amount as regular cigarettes. It was – there was no point.

Area B, pregnant woman 17, smoker

Significant others

Narratives surrounding e-cigarettes were mixed. Some SOs described their own positive experiences of using e-cigarettes. Others described other people’s successful use of e-cigarettes. Costs, flavourings and doing something with your hands were mentioned as benefits. As such, some of the SOs suggested them as an approach for their pregnant partners:

The e-cig[arette]s work . . . they are good, obviously they are not, well they say they are not harmful, it doesn’t smell, it doesn’t bother other people, you get the nicotine, you get the – I think the difference with the e-cig is that you actually get the sensation of smoking a cigarette so it’s a lot, a lot of it is psychological I am sure it is.

Area B, partner 9, smoker

E-cig [would be helpful to pregnant partner], yeah, they do work.

Area B, partner 9, smoker

Some respondents, however, described unsuccessful experiences of themselves or their pregnant partners using e-cigarettes or were cautious and felt uninformed:

Another way that I tried to support her was initially getting her some is it e-cigarettes? . . . But I think unfortunately the one we got, because I didn’t know there are different flavours and different strengths and everything, it made her feel a bit sick.

Area B, partner 8, mother, non-smoker

Pregnant women

Having access to specialist cessation support during pregnancy was important to women in area A. Advantages over generic support included greater understanding and knowledge about pregnancy-related issues and NRT, and more targeted risk information:

I think it is definitely good because then it, it makes you see it fae a different light . . . wi’ you being pregnant you’re limited to what you can get, so it is good tae[to] get the information on why you’re limited on that, and you know, and what can happen tae the child and how it affects the child and you know different things, because it’s just things that you don’t even think aboot[about], when you’re smoking you don’t even gi’e[give] it a thought about the baby tugging on the cord for 20 minutes.

Area A, pregnant woman 11, non-smoker

It was more difficult to know whether or not women in area B saw benefit in a specialist service, most probably because fewer participants had accessed the SSS. Among those who had not engaged with the SSS, there tended to be a negative view of smoking cessation services and women doubted that it would be of benefit to them:

I don’t think I would like that. I just think it’s, I work a lot as it is, so I am quite busy all the time, and I just think having to go to like a stop smoking, unless you are ready you know going to be willing to go . . . I don’t feel that service would help me.

Area B, pregnant woman 4, smoker

Significant others

All responses but one from SOs comparing dedicated SSSs for pregnant woman with generic SSSs were from area A. These SOs felt that a dedicated service for pregnant women was important in terms of pregnant women being able to obtain support specific to their situation, with some SOs relating this explicitly to harms to the baby. Others were less sure that the support should be different:

I think it probably would be more helpful if it was tailored more towards pregnant you know so that they could have their own, because you could obviously talk about different things and offer different things to what you would generally.

Area A, mother-in-law 47, non-smoker

Most of the SOs’ responses were positive towards the pregnant women receiving smoking cessation help. However, it was often unclear in their answers to the question whether or not this was comparing pregnancy-specific help with generic help or if it was comparing it with no help:

I think it is really important to be honest, because a lot of pregnant woman give up with that help. I think they would find it a lot harder if they didn’t have the help if you know what I mean.

Area A, partner 47, non-smoker

I think it’s a fantastic idea, I think it’s a really, really good idea, the person who came up with it needs a medal.

Area B, partner 7, non-smoker

Pregnant women

When interviewees were asked whether or not they could remember any mass media campaigns relevant to smoking in pregnancy, some women could not recall any, although there was some support for this idea:

I mean in pregnancy it should be on TV a lot more because they only advertise it on the smoking packets about the babies. In the adverts they only go on about it can give you lung cancer, not what it can do to your baby, so I think for pregnant women, smoking and drinking everything like that should be a bit more on TV, a bit more out there for them to see.

Area B, pregnant woman 12, non-smoker

Some women, however, could recall and clearly describe advertisements they had seen about passive smoking, either in the home or in the car. The influence of these was apparent when women described plans for not smoking around the baby after delivery. Women appeared to have a high sense of the risks from passive smoking and many considered it more harmful to the baby than smoking in pregnancy. Some women were encouraging of the use of repeated shock tactics in advertising, although one woman reported that she would just avoid exposure to such messages:

I’ve seen adverts, about, adverts where like you are smoking and the child is in the room and the smoke is going all around the child . . . I think they are good, they scare people don’t they, like I’d never, even if I was smoking now, which I am, but when the baby comes along like I’d never smoke in the same house as my baby, I’d always go outside because obviously that’s, it’s my choice me smoking when the baby is inside me, but it’s not the baby’s choice to be around the smoke do you know what I mean I would go outside and smoke.

Area B, pregnant woman 11, smoker

Many interviewees in area B recalled the ‘Stoptober’ stop smoking campaign. This is not surprising, as this campaign ran in England where area B was located but not in Scotland where area A was sited. Several described engaging with Stoptober and the campaign encouraging a quit attempt among themselves or their family members. However, there were mixed perceptions on the usefulness of information provided and whether or not it helped with smoking cessation:

I think they are pretty cool. For people who are trying to stop, that want, they have got that support as well as going to the doctors and stuff. . . . So that is quite cool to be involved in something like that.

Do you know anyone that has stopped smoking as a result of that? Stoptober?

No, I know people that have tried it, but have failed. . . . it would have been a bit of fun to be involved with that [if smoker at that time].

Area B, pregnant woman 12, non-smoker

Significant others

Many interviewees could not recall seeing or hearing smoking cessation campaigns on TV, radio or in printed media. However, of those who were aware of mass media campaigns, recall of the key narrative was quite detailed:

I remember seeing the adverts that were like pretty horrible . . . It was the one that really was pretty creepy, I think they were in the car and there was a wee baby in the back seat and the guy had the window open and it was like a snake or a hand . . . and it went up and round the bairn[baby], oh, it was horrible.

Area A, friend 44, smoker

However, in terms of a focus on the actual effects of smoking on babies, interviewees reported a lack of information:

There is not an awful lot of information out there about the effects of smoking on babies to be fair, I mean we have got four kids ourselves and there is not once that I’ve read a leaflet, seen an advert, anything about what affect it has on a baby, just the effects on the person who is smoking.

Area B, partner 10, non-smoker

When SOs were asked whether or not they thought campaigns would impact on their pregnant partner, responses were mixed. Some thought that campaigns could have a positive impact on a woman’s smoking, whereas others considered that there would be little or no impact. Responses were also mixed in terms of campaigns having an influence on the SO’s own smoking behaviours or on their encouragement/support of the pregnant woman:

No matter what actual advertisement you put on, if you actually want to stop, you’ll stop, if you don’t you won’t.

Area A, partner 24, non-smoker using NRT

You see the adverts ken[you know], like the smoke going into like the kids . . . Yeah it’s a bit mm, it makes you think . . . If you are sitting there around your kids and you are smoking and you see that advert you are kind of, you think well maybe not, so.

Area A, partner 55, smoker

Some SOs suggested that campaigns featuring shocking images, better tailored to the target audience and appearing at certain times of the day could enhance their impact:

I think for them to be successful they have got to be sort of really shocking or have a really powerful message so that that’s it you know I’ve seen it once, I am giving up.

Area A, partner 1, non-smoker

Health-care professionals

Several HPs felt that more government campaigns would increase people’s awareness about stopping smoking in pregnancy. Some referred to awareness campaigns about passive smoking to demonstrate their point that media can have a significant impact:

I think the impact of those sorts of campaigns and the advertising that we are seeing on telly about the impact on passive smoking on children, has changed parenting behaviour in terms of smoking over their children.

