Clinical Practice Guidelines Antenatal care - Module I

Smoking interventions

Page last updated: 02 April 2013

This analysis assesses whether recommending certain smoking cessation interventions to pregnant women who smoke, in addition to a minimum intervention, would be cost-effective compared to this minimum intervention.

The minimum intervention is the intervention likely to be received by pregnant women even if the recommendation is not implemented. Consistent with current Royal Australian College of General Practitioners guidelines (Zwar et al 2007), it is assumed to be the application of the 5As smoking cessation framework by general practitioners and referral to Quitline.

1 Cost-effectiveness

In order to measure the cost-effectiveness of these interventions, this paper calculates incremental costeffectiveness ratios for the relevant smoking cessation interventions. The incremental cost-effectiveness ratio is calculated by dividing the relative cost of an intervention by the relative benefit of the intervention. That is,
ICER = (Costintervention – Costcomparator) / (Health benefitintervention – Health benefitscomparator).

The incremental cost-effectiveness ratio is calculated as the cost per life-year saved (based on an increase in life-expectancy for the mother due to a life-time quit and a reduction in perinatal deaths) and the cost of a woman successfully quitting smoking during pregnancy.

How to Compare the Costs and Benefits: Evaluation of the Economic Evidence (NHMRC 2001) recommends that when the incremental cost-effectiveness ratio is expressed as cost per life-year gained:
  • health care options that fall below the threshold of $A30,000 per life-year gained are considered good value and are recommended;
  • health care options that exceed a threshold of $A100,000 per life-year are not recommended without strong justification; and
  • health care options that fall between $A30,000 and $A100,000 per life-year are given further consideration.
Where several interventions are considered, as is the case with the current study, and more than one is cost-effective (within the thresholds outlined above), the preferred intervention is the intervention offering the highest benefit (NHMRC 2001).

Costs

Consistent with How to Compare the Costs and Benefits: Evaluation of the Economic Evidence (NHMRC 2001), which indicates that economic evaluations should focus on health sector budget impacts, this analysis estimates costs to the health system payer (for example, Medicare and the Pharmaceutical Benefits Scheme [PBS]).

Benefits

Both mother and baby benefit from a pregnant woman ceasing smoking.
  • Long-term smokers are at greater risk of developing diseases such as heart disease, stroke, lung diseases and various forms of cancer. However, as the period of abstinence increases, the risk of developing certain diseases approaches that of a non-smoker (Quit Victoria 2011).
  • Smoking during pregnancy has been associated with, among other things, a higher rate of perinatal death (Laws et al 2006).
This analysis uses findings in the literature to estimate (discounted) life-years gained for mother and baby from each relevant smoking intervention.

Potentially, future health expenditure savings from pregnant women ceasing smoking could be included as benefits. However, there is considerable debate about whether smoking interventions, and preventive measures more generally, reduce long-term health expenditure.1 Accordingly, this analysis does not attempt to estimate future health expenditure savings from pregnant women quitting smoking.

Pregnant women quitting smoking may also generate economic benefits — for example, increased productivity from reduced sick leave. These costs fall outside the scope of this analysis.
Top of page

2 Excluded interventions

Motivational interviewing

Due to a lack of evidence that a combination of motivational interviewing and the minimum intervention is more effective than the minimum intervention alone, motivational interviewing is not further discussed
in this paper. For instance:
  • Ruger et al (2008) found that motivational interviewing was more costly and no more effective than usual care in promoting smoking cessation among pregnant women;
  • only one of the seven studies reviewed by Lumley et al (2009) involving motivational interviewing interventions for pregnant women favoured the intervention over the control; and
  • while Parker et al (2007) found that motivational interviewing provided over the phone could be cost-effective, the methodology employed in their evaluation may have produced biased estimates.
Moreover, telephone-based motivational interviewing is already provided by the Quitline within the minimum intervention.

