Health Education Research Advance Access originally published online on November 17, 2005
Health Education Research 2006 21(3):386-392; doi:10.1093/her/cyh069
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Post-partum intention to resume smoking
1 Institute for Medical Psychology, Ernst-Moritz-Arndt-University Greifswald, Walther-Rathenau-Strasse 48, 17487 Greifswald, Germany
2 Institute of Community Medicine, Ernst-Moritz-Arndt-University Greifswald, Walther-Rathenau-Strasse 48, 17487 Greifswald, Germany
3 Institute of Epidemiology and Social Medicine, Ernst-Moritz-Arndt-University Greifswald, Walther-Rathenau-Strasse 48, 17487 Greifswald, Germany
4 Hospital for Psychiatry und Psychotherapy, University of Lübeck, Ratzeburger Allee 160, 23538 Lübeck, Germany
*Correspondence to: K. Röske. E-mail: roeske{at}uni-greifswald.de
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
Little is known about the motivational background of smoking after a period of nicotine abstinence during pregnancy. The study examines the intention to resume smoking (IRS) in the post-partum period and its predictive value for smoking within 12 months post-partum. In a sample of 301 women recruited from obstetric wards who reported having stopped smoking during pregnancy, data on IRS, sociodemographic variables, recent smoking behaviour and smoking in the social network were collected. Smoking status was assessed 6 and 12 months after pregnancy. Among all formerly smoking women, 39 (13%) intended to resume smoking and 262 (87%) intended to maintain abstinence. Women with IRS returned to smoking more often than women without IRS [77 versus 45%, odds ratio (OR) = 4.1, 95% confidence interval (CI): 1.89–9.05]. In a logistic regression model, IRS (OR = 3.7, 95% CI: 1.51–9.01) and number of months currently abstinent (OR = 0.9, 95% CI: 0.76–0.96) attained statistical significance. IRS proved to be the main predictor for relapse; yet, women with no IRS are at risk to restart smoking again, too. IRS offers a cue for tailoring interventions.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
Smoking cessation during pregnancy was conceptualized as a ‘suspended behaviour’ [1, 2] to explain the high post-partum relapse rates. About half of the women who quit during pregnancy resume smoking within 6 months after delivery and up to 80% within 12 months [3–6]. Quitting smoking in pregnancy mainly results from the care for the baby's sake and social pressure not to smoke when pregnant [1, 7]. Smoking cessation is also linked to pregnancy nausea and loss of taste for tobacco [8, 9]. These external motivational factors result rather in a suspension of smoking behaviour than in permanent cessation. This strongly suggests that many women who stop smoking in pregnancy do not intend to be lifelong abstinent as was already reported in very small qualitative studies [10, 11].
On the other hand, most research for factors triggering post-partum smoking depletes in the search for associated sociodemographic or social variables and disregards motivational aspects. A recent review subsumes variables that have been found to be associated with post-partum smoking as (i) pregnancy related (stress of caring for a newborn, weaning from breast-feeding), (ii) social reasons (smoking partner, social pressures, familiar cues such as drinking alcohol or coffee, exposure to high-risk situations) and (iii) general health of self and others (need time for self, under pressure from children, trying to relax, weight gain) [12]. Women who relapse in the post-partum period tend to be younger, were heavier smokers before pregnancy and quit smoking late during pregnancy [3]. This literature conceptualizes the restart of smoking after pregnancy as unintended context-bound behaviour.
The conception and empirical examination of an intention to resume smoking (IRS) after pregnancy adds a significant predictor to the knowledge about post-partum smoking. Furthermore, it is in line with the practical experiences of the authors and probably that of many other professionals dealing with pregnant and post-partum smokers.