Area B, health visitor 2

It appeared, in some instances, that government campaigns made it easier for HPs to approach the issue of smoking in pregnancy and post partum:

Some o’ the government advertising, I really think it’s a very, very powerful, I would definitely say that my job became easier approaching smoking when the passive smoking adverts were on the TV . . .

Area A, health visitor 1

However, although previous campaigns had, from HPs’ recollections, tended to focus on postpartum risks, some felt that more campaigning around the risks of smoking during pregnancy was needed, including the use of ‘scare tactics’ advertising:

. . . they always talk about babies and children, maybe they need to do something advertising like a fetus you know.

Area A, midwife group 1

The TV advertisements, the packets, the notes on the packets of cigarettes, I think that maybe a scare tactic with regards to baby would be a good thing, because I am not sure that people really do you know . . .

Area B, midwife 5

Pregnant women

Some of the women liked to hold on to leaflets, as a valuable reference source for themselves and others; for other interviewees, the leaflets held a more short-term value, read at the time but ultimately disposable:

The leaflets are good but once you’ve read the leaflet you chuck it out.

Area A, pregnant woman, non-smoker

Some of the women had good recall of the information contained in leaflets and had found the information useful and persuasive. One woman gave quite a detailed evaluation of a stop smoking leaflet she had been given, reporting it as too lengthy and quite technical, and suggesting that it could be better targeted to appeal to all age groups. Other women had similar suggestions:

I mean like certainly I think it’s just how they put it across I mean I think if I got an e-mail just with a big boring like 20-page essay, that’s not going to stick in my head. But if they maybe something like an e-mail with bullet points and interesting facts that would appeal to me.

Area B, pregnant woman 13, smoker

The main area identified for improvement related to the amount and type of smoking information provided in pregnancy, with some women wanting more written information and some wanting a stronger message about the harms to the baby to help motivate them to stop smoking:

Yeah, I think, I think if you shocked them more with what could happen to them while they are pregnant, then what could happen to the baby, then I think, you know just a picture of a pregnant tummy on the back of a cigarette packet is not beneficial. I think if the midwife was giving you leaflets with things in it saying this is what could happen, I think that would be quite good.

Area B, pregnant woman 4, smoker

For leaflets that women could pick up themselves (as opposed to being handed them or receiving them in the post), most women reported that these were available at their GP surgeries and hospitals, although not always to take away with them as some were displayed on the wall. Only a handful of women mentioned seeing posters displayed in their GP surgery, at the hospital and at the chemist:

. . . I think you need a bit more, you only get the occasional poster in a chemist or whatever if you are in there, it would be nice if you seen a wee bit more.

Area A, pregnant woman 18, non-smoker

Significant others

Significant others recalled that leaflets promoting smoking cessation and self-help to quit were usually seen, or handed out, only in primary and secondary health-care settings. The locations for signs, wall charts and posters were the same, although libraries were also mentioned:

Yes, I was aware and I think generally you see signs up in libraries; they are quite often in the doctor’s surgery as well so you know it’s quite prominent . . . as a non-smoker I mean I have noticed the availability of these posters and the courses and things available.

Area B, mother 8, non-smoker

Some SOs thought leaflets had been, or may be, useful to their pregnant partner or friend:

I think now with her being pregnant she might pick it up tae[to] help her stop smoking and what websites tae look at, but before she was pregnant I don’t think she would.

Area A, partner 16, smoker

She [pregnant partner] brought one leaflet with advice on it, how to stop smoking and why you should stop smoking and stuff like that . . . I think it was helpful aye because she read it and she obviously, once she read it she was kind of starting to cut down after she had read what obviously she read.

Area A, partner 46, non-smoker using e-cigarette

One respondent also spoke of how leaflets had been informative for him:

Well getting the reading from the [CO], the leaflets that tell you what’s in it, the carbon and all that, it’s, that put me off quite a lot. About what it can do to the baby when it’s inside you that put me off.

Area A, partner 52, smoker

However, respondents generally perceived leaflets as useful for some smokers to help them stop smoking but not for others. In addition, some were more negative in their assessment of leaflets; they were described as unhelpful, too time-consuming to read or something they would never read:

Em, well usually if I get a leaflet I just bin in, I didn’t [laughs], I never really read it like.

Area A, friend 44, smoker

Significant others

Other types of interventions or approaches that SOs believed would help pregnant women who wanted to quit smoking included use of both graphic imagery to visualise the impact of smoking on the baby and graphic experiences presented by people with lung cancer or smoking-related illness.

Other suggestions offered by SOs included ensuring that pregnant women were well informed to allow them to reflect on a decision to quit, offering a range of alternative interventions and approaches to allow women to tailor support to their individual needs, providing mechanisms to enhance willpower, providing a healthy alternative/replacement to smoking and offering a range of alternative NRT formats compatible with morning sickness and hypnosis:

What will work for one person won’t necessarily work for another, so it just depends on the person [. . .] But if the support is there to sort of try these different things and they are offered these different things then they will find the one that will suit them so trial and error.

Area A, mother 20, smoker

So they are given a list of different options to fit in with their background and their ongoing troubles, and their lifestyles and they should make a choice whether they want to try one.

Area B, partner 10, non-smoker

Health-care professionals

Several HPs and advisors felt that targeting younger women before they become pregnant could be an important preventative intervention. HPs also suggested specific postnatal support, to reduce the risk of women resuming smoking after giving birth. In addition, several SSS advisors suggested that improving the physical space occupied by the SSS could help improve retention, by having more inviting premises in accessible locations. Overall, however, HPs emphasised the need to increase the number of specialist staff members able to deliver smoking cessation in pregnancy and the need to improve existing services through adequate funding and longer-term future planning.

Stress and enjoyment

In the studies included in the systematic review, across countries pregnant women described smoking as a resource for managing stress and suggested that attempting to quit could take away a stress management tool. 110–120 This is a consistent finding in the literature that has been described since members of our team first conducted research with disadvantaged women in the 1970s. 151

The importance of smoking to reduce stress was also described by partners and HPs. For partners, continuing to smoke themselves was described by some in the review studies as a mood management strategy and a ‘time out’. 121,152–155 Some review study participants cited stresses around their partner’s pregnancy, including her changes in mood, with smoking described as ‘. . . the excuse to get out of the house’. 154

Despite these accounts indicating that smoking helps to reduce and manage psychological stress, the wider literature provides more insight into the relationship between smoking and stress than in earlier studies. Some of the stress that women, or indeed partners, feel when trying to reduce their smoking or to stop smoking is likely to be a result of nicotine withdrawal, which creates cravings and feelings of anxiety. Pregnancy can, of course, exacerbate these feelings. However, it is known that adults who stop smoking are far more likely to report lower levels of stress, and that smoking cessation is beneficial for stress management and reduction. 169,170 It is important that this evidence is explained to women; however, it is also known that midwives and other HPs are not always aware of this evidence.

In the review women also talked about the enjoyment of smoking, as did partners who in a number of studies said they could not imagine their lives without cigarettes. 121,156–158 Enjoyment of smoking is often neglected in the smoking cessation literature but it is a key component of continued tobacco use and involves a wide range of factors, not just craving for nicotine. Women and partners in the review studies talked about the benefts of a smoking break by themselves or with others. 111,115,120,122–125 This was mirrored in our interviews. Smoking was a key part of daily routine, relieving boredom and providing a welcome pause and time away from other children, an unemployed partner or other members of the household. Smoking as a break in the workplace was also discussed and is described in Community influences supporting smoking cessation.

