Volunteered, negotiated, enforced: family politics and the regulation of home smoking

Introduction

The investment of resources in tobacco control, including smoking cessation campaigns, advertising restrictions, tax and price increases and the increased provision of services to help smokers quit have led to a significant decline in the rates of cigarette smoking in many western countries (Davis et al. 2007). More recently, concerns over the exposure of people to secondhand smoke have led some countries to introduce population-level policies to restrict smoking in public places. Declining rates of smoking have been supported by changing attitudes towards the practice, partly due to a growing awareness of the risks to health of exposure to secondhand smoke, which has led to a marked shift in the ways in which both smoking and smokers are perceived by the wider population (Bayer and Colgrove 2002, Louka et al. 2006). In addition, research has shown that an increase in negative attitudes towards smoking can reduce rates of smoking and smoking behaviour at a community level (Smith et al. 2008).

However rates of smoking are still unacceptably high in countries such as the UK, with an average prevalence of 21 per cent of people aged over 18 years, rising to 25 per cent for people working in routine and manual occupations (Robinson and Lader 2008). While public health measures have successfully protected people in public places and have been linked to reduced rates of home smoking (Borland et al. 2006), there are currently no statutory restrictions on smoking in private places, leaving home smoking as ‘one of the last frontiers’ of tobacco control (Thomson et al. 2006). Although secondhand smoke exposure in adults is associated with early death and the development of heart disease (Jamrozik 2005), concerns over continued high levels of smoking in the home tend to focus on the exposure of younger children to it, as they are more likely to spend time in the company of their carers (Ferrence and Ashley 2000, Graham 1993b), less likely to move away from the source of the smoke (Robinson and Kirkcaldy 2007a), and are more susceptible to the short-term and long-term health risks of being exposed to secondhand smoke (Cook and Strachan 1999, Muller 2007).

Smoking has been viewed as an individual life-style choice and past studies of mothers suggest that they were aware only of its adverse consequences to their own health rather than to the health of their children (Graham 1993a). However, the changing social context of smoking from being a mainstream behaviour to one practised by a stigmatised minority (Bayer and Stuber 2006), has meant that parents are increasingly aware of how their smoking is perceived by the wider population and may be motivated to change their smoking behaviour (Blackburn et al. 2005, Bottorff et al. 2006a, Wakefield et al. 2000). While family life is influenced by people in the family and may be personal, private and idiosyncratic, the family is also a social institution and both its form and it function is affected by social and economic changes in wider society (Allan and Crow 2001).

The home of a child can be viewed as the site of social and smoking activities that extend beyond the smoking activities and status of the residents. A number of research studies have looked at household dynamics and smoking, focusing on couple dynamics and interactions in nuclear families (Bottorff et al. 2005, Bottorff et al. 2006b, Greaves et al. 2007). Yet sociological perspectives no longer conceptualise families or households solely as nuclear family units and have accepted that identifying members of one’s family or household is as much a personal as it is a social construct (Morgan 1999). Recent research has also highlighted how the boundaries between the private social world of the home and family and the wider external world are increasingly blurred, with new models of families and relationships combined with changes in technology creating a diverse and rich social environment (McKie and Cunningham-Burley 2005, Smart 2007).

Researchers have identified the importance of the social and cultural context of smoking (Chapman and Freeman 2008, Poland et al. 2006), yet few researchers have looked at how family politics can affect smoking behaviour in households. In a study by Kegler et al. (2007) changes to smoking practices involved participants recounting examples of verbal and active resistance by different family members, and the research of Bottorff et al. points to the essentially gendered nature of the negotiation of smoking bans (Bottorff et al. 2006a, 2009, 2006b). This links to wider sociological research that has looked at the ways in which families make economic, social and personal decisions, and which has challenged assumptions that there are any natural or taken for granted patterns of family decision-making (Finch 1989, Finch and Mason 1992). This suggests that research is needed to explore what strategies families need to employ to successfully introduce smoking restrictions in their homes, and how wider social relationships within and between households may contribute to the effectiveness of home smoking restrictions (Bercovitz et al. 2005, Borland et al. 2006, Yousey 2007). This article explores how positive messages about the need to protect children from tobacco smoke are transmitted and discussed by adults, and how they attempt to extend the protection of children outside their own household into that of others.

