Smoking in pregnancy: comparisons between data from the Infant Feeding Survey 2000 and 2005
Evidence Based Midwifery: December
2010
Carmel O’Gorman BSc, RM.
Smoking cessation in pregnancy lead, Heart of England Foundation Trust, Birmingham B9 5SS England. Email: carmel.ogorman@heartofengland.nhs.uk
The author would like to acknowledge the support of Dr Janine Stockdale (RCM research fellow and lecturer, Trinity College Dublin), her mentor for this paper.
Abstract
Background. Although the UK Infant Feeding Survey was first introduced in 1975, data in relation to smoking behaviour was first collected in 1985. It was hoped that the collection of five yearly trend data could facilitate future health promotion activity and effective intervention development.
Aim. The aim of this paper is to review the smoking related findings of the 2005 Infant Feeding Survey with that of 2000 as a means of identifying both evidence of health improvement and health challenges.
Method. The Infant Feeding Surveys completed in 2000 and 2005 were sourced. Each survey report was read and the key findings were extracted and compared.
Findings. Across the UK, the overall incidence of smoking throughout pregnancy appeared to fall between 2000 and 2005; where 20% of pregnant women reported smoking in 2000 compared with 17% in 2005. Similarly, 10% of women in 2005 compared to 11% in 2000 reported spontaneous cessation early in pregnancy. Most women who continued to smoke reduced their consumption 14% in 2000 and 11% in 2005. Among younger mothers smoking increased from 40% in 2000 to 45% in 2005. Smoking among disadvantaged mothers remains much higher than in professional/managerial groups (29% versus 7% respectively). Similar patterns were observed in the data for women who lived with a partner who smoked (39% in 2000 falling to 36% in 2005).
Conclusion. While health professionals work towards the reduction of smoking in pregnancy and early motherhood by 2020 (Department of Health, 2010), comparing the findings from the 2005 survey with that of 2000, confirms the need to provide health promotion activity that is specifically targeted at young and disadvantaged women and those who live with a partner who smokes.
Key words: Infant Feeding Surveys, smoking, pregnancy, partners, smoke free, cessation, evidenced-based midwifery
Introduction
Every five years, the Infant Feeding Survey is commissioned by the UK government. While the main emphasis of the survey is to report on infant-feeding behaviour, important information is provided in relation to other critical factors such as whether or not mothers took dietary supplements, drank alcohol or smoked during pregnancy. While issues such as alcohol consumption during pregnancy are important public health issues, this paper compares the findings of the 2005 Infant Feeding Survey (Bolling et al, 2007; IFS) with that of the 2000 survey (Hamlyn et al, 2000) in relation to maternal smoking behaviour. Particular focus is placed on the initiation and discontinuation of smoking behaviour during pregnancy and early motherhood.
Background
Although the UK IFS was first introduced in 1975, data in relation to smoking behaviour was first collected in 1985. It was hoped that the collection of five yearly trend data could facilitate future health promotion activity and effective intervention development. An important response to the national IFS data related to maternal smoking both during pregnancy and early motherhood was the publication of the Smoking kills legislation in 1998 (Department of Health, 1998). Acting as a landmark public health strategy, this government White Paper emphasised reduction in smoking behaviour among pregnant women as a priority target goal for health professionals. To date, the evidence provided by the 2000 (Hamlyn et al, 2002) and 2005 (Bolling et al, 2007) IFSs suggests that while significant progress has been made, further research into the phenomenon of maternal smoking is required. Therefore, on the brink of the publication of the 2010 IFS, this paper reminds readers of the survey method applied by the Office of National Statistics (Bolling et al, 2007; Hamlyn et al, 2002) and compares the evidence of 2005 with that of 2000, as a brief reminder of the national progress already made and the challenges ahead.
Table 1. Three main categories of smoking behaviour (IFS 2000 and 2005)
1. Smoking before or during pregnancy
This is the proportion of mothers who smoked at all in the two years before they completed stage one of the survey. This roughly covers the period of their pregnancy plus the year before conception.
2. Smoking throughout pregnancy
This is the proportion of all mothers who smoked in the two years before they completed stage one of the survey, and who were smoking at the time of their baby’s birth. This includes mothers who may have given up smoking before or during their pregnancy, but who had restarted before the birth.
3. Gave up smoking before or during pregnancy
This is the proportion of mothers who smoked in the two years before they completed stage one of the survey and who gave up during this period and had not restarted before the birth of the baby.