Bupropion SR

Two bupropion SR medicines, Zyban and Prexaton, are currently available on the PBS. The Therapeutic Goods Administration has rated bupropion as a class B2 medicine, indicating that while the use of bupropion does not appear to be associated with an increase in the frequency of harmful effects on the human fetus, it has only been taken by a limited number of pregnant women and women of childbearing age. Accordingly, the safety of using bupropion SR during pregnancy has not been sufficiently established and is therefore not discussed further in this analysis.

Varenicline

While varenicline tartrate is available on the PBS, the safety of using this medicine during pregnancy has not been established and it has not been rated by Therapeutic Goods Administration. Moreover, the manufacturer of Champix® (the only varenicline medicine available on the PBS) recommends against its use in pregnancy (Champix® Consumer Medicine Information). It is therefore not discussed further in this analysis.

3 Literature summary

Cochrane, Embase and Medline were searched for English-language articles published since 2006. The search terms pregnancy with smoking cessation were combined with cost effectiveness, utility or benefit. Manual searches were undertaken on the United Kingdom National Health Service Evidence website. These searches resulted in one review of the relevant literature (Ruger & Emmons 2008) and three additional economic evaluations of smoking cessation interventions aimed specifically at pregnant women (Dornelas et al 2006; Parker et al 2007; Ruger et al 2008). The search also uncovered a number of articles relating to the effectiveness and costs of smoking cessation interventions for pregnant women, as well as the population more generally.

Economic evaluations

Ruger and Emmons (2008) assessed eight economic evaluations of smoking cessation and relapse prevention programs for pregnant women. The interventions included self-help programs, educational material and counselling. The authors found that the results from all studies ‘indicated favourable outcomes for intervention methods aimed at reducing smoking during pregnancy and, subsequently, improving maternal and infant health outcomes’. Four of the studies also compared the costs of smoking cessation interventions with the potential cost-savings from reduced admissions to neonatal intensive care units and preventing neonatal complications and long-term disability. The cost-savings were found to be larger than the intervention costs for all four studies.

Dornelas et al (2006) conducted a randomised trial comparing usual care with a counselling intervention (which employed cognitive behavioural techniques) with telephone follow-up, in additional to usual care, among low-income pregnant women. The authors concluded that the intervention was a cost-effective way of yielding a non-smoker by the end of pregnancy.

A number of other studies have demonstrated the cost-effectiveness of smoking cessation interventions for the general population. For example, Cornuz et al (2006) estimated the cost-effectiveness of various forms of NRT (including patch, gum, spray and inhaler) and bupropion SR in addition to brief counselling by a GP, compared with counselling alone. The authors found that pharmacotherapies would be costeffective potentially doubling the changes of achieving cessation, with bupropion followed by the NRT patch being the most favourable treatments.

Clinical literature

The Cochrane Collaboration published a review of 72 trials relating to the effectiveness of smoking cessation interventions during pregnancy (Lumley et al 2009). This review found, among other things:
  • a significant reduction in smoking in late pregnancy attributable to the combined effect of various interventions with an absolute difference of six in 100 women (95% CI 4–7) who stopped smoking during pregnancy, compared with the control;
  • a significant increase in smoking cessation from CBT, rewards and NRT, compared with the control; and
  • a significant reduction in low birth weight and preterm births for women in the intervention group.
On the other hand, no significant difference was found for very low birth weight, stillbirths, neonatal deaths, neonatal intensive care admissions or total perinatal mortality. In addition, Lumley et al (2009) reviewed only a limited number of studies involving motivational interviewing interventions for pregnant women and only one that favoured the intervention over the control.
Top of page
An Australian study examined the effectiveness of nicotine patches and counselling, versus counselling alone, for pregnant women who smoke more than 15 cigarettes per day (Hotham et al 2006). The study found that 35% of the treatment group recorded a 50% reduction in their cotinine level from the start of the study, compared with 25% for the control group. In addition, 15% of the treatment group reported that they had achieved cessation (confirmed by cotinine analysis) compared with none of the control group. The authors noted that ‘nicotine metabolism is accelerated in pregnancy and higher than usual doses of NRT may be necessary to achieve nicotine levels that ease withdrawal and thus optimise efficacy’.

