Health Promotion International, Vol. 14, No. 3, 231-240,
September 1999
© Oxford University Press 1999
Patterns and predictors of smoking cessation among British women
Department of Applied Social Science, Cartmel College, Lancaster University, Lancaster LA1 4YL and 1 MRC Medical Sociology Unit, 6 Lilybank Gardens, Glasgow G12 8RZ, UK
Address for correspondence: Hilary Graham Department of Applied Social Science Cartmel College Lancaster University Lancaster LA1 4YL UK
 |
SUMMARY |
|---|
 TOP SUMMARY INTRODUCTIONSOCIAL AND INDIVIDUAL INFLUENCES...THE STUDY: METHOD AND...RESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Reducing the prevalence of cigarette smoking among adults is a key health promotion target in counties where tobacco consumption is the major preventable cause of ill health. This study describes the patterns and predictors of smoking cessation in a representative British survey of women aged 16 to 65. It examines the influence on quitting of five factors: socio-economic status; domestic circumstances; psychological health; tobacco dependence; and pregnancy status. Smoking status was measured at the point of recruitment to the study and at 1 year and 2 year follow-up interviews. Quit rates at both follow-up points were significantly related to three of the factors: socio-economic status; tobacco dependence; and pregnancy status. Tobacco dependence, as measured by average daily cigarette consumption, was the most powerful predictor of a woman's chances of being an ex-smoker at both the 1 year and 2 year follow-up interviews. The implications of the findings for health promotion policy are discussed.
|
INTRODUCTION |
|---|
|
TOPSUMMARYINTRODUCTIONSOCIAL AND INDIVIDUAL INFLUENCES...THE STUDY: METHOD AND...RESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Reducing the prevalence of cigarette smoking is a central goal of health promotion strategies in the UK and other countries where tobacco consumption is the major preventable cause of morbidity and premature mortality. These health strategies have followed a two-track approach, seeking both to prevent initiation into smoking and to encourage cessation. Because most smokers take up smoking in adolescence, strategies targeted at preventing initiation have little impact on smoking prevalence among adults. In adulthood, it is cessation which therefore holds the key to reducing smoking prevalence.
The majority of adult smokers want to give up smoking (Bennett et al., 1996
). In Britain, nearly 80% of current smokers have made at least one quit attempt, while US data suggest that a third of smokers quit smoking for at least a day each year (Hatziandreu et al., 1990; Bridgwood et al., 1996). Most of these quit attempts are self-initiated and rely on self-help and informal sources of support (Fiore et al., 1990; Bridgwood et al., 1996). In Britain, less than 5% of smokers report using formal treatment programmes when attempting to stop smoking (Bridgwood et al., 1996). Despite the desire to give up, the majority of quit attempts are unsuccessful. British surveys report that 75% of current smokers who have tried to quit, resume smoking again within 6 months (Bridgwood et al., 1996). While early reports pointed to gender differences in cessation in the general population, the weight of evidence now suggests that socio-economic circumstances, age and patterns of tobacco use are more important influences on cessation than gender (Jarvis, 1994; Bjornson et al., 1995; Jarvis, 1997). Among women, as among men, the challenge for health promotion is to increase the rate of cessation, a rate measured by the proportion of former smokers to ever smokers.
Designing and delivering effective policies requires an understanding of the complex factors which shape risk-related behaviours. The psychological processes and stages of change have received increasing attention within health promotion, and theoretical models of behavioural change are an important resource for interventions to encourage cessation (Ajzen, 1990; DiClemente et al., 1991). But it is increasingly recognized that whether and how individuals move through the quitting process is patterned by their social and individual circumstances (Jarvis, 1997). It is these broader social influences on cessation which are the focus of the current study. It draws on a large-scale longitudinal survey, the British Household Panel Survey (BHPS), to examine the patterns and predictors of smoking cessation among women in Britain in the 1990s. Subsequent stages of the study will extend the analysis to include men.