Survey method
As already highlighted, the IFS is a national survey that is carried out every five years across the UK. Although the title emphasises the role of infant feeding, the data collected concerning maternal smoking behaviours is of equal importance to the health of the nation. For example, in 2000, a sample of 13,112 births was extracted from all the live births that were registered between August and October that explored smoking behaviour in addition to infant-feeding behaviour. To achieve a more comprehensive picture of smoking behaviour across the four countries of the UK, an unclustered sample of births was selected from all live births registered in specific months during 2005. Likewise in 2005, 19,848 births were sampled from all registered live births during August and October. Data in both surveys were collected from mothers via postal questionnaires using a three-stage process:
• Data collection phase one: four to ten weeks postnatal in 2000 and 2005
• Data collection phase two: four to five months in 2000 and four to six months in 2005 postnatal
• Data collection phase three: eight to nine months in 2000 and eight to ten months in 2005 postnatal.
In each survey, the findings are reported in relation to past and current maternal smoking behaviour as outlined in Table 1.
Similar response rates were satisfactory across all three phases in 2000 and 2005; for example, in 2000, the response rates were 72% (phase one), 87% (phase two) and 88% (phase three). In 2005 the response rate although initially lower than in 2000 at 62% (phase one), increased to 88% in phase two and 87% in phase three. As a result, in 2000, the overall response rate for all three questionnaires was recorded as 55%, while in 2005 this had decreased to 47.4%; indicating an overall reduced cumulative response as illustrated in Table 2.
Table 2. Cumulative response rates (IFS 2000 and 2005)
Cumulative response rates 2000 2005
Response rate to first questionnaire 72% 62%
Response rate to first and second questionnaire 63% 54%
Response rate to first, second and third questionnaire 55% 47%
The surveys have demonstrated that factors such as age and educational attainment are also determinants of breastfeeding behaviour; it therefore must be noted that women who are most likely to smoke are also those least likely to breastfeed. Collecting data in an infant-feeding survey may therefore have had an effect on women’s willingness to participate in relation to their smoking behaviour and vice versa. Analysis of the 2005 survey showed a lower response rate among young mothers and areas of deprivation; therefore the final analysis was adjusted to control for such variation.
Table 3. Summary of the categories of women's smoking behaviour in 2000 and 2005 (UK)
Although
the overall structure and methods applied in the 2000 and 2005 surveys
accommodates important comparison, certain differences in the survey
methods exist. For example, in 2000 as a nationwide survey, data was not
available for the four countries within the UK, while in 2005 the
findings are reported across and within the four countries, enabling a
more regional and targeted response. Prior to discussing the main
differences and similarities presented in the 2000 and 2005 surveys,
details concerning the method of comparing the data presented in the
2000 survey and 2005 survey is provided.
Method
To enable comparison, the complete
survey report of the 2005 IFS (Bolling et al, 2007) and the 2000 IFS
(Hamlyn et al, 2002) were sourced. The relevant chapter in each survey
was read and re-read, marking all the important data related to smoking
behaviour. A list of results reported in the 2000 survey was entered
into a table, for example the percentage of women who reported living
with a smoker; the process was then repeated in a second table for the
2005 survey. Differences in the nature of the data collected were also
entered into the relevant table, such as maternal smoking within the
home was entered into the 2005 survey table, but omitted from the 2000
survey table as this data were not collected. The tables were then
amalgamated and increases and decreases in the overall percentages of
maternal smoking behaviours noted. An additional column was then added
that enabled increases and decreases in the percentages to be easily
identified. The results across the studies were then categorised to
enable further examination. Categories included: general maternal
smoking behaviour, sub-group results such as incidence by country,
influence of a woman’s partner, impact of health promotion advice during
pregnancy and behaviour following birth.
Results
Demographics
Some differences in the
socio-demographics of the sample were noted. Women in 2005 were more
likely to be having their first baby (51% and 47% respectively), be
slightly older (there was an increase in mothers aged 35 or over from
16% in 2000 to 19% in 2005) and more educated (in 2000, 28% of mothers
were educated to higher education age (19 or over) compared to 38% in
2005). It was noted that across the UK in 2005 smoking throughout
pregnancy was over four times more likely among routine and manual
groups compared to professional and managerial groups (29% and 7%
respectively). Compared with 2000, smoking throughout pregnancy in
routine and manual groups was unchanged but there was a reduction among
professional and managerial groups (29% and 8% respectively). But in
England, there was an increase in smoking throughout pregnancy in
routine and manual groups from 28% in 2000 to 29% in 2005 as compared to
professional and managerial groups where the rate was unchanged (7% in
2000 and 2005). In 2005, across the UK, smoking throughout pregnancy was
five times more likely than in older mothers (45% and 9% respectively)
compared to 2000 where there was an increase in teenagers and a decrease
in older mothers smoking (40% and 13% respectively).