Beliefs and risks

Women’s (and partners’) beliefs about the risks of smoking were a central theme in this study. These beliefs could be modified or changed by interventions including contact with HPs and, in particular, SSSs. However, in the review studies and in our interviews, many of these beliefs stemmed from norms (discussed in Community and societal factors) or knowledge influenced by mass media campaigns or, arguably, earlier interventions including health information in childhood or early adulthood. Risk perceptions were often described as relating to smokers in the population, therefore a disembodied risk, rather than a personal one. Some women in the review studies questioned the applicability of these risks to them or even if these risks were ‘real’ at all. 113,115–117,126–128 In addition, the partner’s review illustrated a particular and widespread scepticism among those interviewed in the studies. They accepted there were some risks to smoking in pregnancy but felt that what was missing was more specific information on how smoking was damaging to the baby or indeed to themselves. 53,159,160 Partners in the review studies also felt that there were inconsistencies in government advice, and smoking partners had negative views about tobacco control policies.

In our own interviews with partners and SOs, there was a broader acceptance of the risks of smoking in pregnancy than reported in the review studies, perhaps reflecting more current knowledge and attitudes. Awareness of the links between smoking and lower birthweight and/or miscarriage, for example, was evident. However, even in our interviews, doubts were raised about the consistency, severity or magnitude of risks.

Beliefs about risks were shaped by personal experience. In the partners’ review, first-time fathers were more concerned than those who already had children, where the mother had smoked during her pregnancy and the baby appeared healthy. As outlined in Community and societal factors, for women, both their own experience and that of their mothers or other women in the community served to challenge HPs’ advice about the dangers of smoking or the benefits of quitting. Partners interviewed for our study also provided examples of healthy babies born to mothers who smoked, pointing out that known evidence on risks may not apply to all.

There was some distinction drawn, however, between risks in utero and once the baby was born, with pregnant women perceived by some to act as a ‘filter’ for the worst effects of smoking. In contrast to smoking in pregnancy, in which doubts were expressed about the risks to the unborn baby, there was an almost universal agreement among women and partners in both the systematic review studies57–59 and our interviews that babies should be protected from exposure to SHS. There was, therefore, widespread acceptance that exposing a baby or child to SHS was unacceptable. 53,121,152,154,157,159

Beliefs and cutting down

In our interviews with pregnant women in particular, interviewees described avoiding thinking or learning about the risks of smoking during pregnancy. For those who did engage with this evidence, a common response was to talk about the advantages of cutting down, which many perceived as conferring benefits. Some partners and SOs interviewed reported similar beliefs. In addition, a small number of women reported that they perceived that HPs encouraged cutting down.

The evidence on smoking reduction is now much clearer, including in pregnancy. Women who cut down without using NRT are more likely to engage in compensatory smoking, smoking more of each cigarette and taking longer and harder drags on each cigarette to get the nicotine they need. 171,172 This results in exposure to more toxins per cigarette, resulting in few, if any, health benefits to cutting down. 173 As with stress reduction, midwives and others may not know or understand this evidence, which may result in mixed messages being given to women. In contrast, research suggests that smoking reduction with NRT does not result in the same compensatory smoking and can also represent a step on the pathway to quitting. 174,175 Women who cut down their smoking in pregnancy should be supported to do so with NRT or other nicotine-containing devices, and the importance of stopping all tobacco use should be emphasised. 176

A finding of the review was that low birthweight was understood as a risk of smoking in pregnancy but also seen as an advantage, with a smaller baby perceived to make labour and delivery easier. 115,123,129,130 This view was still present in our interviews but from only a small number of women across both study sites, perhaps reflecting better current understanding, including via advice from HPs. This could be emphasised, however, in future training.

Identity

Women’s perceptions that smoking as a behaviour was part of their identity came through in the reviews of studies with women and also in our interviews with pregnant women and new mothers in both study sites, although not as prominently as in other qualitative research with non-pregnant populations who smoke. In terms of this discussion, it is perhaps best described as a set of beliefs that played a powerful role at different stages in women’s lives. Being pregnant represented a break for most women from the ‘acceptability’ of a smoker identity. This perception served as part of their motivation to stop or at least to try. Women interviewed in both sites, particularly those who managed to quit at least at some point during pregnancy, described willpower and determination as very important in the success of a quit attempt (reflecting any smoker’s experience of stopping), but with the added impetus of pregnancy-specific motivators as highly salient. Some of these motivators were tangible, such as nausea or changes in taste and smell, and others were sensed or perceived motivators, including the importance of stigma, which is discussed below in community and societal factors.

There were distinctions in smoking identity between being pregnant and having had the baby, and this emerged in both the reviews and in our interviews. During the postpartum period, women in the review studies in particular talked about ‘reclaiming’ part of their former self that they associated with smoking alongside specific more tangible benefits that they associated with tobacco use (e.g. weight control). Smoking cessation was perceived by some women as a temporary change undertaken for the baby in utero and while breastfeeding, particularly in the older studies in the reviews. 115,126,128,131 As such, any return to smoking in the postpartum period was seen as an assumed resumption and not a relapse.

In the systematic review, for men who contined to smoke, being a smoker was perceived as conveying autonomy and independence and a positive masculine identity. 121,155 Some men felt that they had a ‘right’ to smoke that was curtailed by tobacco control measures such as banning smoking in public places. 53 Some partners in the review studies talked about the fact that being forced to agree to quit could compromise parts of their (masculine) identity as well as causing tension or arguments. 152,155 However, a partner’s smoking identity could also be challenged by the fact that they were, or were about to become, a father. The review studies showed that some men recognised that their continued smoking would undermine their capacity to be a positive role model for their children and also, fundamentally, could harm their capacity to be around as their children grew up, given the real risks that continued smoking would pose to their life expectancy. 121,153,154,159,160

Support from partners and significant others

It was identified, in a number of studies included in the review, that partners helped to encourage and motivate women. There was a shared understanding that smoking cessation was not necessarily easy but it was important. 115,121,132 Some men who were smokers recognised that if they wanted their pregnant partner to stop, then they should attempt smoking cessation as well. 53 One study in the systematic review of partners explored this issue in depth and outlined how men proactively chose not to smoke around their partner in order to assist her quit attempt, and once the baby was born chose not to smoke around their children. 152 Men described undertaking activities together with their partner that involved not smoking and also offering praise and encouragement to women when they cut down or stopped. Men’s own attempts to quit, however, were often about being a good role model for their children and/or recognising the harm caused by SHS exposure, or at least its smell or even concerns about third-hand smoke on clothes or furnishings. 154

Partners who were non-smokers could serve as important supporters of a quit attempt. In the review of studies of partners’ views, just 10% were non-smokers, so few data were available on this aspect. This small group of men did find it challenging if their wives or girlfriends continued to smoke but also recognised that if they applied pressure on them to quit, it could in some circumstances ‘backfire’ and result in increased levels of smoking. 152

More information on non-smoking partners and SOs and their support in smoking cessation was available from our interviews, as just over one-third were ex-smokers or never-smokers. Particularly in discussing this with pregnant women, living with a non-smoking partner (or a mother who did not smoke) was perceived as helpful, normalising non-smoking in the household. In the few instances in which partners stopped alongside the pregnant woman, this was described as very helpful, a joint effort recognising the importance of the unborn child. Even the partners or SOs who were still smoking when interviewed reported actively trying to stop or cut down to support the pregnant woman and the unborn child. They talked about not wanting to be a ‘hypocrite’ for urging the pregnant woman to stop while not changing their own behaviour and therefore at least made attempts to stop, including some partners engaging with SSSs.