|
SOCIAL AND INDIVIDUAL INFLUENCES ON CESSATION |
|---|
|
TOPSUMMARYINTRODUCTIONSOCIAL AND INDIVIDUAL INFLUENCES...THE STUDY: METHOD AND...RESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Five factors have been identified as influencing the quitting process. Firstly, there is a sharp socio-economic gradient in cessation among both men and women (Hatziandreu et al., 1990
; Bjornson et al., 1995; Bennett et al., 1996). Among British adults in the most affluent circumstances, 60% of those who have ever smoked are now ex-smokers. Among those in the poorest circumstances, 15% are ex-smokers (Jarvis, 1997). The socio-economic differentials in cessation reflect a broader cluster of factors including educational disadvantage and poor current circumstances (Graham, 1993; Ockene, 1993; Jarvis, 1997).
Secondly, the domestic circumstances in which smokers live affect their chances of quitting. Smokers whose partners are also smokers have significantly lower rates of cessation (Jones, 1995; Nafstaad et al., 1996; Jarvis, 1997), and married/co-habiting smokers having higher quit rates than smokers who are not in a co-habiting relationship (Bjornson et al., 1995; Jones, 1995; Nafstaad et al., 1996; Jarvis, 1997). A number of studies has also identified lone parenthood as a barrier to quitting (White et al., 1992; Jones, 1995; Jarvis, 1997). In studies which have taken account of the confounding effects of SES, the influence of co-habitation status and lone parenthood remains (Bjornson et al., 1995; Jones, 1995; Jarvis, 1997).
Psychological factors represent a third cluster of influences on quitting. Lower rates of cessation are reported among smokers with psychiatric problems and high levels of dependence on alcohol and illegal drugs (Hughes, 1996), while good psychological health is associated with higher rates of cessation (Jones, 1995; Jarvis, 1997). Tobacco dependence is the fourth, and closely related, set of influences. Physiological addiction, psychological need and social reinforcement are recognized to be elements of tobacco dependence, with about 20 cigarettes a day needed to maintain steady-state nicotine levels through the day (Heatherton et al., 1989; Shiffman, 1991). Along with time to first cigarette of the day, average daily consumption is the most widely used measure of dependence. Mean daily consumption exhibits a strong socio-economic gradient. Among female smokers in Britain, the proportion smoking 20 or more cigarettes a day increases from 23% in the highest socio-economic group to 41% in the lowest group (Bennett et al., 1996). In British studies which have examined the effects of socio-economic status and cigarette consumption, both remain significant predictors of cessation (Jones, 1995; Jarvis, 1997).
Finally, key life transitions and events provide an important trigger to smoking cessation, prompting a re-evaluation of taken-for-granted behaviours, e.g. smoking (Stott and Pill, 1990). Pregnancy has a particularly marked cessation effect, with significantly higher rates of quitting reported among expectant mothers than among the general population of women. In Britain, approximately one in four smokers give up smoking in pregnancy; in the US, cessation rates of around 40% are reported among pregnant smokers (White et al., 1992; Ockene, 1993). Giving up smoking in pregnancy is strongly patterned by the broader influences on cessation. Cessation rates are higher among pregnant smokers in higher socio-economic groups, among women with a partner who does not smoke, and among light smokers, with each of these influences exerting its own and independent effect on the chances of quitting (Ockene, 1993; Olsen, 1993; Wakefield et al., 1993; Severson et al., 1995). The cessation effect of pregnancy is parity related, with primiparous mothers exhibiting significantly higher rates of quitting in pregnancy than multiparas (Waterson and Murray-Lyon, 1989; Nafstaad et al., 1996). Despite the high rate of postpartum relapse, pregnancy has a long-term cessation effect (McBride et al., 1992; Severson et al., 1995). However, its effect does not explain why year-on-year cessation rates in Britain are higher among women than men between the ages of 16 and 34 (Buck et al., 1994; Jarvis, 1994).