There has been only one approach including IRS in predicting post-partum smoking. Stotts et al. [2] assessed IRS (as part of a staging algorithm for post-partum smoking abstinence according to the Transtheoretical Model of behaviour change [13]) in the 28th week of pregnancy. In a sample of 256 pregnant women who had quit, 9% reported ‘to control where and when smoking’ or ‘to go back to smoking’ would best describe their personal smoking goal with regard to smoking after delivery and 74% of these had returned to smoking 12 months after pregnancy. However, in this study, IRS was measured during the time of pregnancy, in which smoking is highly socially restricted. The tendency for underreporting IRS might be exceeding that after the birth of the baby. Furthermore, for the investigation of IRS after having given birth, a social climate is needed which makes it likely that women disclose their IRS. Germany is suited for such an investigation for the following reasons: (i) until the last 2 years no mentionable activities in the prevention of tobacco attributable diseases have been undertaken [14], (ii) the anti-smoking climate is lowest among European countries (together with Austria) [15] and (iii) the proportion of smokers who intend to quit is generally lower than that in countries with more prevention activities [16].
The objective of this analysis was to examine post-partum return to smoking as an intentional process. First, we hypothesized that there is either an intention to stay abstinent or an IRS in women post-partum and that this intention prospectively corresponds to smoking. Second, in comparison to women without IRS, we expected women with IRS to be of lower socioeconomic status, to report a higher degree of nicotine dependence before pregnancy, to report a shorter period of abstinence and to more often have a household member who smokes.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
Study design
Data of a subsample of the prospective randomized controlled trial ‘Smoking cessation and relapse prevention in women post-partum’ were analysed. The sample under examination consisted of 301 women, who were identified as follows.
Between May 2002 and March 2003, 3343 women gave birth in the maternity wards of six hospitals in Mecklenburg West-Pomerania. The study region is a rural area in the northeast of Germany. A sample of 2790 women were screened about their smoking behaviour before pregnancy. Of these, 1140 women indicated that they had smoked before pregnancy and 871 gave informed consent to participate in the randomized controlled trial. These women were randomized to either an intervention or a control group. The intervention consisted of a home counselling session and two telephone counselling sessions 4 and 12 weeks later according to the principles of motivational interviewing [17, 18]. Both groups received (i) a self-help manual addressing maternal smoking, smoking cessation and relapse prevention and (ii) a manual addressing the partner of the participating women, in order to initiate a tobacco smoke-free environment for mothers and newborns. Data of the baseline assessment (T0) 
4 weeks after giving birth are available for 642 women. Inclusion criteria for the following analysis were (i) having been abstinent 4 weeks prior to assessment and (ii) having provided information about IRS. A sample of 301 women fulfilled these criteria and was considered for this analysis.
Data about the smoking status from the first follow-up 6 months after delivery (T1) were available for 285 (94.7%) women. For the second follow-up 12 months after giving birth (T2), data were available for 276 (91.7%) women. The study procedure has been described in detail elsewhere [19] and followed the ethical guidelines for research of the University of Greifswald.
Assessments
The IRS was defined as agreement either to the item ‘Do you ever intend to start smoking again?’ or to the item ‘Do you intend to start smoking again after weaning?’ Items that measured the sociodemographic characteristics included age, school education (<10 years, 10 years, >10 years), currently living in a steady partnership (yes/no), monthly household income (Euro), number of pregnancies and having further children (yes/no). Items measuring variables according to the smoking behaviour of the women included the age of onset of regular smoking of the woman, the number of months currently abstinent and the number of cigarettes smoked per day before pregnancy. The latter was assessed as part of the German version of the Fagerström Test for Nicotine Dependence (FTND) [20, 21] as an indicator for the level of nicotine dependence prior to pregnancy. We asked whether further smokers lived in the household (yes/no), whether the partner smoked (yes/no) and what the percentage of daily time of environmental tobacco smoke (ETS) exposure was, which then was dichotomized into exposure (>0%) versus no exposure (0%).
Women who agreed to the item ‘Are you a current smoker?’ either at 6 (T1) or 12 months (T2) after pregnancy were considered to have resumed smoking within 12 months post-partum. Women not reached for the survey were considered as current smokers for the respective survey, too.