Woolacott et al (2002) performed a review of 157 studies relating to the clinical effectiveness and cost-effectiveness of bupropion SR and NRT as smoking cessation interventions for the general population. This study found that ‘irrespective of the methods used or the assumptions involved, the results of existing economic evaluations consistently suggest that smoking cessation interventions are relatively cost-effective in terms of the cost per life-year saved’. The potential negative side-effects of NRT on an unborn child were not taken into account, as the studies reviewed by Woolacott et al were not aimed at pregnant women specifically.

Adams et al (2002) found that smoking during pregnancy increases the risk of admission to a neonatal intensive care unit by almost 20 per cent and increases the length of stay, compared to admitted babies with non-smoking mothers. The authors found that in 1995, 2.2% of neonatal costs across 13 states in the United States were attributable to smoking during pregnancy.

4 Methods

This section:
  • estimates the cost-per-treatment of each intervention (costs are in Australian dollars and no discounting was applied, given that they would all occur within a nine-month period);
  • estimates the likely quit-rate from each intervention; and
  • estimates the consequential benefits to mother and baby from quitting (in avoided perinatal deaths and discounted life-years gained).

Cost-per-intervention

Minimum intervention: Advice, education and referral to the Quitline
The following assumptions were made in estimating the cost for the minimum intervention:
  • the minimum intervention is provided to pregnant women over two Level B consultations2 with their GP during which a range of standard antenatal care matters are addressed, with the minimum intervention taking up one-third of the consultation; and
  • patients are bulk-billed.
Under these assumptions, the per patient cost (to Medicare) of the minimum intervention is $A23.27.

Cognitive behavioural therapy
It is unclear how many CBT sessions would be required for a successful quit — it is likely that the number of sessions would differ by patient.

This analysis initially assumes that CBT involves one session with a clinical psychologist, lasting between 30 and 50 minutes,3 similar to the intervention studied by Dornelas et al (2006). Under this assumption, the cost (to Medicare) per patient for CBT, in addition to the cost of the minimum intervention, is $A81.60, with an additional $A14.40 borne by the patient.

NRT
It was assumed that the treatment protocol used by Hotham et al (2005) would be prescribed. This protocol was continuous (15 mg per 16 hours) patch use for up to 12 weeks, with an option to step down to lower strength patches over time. It was also assumed that pregnant women receive the prescription during the GP appointments included in the minimum intervention. Under these assumptions, the cost to the PBS per intervention for NRT was $A63.06 with an additional $A102.60 borne by the patient.

Quit rates

Minimum intervention and spontaneous quitting
The available data suggest that 16.6% of pregnant women smoked during pregnancy in 2007. Further, approximately 19% of women between 20 and 39 were smokers in 2007. This analysis assumes that the resulting 12.9% quit rate represents those women who spontaneously quit (that is, who quit without any assistance prior to their first appointment with their GP) and those who quit as a result of the minimum intervention.

To estimate the cost-effectiveness of the minimum intervention, it is necessary to estimate the spontaneous quit rate. Spontaneous quit rates of between 1% and 8% have been used in existing economic evaluations of smoking cessation interventions for the population in general, with a rate of 1% being used in most of the United Kingdom studies that were reviewed by Woolacott et al (2002). It was assumed, however, that the spontaneous quit rate for women who are pregnant would be higher than that for the total population. Accordingly, it was assumed that the spontaneous quit rate for Australian women who are pregnant was 4%.

Cognitive behaviour therapy
Lumley et al (2009) found an additional 5% quit rate from CBT (95% CI 3–7%), compared with the control. The control differed between studies, but was typically the provision of advice and education at a standard antenatal appointment. The Lumley finding was adopted in this analysis, implicitly assuming that the control in Lumley was equivalent to the minimum intervention.
Top of page
NRT
Lumley et al (2009) found an additional 5% quit rate from pharmacotherapies (95% CI 2–8%), after examining five studies that included the use of NRT compared with the control.