The study

In 2006, NHS Health Scotland commissioned a research project, ‘Qualitative Study of Smoking in the Home’, to explore any changes to smoking behaviour in Scotland after the introduction of the smoking legislation on 26 March 2007. The legislation made it an offence to smoke in enclosed (more than 50% covered) public spaces and in vehicles owned by companies. The participants were purposively recruited from 106 people who had taken part in Wave 10 of the Health Education Population Survey (HEPS) in Scotland in 2005, and had indicated that they would be interested in future research. The first contact was made by the British Marketing Research Bureau who had carried out HEPS, and 54 people who were eligible and willing to take part responded. The final sample of 50 men and women over 18 years old was based on three characteristics: the pattern of household smoking; their socioeconomic group and their gender. As smoking prevalence is associated with poverty and disadvantage the sample was weighted towards people from lower socioeconomic groups. Socioeconomic status was designated using the National Readership Survey demographic classification, where occupation, education and income are designated by letters A to E, with A representing the most affluent socioeconomic group.

In Phase 1 of the study, 50 smokers and non-smokers living with smokers were interviewed in their homes using a semi-structured interview schedule. The interviews lasted between 40 and 90 minutes and included questions on home smoking restrictions and when, why and how any such changes had been made. We were aware that the view of any individual family member of the dynamics of family life is likely to be partial, and our picture of home smoking depends on who is addressed and their status in the family (Allan and Crow 2001). We therefore carried out a second phase of interviews (Phase 2) with other family members to explore the possibility of divergent accounts of home smoking. Based on the sampling strategy for Phase 1, we selected three smokers, three non-smokers and three adults living with children from each socioeconomic group from the 14 people who were willing to take part. Eight of these people were the partners of Phase 1 participants and one was the mother of a Phase 1 respondent (see Table 1).

The data were analysed thematically using NVivo to assist in the development of analytical codes and aid constant comparisons between transcripts. The accounts of home smoking from the nine families who had taken part in Phase 1 and Phase 2 were found to be remarkably congruent. A key finding of the initial analysis was that, with some exceptions, the participants had moderated their home smoking behaviour in a variety of ways, although patterns of home smoking restrictions were common across socioeconomic groups and home smoking profiles (Phillips et al. 2007). To deepen and to further account for the differences between households, the original research team was joined by two new members, JR and LG, to apply aspects of social theory to see whether they had any explanatory power. Using a collaborative and consultative approach, the transcripts were read and re-read and existing codes were re-examined by JR, with a particular regard to any gendered nuances of the data. Any new themes or insights in the data were discussed with the whole team through e-mail, telephone conferencing and face-to-face meetings. The original team members were able to brief the new collaborators as to the context of the data and were able to comment on the scope and focus of the new interpretations of the data (Mason 2007).

Overview of the findings

The findings suggest that people had been modifying their smoking behaviour over the years prior to the introduction of the legislation in Scotland, and most participants describe experiencing a gradual awareness that smoking near children was ill-advised. However the mechanisms for transforming this awareness into actual decisions to modify their smoking behaviour in home settings to protect children, and the nature of the decisions made, varied in three ways. For some participants the process was imperceptible, apparently the result of internalising a social norm, and decisions were taken without any memorable formal discussions and without conflict. We have described this as voluntary decision-making. For other people, such decisions were discussed with others over time and on a number of occasions with friends and family, which we have called negotiated decision-making. However a minority of people in the study described how either they or someone in their family had imposed smoking sanctions with the threat of some form of social sanction if they were not complied with.