Barriers from partners and significant others

On the other hand, partners or SOs also acted as a barrier to smoking cessation during pregnancy. This finding was particularly prominent in the studies reviewed 112,128,133 but was also reported by some of our research participants. This negative influence operated in one of two ways. The first, and most common, was when the partner smoked and their continued smoking affected the couple’s interpersonal relationship in a variety of ways. The second was when the partner applied pressure on the woman to quit or maintain abstinence and this pressure had negative consequences.

In our interviews, women whose partners continued to smoke expressed disappointment or annoyance that either their partner had not tried or their attempts were unsuccessful, or their partner’s smoking behaviour somehow mattered less because they were not carrying a child. Women reported how some smoking partners blamed them for potential harm to the baby while simultaneously exposing them to SHS by continuing to smoke in their presence. They also expressed regret about a loss of shared rituals and time together that had been enjoyed when both smoked or at least when the woman was not trying to stop. Reported cravings, stress and mood swings combined with bickering and less time spent together could sometimes create a difficult situation.

In the review, women talked about partners who monitored their smoking, sometimes in ways that were controlling or abusive. 111,115,117,121,132,134,135 Examples of confrontation, removing cigarettes or withholding money were provided. 121,132,134 Some women who were struggling to quit felt forced to reduce their smoking or at least pretend to be doing so. Women and partners in the studies also expressed regret about the loss of a shared ritual and time together that smoking involved. 134,159 Unfortunately, however, the review also showed inconsistent or manipulative behaviour by some smoking partners, who might offer cigarettes during stressful times and then reprimand women for smoking on other occasions. 135

Support from health-care professionals

Although HPs’ roles in smoking cessation often operated at the organisational level, the qualitative data also highlight the importance of interpersonal relationships between women and midwives or doctors, for example in discussing smoking and supporting cessation.

In the reviews, women related how midwives, nurses or doctors treated them with respect and in this context they felt supported to talk about and try to change their smoking behaviour. HPs who were ‘facilitating’ were described as friendly, non-judgemental and applying a positive and systematic approach to supporting cessation. 115,116,122,123,136 In our interviews with women, there was a strong focus on the midwife or, for those who accessed cessation services, the SSS advisor. Women often described their midwife as supportive and reassuring, and they viewed this as an important and positive relationship. Overall, they did not feel pressurised by midwives in relation to smoking cessation.

In the review of HPs’ views, the value of a positive relationship with the pregnant women was also emphasised. A trusting relationship was described as a prerequisite to fulfilling their responsibilities to the woman and her baby, including around smoking cessation. 162 It was also considered important that, as HPs, they adopted positive and non-judgemental approaches, encouraging women to take small steps towards quitting, encouraging cutting down to quit and using positive messages around the health of the baby. There was also recognition in the review studies and interviews that HPs could take a more assertive position, drawing attention to the negative effects of smoking on the baby in pregnancy and regarding the risks of SHS after birth. 133,162–165 The systematic review of HPs found that professionals recognised the importance of involving women’s partners in smoking cessation advice but this type of engagement was described as difficult – perhaps more difficult in the review studies than in the experiences of partners and SOs interviewed for this study.

Community influences hindering smoking cessation

A prevalent theme to emerge from the reviews was that smoking was still considered a normal part of life for many in the communities where the studies took place. It was part of relationships in the home, workplace and community. 53,112,117,120,123,132 HPs widely understood that, despite their own role in promoting smoking cessation, they were pushing against community or societal (‘social determinants’) factors that influenced women’s smoking and that these factors could often outweigh any attempts they made to help women stop. 113,164–166 Risks of smoking in pregnancy were downplayed through the experience of others who had smoked during pregnancy but had healthy children, and stopping smoking was perceived (particularly in the studies in the systematic review of pregnant women) as severing important social connections. 111,137

The normalcy and commonplace nature of smoking in the local community was also mentioned by our interviewees, both pregnant women and SOs. Some listed those in their social network who smoked, often the majority. At least one partner mentioned that all of the pregnant woman’s friends smoked. Some examples of friends or family continuing to smoke in front of the pregnant woman were also given. There was a general recognition that these high smoking rates in the (often) disadvantaged communities where women lived made stopping more difficult.

Community influences or norms also undermined beliefs about the risks of smoking in pregnancy, as discussed above. In our interviews with pregnant women and some members of their social network (e.g. mothers or friends), the theme of previous healthy babies born to smoking mothers was also prevalent, and in some instances the pregnant woman herself had a mother who had smoked and she did not see this as having impacted on her own health in child- or adulthood. This ‘experiential’ evidence was presented to counter or refute the risks described by HPs. However, in the interviews a small number of women also provided concrete examples of when they thought that smoking had harmed the baby of family or friends, with miscarriage, low birthweight or asthma/respiratory conditions mentioned.

Norms also had a role to play in the smoking status of women after the baby was born. The accepted nature of smoking in some communities where the study took place did not extend to seeing women smoking during pregnancy (or when the pregnancy was visible) as ‘normal’. It was widely understood that it was not or at least should not be normal. However, returning to smoking after the baby was born was more acceptable in some cases; this was apparent in a number of interviews. In addition, in the review studies with pregnant women there was a widespread expectation – among women and their social networks – that they would resume smoking once the baby was born. 112,115,137,138

A final theme relating to community or societal views evident in the reviews (but not in the interviews) was that of status. Smoking was potentially a tool to conform to a desired female form in that it was also perceived as controlling weight gain. For younger women in particular, this was seen as particularly important in maintaining friendships and social standing. 123,124,129,130,136,139

Community influences supporting smoking cessation

Community influences also supported women’s efforts not to smoke, although less often than they supported continued smoking. In the reviews, smoking in pregnancy – and, to a lesser extent, resuming smoking post partum – aroused disapproval from some peers, particularly when the pregnancy was visible. 110,112,117,119,121,122,133,135 For some women, disapproval could facilitate positive changes in their smoking behaviour, although this was not the case for all. 112,115–117,121

The pregnant women interviewed in both study sites described negative views conveyed by colleagues or strangers. Some also talked about how these social norms had influenced their own negative view of women who continued to smoke during pregnancy prior to finding themselves in the same situation. Workplaces that were non-smoking could help women to stay away from cigarettes, although those working with other smokers found losing their cigarette break difficult, as discussed above regarding interpersonal relationships and continued smoking. This was also the case for partners, which was demonstrated in both the systematic review of partners’ views and the interviews with partners. 158 Recent research on smoking in pregnancy by members of the team showed the importance of avoiding time with other smokers as an important relapse prevention strategy, and this was a theme supported by the interviews in this study. 177

Organisational factors supporting smoking cessation

In the interviews, smoking was normally discussed with women by midwives (or in some cases their GP) early in pregnancy and in all cases at the booking appointment, no matter how briefly. In the interviews with women, partners and HPs, there was a general expectation that smoking would be raised, reflecting policy in both study sites and nationally in the UK. This expectation was helpful, and in general women understood that the midwife’s role in particular was to gather information and, where appropriate, refer on to a SSS.

In the review studies, in which 190 HPs’ views were included, and in the interviews with 48 professionals, supporting smoking cessation in pregnancy was widely recognised as needed and appropriate, but this could depend on the HP’s background and the organisational context in which they worked. The importance of training covering the risks of smoking, benefits of cessation and effective approaches to support cessation was emphasised, particularly in the HP systematic review. Preregistration education and postregistration professional development on these issues was described as very important but commonly insufficient.