Longitudinal studies of smoking cessation have typically focused on particular sub-groups within the smoking population, e.g. expectant mothers and those enrolled in treatment programmes, where monitoring occurs across relatively short time periods (McBride et al., 1992; Bjornson et al., 1995; Susser, 1995). Longitudinal data on the wider population are provided by the British birth cohort studies which follow individuals born in the same week through childhood and adulthood (Mann et al., 1992; Jarvis, 1997). The 7—10 year gap between data sweeps enables the long-term predictors of cessation to be identified but not the shorter-term changes which characterize quitting behaviour. Their restriction to one age group also makes it difficult to separate out cohort-specific factors from those operating across the population as a whole.
|
THE STUDY: METHOD AND MEASURES |
|---|
|
TOPSUMMARYINTRODUCTIONSOCIAL AND INDIVIDUAL INFLUENCES...THE STUDY: METHOD AND...RESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
The study examines the 1 year and 2 year patterns of cessation among women in Britain. It is based on the British Household Panel Survey (BHPS), a representative longitudinal survey of the adult population of Britain living in private households (Buck et al., 1994
). The 1991 baseline survey collected information on personal and household circumstances, with smoking status measured both at the baseline survey and in the annual follow-up interviews. The BHPS thus permits analysis of the patterns and predictors of short-term quitting (between one year and the next) and longer-term quitting (across more than 1 year).
The BHPS is based on a two-stage stratified clustered design, the standard design used in household surveys in Britain. All resident members of the household aged 16 and over were eligible for inclusion. The 1991 baseline survey achieved a 95% response rate and a household sample of 5538 households (Freed Taylor, 1995). The data reported here are drawn from the first three waves of the survey (1991—1993). With little change in the patterns of women's smoking in Britain since then, these data are likely to provide a reliable base for the analysis of patterns and predictors of cessation.
The present study includes women aged 16—65 who were smokers in 1991 and for whom data were collected in the two follow-up interviews in 1992 and 1993. Because tobacco use is influenced by the smoking behaviour of other household members, only one respondent per household was selected from households containing two or more women. This yielded a sample of 3060 women.
Current cigarette smokers were those who stated that they smoked one or more cigarettes a day at the baseline survey. This measure produced a baseline group of 934 smokers. Repeating the measure at the 1 year follow-up permitted the identification of continuing smokers and quitters, and an analysis of 1 year cessation. Repeating the measure at the 2 year follow-up permitted a further breakdown of quitters, into those who had resumed smoking and those who were still ex-smokers.
At the baseline survey, data were collected on socio-economic status, domestic circumstances, psychological health and tobacco dependence. The BHPS includes more extensive and detailed questions on socio-economic and domestic circumstances than on psychological health and tobacco dependence. Nonetheless, the range enabled each dimension to be measured, using indicators recognized to be valid and reliable. Because the quitters made up a relatively small proportion of the sample, dichotomous measures were used wherever possible. Even so, cell sizes for some analyses were small, with the result that the statistical power to detect influences on cessation is reduced.
Four measures of socio-economic status were included in the study: school-leaving age (
17/ 
16), educational qualifications (yes/no), own social class based on current/last occupation (non-manual/manual/no occupation) and housing tenure (owner-occupied/rented). Four dimensions of domestic circumstances were examined: co-habitation status, partner's smoking status, the presence of children in the household and being a lone mother. Co-habitation status and partner's smoking status were measured through a composite indicator (non-smoking partner/smoking partner/no partner). The influence of having children under the age of 16 in the household (yes/no) and being a lone parent (yes/no) were examined both separately and through a composite measure (no children/co-habiting mother/ lone mother).
Psychological health was assessed by the 12-item General Health Questionnaire, designed to detect psychiatric morbidity in the general population. It consists of a suite of 12 questions about general levels of happiness, depression, anxiety and sleep disturbance over the previous 4 weeks. Responses are scored and a score of 3 or more can be taken to indicate poor psychosocial health (Goldberg and Williams, 1988).