Smoking behaviours
Across the UK, the
percentage of women who smoked either in the year before or during their
pregnancy fell from 35% in 2000 to 33% in 2005. As illustrated in Table
3, variations in rate of smoking reported varied in relation to the
timing and nature of smoking behaviour.
A number of important factors were measured in 2005 for the first time:
•
Women who were only advised to stop were much more likely to do so than
women who were advised to cut down (36% and 8% respectively)
• Women reported that between the age of four-six months, 9% of infants were exposed to cigarette smoke in the home
• Women reported that between the age of 8-10 months, 7% of infants were exposed to cigarette smoke in the home.
Table 4. Progress and challenges associated with smoking behaviour
Key: Progress in reducing smoking behaviour
Key: Ongoing challenges
Table 4. Progress and challenges associated with smoking behaviour
Table 4 provides further details concerning the
progress and challenges associated with smoking behaviour. Focusing
primarily on comparisons noted between smoking behaviour in 2000 and
2005, the following two aspects are discussed as challenges associated
with further health promotion activity and research:
• Pregnancy as motivation to stop smoking
• Smoking cessation and smoking relapse.
Discussion
It is known that smoking in
pregnancy is associated with an increased risk of infant mortality
(Department of Health, 2007). Likewise, the financial cost to the NHS
(UK) is also reportedly high; ranging from £8.1m to £64m per annum
(Godfrey et al, 2010). By comparing smoking behaviour reported in the
2000 IFS with that of the 2005 IFS, two aspects of smoking in pregnancy
and successful smoking cessation come to the fore; pregnancy as
motivation to stop smoking and smoking cessation and relapse.
Pregnancy as motivation to stop smoking
At
first glance, the results presented in Table 3 indicate that little
progress has been made in relation to the cessation of smoking in
pregnancy. For example, overall the same percentage of established
smokers reported giving up in 2000 as in 2005 (16%). In fact, smoking
cessation as an overall response to the confirmation of pregnancy
appeared to be lower in 2005 than in 2000. However on closer exploration
of the results related to sub-groups (Table 4), progress is noted. For
example, there was a 2% increase in the number of women who decided to
stop at some point prior to the birth of their baby. Likewise, a slight
reduction in smoking behaviour among professional/managerial women was
reported.
According to Fang et al (2004), women’s desire
to stop smoking during pregnancy should not be surprising; the authors
point out that the main motivation underpinning women’s decision to stop
or reduce the amount of cigarettes they smoke, is their desire to
protect their unborn baby. Recognising the importance of smoking
cessation during pregnancy is beneficial. Evidence shows that quitting
smoking before 15 weeks’ gestation can significantly reduce the serious
risks (McCowan et al 2009). However, during the last trimester of
pregnancy when rapid fetal growth normally occurs, smoking is related to
placental insufficiency and a subsequent low birthweight (British
Medical Association, 2004). Although the results indicated a 1% increase
in the number of women who were advised about smoking during their
pregnancy (of which 89% reported midwifery advice), Lawrence (2002)
points out that there is less than a 50% chance that health
professionals will routinely discuss smoking advice during this critical
period.
Interestingly, more women (3%) appeared to find
the motivation to reduce the amount of cigarettes they smoked during
their pregnancy. Although the IFSs demonstrated that the majority of
smokers cut down their cigarette consumption in pregnancy; studies using
cotinine measures of exposure to tobacco smoke suggest that women who
cut down, change the way in which they smoke so as to obtain their usual
level of nicotine. Furthermore, women are still exposed to the same
level of toxins and the reduction in the number of cigarettes smoked is
only temporary and unlikely to last into the third trimester (Lawrence
et al, 2003). Consistent with the evidence to date, the data collected
in 2005 demonstrated that women, who were advised to stop smoking
completely, rather than reduce smoking, were more likely to actually
stop (36% and 8% respectively). It is therefore on the strength of such
evidence that the smoke-free pregnancy approach aims to strengthen the
message that every cigarette smoked when pregnant is harmful to the
unborn baby (NICE, 2010; Department of Health, 2007; Dempsey et al,
2001).