The voluntary introduction of smoking restrictions

For many participants the decision to restrict smoking near their children was not made as the result of a direct request or through discussions but appeared to reflect the gradual internalisation of the need to protect children from tobacco smoke. A non-smoking woman with a 35-year-old husband who smokes, who lives with their children aged 5 and 7 years, described how her husband came to smoke outside their home:

This suggests that for some people, the decision not to smoke near children is undertaken tacitly by the smoker(s), and was not discussed or questioned. This 75-year-old man, who lives with his non-smoking wife and has four children, also described deciding to smoke outside without a direct request from them, although he does cite comments from his grandchildren as a possible motivating factor:

Such comments may have induced this grandparent to change his behaviour by making him feel guilty about the response of his grandchildren to his smoking. However, the grandchildren’s reaction to his smoking does not seem to have been reinforced by any direct requests from their parents (his children), and so it is worth noting that this grandparent volunteered to take his smoking outside in response to his grandchildren’s negative comments.

Previous research suggests that home smoking bans are easier to introduce in houses where only one person smokes (McMillen et al. 2003, Pizacani et al. 2003), and in this study, in 21 out of the 24 households where there was a non-smoker living with a smoker, there were either some smoking restrictions (12) or a total ban on home smoking (9). The internalisation of the need to stop smoking near children did not extend only to parents of children but also to grandparents. This respondent, who lives with her smoking husband and has a ban on smoking in their home, noted that her parents-in-law’s decision not to smoke in front of her 1-year-old daughter required no prompting or discussion from them:

Although there were a number of factors that influenced their decision to ban smoking indoors, for this grandparent the arrival of his first grandchild was cited as the deciding factor:

Past research suggests that knowledge about child rearing tended to descend rather than ascend through the generations, accompanied by the transmission of goods, money, emotional support and even housing (Attias-Donfut and Arber 2000). However information about the need to keep children smoke-free appears to be passed from parents to grandparents, suggesting that such arguments may have the power to invert other normative social arrangements (Brannen 2006). Despite the apparent lack of need for any direct verbal communication between some family members about reasons to stop smoking when children are present, any changes to behaviour appear to be real and sustained in other settings. There is also some evidence of a positive cascade effect, where the people who have made the decision not to smoke near their own (grand) children, extend this to avoiding smoking near all other children:

This suggests that once people have started not to smoke in front of their children or grandchildren they are less likely to smoke in front of other children, which could not only protect other children from smoke but act as a positive example to people in other settings. There is also evidence from this study that the need to stop smoking near children is accepted in the wider population, even by people with little or no contact with children, suggesting that such parents are acting as positive role models for others by enacting the new social smoking norms:

Actively negotiating home smoking restrictions

Further evidence of the changing social norms around home smoking and children came from parents who had previously smoked in front of their older children but intended to change their behaviour to avoid smoking near their new baby. This 53-year-old father describes how he and his wife, both of whom smoke, decided not to smoke in their home after the birth of their child:

This respondent uses the term ‘collective’ to describe their decision-making, suggesting that there is equity in their relationship in terms of decision-making. While they decided to discuss the issue, it seems that there was agreement about the outcome and both modified their behaviour from that time. Patterns of decision-making between partners have been linked to how money and other resources flow through families (Finch and Mason 1992). In this family, the man has retired to care for their child while his wife (in the same profession as him) carries on working. While the wife is currently the principal wage earner they have shared a professional background and status for a number of years, suggesting financial and social equity. While men and women may appear to share a way of life as they live in the same household and eat the same food and so on, imbalances in employment and education and gendered expectations may mean that one partner is able to exert power and authority over another (Daly and Rake 2003: 987).

Generally, in the UK, despite changes in employment and educational opportunities, women remain disadvantaged in terms of controlling material resources and making major decisions, although this does not mean they do not exert power in the day-to-day running of the household and decisions around children (Daly and Rake 2003, Pahl 1989). This 19-year-old woman, who was pregnant with her first child and gave up work and smoking 6 months previously once she realised she was pregnant, recounted how, after some negotiation, her (working) partner modified his smoking:

This example suggests that people can adapt their smoking behaviour during relatively short periods of time, although this woman describes how she had to assert herself repeatedly to achieve what she wanted. In addition, her account suggests that her partner was motivated by a concern for their future child’s health rather than for hers. Becoming a parent is known to motivate people to modify their smoking behaviour and although fathers are not always targeted by healthcare professionals to give up smoking during their partner’s pregnancy or to change their smoking, having a supportive partner is known not only to improve the health of the woman and her child, but to decrease her chances of starting to smoke after the birth (Blackburn et al. 2005, Wakefield et al. 1998).