Tobacco dependence was measured by reported average daily consumption. When asked how many cigarettes they smoke, smokers tend to round their estimates of consumption to pack sizes (10, 20), with further peaks in reported consumption at 5, 15 and 20. This patterning of reported consumption was evident in the study, and was accommodated within the standard classification of daily consumption levels employed in the analysis (1—9/10—19/20).
The study also sought to monitor the cessation effect of being pregnant on quitting behaviour. Because data on pregnancy are not collected in the BHPS, a measure of pregnancy status at the 1 year follow-up interview was therefore constructed from data on new household members collected at the subsequent wave of data collection. It should be noted that this derived measure of pregnancy status excluded expectant mothers for whom, because of termination, perinatal death or adoption, no births were recorded at the follow-up interview.
The statistical analysis proceeds in two stages. Firstly, the patterns of 1 year and 2 year quitting are described, and their association with socio-economic circumstances, domestic circumstances, psychological health, tobacco dependence and pregnancy status is examined using the Pearson Chi square test (p < 0.05). Secondly, logistic regression tested the influence of the five key factors, with the most significant factors identified through a stepwise procedure. The variables omitted through this procedure are those whose effects are no longer significant after adjustment for the other factors in the model. For each factor, a baseline odds of 1.0 is set for those in the less advantaged group, with the odds ratio providing an estimate of how much a woman's chances of quitting are increased by being in a more advantaged group.
|
RESULTS |
|---|
|
TOPSUMMARYINTRODUCTIONSOCIAL AND INDIVIDUAL INFLUENCES...THE STUDY: METHOD AND...RESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Patterns of short-term and longer-term quitting
At the point of recruitment to the study in 1991, 29% of the panel sample were smokers. Of this group, 12% were non-smokers 1 year later (short-term quitters) and 8% were non-smokers both 1 and 2 years later (longer-term quitters). The patterns of cessation are summarized in Table 1
. For completeness, the table includes information on the 821 women who were smokers at both the baseline survey and the 1 year follow-up interview.
|
Table 2
identifies the factors which were significantly related to cessation, describing in turn the association with socio-economic status, domestic circumstances, psychological health, tobacco dependence and pregnancy status. All these measures relate to the 1991 baseline survey, with the exception of pregnancy status, which relates to whether or not a woman was pregnant at the time of the 1 year follow-up interview.
|
Socio-economic disadvantage was consistently related to lower rates of cessation, both at the 1 year and 2 year follow-up interview. Of the socio-economic measures, school-leaving age was the most strongly associated with cessation. Women who stayed on at school beyond the age of 16 had a quit rate of 22% at the first year follow-up and 20% at the second year follow-up; the rates among women leaving school at 16 or under were 11 and 8%, respectively (p < 0.001 and p < 0.0001). There was also a significant association between educational qualifications and quit rates at the 1 and 2 year follow-up (p < 0.01), and between social class and 2 year quit rates (p < 0.05). The association between housing tenure and quitting did not reach statistical significance.
There were no significant differences in cessation rates among women whose domestic circumstances differed with respect to partner's smoking status and being a lone mother, although the trend was in the expected direction. However, with respect to the presence of children in the household, it was those who were living with children at the baseline survey who were less likely to be non-smokers both 1 year (not significant) and 2 years later (p < 0.05). When the influence of children was examined through the composite measure (no children/co-habiting mother/lone mother), there were no significant differences at either the first year or second year follow-up.
Poor psychological health was not significantly associated with lower rates of cessation, but the trend was in the expected direction. Tobacco dependence and pregnancy status were significantly related to short-term and longer-term quitting. Cessation rates at the 1 year follow-up were over five times higher among light smokers (31%) than among those who were smoking 20 or more cigarettes a day (6%, p < 0.0001). By the second year follow-up, the relative difference in cessation rates had widened further, to 27% and 4%, respectively (p < 0.0001). Pregnancy also had a powerful cessation effect. Of the 19 women who were pregnant at the 1 year follow-up interview, nine (47%) had given up smoking. While the absolute numbers were small, the association is highly significant (p < 0.0001). Pregnancy had a marked longer-term effect on cessation as well, with a cessation rate at the 2 year follow-up which was four times higher among the women who were pregnant at the 1 year follow-up interview (p < 0.0001).