Women who were struggling to stop smoking viewed advice on cutting down (vs. quitting) as useful and constructive. This was also evident in the review of pregnant women’s views, in which advice on cutting down was perceived as being understanding of difficult life circumstances and of the perceived stress that ‘abrupt’ stopping might cause. 116–119,123,124,127,136,140

The referral process from midwives or GPs to a SSS was a focus of the interviews and an important organisational element. Area A had an opt-out pathway and, despite some controversy about this (e.g. lack of choice), the women interviewed described this process as straightforward and not unexpected. In area B, the referral process was not opt-out and was more variable. As a result, in some cases those who accepted a referral waited for weeks for a SSS advisor to contact them. Some who thought they had been referred did not hear from the service, or had received an initial telephone call that had not been followed up.

Overall, CO testing at booking (in place in area A) and during appointments with SSS advisors (in both areas) were received positively by the women interviewed. Some expressed shock at seeing their high CO reading as a result of smoking, and in general this information was described as helpful. Seeing CO levels go down was motivating (e.g. described by one woman as ‘a confidence boost’) and advisors themselves (and others involved in CO testing, such as pharmacists) described CO monitoring as an important motivational tool.

In terms of experiences with SSSs, those women who accessed the services were, overall, positive in the interviews about what they had received. Chapter 6 provides detailed findings on the positive elements that will be of interest to SSS commissioners, managers and advisors working with pregnant women. To some extent this was also the case for partners and SOs in relation to their own smoking (see Chapter 8), particularly those who had contact with a SSS. This also came through in the interviews with women in the postpartum period, when they were able to reflect on what support they had received. For those that had been successful in quitting during pregnancy, SSS support had, in many cases, been key. SOs were generally also very positive about the cessation support received by pregnant women. Women and partners expressed preference for one-to-one support (often described as better than expecting women to attend groups) and particularly for the personal support provided by smoking cessation advisors. Partners and SOs indicated an appreciation of the flexibility and responsiveness of the personal support provided by smoking cessation advisors.

Nicotine replacement therapy was an additional element of the organisational context for smoking cessation, although, in the interviews, pregnant women’s perceptions of the value of NRT were mixed. In previous quit attempts (including pre pregnancy), NRT had not always worked. This was also the case for partners, some of whom described using NRT to stop, either successfully or, more commonly, in previous quit attempts that had not been sustained and had resulted in relapse. In the interviews, some women (and partners) reported that they were now using e-cigarettes which were, overall, described as helpful, although a lack of evidence about these devices was acknowledged.

Some women who were interviewed raised concerns about the safety of NRT in the context of a quit attempt, but these concerns had been allayed in some instances by HPs. However, no real evidence was found of partners or SOs expressing worry about NRT safety in pregnancy. The cost of NRT was a concern for some pregnant women, particularly in area B, which suggested that some were unaware that it was available free on prescription. In area A, women were positive about the offer of free NRT. In relation to individual NRT products, the nicotine inhalator was mentioned as helpful by some, particularly in offering a hand-to-mouth action similar to that provided by smoking.

Organisational factors hindering smoking cessation

In both the reviews and the interviews, the health-care context did not always help or promote cessation (or relapse prevention) to the extent that it might have done. Some of the barriers were attitudinal factors but this did vary between the review studies and the interviews, probably because some studies in the reviews were conducted some years ago or in other national contexts.

In the review of pregnant women’s views, although some HPs were described as facilitating quitting, most were seen as operating within systems that permitted ambivalent or negative views. Half-hearted support, combined with insufficient practical help and advice, was a barrier to quitting. 116,119,124,127,129,140 This was also apparent in the systematic review of HPs, in which some interviewees expressed a lack of confidence about what they could meaningfully do to help women quit. In those circumstances, promoting harm reduction (cutting down) was prominent and still practised. 162,164,165,167 There was less evidence found of promoting cutting down in the interviews we conducted with UK HPs, although some who were not trained smoking cessation advisors still regarded this as acceptable or helpful, or in some cases inevitable, particularly when dealing with women or families living in complex and disadvantaged circumstances.

In all of the evidence in the review, although the health-care system primarily aimed to support smoking cessation, there were also structural or systems factors that were seen as unhelpful. For HPs in the review, a lack of training was a barrier to providing smoking cessation in pregnancy, resulting in poor knowledge and skills, and lack of awareness of confidence around effective interventions, including the prescribing of NRT. Procedures and time pressures were also frequently cited, such as a lack of time during midwifery appointments to discuss smoking cessation.

There were also issues specific to SSSs or referrals to these services in the two UK study sites. These are described in detail from the HP perspective in Chapter 10. This included concerns about the viability and communication of routine CO screening and opt-out referral, the latter of which was operating fairly well in area A but was not universally in place in area B and so was still questioned. Communication problems or lack of information in referrals were mentioned by both SSS advisors and other HPs in both areas. As in the studies in the systematic review of HPs, problems around NRT provision during pregnancy were also raised by the professionals interviewed. Professionals who were not SSS advisors still expressed caution about the safety or benefits of NRT in pregnancy, despite it being available on prescription and almost universally provided by SSSs. From the perspective of the pregnant women interviewed, some issues raised that related to SSSs included women providing examples of not being offered home appointments and of negative experiences with some advisors who were viewed as judgemental, resulting in a declining of future appointments or a disinclination to engage.

Organisational factors were also key to the issue of addressing postpartum relapse. The systematic review of HPs showed that interviewees were very unsure about how they could prevent relapse, and in one study it was described as almost inevitable. 133 HPs correctly pointed out that the research evidence on how to support women to avoid relapse was very thin.

Improving the organisational context for smoking cessation

The studies included in our systematic reviews57–59 were from a range of countries and some were conducted one or two decades ago, so their relevance for informing current organisational practice in the UK may be limited. In contrast, the interviews with all three groups of participants (pregnant women, partners/SOs and HPs) provide practical suggestions for improving current provision. This concerns the organisational context for addressing smoking in pregnancy in general (i.e. contact with any HP the woman and/or SOs have) and more specifically for SSSs.

The interviews revealed an ongoing reluctance or concern among some HPs about discussing smoking and smoking cessation in detail with pregnant women because of the perception that it could damage a valuable relationship. This is a prevailing theme in the literature including in the systematic reviews. 57–59 It is somewhat dispiriting, therefore, that concern about the negative impact on the HP’s relationship with the pregnant smoker continues to persist when the risks of continued smoking in pregnancy and the benefits of cessation are so well established. Nevertheless, this concern remains, and it was evident in both study sites in the UK. On this topic, as outlined above, support and training for HPs are key, as are time and space during an appointment with a client/patient to raise and explore this important issue. It is also worth noting that, among midwives, there was some evidence that the introduction of an opt-out referral pathway to a SSS at maternity booking can help. The interview data indicated that midwives in area A, where this policy was in place, better understood smoking cessation as a priority and understood that there were steps in place for access to specialist support beyond the time they could afford during the appointment.

For HPs in general, other practical steps could be taken to improve how the issue of smoking cessation in pregnancy is managed. From the interviews, the most obvious is the importance of staff availability and capacity to deal with the many issues that they are responsible for, including advice on alcohol and weight management. There were also examples in both areas in which improving protocols to include more effective prompts, and providing clear information for HPs regarding smoking, could effectively elevate its perceived priority. This included clear information about NRT, greater confidence in which could be built among midwives, pharmacists and others by improving access to information about its safety in pregnancy. In addition, despite limited evidence of NRT’s effectiveness in clinical trials, its value in routine care could be useful, as illustrated by the positive outcomes for women who used it in both areas A and B (from the available routine SSS data from both areas).