Caveats
Many of the studies that informed this analysis relied on self-reported smoking status, which may mean that quit rates are overstated (Lumley et al 2009).

The quit rates assumed above are the average quit rates across pregnant women who smoke differing numbers of cigarettes per day. It is conceivable, however, that quit rates may vary by the level of nicotine addiction.

In addition, as smoking rates across the general population fall, the marginal effectiveness of interventions aimed at assisting smoking cessation may decline (Woolacott et al 2002). That said, the quit rates assumed above are, for the most part, calculated based on studies conducted in the United States where the proportion of pregnant women that smoke (around 12% [Dornelas et al 2006]) is below that of Australia (around 16% [Laws & Sullivan 2009; Laws et al 2010]). Accordingly, the quit rates used here may underestimate the effect of these interventions on pregnant women in Australia.

Given these caveats, the sensitivity of the analysis in this analysis to the assumed quit rates is tested in Section 5.

Relapse rate
It was assumed that pregnant women will only realise benefits from quitting smoking if they remain a non-smoker for the rest of their lives. Relapse rates vary in the literature between 10% and 50%. For example, Cornuz et al (2006) used a 35% lifetime probability of relapse after one year of abstinence and varied this between 10% and 50% in their sensitivity analysis (see also Woolacott et al 2002; Etter & Stapleton 2006; Gordon et al 2007).

The conservative assumption was adopted that 50% of women who quit during pregnancy would relapse after birth (varied to 30% and 100% in the sensitivity analysis).

Health benefits for mother and baby


Mother
The median age of women who gave birth in 2007 was 30.0 years (Laws & Sullivan 2009). Doll et al (a 50year British study on the health impacts of smoking on male doctors cited in Woolacott et al [2002]) suggests that the number of life-years saved from quitting smoking could be as high as 7.1 for those who quit by the age of 35. Woolacott et al (2002) stated that, for all quitters of all ages, an average of 1 to 3 undiscounted life-years saved per long-term quitter seemed a reasonable assumption, based on their literature review.

Given the uncertainty about this issue, it was conservatively assumed that a life-time quit by a 30-year old pregnant woman would result in her gaining 5 life-years and living to an age of 84.6 years, this being the life-expectancy of all 30 year old women in 2009, including smokers and non-smokers (ABS 2010). This figure was varied in the sensitivity analysis between 2 and 7 years.

Discounting the additional 5 life-years by 5% per annum4 gives 0.3377 additional life-years per successful lifetime quit (noting that the additional life-years come at the end of the woman’s life).

Baby
Smoking during pregnancy has been associated with an increased risk of low birth weight, preterm birth, admission to a neonatal intensive care unit or special care nursery, perinatal death5 and sudden infant death syndrome (SIDS) (Laws et al 2006). Data are available on the size of many of these increased risks. However, other data limitations mean that, in many cases, it is not possible to estimate the size of the benefits arising from eliminating the additional risk by quitting smoking.6 Consequently, this analysis is limited to estimating the benefits arising from reducing perinatal deaths.

Laws et al (2006) reported that babies of mothers who smoked during pregnancy were 46% more likely to die during the perinatal period.7 However, Lumley et al (2009) found no statistically significant difference between the rate of perinatal death for smoking mothers who received a cessation intervention and the control group. Lumley et al (2009) note, however, that women who had a perinatal death were often excluded from the outcome measurements of many of the studies they reviewed. Accordingly, it was concluded that the finding in Lumley et al (2009) does not rule out an increased risk of perinatal death from smoking.

Adopting the Laws et al finding, one successful quit during pregnancy would result in a reduction of 0.0044 perinatal deaths8 — that is, for every 1,000 women that quit smoking during pregnancy, approximately four perinatal deaths would be avoided.