As a check on the potentially distorting effects of pregnancy on the patterns of cessation, the analyses summarized in Table 2
were repeated, excluding women who were pregnant at the follow-up interview. With one exception, the same factors were significantly associated with short-term and longer-term cessation. However, the relationship between presence of children in the household and a lower rate of longer-term cessation was no longer significant (data not given).
Predictors of short-term and longer-term quitting
The range of factors examined in Table 2 were included in the multivariate analysis, together with age. The factors which emerge as independent predictors of giving up smoking are: socio-economic status (school-leaving age and own social class); tobacco dependence; and pregnancy status (Table 3). With respect to 1 year cessation, only tobacco dependence and pregnancy status contributed to the prediction. The most powerful predictor of giving up smoking was being a light smoker; smoking less than 10 cigarettes a day was associated with an eightfold increase in the odds of being a non-smoker 1 year later. Smokers who were pregnant at the first follow-up interview had odds of cessation six times higher than non-pregnant smokers.
|
The multivariate analysis of the predictors of longer-term cessation confirmed the importance of tobacco dependence and pregnancy status. Smoking less than 10 cigarettes a day at the baseline survey increased ninefold the odds of being an ex-smoker at the 1 year follow-up and still being an ex-smoker at the second year follow-up. The odds of a woman who was pregnant at the 1 year follow-up being a non-smoker both at that point and 1 year later was six times greater than a non-pregnant woman. Socio-economic status also contributed to the likelihood of longer-term cessation. Staying on at school doubled the odds of being a long-term quitter, while having a non-manual job increased the odds by a further 74%. There were no significant interactions between the main predictors of cessation.
It should be noted that the presence of children in the household, a factor associated with lower longer-term cessation in the bivariate analysis (Table 2) did not emerge as an independent predictor of cessation. Further analysis confirmed that this was the result of the multivariate analysis controlling for the effect of pregnancy status. The effect of pregnancy was confined to women having their first baby (and who were therefore living at the baseline survey in households with no dependent children). None of the smokers expecting a second child were long-term quitters.
The multivariate analysis was repeated, excluding the women who were pregnant at the second interview. The results were consistent with those presented in Table 3: only cigarette consumption emerged as independently predictive of short-term cessation, while both consumption and school-leaving age exerted an independent effect on the chances of giving up smoking prior to the first year follow-up interview and still being a non-smoker 1 year later (data not given).
|
DISCUSSION |
|---|
|
TOPSUMMARYINTRODUCTIONSOCIAL AND INDIVIDUAL INFLUENCES...THE STUDY: METHOD AND...RESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
A number of features of the study design need to be noted before conclusions are drawn from the results.
Firstly, the study relies on self-reported measures of tobacco use. While self-reports are known to underestimate both prevalence and consumption (Patrick et al., 1994), this will only bias the results if the degree of underestimation varies in a systematic way with the factors identified as predictors of cessation. Cotinine validation of self-reported smoking status for British women provides no evidence of a socio-economic differential in under-reporting (Prescott-Clarke and Primatesta, 1998); a cotinine validation study of self-reported quitting among pregnant women also showed a high level of agreement between the two measures (Nafstaad et al., 1996). Secondly, the relatively small number of quitters limits the confidence that can be attached to the findings. The associations and effects that are detected are likely to be robust; however, others may have been missed because of small cell sizes. As a result, influences on cessation may fail to be detected in the statistical analyses.