Another practical step, among SSS staff in particular, is to enable staff members to be more fully involved in service development, potentially allowing the service to make better use of staff expertise, to increase staff buy-in and to reduce the feeling among some staff that changes are imposed from elsewhere. Other practical changes revealed in the interviews to improve the organisational context for smoking cessation in pregnancy were also important. Although the opt-out pathway was in place in area A, there were still reports that referrals could be low; however, it is important to note that, at the time of the research, this new process might not have been fully embedded. Delays in prescribing NRT (in the pathway between services and GPs) were also identified in both areas, and examples from elsewhere in the UK suggest that these can be addressed.

A very positive finding from this study was the value and, in most cases, reported acceptability of routine CO screening in pregnancy. This was useful both to identify pregnant smokers without stigma and as a motivational tool for those trying to quit. Routine CO monitoring is recommended in NICE guidance40 and other recent policy documents,91 and represents a valuable development in the UK’s approach to helping women stop smoking in pregnancy. The lack of reference to CO monitoring in the reviews reflects the relatively recent introduction of this tool.

A remaining unresolved area for service development is how to best support women not to relapse to smoking in the postpartum period. More research is needed that can inform practice, particularly to challenge women’s views that smoking cessation in pregnancy is only a temporary measure and that postpartum relapse is assumed. One potentially important finding from this study’s interviews with women post partum, however, was that those who had stopped attending cessation services and had relapsed to smoking would have tried again if they had managed to maintain face-to-face contact with an advisor. This may point to the potential for services to do more to help women regain confidence following a lapse, including increased efforts to keep in contact with women and offer alternative appointments or home visits for those who have relapsed.

Key overarching issues for future research

A somewhat disappointing finding from our interviews was that many of the barriers to smoking cessation in pregnancy identified in our systematic reviews remain and have not been addressed by existing services or interventions, or indeed by changes in social norms around smoking or support for pregnant women in general. Some of these merit exploration in further qualitative research for which the literature is still limited, for example around the role of partners in facilitating or hindering cessation in pregnancy. Others, however, should be the focus of empirical research designed to inform interventions or improve existing practice.

Identifying the best ways to enhance the motivation, capability and confidence of HPs around smoking cessation in pregnancy remains key. Some of the HPs we interviewed reported that they were not always confident about the best way to approach the issue, how to prioritise it above or alongside other aspects of maternal health, or what to recommend. This can potentially be addressed through the improved training of relevant groups and better access to evidence, such as on the benefits (or lack of) of cutting down or on the use of NRTs. Data are now available from two recent trials of NRT in pregnancy, one of which followed up infants of women using NRT for up to 2 years (to provide evidence on safety), and these results are possibly not well known. Likewise, two recent studies illustrating that routine CO screening and opt-out referral increases the number of pregnant women who stop smoking in the UK is important evidence to convey to HPs who are unsure about this type of system change (as observed in area B in our study).

Some pregnant women in our study reported that, although they were told about the risks of smoking in pregnancy, they were offered limited explanation of the mechanisms of harm or how smoking was harmful to the fetus and infant. Evidence on these issues exists, and further research could explore how best to convey this to HPs and, in turn, how they can develop the confidence to discuss it with pregnant women and others, including through the use of visual aids and tools.

Within existing interventions, concerns about weight both during pregnancy and, particularly, post partum need more attention and, potentially, further research. Women who smoke are aware that smoking helps them to stay slim (through both nicotine’s effects on metabolism and smoking instead of snacking) and that weight gain with smoking cessation is common. Although this is an issue for cessation interventions in general and is not limited to pregnancy, it merits further exploration, as it is a real and persistent barrier to cessation and a contributor to relapse.

Likewise, more research on relapse prevention is urgently required. We have already made this point elsewhere in this report, but the evidence base remains very poorly developed and, perhaps as a result, services and interventions are not well developed.

It is worth noting also that these gaps in research and priorities for improving practice are consistent with recent recommendations for improvement in this area published by the multiagency Smoking in Pregnancy Challenge Group cochaired by the PI of this study (along with the chief executive of a leading baby charity) and co-ordinated by ASH. The Challenge Group report from 201391 and its 2015 update92 should be consulted alongside the results of this study, as we believe that the two together provide important insights into how we can support pregnant women to stop smoking and reduce the harm that maternal smoking causes to mothers and children.

Financial incentives

In addition to priorities for research and improved practice discussed, newer approaches merit further study. The first of these is financial incentives. Adding financial incentives to behaviour change interventions and services is controversial, but in the context of this study it may be directly relevant to the SEF. Incentives arguably operate to modify some of the wider determinants of smoking and cessation, providing more money for the household and potentially easing income-related sources of individual and interpersonal stress. They provide women from smoking communities with an additional ‘reward’ for engaging in a behaviour change process that may be difficult to achieve. In the interviews, women, partners and HPs expressed an interest in this topic, although some were cautious about the acceptability of ‘paying’ women to stop smoking.

The study took place at a time when incentive schemes were in place in areas relatively near the study sites (one long-standing programme in Tayside in Scotland and a scheme established 3 years ago across the north-west of England). Participants had heard about these schemes either directly or in the media. In addition, while some of the postpartum interviews were taking place, members of our team published the world’s largest trial to date of incentives for smoking cessation in pregnancy, which built on earlier studies in the USA and involved a Phase II trial with just over 600 women in Glasgow93 that was extensively reported in the media at the time. Awareness of the principles of incentives, sometimes described as a ‘nudge’ intervention, was high.

Incentives are a valid and important area for future research. At the time of completing this report, members of our team had submitted funding proposals for a larger Phase III multicentre trial based on the earlier study and the majority of funding for this research has now been secured. This would combine cessation services support with incentives delivered at four time points during pregnancy, in the form of shopping vouchers up to a total value of £400. Results from the earlier trial supported those from US studies that showed that incentives were both clinically effective and cost-effective. 93,178 In fact, the Phase II trial showed that this combination of NHS support and vouchers resulted in biochemically validated abstinence rates from smoking at the end of pregnancy that were 2.5 times higher in the intervention group than in the control group. There is a need for a multicentre trial to test whether or not this can work as well in different health and social care settings and communities in the UK.

Self-help interventions

Interviewees also expressed an interest in new approaches to allow women and family members to ‘help themselves’ in the context of smoking cessation. On this topic, future interventions could focus on self-help approaches. A previous systematic review96 conducted by some members of the team found that self-help interventions for pregnant smokers almost doubled the odds of smoking cessation compared with usual care. Following this, we investigated the potential of using text messaging to deliver tailored self-help support to pregnant smokers,141 and designed and evaluated the feasibility and acceptability of an intervention using tailored written and text message support for pregnant smokers recruited via antenatal care (MiQuit). 179

In terms of current work, after refining the intervention and restricting delivery to tailored text message only, two parallel streams of evaluation research are being undertaken: evaluating the impact of MiQuit on smoking behaviour and assessing its uptake when offered for use in real-world contexts. For impact evaluation, a further trial of MiQuit [Coleman T, Naughton F, Cooper S, Sutton S, Parrott S, Hewitt C, et al. The MiQuit study: RCT and meta-analysis testing effectiveness and cost effectiveness of a tailored text message programme for smoking cessation in pregnancy (CRUK Population Research Committee)] is being undertaken. The results from this trial will then be combined with two previous trials in a pooled analysis to estimate the intervention effect on abstinence compared with a standard cessation leaflet comparison group. 94 For the uptake evaluation stream, we have assessed or are currently assessing smoking cessation in the following settings: (1) when promoted via a leaflet in maternity booking notes (without HP input),95 (2) when promoted by HPs via a leaflet in antenatal care, (3) when promoted on pregnancy test kits in 10 ASDA supermarkets180 (ASDA Stores Ltd, Leeds, UK) and (4) when promoted via paid for and free online advertising/links.