For some parents the decision to introduce smoking rules and restrictions to protect their children from smoke had to be communicated explicitly to other people, and most parents extended these restrictions to other people visiting their home if their children were present:

Although many parents did still permit some smoking in their houses at times when their children were not in the room and so had not introduced a ban on household smoking, they were adamant that they would not relax their particular rules for visitors. Contrary to previous research findings, which suggested that some parents felt it was inhospitable to ask someone to go outside to smoke (Robinson, 2008: 1093), most parents said that they would ask a visitor to leave or go outside if they did not conform:

Accounts of the use of more direct measures to ask for a smoke-free environment for children included ongoing arguments with relatives to enforce their compliance, suggesting that even those who have successfully protected their children in the past may still encounter incidents where they need to assert themselves and even times when their actions are ineffective. Such incidents suggest that parents’ ability to positively intervene to protect their children is affected by their own personal qualities (in terms of motivation and assertiveness), which is likely to be mediated both by their gender and relative power in that specific social context, which in turn relates to the quality of a particular social relationship (Jarrett and Jefferson 2003, Mumford and Power 2003). There is evidence in this study that some parents had tried to implement changes but were less successful. For example, this women describes how even if she moved her smoking away from her children, her partner continued to smoke in front of them, and her description of her withdrawal rather than an engagement in active confrontation suggests that she believed that her partner would not change his behaviour and she believed herself powerless to change it:

Given the scope of this study and the negative publicity around secondhand smoke in Scotland post-legislation, it is possible that participants were more likely to give accounts of how they successfully achieved a smoke-free environment for their children than instances when they failed to act or were unable to achieve this. However, what the positive accounts do provide is the evidence that some parents and grandparents have developed some forms of intervention on behalf of their (grand) children to protect them from secondhand smoke and in most cases, these restrictions would reduce or eliminate the exposure of their (grand) children to secondhand smoke while in their care. In the following account, this mother describes how her decision not to smoke near her child positively impacted on the behaviour of her sister with her own children and extended to her brother-in-law and also her cousins. Although she describes herself as having to be relatively direct, she does not report any real conflicts, as she maintains that she acted in the best interests of her child:

Enforcement of non-smoking environments with relatives

This type of positive assertion for the need for other people to moderate their smoking near children, even in their own homes, was more often made by a mother than a father and is found in different accounts in the study. Many of the people in this study who were aged 50 years and older were grandparents, and although few had given up smoking for their grandchildren, a higher number had moderated their smoking in some way in response to a request from their children. This 60-year-old grandmother had imposed strict rules on herself and her husband when her grandson visited:

That this woman talks about her and her husband not being ‘allowed’ to do something in their own home suggests that an external authority, possibly the child’s parents, has actively asked them to restrict their smoking while the child visits their home. This action resonates with accounts from other grandparents in the study, who describe how having a child motivated and empowered their children to demand a smoke-free environment for them while they visited. This 61-year-old grandmother described how her children told her that both she and her husband could not smoke in the house while her grandchildren were there:

Like other grandparents in the study, this grandmother looks after her grandchildren on a regular basis, and at the time of the study, looked after her 19-month-old granddaughter once a week for a full day. Many parents access informal care for children but it is used most frequently by parents, typically lone parents, of whom 80 per cent are women on low incomes (Fergusson et al. 2008, Skinner and Finch 2006). While the reasons mothers give for their preference for informal care are complex, informal care is usually free, non-regulated and often available at times when there is no statutory (and also non-smoking) childcare available, such as early in the mornings, after school, evenings, overnight or at weekends (Shivers 2006). The transmission of non-smoking norms into other houses is therefore an important social development for the future health of such children, particularly those living in disadvantaged areas where rates of smoking remain high (Holdsworth and Robinson 2008).

