Demographic factors associated with smoking cessation during pregnancy in New South Wales, Australia, 2000–2011

This study examined demographic characteristics associated with smoking cessation in pregnancy and trends in these demographic characteristics between 2000 and 2011. Demographic characteristics associated with both higher rates of smoking in pregnancy and lower rates of smoking cessation during pregnancy included being a teenage mother, living outside major cities, being an Aboriginal person, and having an increasing number of previous pregnancies. These associations were observed in the first and last year of the study period. The study findings are consistent with previous findings from Australia [16,17] and internationally [18] that rates of smoking cessation during pregnancy are lower among teenage mothers, Aboriginal women, and women with more children.

This study indicates that smoking during pregnancy decreased steadily between 2000 and 2011, and smoking cessation during pregnancy was stable between 2000 and 2006 but increased markedly from 2007 onward. Many factors may have contributed to the changes in smoking behaviour observed in this study. At the population level, the adult smoking rate in NSW has been declining for over 30 years [25]. This decline is likely a result of sustained tobacco control strategies including taxation, advertising bans, mass media campaigns, and smoke-free environment legislation [26]. Recent initiatives include increases in the tobacco excise in 2010 and 2013, the introduction of bans on point-of-sale tobacco product displays in 2011, and the introduction of plain packaging of tobacco products in 2012 [26]. These population-level initiatives are likely to have raised general awareness of the risks of smoking among pregnant women, and contributed to the decline in smoking in pregnancy and increase in cessation during pregnancy observed in this study. However, the reason for the marked increase in rates of smoking cessation during pregnancy since 2007 is unclear. Although antenatal care providers are likely to provide some smoking cessation support as part of routine care, to the authors’ knowledge there have been no specific interventions targeting smoking in pregnancy, or changes in antenatal care practices since 2007, that would explain the sudden marked increase in cessation.

Despite the overall decline in smoking during pregnancy, demographic disparities in rates of smoking and smoking cessation in pregnancy remain substantial. Demographic groups including Aboriginal women, teenage mothers, women living outside major cities and women with three or more previous pregnancies remain significantly less likely to quit smoking during pregnancy. Moreover, although smoking cessation during pregnancy has increased in all demographic subgroups since 2000, the improvement in smoking cessation in these demographic groups is less than in other groups. To address these inequities, future efforts to reduce smoking in pregnancy may need to target specific population groups. Ongoing monitoring and evaluation is also valuable to assess the effectiveness of smoking cessation interventions in specific population groups, and the impact on pregnancy and birth outcomes.

Of particular concern are the high rates of smoking and low rates of smoking cessation among Aboriginal women. Aboriginal women continue to have the highest rates of smoking and lowest rates of smoking cessation during pregnancy of any demographic group, and the gap between Aboriginal and non-Aboriginal women is substantial. Reducing rates of smoking during pregnancy among Aboriginal women, and closing the gap in rates of smoking in pregnancy between Aboriginal and non-Aboriginal women, are key priorities in New South Wales [27]. The New South Wales Government state plan for the period 2011–2021 sets targets to reduce rates of smoking in pregnancy by 0.5% per year for non-Aboriginal women and by 2% per year for Aboriginal women. To achieve these targets, the New South Wales Ministry of Health is currently implementing Quit for New Life, a program facilitating improved provision of smoking cessation support to pregnant Aboriginal women. The impact of this program will be assessed and reported.

The observed associations between demographic characteristics and smoking cessation may in part reflect differential access to smoking cessation support. To the authors’ knowledge there are no Australian studies exploring factors associated with access to smoking cessation support during pregnancy, however overseas evidence suggests smoking cessation support is less likely to be provided to older women and poorer women [28]. In addition, there is evidence access to smoking cessation support varies by ethnicity, with one American study of pregnant women finding African American women were more likely, and Native American women less likely, to report receiving smoking cessation support compared to other women [29]. Several studies have also explored whether the effectiveness of smoking cessation support during pregnancy varies by demographic characteristics, however evidence is mixed. A recent systematic review concluded that psychosocial smoking cessation interventions during pregnancy are effective in reducing smoking in pregnancy. Sub-group analyses indicated psychosocial interventions have similar effectiveness among women from lower socioeconomic groups compared to other women, however trials in specific ethnic groups (African-American and Hispanic women) and Aboriginal women have found no treatment effect [30].

The main limitations of this study relate to the use of administrative data. First, the self-reported smoking status recorded in this dataset has not been validated. Self-reported data may underestimate the true rates of smoking in pregnancy as some women may choose not to disclose their smoking status, or unsuccessful attempts to stop smoking, to a health care provider. Second, the administrative dataset used for this analysis contains very limited information about smoking cessation. Using the available data, smoking cessation is defined as smoking during the first half of pregnancy but not in the second half. Smoking cessation which occurs during the second half of pregnancy is not recorded, as women who smoke at any time during the second half of pregnancy are recorded as smokers. Therefore it is likely that the Perinatal Data Collection underestimates rates of smoking cessation during late pregnancy. Further research is required to develop and validate robust measures of smoking cessation that could be incorporated into routine maternity data collection processes.