Formative work is soon to start to inform the design and development of a NRT adherence intervention for pregnant smokers provided with or using NRT, as part of a National Institute for Health Research (NIHR) Health Technology Assessment (HTA)-funded programme. A major component of this proposed intervention is likely to be delivered via digital self-help. Overall, therefore, self-help interventions represent an important area for further research on smoking cessation in pregnancy and may be promising for the future.

Social networks and other interventions

The findings of this study illustrate that smoking in pregnancy is heavily influenced not only by the individual characteristics of women but also by women’s interpersonal relationships and the communities they live in. It therefore seems a natural extension to suggest that services and interventions should also be rooted in, or at least take account of, this context. Good-quality behavioural support (of the type provided by trained advisors on smoking in pregnancy) should aim to include discussion of all these aspects with women and may involve practical help such as signposting to other local services that can assist with social or financial problems. Some services also actively aim to engage the pregnant woman’s partner or SOs (such as a parent or close friend) in the interventions offered, particularly if that individual also smokes; however, interventions that actively involve partners are underdeveloped in the UK and elsewhere. The women interviewed in this study still experienced support to stop smoking as an intervention focused on them rather than one involving their wider social network.

Valente181 described social network interventions as ‘purposeful efforts to use social networks or social network data to generate social influence and/or accelerate behaviour change among individuals, communities, organisations, or populations’. Studies that proactively involve social networks in smoking cessation have mainly centred on the general adult population182 and remain under-researched in pregnancy. Only five such studies were identified in the most recent Cochrane review183 of psychosocial interventions to support women to stop smoking in pregnancy. Evidence from this review suggested that social support interventions appeared effective when provided by peers (five studies: average risk ratio 1.49, 95% confidence interval 1.01 to 2.10), but the effect was unclear in a single trial of support provided by partners.

Determining precisely how a social network intervention for smoking cessation in pregnancy would be delivered and what it would involve is complicated. Identifying social network members and ascertaining whether or not and how they could support the pregnant woman who is smoking are not straightforward. For these reasons, some of the authors of this report have already undertaken a small feasibility study (funded by CRUK) that aims to address the following questions:

• Can social networks be used to help pregnant women stop smoking? If yes, how can they be used, what support can they offer and how should they be supported?

• What stage of pregnancy should the intervention be aimed at and should prevention of postpartum relapse be included or not?

• How will social network members be defined, identified, approached and positively engaged (i.e. how can they be encouraged to take part)?

• What will the intervention components be, and what level of resource will be required for a future pilot?

• What are the barriers to using social networks for health behaviour change and how can they be overcome?

• What capacity is there for social network approaches to be used to address other public health issues (obesity, alcohol consumption, diet and exercise)?

This study will report in 2017.

It is also worth noting that these suggestions for new interventions should sit alongside efforts to improve or further investigate what is already happening. We touch on this theme above, but it is clear that more research on the efficacy and safety of NRT in pregnancy is required that will help to improve the confidence of staff in prescribing NRT or providing information about these products to women. The awarded NIHR programme grant mentioned above includes a study that tests higher-dose (combination therapy) NRT with pregnant women (already used by some SSSs, as nicotine is metabolised more rapidly in pregnancy and standard doses do not have good evidence of efficacy) alongside the adherence research.

Finally, there is evidence that pregnant women who smoke are using e-cigarettes, which are currently the most popular aid to stopping smoking in the UK among the general adult population of smokers and are used in around one in three quit attempts. Members of the team have just completed a qualitative study of e-cigarette use among pregnant women in England and Scotland and are now beginning a UK-wide survey on this issue; both of these have been funded by CRUK. In addition, we are about to begin (with colleagues) a trial of e-cigarette use in pregnant women in England and Scotland, funded by NIHR HTA. Results from these new studies, along with studies under way in the USA, will provide much-needed evidence on e-cigarettes for smoking cessation among pregnant women.

Systematic reviews and interviews

The contemporary accounts of women, their SOs and HPs in our interviews reflected themes that are prevalent in the international literature and were identified in our three systematic reviews. 57–59 Overall, there was a great deal of consistency between the systematic reviews and interviews, with some important exceptions.

Although the review studies were conducted in previous years and in both the UK and other countries, findings, particularly at the individual and interpersonal levels of the SEF, were similar between the review studies and our interviews. In high-income countries, there are marked socioeconomic inequalities in smoking in pregnancy and, therefore, it is not surprising that, for the most part, the role of disadvantage in women’s lives was relevant to their smoking. Women who smoke in pregnancy are more likely to have started in their early to mid-teens, to live in areas with many other smokers and to suffer financial hardship. These factors shaped the prevalent accounts from women and partners about smoking as a coping mechanism, a rare (and shared) break and pleasure. Any future interventions should acknowledge the context in which the majority of smoking in pregnancy occurs. These factors also influenced how HPs approached discussing smoking with pregnant women, often recognising the other difficulties they faced, prioritising developing a good relationship with their clients/patients and continuing to imply either directly or indirectly that cutting down was ‘good enough’. Women who managed to engage with SSS staff benefited from the staff’s knowledge and expertise in supporting women with challenging life circumstances and from techniques, tools and aids to promote complete abstinence from smoking as a realistic goal.

Areas in which the review findings differed from the interview findings were specific aspects of the influence of HPs, the viewing of cessation as a temporary change and risk perceptions. In the first area, the review found that women and partners often described HPs as ambivalent or negative about trying to stop smoking, perhaps reflecting previous social norms on smoking, less understanding of risks or local contexts in which cessation services were not in place. This perception was not prevalent in our interviews, but there were exceptions. Women expected to be asked about smoking and, for the most part, they were indeed asked and HPs aimed to be facilitative and understanding. In terms of SSS staff in particular, most women who accessed the services were positive about the support of their smoking cessation advisor.

Second, the reviews also indicated that cessation in pregnancy was widely regarded as a temporary change, with relapse post partum seen as largely inevitable. This may reflect different contexts and study populations from those in our study. The small number of women interviewed post partum cautions against firm conclusions being drawn. However, although relapse post partum was a theme in these interviews, women were clear that they wanted to remain non-smokers. It is also worth noting that we did not have any interviewees who had successfully stopped during pregnancy and who then started smoking again after birth. All of these non-smoking women expressed a strong desire to keep away from tobacco.

Finally, in the systematic review of pregnant women’s views, risk perceptions of smoking in pregnancy often operated as barriers to, rather than facilitators of, quitting. We certainly found some evidence of this in our interviews, not only with women but also with SOs. However, there were other examples of women who genuinely believed that smoking was harmful and who gave concrete instances of children they knew who had suffered ill effects, although these were few. Accurate perceptions of real harm did seem to help foster determination to quit, and this was also recognised by HPs, some of whom explicitly expressed the importance of being very truthful with women and family members regarding the potential impact of smoking in pregnancy.