However, the requests made for people within the family not to smoke can become more confrontational and direct, and even coercive. This woman was told by her daughter that she would not be able to see her grandchild if she did not stop smoking, although this was later relaxed to not smoking in the same house:

This account is supported by the Phase 1 interview with her (other) daughter, who confirms that both she and her sister have always put pressure on their mother to stop smoking, and saw the birth of her first grandson as the ideal opportunity to motivate their mother to quit:

Despite the very real pressure that this woman has from her daughters, like other people in this study, she sees it ultimately as a reasonable, understandable request, at least in terms of not smoking near her grandchild:

This suggests that the social context of smoking has changed sufficiently for the demands made by parents on behalf of their children to be seen as ultimately reasonable, and so in this study at least, all the relatives and grandparents who were asked to provide a smoke-free environment for their children complied with the request, even if some parents had argued and enforced their case by using access to their own children as a lever.

Discussion

In line with the findings from other studies, we found that parents are increasingly aware of the risks that their smoking poses to the health of their children (Coxhead and Rhodes 2006, Robinson and Kirkcaldy 2007b). A key finding from the study is that few of the restrictions on home smoking were applied solely to family members or visitors to the home, or only to homes where children were normally resident. In this study grandparents and adults, including young adults without children, were mindful of the dangers of exposing children to secondhand smoke. This resonates with the finding by Borland et al. (1999) that people who smoked and occasionally spent time with children were more likely to make radical changes to their smoking behaviour than smokers who were frequently with children.

The ability for parents to utilise existing familial and social relationships to influence the environment of their child outside their immediate household is crucial, as research in the UK and the USA has shown that the younger children and school-aged children of non-smoking parents may also be exposed to tobacco smoke, and some households with non-smokers permit family and friends to smoke when they visit their homes (Borland et al. 1999, Hopper and Craig 2000, Jarvis et al. 2000, 1992). Furthermore, some children are exposed to tobacco smoke in homes of other people; hence a child’s home may not be their only site of exposure to secondhand smoke and their parents may not be the only source of such exposure. The development of strategies to protect all children from potential exposure to smoke in home settings therefore needs to go beyond the homes of parents who smoke to the homes of their friends, caregivers and relatives where they spend time.

The findings from this research suggest that a child’s home should be viewed as the site of wider social and familial activity rather than as an isolated setting where negotiated decisions around smoking affect only the people living in that home. Our research suggests that decisions about home smoking may affect visitors, and as these visitors are part of wider social networks, decisions not to smoke near children can transfer to other home settings where children visit. Other changes to smoking behaviour may be the result of discussions, negotiations and even conflicts between mothers and fathers, which can produce some form of resolution, but not necessarily consensus (Bercovitz et al. 2005, Greaves et al. 2007, Kegler et al. 2007). The results of this research suggest that successful negotiation depends heavily on the personalities of both those making the request and those they are making it to, making the outcome of such decision-making far from inevitable or predictable (Finch and Mason 1991). Issues around the relative power between men and women in households cannot be ignored, as there is evidence that some women are unable to change the smoking behaviour of their partners (Greaves 2005). However, these results suggest that some parents, particularly mothers, may be in a powerful position to ask for a smoke-free environment for their children, as access and sustained contact with their grandchildren was clearly important to a number of grandparents in this study, which is consistent with findings from research into ageing and quality of life (Bowling and Gabriel 2007).

Therefore, rather than relating ‘household’ smoking bans to homes as places, the household should be conceptualised in relation to the people living in those houses, as the members of any one household are likely to be members of a wider familial and social networks through which behaviour such as restricting smoking near children, can be transmitted. Unlike previous studies (Green et al. 2003, Robinson 2008), the accounts from this study suggest that few exceptions were made for visitors and there was generally a willingness to ask and even challenge people about the need to maintain a non-smoking environment for children. The findings from this study suggest that there needs to be further work to explore the social context of smoking and exposure to smoke in the home and the components and factors affecting normative change in the home, as the transmission of home smoking restrictions to all homes is a crucial dimension to the creation of smoke-free home environments for all children.