Thirdly, the study relies on measures which are likely to be relatively crude measures of the factors influencing quitting behaviour. Differences in the precision with which the measures capture these underlying processes may be contributing to differences in the predictive power of different factors. For example, school-leaving age emerges as a significant predictor of longer-term cessation (Table 3). This may be because staying at school is a key determinant of quitting behaviour in adult life or because the measure is capturing broader socio-economic processes with greater accuracy than alternative markers of SES. Similarly, the powerful statistical effect of cigarette consumption on the odds of cessation may reflect the influence of nicotine addiction on the frequency and outcome of quit attempts. It is also likely to be a marker of other and unmeasured factors, including self-efficacy and motivation to quit, which have been identified as influencing the process of cessation and relapse (DiClemente et al., 1991; Gulliver et al., 1995).
These features are not unique to this study. While they underline the need for caution in drawing policy implications from research results, the findings underline some key issues for the theory and practice of health promotion.
|
CONCLUSIONS |
|---|
|
TOPSUMMARYINTRODUCTIONSOCIAL AND INDIVIDUAL INFLUENCES...THE STUDY: METHOD AND...RESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Two conclusions can be drawn from this study which reinforce and refine the findings of previous research. Firstly, this study confirms that cigarette smoking among adult women is a stable dimension of identity and lifestyle, a habit which exhibits relatively little short-term change. Eighty per cent of those who were regular smokers at the beginning of the study were still smoking regularly 2 years later. Only a small minority gave up across this period: 8% of those who smoked at the point of recruitment to the study were ex-smokers both 1 and 2 years later. As this suggests, increasing cessation rates among women is likely to be a slow and uphill task. It argues for the setting of modest targets with long time-frames for smoking cessation policies, both with respect to population-wide strategies and for targeted treatment programmes.
Secondly, while infrequent, giving up smoking between one year and the next is not a random event. Smoking cessation among women is strongly patterned by individual and social circumstances. Of the range of influences examined in the study, smoking less than 10 cigarettes a day was the strongest predictor of both short-term and longer-term cessation. The important policy message of this finding is not only that ex-smokers are disproportionately drawn from those who are less entrenched in, and dependent on, their smoking habit. It is also that this group is atypical of the broader population of female smokers. Less than one in five of the smokers in the study smoked less than 10 cigarettes a day. The implication for national anti-tobacco policies is clear: because of differential cessation rates between lighter and heavier smokers, the remaining smokers are increasingly those who are highly nicotine dependent. They are also those who are less responsive both to broad public health programmes and more focused interventions (COMMIT, 1995; Susser, 1995; Hughes, 1996). Increasing cessation rates among heavier smokers represent a major health promotion challenge and one that points to the need for primary prevention. It requires investment in programmes directed at young people most at risk of becoming and remaining heavy smokers in adulthood (Whitehead, 1989; Raw et al., 1990). It also requires interventions which are targeted at heavy smokers for whom addiction, need and reinforcement play a key role in maintaining their long-term dependence on cigarettes (Shiffman, 1991; Bott et al., 1997).
Becoming pregnant was the second most important predictor of giving up smoking, both in the shorter and longer term. While only a small minority of female smokers are pregnant in any one year, most will become pregnant and become mothers at some point in their reproductive lives. Pregnancy is already recognized as providing unrivalled opportunities for smoking cessation, and this study underlines its importance in women's smoking careers. In so doing, the findings confirm the need for perspectives and policies on cessation which take account of the gender-specific factors shaping smoking behaviour (Ockene, 1993). Other research makes it clear that it is those in higher socio-economic groups who are most likely to quit, while those in poorer circumstances are more likely to remain smokers as they pass through this key life event. Developing strategies which equalize the chances of quitting and remaining a non-smoker could make a major contribution, both to the reduction of smoking prevalence and to the health of women and children.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPSUMMARYINTRODUCTIONSOCIAL AND INDIVIDUAL INFLUENCES...THE STUDY: METHOD AND...RESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
This study was supported by an ESRC Senior Research Fellowship (H52427504395) and was undertaken at the MRC Medical Sociology Unit, Glasgow. The data used in the paper were made available through the ESRC Data Archive. The data were originally collected by the ESRC Research Centre on Micro-social Change at the University of Essex. The ESRC, the original collectors of the data and the Archive bear no responsibility for the analyses or interpretations presented here.
|
REFERENCES |
|---|
|
TOPSUMMARYINTRODUCTIONSOCIAL AND INDIVIDUAL INFLUENCES...THE STUDY: METHOD AND...RESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Ajzen, I. (1990) The theory of planned behaviour. Organisational Behaviour and Human Decision Processes, 50, 179—181.