Smoking cessation and smoking relapse
While
the evidence focuses on total smoking cessation rather than smoking
reduction, the findings of the surveys highlight the reality of smoking
relapse. A 4% increase was found in the number of women relapsing within
a year of cessation between 2000 and 2005.
Recognising the risk of relapse, authors such as
Lumley et al (2009) recommend that support and relapse prevention
strategies should be a routine part of antenatal care as the recording
of women’s blood pressure measurement. Although not reported in the
infant-feeding surveys, evidence shows that some pregnant women on
occasions increase their smoking behaviour due to stress or boredom
(Department of Health, 2007). As noted by Cope et al (2003) in the last
trimester issues such as anxiety, immobility and boredom may motivate
women to increase smoking among women who initially cut down their
smoking behaviour when first pregnant. Even though the reported data
illustrates the reality of smoking relapse, it is worth pointing out
that women who stopped smoking before their first antenatal booking
appointment are typically categorised as ex-smokers/non-smokers and
generally not offered cessation support. Although their cessation rates
throughout their pregnancy are high, many will relapse within six months
of delivery. NICE recommends the need for research into relapse
prevention during or after pregnancy (NICE, 2010).
Even though published evidence points to
motivators such as boredom and anxiety in late pregnancy, the results of
both surveys highlight the cultural influences related to smoking
behaviour. In 2005, the number of women who reported living with someone
who smoked dropped to 36% in comparison to 39% reported in 2000. While
this slight change in smoking behaviour among women’s partners is
encouraging, the reality remains that 36% of pregnant women are living
in a smoke-filled environment. Inferences as to the motivational impact
of living with a smoker on women’s smoking behaviour cannot be drawn
from this level of data, nonetheless the surveys indicate a lower
incidence of smoking behaviour among women who did not report living
with a smoker. Recognising that women who live with partners who smoke
find it harder to stop and are more likely to relapse (Fang et al,
2004), the Department of Health launched the national Smoking and
Pregnancy Partner’s Campaign (MIDIRS Midwifery Digest, 2002). Success of
that campaign may not be strongly evident in the 2005 data; however the
survey does provide some early indication of its success.
In 2005, for the first-time mothers were asked
about smoking in the home indicating the proportion of young infants who
are exposed to tobacco smoke at home. Only 9% of infants at stage two
and 7% at stage three lived in a household where at least one person
smoked in the home (Bolling et al, 2007). Considering that parental
smoking has major implications for infant wellbeing (Burgess, 2008),
evidence from the surveys emphasise the importance of the role of the
midwife in promoting smoking cessation in both women and their partners
(Bull, 2007).
The introduction of smoke-free legislation from
March 2004 in the Republic of Ireland, from March 2006 in Scotland;
April 2007 in Wales, April 2007 in Northern Ireland and from July 2007
in England has been popular and effective and has achieved high rates of
compliance (Department of Health, 2010). Despite concerns that smoking
bans may increase in smoking in the home (Flemming et al, 2008) the
amount of households that allow smoking is falling. The Office for
National Statistics reports a statistically significant increase in the
number of smoke-free homes in the UK from 61% in 2006 to 69% in 2008/09
(Lader, 2008). A welcoming development is therefore the publication of
national best practice guidance in relation to the extension of the NHS
stop-smoking services to address partners and other household smokers
(NICE, 2010).
Conclusion and recommendations for practice
The key messages to be taken are:
•
Comparison of the data between 2000 and 2005 demonstrates how young and
disadvantaged women are more likely to smoke throughout pregnancy
• Extend interventions and support to partners and other smokers living with the pregnant woman
• Pregnant women need clear and consistent messages to stop smoking rather than cutting down
• Stop smoking messages should be sensitive, non-judgemental and relevant to the circumstances of women’s daily lives.
Although the surveys report a decrease in the UK
prevalence of smoking in pregnancy, the findings still translates to
17% (one in six) women in 2005 who admitted to smoking throughout their
pregnancy. Areas of slight improvement have been noted since 2000
however the challenge of promoting smoke-free pregnancies and home
environments should not be underestimated. While health professionals
work towards the reduction of smoking in pregnancy and early motherhood
by 2020 (Department of Health, 2010), comparing the findings from the
2005 survey with that of 2000, confirms the need to provide health
promotion activity that is specifically targeted at young and
disadvantaged women and those who live with a partner who smokes.
Greaves et al (2003) reported smoking was not
the only health and wellbeing challenge that pregnant women faced;
issues such as nutrition, financial security and domestic violence may
be as equally pressing and could lead to increased smoking behaviour.