For each avoided perinatal death, it was assumed the baby would live to the average of the current life expectancy at birth for males of 79.3 years and females of 83.9 years (ABS 2010). After discounting, one avoided perinatal death was equivalent to 19.7089 life-years. One successful quit during pregnancy would therefore result in 0.0859 life-years saved.9
Top of page

5 Results

This section:
  • calculates the cost per successful quit during pregnancy and the cost per life-year saved, using the relative costs and outcomes from Section 4;
  • draws conclusions about whether the proposed additional interventions are cost-effective compared to the minimum intervention alone; and
  • assesses whether the minimum intervention is cost-effective compared with doing nothing.

Cost-effectiveness of the minimum and additional smoking interventions

CBT and NRT

Table E3 provides the cost per quit during pregnancy and the cost per life-year saved due to CBT and NRT, compared with the minimum intervention alone. In terms of cost per life-year saved, both interventions are potentially cost-effective methods of reducing the number of pregnant women who smoke, with a cost per life-year saved of well under $A30,000.

Table E3: Baseline estimates of the cost-effectiveness of additional smoking cessation interventions compared with the minimum intervention

Intervention in addition to the minimum
Additional cost per quit attempt (A$)
Cost per additional quit during pregnancy due to the intervention (A$)
Cost per additional life-year saved due to the intervention (A$)
NRT
$63.06
$1,261
$4,951
CBT
$81.60
$1,632
$6,407

As discussed, this analysis was not able to account for all the potential health benefits of quitting smoking. However, the benefits of quitting smoking are well known and quitting smoking, or even cutting down, during pregnancy will always result in at least some small health benefit to the mother or the baby. Accordingly, with a cost per successful quit during pregnancy of less than $A2,000,10 offering NRT or CBT to pregnant women who smoke, in addition to the minimum intervention, seems like a reasonable proposition.

How to Compare the Costs and Benefits: Evaluation of the Economic Evidence (NHMRC 2001) notes that when more than one health care option is deemed cost-effective, the preferred option is the one that is the preferred clinical practice which is generally that which offers the highest benefit. In this case, all three interventions are assumed to deliver the same quit and relapse rates and, therefore, deliver the same health benefit.

Given this, it is noted that the addition of NRT to the minimum intervention is less expensive than CBT per life-year saved, compared with the minimum intervention alone. That said, the addition of NRT may not always be the preferred clinical practice.

Minimum intervention
Based on the assumptions in Section 4, the cost per life-year saved of the minimum intervention, compared with doing nothing, is $A1,031. Given that this figure falls well below the threshold of $A30,000 indicated in How to Compare the Costs and Benefits: Evaluation of the Economic Evidence (NHMRC 2001) (see Section 1), the minimum intervention is very cost-effective.

Sensitivity analysis

Three types of sensitivity analysis were conducted:
  • a one-way sensitivity analysis, under which key parameters were varied one at a time keeping all else equal to the baseline case;
  • determination of the ‘best’ and ‘worst-case’ outcomes using sets of optimistic and pessimistic parameters from Table E3; and
  • a comparison of the minimum quit-rates required for additional interventions to be cost-effective with findings in the literature.
One way sensitivity analysis
Table E4 provides the results of the one-way sensitivity analysis.
Top of page
Table E4: One-way sensitivity analysis cost per life-year saved per intervention in addition to the minimum, compared with the minimum intervention alone
Note: Table E4 is spilt into three tables and organised in proper way to meet accessibility requirements.