Bennett, N., Jarvis, L., Rowlands, O., Singleton, N. and Haselden, L. (1996) Living in Britain: Results from the 1994 General Household Survey. HMSO, London.
Bjornson, W., Rand, C., Connett, J. E., Lindgren, P., Nides, M., Pope, F., Buist, A. S., Hoppe-Ryan, C. and O'Hara, P. (1995) Gender differences in smoking cessation after 3 years in the Lung Health Study. American Journal of Public Health, 85, 223—230.
Bott, M. J., Cobb, A. K., Scheibmeir, M. S. and O'Connell, K. A. (1997) Quitting: smokers relate their experiences. Qualitative Health Research, 7, 255—269.
Bridgwood, A., Malbon, G., Lader, D. and Matheson, J. (1996) Health in England 1995. HMSO, London.
Buck, N., Gershuny, J., Rose, D. and Scott, J. (1994) Changing Households: the British Household Panel Survey. ESRC Research Centre on Micro-Social Change, University of Essex, London.
COMMIT Research Group (1995) Community Intervention Trial for Smoking Cessation (COMMIT): 1. Cohort results from a four-year community intervention. American Journal of Public Health, 85, 183—192.
DiClemente, C. C., Prochaska, J. O., Fairhurst, S. K., Velicer, W. F., Velasquez, M. M. and Rossi, J. S. (1991) The process of smoking cessation: an analysis of pre-contemplation, contemplation and preparation stages of change. Journal of Consulting and Clinical Psychology, 59, 295—304.[ISI][Medline]
Fiore, M. C., Novotny, T. E., Pierce, J. P., Giovino, G. A., Hatziandren, E. J., Newcomb, P. A., Sarawicz, T. S. and Davis, R. M. (1990) Methods used to quit smoking in the United States. Do cessation programs help? JAMA, 263, 2760—2765.[Abstract]
Freed Taylor, M. (1995) British Household Panel Survey User Manual, Volume A. ESRC Research Centre on Micro-Social Change, University of Essex, London.
Goldberg, D. and Williams, P. (1988) A User's Guide to the General Health Questionnaire. NFER-Nelson, London.
Graham, H. (1993) When Life's a Drag: Women, Smoking and Disadvantage. HMSO, London.
Gulliver, S. B., Hughes, J. R., Solomon, L. J. and Deg, A. N. (1995) An investigation of self-efficacy, partner-support and daily stresses as predictors of relapse to smoking in self-quitters. Addiction, 90, 767—772.[ISI][Medline]
Hatziandreu, E. J., Pierce, J. P., Lefkopoulou, M., Fiore, M. C., Mills, S. L., Novotny, T. E., Giovino, G. A. and Davis, R. M. (1990) Quitting smoking in the United States in 1986. Journal of the National Cancer Institute, 82, 402—406.
Heatherton, T. F., Kozlowski, L. T., Frecker, R. C., Rickert, W. and Robinson, J. (1989) Measuring the heaviness of smoking: using self-reported time to the first cigarette of the day and number of cigarettes smoked per day. British Journal of Addiction, 84, 791—800.[ISI][Medline]
Hughes, J. R. (1996) The future of smoking cessation therapy in the United States. Addiction, 91, 1797—1802.[ISI][Medline]
Jarvis, M. J. (1994) Gender differences in smoking cessation: real or myth? Tobacco Control, 3, 324—328.
Jarvis, M. J. (1996) The association between having children, family size and smoking cessation in adults. Addiction, 91, 427—434.[ISI][Medline]
Jarvis, M. J. (1997) Patterns and predictors of smoking cessation in the general population. In Bolliger, C. T. and Fagerstrom, K. O. (eds) The Tobacco Epidemic. Karger, Basel.