While particular women groups may require targeted and tailored
intervention, such as stopping young people from starting smoking
(Department of Health, 2010), effective intervention demands a
sensitive, non-judgemental women-centred approach (Greaves et al, 2003).
In order for midwives to meet the challenges laid down by the
infant-feeding surveys in relation to smoking behaviour in pregnancy and
early motherhood, they require ongoing educational input (NICE, 2010;
Fyle and Kaufmann, 2002).
References
Bolling K, Grant C, Hamlyn B, Thornton A. (2007) Infant
Feeding Survey 2005. A survey conducted on behalf of the Information
Centre for Health and Social Care and the UK Departments by BMRB Social
Research. The Information Centre for Health and Social Care: London.
British Medical Association. (2004) Smoking and reproductive life. The impact of smoking on sexual, reproductive and child health. Board of Science and Education and Tobacco Control Resource Centre.
Bull L. (2007) Smoking cessation intervention
with pregnant women and new parents (part two): a focus group study of
health visitors and midwives working in the UK. Journal of Neonatal Nursing 13: 179-85.
Burgess A. (2008) Maternal and infant health in the perinatal period: the father’s role. See www.fatherhoodinstitute.org/download.php?pID=6379.4 (accessed 14 July 2010).
Cope G, Nayyar P, Holder R. (2003) Feedback from a point of care test for nicotine intake to reduce smoking during pregnancy. Annals Clinical Biochemistry 40: 674-9.
Department of Health. (2007) Smoking and pregnancy qualitative research report. Research Works: Hertfordshire.
Dempsey DA, Benowitz NL. (2001) Risks and benefits of nicotine to aid smoking cessation in pregnancy. Drug Safety 24(4): 277-322.
Department of Health. (2010) A smokefree future: a comprehensive tobacco control strategy for England. HMSO: London.
Department of Health. (2007) Review of the health inequalities infant mortality target PSA target. HMSO: London.
Department of Health. (1998) Smoking kills: a White Paper on tobacco. HMSO: London.
Fang W, Goldstein OA, Butzen AY, Hartsock AS,
Hartmann KE, Helton M, Lohr JA. (2004) Smoking cessation in pregnancy: a
review of postpartum relapse prevention strategies. Journal American Board Family Practice 17: 264-75.
Fleming P, Blair PS. (2007) Sudden infant death syndrome and parental smoking. Early Human Development 83(11): 721-5.
Fyle J, Kaufmann T. (2002) Helping women stop smoking: a guide for midwives. RCM: London.
Godfrey C, Pickett KE, Parrot S. (2010) Estimating the costs to the NHS of smoking in pregnancy for pregnant women and infants. Department of Health Sciences, The University of York: York.
Greaves L, Cormier R, Devries K, Bortoff J, Johnson J, Kirkland S and Aboussafy D. (2003) A best practices review of smoking cessation interventions for pregnant and postpartum girls and women. British Columbia Centre of Excellence for Women’s Health: Vancouver.
Hamlyn B, Brooker S, Oleinikova K, Wands S. (2002) Infant Feeding Survey 2000. See www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsStatistics/DH_4079223 (accessed 4 August 2010).
Lader D. (2008) Smoking related behaviour and attitudes 2007. Office for National Statistics: London
Lawrence T, Aveyard P, Croghan E. (2003) What
happens to women’s self-reported cigarette consumption and urinary
cotinine levels in pregnancy? Addiction 98(9): 1315-20.
Lawrence T. (2002) Smoking and pregnancy. An open learning pack for midwives. Birmingham Health Authority: Birmingham.
Lumley J, Chamberlain C, Dowswell T, Oliver S,
Oakley L, Watson L. (2009) Interventions for promoting smoking cessation
during pregnancy. Cochrane Database of Syst Reviews 3: CD001055.
McCowan L, Dekker GA, Chan E, Stewart A,
Chappell LC, Hunter M, North RA. (2009) Spontaneous preterm birth and
small for gestational age infants in women who stop smoking early in
pregnancy: prospective cohort study. British Medical Journal 338: b1081.
MIDIRS Midwifery Digest. (2002) A health
education campaign designed to encourage the partners of pregnant women
to stop smoking. MIDIRS 12(3): 342-6.
NICE. (2008) Smoking cessation services in
primary care, pharmacies, local authorities and workplaces, particularly
for manual working groups, pregnant women and hard to reach
communities. NICE: London.
NICE. (2010) How to stop smoking in pregnancy and following childbirth (guidance 26). NICE: London.