Baseline case
NRT (A$)
$4,951
NRT (A$) $4,951
CBT (A$) $6,407
CBT (A$) $6,407
Quit rates during pregnancy
(Lower bound)
(0.02)
$12,378
(0.03)
$10,678
Quit rates during pregnancy
(Upper bound)
(0.08)
$3,095
(0.07)
$4,576
Baseline case
NRT (A$) $4,951
CBT (A$) $6,407
Life-time relapse rate - 100%
$14,686
$19,004
Life-time relapse rate - 30%
$3,914
$5,064
Perinatal deaths avoided per quit - 0a
$7,470
$9,666
Perinatal deaths avoided per quit - 0.0035
$5,319
$6,883
Perinatal deaths avoided per quit - 0.0054
$4,583
$5,931
Undiscounted life-years saved for the mother per lifetime quit - 7
$3,761
$4,867
Undiscounted life-years saved for the mother per lifetime quit - 2
$8,506
$11,007
Number of CBT sessions - 4
na
$25,628
Number of CBT sessions - 10
na
$64,070
Discount rate applied to health benefits - 1.50%
$944
$1,221
Discount rate applied to health benefits - 3.50%
$2,511
$3,250
Baseline case
NRT (A$) $4,951
CBT (A$) $6,407
Direct patient out-of-pocket costsb are included in health system costs
$13,007
$7,538

a. This extreme case relates particularly to pharmacotherapies – that is, it effectively assumes that the potential negative side-effects of pharmacotherapies on babies cancel out all the health benefits of smoking cessation.
b. That is, PBS co-payments and health professional fees above the relevant Medicare rebate.

Overall, this analysis indicates that NRT and CBT remain cost-effective even when key assumptions are significantly varied.
  • The cost per life-year saved does not exceed $30,000 for NRT in any of the scenarios examined.
  • The incremental cost-effectiveness ratios exceed $30,000 for CBT only when the number of sessions involved is greater than four. That said, even if 10 CBT sessions were required to produce a quit during pregnancy, the cost per life-year saved remains below $A100,000.
Best and worst case outcomesBest and worst-case outcomes are shown in Table E5.
Top of page

Table E5: Best and worst case scenarios of cost per life-year saved per intervention in addition to the minimum, compared with the minimum intervention alone



NRT
CBT
Best case scenarioc
$A1,730
$A2,559
Worst case scenariod
$A46,191
$A39,848

c. The best case scenario assumed here is the upper bound for the 12-month quit rate, a relapse rate of 30 per cent, 0.0054 perinatal deaths avoided per successful quit and 7 undiscounted life-years saved for the mother per lifetime quit, using a discount rate of 5 per cent.
d. The worst case scenario assumed here is the lower bound for the 12-month quit rate, all women start smokingagain after giving birth and 0.0035 perinatal deaths avoided per successful quit, using a discount rate of 5 per cent.

Even in the worst-case outcome, the use of CBT or NRT with the minimum intervention, compared with the minimum intervention alone, may still be cost-effective. Importantly though, the worst-case scenario assumes that only a small percentage of pregnant women (3% for CBT and 2% for pharmacotherapies) quit as a result of the additional intervention, and that they all relapse — that is, it assumes the additional interventions are highly ineffective, a scenario that is inconsistent with clinical findings.

Minimum quit rate needed to make interventions cost-effective
The quit rate during pregnancy that would be required to result in an incremental costeffectiveness ratio of $A30,000 was calculated, assuming the parameters used in the main analysis above. These quit rates were compared with the findings in Lumley et al (2009). The results are set out in Table E6.

Table E6: Minimum additional quitters per quit attempt required for each additional intervention to be considered cost-effective (compared with the minimum intervention alone)

Intervention
Additional cost to health system per quit attempt
Number of additional quitters required per 1000 quit attempts
Number of additional quitters per 1000 quit attempts due to the intervention found by Lumley et al (2009)
NRT
$A63.06
12
50 (CI 20–80)
CBT
$A81.60
16
50 (CI 30–70)

The minimum number of additional quitters required such that the interventions are considered to be cost-effective compare favourably with the findings of Lumley et al (2009) for CBT and NRT.

6 Discussion

CBT and NRT are likely to be cost-effective interventions for smoking cessation during pregnancy. Both were assumed to result in the same quit rate above that achieved by the minimum intervention alone. Accordingly, the preferred option is the least costly intervention, which for most scenarios considered is NRT. That said, cost-effectiveness should not be the only consideration in choosing the preferred intervention as NRT may not be a clinically appropriate option for women who, for example, only smoke a few cigarettes per day. Furthermore, if the direct costs to the patient are taken into account, CBT appears to be the most cost-effective intervention.