Jones, A. M. (1995) A Micro-econometric Analysis of Smoking Cessation in the General Population. University of York, York.
Mann, S. L., Wadsworth, M. E. J. and Colley, J. R. T. (1992) Accumulation of factors influencing respiratory illness in members of a national birth cohort and their offspring. Journal of Epidemiology and Community Health, 46, 286—292[Abstract]
McBride, C. M., Pirie, P. L. and Curry, S. J. (1992) Postpartum relapse to smoking: a prospective study. Health Education Research, 7, 381—390.
Nafstaad, P., Botten, G. and Hagen, J. (1996) Partner's smoking: a major determinant for changes in women's smoking behaviour during and after pregnancy. Public Health, 110, 379—385.[ISI][Medline]
Ockene, J. K. (1993) Smoking among women across the lifespan: prevalence, interventions and implications for clinical research. Annals of Behavioural Medicine, 15, 135—148.
Olsen, J. (1993) Predictors of smoking cessation in pregnancy. Scandinavian Journal of Social Medicine, 21, 197—202.[ISI][Medline]
Patrick, D. L., Cheadle, A., Thompson, D. C., Diehr, P., Koepsell, T. and Kinne, S. (1994) The validity of self-reported smoking: a review and meta-analysis. American Journal of Public Health, 84, 1086—1093.
Prescott-Clarke, P. and Primatesta, P. (1998) Health Survey for England. HMSO, London.
Raw, M., White, P. and McNeill, A. (1990) Clearing the Air: A Guide for Action on Tobacco. World Health Organisation Regional Office for Europe and British Medical Association, London.
Severson, H. H., Andrews, J. A., Lichtenstein, E., Wall, M. and Zoref, L. (1995) Predictors of smoking during and after pregnancy: a survey of mothers of new-borns. Preventive Medicine, 24, 23—28.[ISI][Medline]
Shiffman, S. (1991) Refining models of dependence: variations across persons and situations. British Journal of Addiction, 86, 611—615.[ISI][Medline]
Stott, N. C. H. and Pill, R. M. (1990) Making Changes. University of Wales College of Medicine, Cardiff.
Susser, M. (1995) The tribulations of trials—intervention in communities. American Journal of Public Health, 85, 156—158.
Wakefield, M., Gillies, P., Graham, H., Madeley, R. and Symonds, M. (1993) Characteristics associated with smoking cessation during pregnancy among working class women. Addiction, 88, 1423—1430.[ISI][Medline]
Waterson, E. J. and Murray-Lyon, I. M. (1989) Drinking and smoking patterns amongst women attending an antenatal clinic—II during pregnancy. Alcohol and Alcoholism, 24, 163—173.
White, A., Freeth, S. and O'Brien, M. (1992) Infant Feeding, 1990. HMSO, London.
Whitehead, M. (1989) Swimming Upstream: Trends and Prospects in Education for Health. Research Report No 5, King's Fund, London.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Graham, H. M Inskip, B. Francis, and J. Harman Pathways of disadvantage and smoking careers: evidence and policy implications J. Epidemiol. Community Health, September 1, 2006; 60(suppl_2): ii7 - ii12. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Graham, B. Francis, H. M Inskip, J. Harman, and SWS Study Group Socioeconomic lifecourse influences on women's smoking status in early adulthood. J. Epidemiol. Community Health, March 1, 2006; 60(3): 228 - 233. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M Siahpush, R Borland, and M Scollo Smoking and financial stress Tob. Control, March 1, 2003; 12(1): 60 - 66. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Sykes and D. F. Marks Effectiveness of a cognitive behaviour therapy self-help programme for smokers in London, UK Health Promot. Int., September 1, 2001; 16(3): 255 - 260. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Wiltshire, A. Bancroft, A. Amos, and O. Parry ""They're doing people a service""{---}qualitative study of smoking, smuggling, and social deprivation BMJ, July 28, 2001; 323(7306): 203 - 207. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||