Strengths and limitations of the research

This study had a number of strengths. These included the production of three new qualitative systematic reviews57–59 on topics of current importance for policy and practice. Qualitative reviews have a valuable role to play in drawing out key themes on topics of contemporary interest from the perspective of people’s lived experiences, providing a rich source of information that can capture common issues across time and different national and organisational contexts. We updated a previous systematic review, conducted by members of our team, that examined pregnant women’s views, and then carried out two new reviews on topics not previously examined systematically: SOs’ and HPs’ views on smoking and smoking cessation in pregnancy. The systematic review of partners and SOs was particularly useful, as it demonstrated the very considerable lack of evidence supporting the views of (primarily) male partners regarding smoking in pregnancy. In the systematic review of partners’ views, most of the studies were from Canada, with only one UK study included that involved only five male partners. By conducting interviews with many more male partners in our study (as well as mothers, friends and a mother-in-law of the pregnant women), we have significantly expanded the available qualitative literature on this topic.

A second strength of our study was the in-depth nature of the research we conducted with a reasonably large sample of pregnant women, their SOs and HPs. Each of these samples aimed to capture a range of individuals in different circumstances (i.e. by smoking status) and, for the HPs in particular, from a number of different occupations. Our interviews were, in many cases, long and detailed. They also aimed to provide constructive information on current and future services, and interventions to inform policy and practice in the UK context. A final strength was the detailed synthesis of findings between the systematic reviews and our qualitative data, examining areas of similarity and difference and structured around a theoretical model, the SEF, which provided a ‘lens’ through which to view and interpret our data.

In terms of limitations, the design of the research meant that our findings are grounded in qualitative studies; this enabled a sustained focus on perceived barriers and facilitators and not on outcomes relating to smoking or smoking cessation in pregnancy. Although the in-depth nature of the work is one of its strengths, it meant that sample sizes, both in the systematic reviews57–59 and in our primary studies, were smaller than those in quantitative studies, although they were still large for a qualitative study. Similar to other studies, we had particular difficulty recruiting partners and SOs in both sites (see Chapter 2), initially because of the requirements of the study set by the research ethics approval process and latterly simply as a result of men’s reluctance to be interviewed. We aimed to address this by alternative referral routes, again described in Chapter 2, but this did mean that some of the partners we interviewed in area A were not those of women in the study, limiting our ability to look at couples together during our data analysis.

An additional limitation relates to our study sites and how a combination of local service configuration and our recruitment processes may have produced a skewed sample. In both areas, we knew from previous experience that the best way to recruit research participants was to work with existing services and not try to disrupt or replicate local systems. We therefore tailored our approach to these established systems. In area B, members of our team were already working with a research midwife who had links to the maternity booking setting and therefore women could be recruited through that route. In area A this was not the case, but our team had an established relationship with the local SSS; this meant that we recruited women who already had some contact with the service, no matter how brief, building on the opt-out pathway in place in that site. In contrast, some women in area B had no contact with the SSS. Looking at our data now, we believe that this has resulted in a useful cross-section of views emerging in the findings across the two sites. However, it does mean that we have not captured the views of those women in area A who smoke or smoked and did not have any contact with the local SSS.

Further limitations relate to the methods used in some instances. We conducted one-to-one in-depth interviews with HPs, except for some smoking cessation service staff and midwives. In those instances, we chose to conduct a small number of focus groups, as we had a number of staff who were interested in participating, and conducting individual interviews with all of them would have been logistically challenging and time-consuming. We believe that the focus group discussions were useful and insightful, but it is possible that discussing the issues in a group meant that not all divergent views were captured. In addition, we chose from the outset not to follow up all pregnant women in the study postnatally but instead followed up only a subset of women who agreed to this at recruitment. This represented a good cross-section of women in different circumstances but means that we did not capture all women’s views in the postpartum period. These and other potential limitations should be kept in mind when reviewing our results.

Social–ecological framework

As we discussed in Chapter 2, in the initial months of this study we decided to draw on the SEF as a theoretical model for the research, as our initial scoping for the systematic reviews emphasised that environmental barriers and facilitators, operating at a range of interconnecting levels, were key to understanding smoking cessation in pregnancy. This approach has drawn attention to some overarching issues that have wider implications beyond our own study and are worth highlighting in the concluding paragraphs of this report. Two main issues emerge. The first is that it is helpful to locate interventions within the SEF and consider which aspects of the interventions offered address (or do not address) different elements of the framework. Different types of interventions will primarily operate at one or two levels but not all. For example, self-help materials or e-cigarettes are located at the individual level, family-based approaches are located at the community level and SSSs are located at the organisational level. The success of interventions at one level could be undermined by factors operating elsewhere in the framework; the assessment, measurement or attempted modification of these could be a viable area for research, and identifying the gaps could be constructive.

The second issue is that this framework recognises and encourages thinking beyond the individual domain. Most studies of behaviour change interventions, including smoking cessation studies, have a strong focus on the individual. Treatments, services or even policies are offered or introduced to smokers, and outcomes are then observed, most commonly in controlled studies in which one group receives one or more interventions and another receives something different. Even the ‘future interventions’ we identified in our study, and that we asked interviewees about, focus fairly heavily on the individual. Future research, and future service development, needs to go beyond a focus on the pregnant woman and her baby to wider networks and contexts. One approach may be to focus on thinking of organisations as particular settings for interventions, such as the workplace for pregnant women, their partners and SOs, and NHS settings for HPs.

Overall, our findings suggest that, to be effective in supporting smoking cessation in pregnancy, consideration needs to be given to all layers of the SEF and how they relate to individual women’s needs and experiences. Ideally, interventions, or connected sets of interventions, should simultaneously operate at multiple levels. This is challenging but provides a basis for thinking about how best to approach future studies on behaviour change in pregnancy and, indeed, during other stages of the life course and in different populations.

Cumulative Index to Nursing and Allied Health Literature (CINAHL)

Searched via EBSCOhost interface on 26 April 2013.

Limiters – published date: 20120101–20131231; English Language.

Notes: update of original Public Health Research Consortium (PHRC) review49 search strategy, minus the social disadvantage terms, limited to 2012–13 only.

Social Sciences Citation Index

Searched via Web of Knowledge interface on 26 April 2013.

Databases = SSCI; timespan = 2012-01-01 – 2013-04-26; language = (English).

Notes: revised update of original PHRC review49 search strategy, minus the social disadvantage terms, limited to 2012–13 only.

Economic and Social Research Council

Searched via www.esrc.ac.uk/search/advanced-search.aspx on 26 April 2013.

Notes: the Economic and Social Research Council website’s advanced search engine is configured to use the Porter Stemming plugin, which reduces search words down to their base word. Update of original PHRC review49 search strategy, minus the social disadvantage terms, no date limitation.

PubMed (including MEDLINE)

Searched via the National Center for Biotechnology Information (NCBI) interface, on 9 May 2013.

Limited to Ahead of Print Citations.

Notes: update of original PHRC review49 search strategy, minus the social disadvantage terms.

Google Scholar

Searched via http://scholar.google.co.uk on 8 May 2013.

Limited to English language pages only and 2012–13 to capture ‘ahead of print’ but ‘published online’ articles excluding patents; cookies deleted between searches.

Cumulative Index to Nursing and Allied Health Literature (CINAHL)

Searched via EBSCOhost interface on 9 January 2014.

Limiters – published date: 19900101–20140131; English language.

PubMed (including MEDLINE)

Searched via NCBI interface on 10 January 2014.

Limited to Ahead of Print Citations and articles published in the last 2 months.

Cumulative Index to Nursing and Allied Health Literature (CINAHL)

Searched via EBSCOhost interface on 26 February 2014.

Limiters – published date: 19900101–20141231; English language.

Social Sciences Citation Index

Searched via Web of Knowledge interface on 25 February 2014.

Databases = SSCI; timespan = 1990–2014; language = (English).