7 Limitations

The sensitivity of this study to the key assumptions upon which it relies was tested in Section 5. Specific limitations of the study have also been noted — for example, data limitations meant that a range of potential benefits to the baby from the mother quitting smoking were not able to be considered. More general limitations of the study include:
  • The study focuses on cost per life-year saved and does not take the quality of additional life-years into account.
  • The analysis does not take into account benefits to the pregnant woman and fetus when the pregnant woman reduces her smoking rather than quitting altogether. Hotham et al (2006) found that 35% of their intervention group achieved at least a 50% reduction in cotinine, but only 15% of the intervention group quit smoking.
  • The data do not take into account health benefits to the baby (and others) from a reduction in exposure to passive smoking as a result of the mother remaining abstinent after birth as a result of the smoking cessation intervention she received during pregnancy.
As a result of potentially significant benefits being excluded from this study, the incremental cost-effectiveness ratios may be over-estimates.
Top of page

8 Conclusion

It is likely that offering CBT or NRT to pregnant women who smoke in addition to the minimum intervention would be cost-effective, compared with the minimum intervention alone.

References

ABS (2010) Life Tables, Australia 2007-2009. Cat. No. 3302.0.55.001. Canberra: Australian Bureau of Statistics.

Adams EK, Miller VP, Ernst C et al (2002) Neonatal health care costs related to smoking during pregnancy. Health Econ 11: 193–206.

Adhikari P & Summerill A (2000) 1998 National Drug Strategy Household Survey: Detailed findings. Drug Statistics Series No. 6. AIHW cat. no. PHE 27. Canberra: Australian Institute of Health and Welfare.

AIHW (2008) 2007 National Drug Strategy Household Survey: Detailed Findings. Drug Statistics Series no. 22. Cat. no. PHE 107. Canberra: Australian Institute of Health and Welfare.

Champix® Consumer Medicine Information. December 2010.

Cornuz J, Gilbert A, Pinget C et al (2006) Cost-effectiveness of pharmacotherapies for nicotine dependence in primary care settings: a multinational comparison. Tobacco Control 15: 152–59.

Dornelas EA, Magnavita J, Beazoglou T et al (2006) Efficacy and cost-effectiveness of a clinic-based counselling intervention tested in an ethnically diverse sample of pregnant smokers. Patient Ed Counsel 64: 342–49.

Etter JF & Stapleton JA (2006) Nicotine replacement therapy for long-term smoking cessation: a meta-analysis. Tobacco Control 15: 280–85.

Gordon L, Graves N, Hawkes A et al (2007) A review of the cost-effectiveness of face-to-face behavioural interventions for smoking, physical activity, diet and alcohol. Chronic Illness 3: 101–29.

Hotham ED, Gilbert AL, Atkinson ER (2006) A randomised-controlled pilot study using nicotine patches with pregnant women. Addict Behav 31: 641–48.

Laws PJ & Sullivan EA (2009) Australia’s Mothers and Babies 2007. Perinatal statistics series no. 23. Cat. no. PER 48. Sydney: AIHW National Perinatal Statistics Unit.

Laws PJ, Grayson N & Sullivan EA. 2006. Smoking and Pregnancy. AIHW Cat. No. PER 33. Sydney: AIHW National Perinatal Statistics Unit.

Laws PJ, Li Z, Sullivan EA (2010) Australia’s Mothers and Babies 2008. Perinatal statistics series no. 24. Cat. no. PER 50. Canberra: Australian Institute of Health and Welfare.

Lumley J, Chamberlain C, Dowswell T et al (2009) Interventions for promoting smoking cessation during pregnancy. Cochrane Database of Systematic Review, Issue 3.