Statistical analyses
The univariate association between IRS and relapse to smoking was tested using 
2-test. For assessing the predictive value of IRS for relapse in the context of other variables logistic regression (forced entry modelling) was applied. Because of single missing data, 258 (86%) subjects were included in the analysis. The proportion of missing data on single variables was <5% except for household income (8%). In order to control for effects of the intervention, experimental group status was included. The variables ‘further smokers in the household’ and ‘exposure to ETS’ were excluded in favour of ‘smoking partner’ to avoid multicollinearity due to the high overlap of these variables. For the same reason the number of cigarettes per day was not analysed separately since it is part of the FTND score. The FTND score delivers a better description for pre-pregnancy smoking since it additionally asks for further smoking habits than the mere amount of cigarettes smoked. Odds ratios (ORs) and their 95% confidence intervals (CIs) are reported. To identify differences of women with IRS and women without IRS univariate between-group comparisons were tested using 2-test and t-test. Effect sizes were estimated and interpreted according to Cohen [22]. Data were analysed using SPSS 13.0 for Windows.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
IRS and post-partum smoking rates
A total of 39 women (13%) reported their IRS. The proportion of women who reported having resumed smoking within 12 months after giving birth was 49% (Table I). More women with IRS returned to smoking (77%) than women without IRS (45%). The OR for smoking was 4.1 (95% CI: 1.89–9.05).
|
Multivariate analysis of IRS
The results of the logistic regression analyses show that, except for the IRS and the number of months currently abstinent, no beta weights attained statistical significance (Table II). The OR for IRS was 3.7 (95% CI: 1.51–9.01). The OR for the number of months currently abstinent was 0.9 (95% CI: 0.76–0.96).
|
Variables related to IRS
Women with IRS did not differ significantly (P > 0.05) from women without IRS on the variables under consideration (Table III). According to the effect sizes, differences might exist on variables of the personal smoking history before pregnancy. Women with IRS were slightly older when they started smoking regularly, resulting in less years of smoking. They smoked fewer cigarettes per day and stopped smoking later in pregnancy. The percentage of smokers with a low level of nicotine dependence was higher in this group.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
The main finding of our analysis is that IRS after delivery is the factor most strongly associated with return to smoking within 12 months post-partum. More than one in 10 women who succeeded in quitting during her pregnancy intends to start smoking again after delivery. The odds for smoking for these women were about four times higher than for women who reported an intention to stay abstinent. Our results correspond to those of Stotts et al. [2], although we used a different item formulation and assessed IRS not in pregnancy but in the post-partum period. Apart from replicating the findings of another independently working group, the results are of practical relevance.
The smoking rates of either women post-partum with IRS or those who do not intend to restart smoking indicate that both are at risk. This implies the necessity for proactive relapse prevention interventions for all women who successfully stopped smoking during pregnancy. Women with IRS will not seek for help to stay smoke free. On the other hand, women who want to stay smoke free and were abstinent for several months are likely to smoke within the first year after pregnancy, too. Overall, half of the women restarted smoking within 12 months after pregnancy. This rate is relatively low but a comparison with the international literature is problematic because of different prevalences of women in childbearing age and different national tobacco control activities. There are no comparable data available for Germany so far.
The IRS is a useful measure to tailor interventions. First, the single question for IRS helps professionals to identify the subjects at high chance for a return to smoking. Especially in time-limited health care settings, IRS might be a marker for an at-risk group that suggests a priority for interventions respectively for a more intensive intervention, i.e. a higher number of counselling sessions. Second, abstinent women post-partum stopped smoking to prevent their babies from the risks of smoking during pregnancy. Therefore, we assume that women with IRS after giving birth may not be aware of the adverse health effects of exposure to ETS on the newborn child [12]. Interventions should include information about ETS as one aspect for women who succeeded in smoking cessation in the course of pregnancy but intend to start smoking again in the post-partum period. This information must be offered in a sensitive non-confronting way to avoid psychological resistance. Third, most women reported an intention to remain smoke free but almost half of them restarted smoking. For these women, relapse prevention interventions are necessary in which psychological reinforcement of their intention to maintain abstinence is combined with increasing their awareness of a high chance for a return to smoking and with assistance in developing strategies to manage high-risk situations like stressful incidents or the contact to a smoking social network.