Mortimer D, Segal L, Dalziel K (2005) Risk Factor Study – How to Reduce the Burden of Harm from Poor Nutrition, Tobacco Smoking, Physical Inactivity and Alcohol Misuse: Cost-utility Analysis of 5 Interventions to Discourage Tobacco Smoking. Monash University, Centre for Health Economics Research Paper April 2005 (5).

NHMRC (2001) How to Compare the Costs and Benefits: Evaluation of the Economic Evidence. Handbook series on preparing clinical practice guidelines. Canberra: National Health and Medical Research Council.

NICE (2004) Guide to the Methods of Technology Appraisal. London: NHS National Institute for Clinical Excellence.

OECD (2010a) Obesity and the Economics of Prevention, Fit not Fat. Franco Sassi, OECD.

OECD (2010b) Value for Money in Health Spending. OECD.

Oncken CA & Kranzler HR (2009) What do we know about the role of pharmacotherapy for smoking cessation before or during pregnancy? Nicotine & Tobacco Res 11(11): 1265–73.

Parker DR, Windsor RA, Roberts MB et al (2007) Feasibility, cost, and cost-effectiveness of a telephone-based motivational intervention for underserved pregnant smokers. Nicotine & Tobacco Res 9(1): 1043–51.

Quit Victoria (2011) For more information please visit Quit Victoria website.
Top of page
Ruger JP & Emmons KM (2008) Economic evaluations of smoking cessation and relapse prevention programs for pregnant women: A systematic review. Value in Health 11(2): 180–90.

Ruger JP, Weinstein MC, Hammond SK et al (2008) Cost-effectiveness of motivational interviewing for smoking cessation and relapse prevention among low-income pregnant women: A randomized controlled trial. Value in Health 11(2): 191–98.

TGA (1999) Prescribing Medicines in Pregnancy, 4th edition. Canberra: Therapeutic Goods Administration.

TGA (2006) Amendments to the Prescribing Medicines in Pregnancy, 4th edition, 1999. Canberra: Therapeutic Goods Administration.

Woolacott NF, Jones L, Forbes CA et al (2002) The clinical effectiveness and cost-effectiveness of bupropion SR and nicotine replacement therapy for smoking cessation: a systematic review and economic evaluation. Health Tech Assess 6(16).

Zwar N, Richmond R, Borland R et al (2007) Smoking Cessation Pharmacotherapy: an Update for Health Professionals. Melbourne: The Royal Australian College of General Practitioners.

1 See for example, OECD 2010a, p20-1; OECD 2010b, p72. In addition, many of the studies reviewed by Woolacott et al (2002) (a key study outlined in section 3 of this analysis) did not include future health expenditure costs.
2 January 2011 Medicare Benefits Schedule (MBS), item number 23. The MBS fee for this item is $34.90.
3 January 2011 MBS, item number 80000.
4 As per NHMRC (2001, page 54) recommendations.
5 Fetal or neonatal deaths of at least 400 grams birth weight or at least 20 weeks gestation.
6 For example, we do not have access to reliable weights to apply to life-years that would represent the health benefits associated
with reduced low birth weight or preterm births. Reliable data on the risk of SIDS deaths due to smoking during pregnancy are lacking and it would be difficult to separate out the impact of second-hand smoke, from either parent, after birth. In addition, we have not included potential cost-savings due to reduced admissions to special care nurseries or neonatal intensive care units given a lack of suitable data and the uncertainty around competing factors (eg the longer median length of stay for babies of nonsmoking mothers).
7 Laws et al notes, however, that this result has not been controlled for potential confounding factors, such as the age, educational level, socioeconomic status or medical conditions of the mother.
8 Or between 0.0035 and 0.0054 using the upper and lower bounds of the 95% CI of the risk ratio.
9 That is, 0.0044*19.7089.'
10 The cost per life-year saved is greater than the cost per quit during pregnancy because of the small probably of a perinatal death for both smokers and non-smokers and because we assumed that only women who remain abstinent after birth will accrue health benefits due to quitting smoking.