There is one question of interest for future research in the field of IRS in pregnant women and women post-partum. Our data revealed no significant differences between women with IRS compared with women who want to stay smoke free. This lack of significance might be due to different and small sample sizes. On the basis of the effect sizes, we assume that women intending to resume smoking do not seem to be a special subgroup according to the assessed sociodemographic, pregnancy-related or smoking social network characteristics. But IRS might be related to the smoking behaviour before pregnancy and the time point of smoking cessation during pregnancy. It seems surprising that women with IRS started smoking later in their life and smoked fewer cigarettes before pregnancy compared with women without IRS. These women stopped smoking later in pregnancy, too. Since this is the first study to examine IRS post-partum, we cannot compare this finding with current research. However, we assume that women with IRS could be light smokers, who perceive smaller health risks due to their light smoking than women with no IRS. Women with no IRS might be heavier smokers, happy to have finally quit. However, these assumptions need further examination, the observed effect sizes were small and therefore of small practical relevance.
Limitations
Some limitations of the analysis must be considered. First, reports of the IRS might underlie effects of social desirability resulting in an underestimation of the proportion of women with IRS. Second, the definition of post-partum return to smoking on grounds of the point prevalence at 6 or 12 months after delivery is very liberal. It might underestimate the proportion of smokers due to disregarding temporary smoking in the interim. Third, only self-reports about smoking behaviour were provided without biochemical validation.
|
Conclusion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
During pregnancy and after delivery, IRS ought to be the focus of several brief intervention contacts. Smoking relapse prevention counselling should be part of the care as soon as abstinence is achieved and must proceed beyond delivery. By addressing all women who quit smoking during pregnancy the highest impact on relapse rates is expected, resulting in a reduction of health risks for the mother and the newborn child. Consequently, all professionals in the field of health care for pregnant women, women post-partum and their newborn children (e.g. midwives, gynaecologists and paediatricians) should make use of the possibility to easily address the IRS and to offer adequate interventions.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
The study, as part of the Research Collaboration in Early Substance Use Intervention (EARLINT), has been funded by the German Federal Ministry of Education and Research (grant no. 01EB0120), the Social Ministry of the State of Mecklenburg-West Pomerania (grant no. IX311a 406.68.43.05 [EC] ) and the Krupp von Bohlen and Halbach-Foundation.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
1. Stotts AL, DiClemente CC, Carbonari JP, Mullen PD. Pregnancy smoking cessation: a case of mistaken identity. Addict Behav 1996 21:459–71.[CrossRef][Web of Science][Medline]
2. Stotts AL, DiClemente CC, Carbonari JP, Mullen PD. Postpartum return to smoking: staging a "suspended" behavior. Health Psychol 2000 19:324–32.[CrossRef][Web of Science][Medline]
3. Fingerhut LA, Kleinman JC, Kendrick JS. Smoking before, during, and after pregnancy. Am J Public Health 1990 80:541–4.
4. Mullen PD, Richardson MA, Quinn VP, Ershoff DH. Postpartum return to smoking: who is at risk and when. Am J Health Promot 1997 11:323–30.[Web of Science][Medline]
5. Lelong N, Kaminski M, Saurel-Cubizolles MJ, Bouvier-Colle MH. Postpartum return to smoking among usual smokers who quit during pregnancy. Eur J Public Health 2001 11:334–9.
6. Colman G-J and Joyce T. Trends in smoking before, during, and after pregnancy in ten states. Am J Prev Med 2003 24:29–35.[CrossRef][Web of Science][Medline]
7. Edwards N and Sims-Jones N. Smoking and smoking relapse during pregnancy and postpartum: results of a qualitative study. Birth 1998 25:94–100.[CrossRef][Web of Science][Medline]
8. McLeod D, Pullon S, Cookson T. Factors that influence changes in smoking behaviour during pregnancy. N Z Med J 2003 116:U418.[Medline]
9. Pletsch PK and Kratz AT. Why do women stop smoking during pregnancy? Cigarettes taste and smell bad. Health Care Women Int 2004 25:671–9.[CrossRef][Medline]
10. Abrahamsson A, Springett J, Karlsson L, Ottosson T. Making sense of the challenge of smoking cessation during pregnancy: a phenomenographic approach. Health Educ Res 2005 20:367–78.
11. Bottorff JL, Johnson JL, Irwin LG, Ratner PA. Narratives of smoking relapse: the stories of postpartum women. Res Nurs Health 2000 23:126–34.[CrossRef][Web of Science][Medline]
12. Fang WL, Goldstein AO, Butzen AY, et al. Smoking cessation in pregnancy: a review of postpartum relapse prevention strategies. J Am Board Fam Pract 2004 17:264–75.
13. DiClemente CC, Prochaska JO, Fairhurst SK, Velicer WF, Velasquez MM, Rossi JS. The process of smoking cessation: an analysis of precontemplation, contemplation, and preparation stages of change. J Consult Clin Psychol 1991 59:295–304.[CrossRef][Web of Science][Medline]
14. Poetschke-Langer M and Schunk S. Germany: tobacco industry paradise. Tob Control 2001 10:300–3.
15. Boyle P, Gandini S, Robertson C, et al. Characteristics of smokers' attitudes towards stopping. Survey of 10,295 smokers in representative samples from 17 European countries. Eur J Public Health 2000 10:S5–14.
16. John U, Meyer C, Rumpf H-J, Hapke U. Relation among stage of change, demographic characteristics, smoking history, and nicotin dependence in an adult German population. Prev Med 2003 37:368–74.[CrossRef][Web of Science][Medline]
17. Miller WR and Rollnick S. Motivational Interviewing. Preparing People to Change Addictive Behavior. 2nd edn New York: Guilford Press 2002.
18. Thyrian JR, Hannöver W, Grempler J, Röske K, John U, Hapke U. An intervention to support postpartum women to quit smoking or remain smoke-free. J Midwifery Womens Health. in press.
19. Thyrian JR, Hannöver W, Kelbsch J, Röske K, John U, Hapke U. Velicer WF and Keller S. Smoking cessation and relapse prevention for women postpartum: design of a study to evaluate theoretical implications of the TTM. Research on the Transtheoretical Model—Where Are We Now, Where Are We Going? Lengerich: Pabst Science Publishers 2004.
20. Heatherton TF, Kozlowski LT, Frecker RC, Rickert W, Robinson J. Measuring the heaviness of smoking: using self-reported time to the first cigarette of the day and number of cigarettes smoked per day. Br J Addict 1989 84:791–9.[CrossRef][Web of Science][Medline]
21. Schumann A, Rumpf H-J, Meyer C et al. Fagerström-test for nicotine dependence (FTND) (FTND-d) and the heaviness of smoking index (HSI) (HSI-d). In: A Glöckner-Rist FR, Küfner H (eds). Electronic Handbook of Data Collection Instruments in Addiction, version 3.00 (German version). Mannheim: Zentrum für Umfragen, Methoden und Analysen. Available at: http://www.gesis.org/Methodenberatung/ZIS/index.htm, 2003.
22. Cohen J. Statistical Power Analysis for the Behavioral Sciences. Hillsdale, NJ: Lawrence Erlbaum 1988.
Received on July 29, 2005; accepted on October 5, 2005
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  K. Roske, A. Schumann, W. Hannover, J. Grempler, J.R. Thyrian, H.-J. Rumpf, U. John, and U. Hapke Postpartum Smoking Cessation and Relapse Prevention Intervention: A Structural Equation Modeling Application to Behavioral and Non-behavioral Outcomes of a Randomized Controlled Trial J Health Psychol, May 1, 2008; 13(4): 556 - 568. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||