The prevalence and factors associated with smoking among lesbian and bisexual women: Analysis of the Australian National Drug Strategy Household Survey

Abstract

Background

In many countries, sexual minority women smoke at higher rates than their heterosexual counterparts. Analyses tend to combine lesbian and bisexual women, preventing an understanding of relevant factors associated with smoking for each group. This analysis used a representative sample of the Australian population to compare tobacco use between heterosexual, lesbian and bisexual women, and examine factors associated with smoking among these groups.

Methods

In a secondary analysis of data from the National Drug Strategy Household Survey (N = 23,855), descriptive statistics were produced for heterosexual (n = 11,776), lesbian (n = 135) and bisexual (n = 167) women. Multivariate logistic regression modelling was undertaken to assess which factors were associated with current smoking among the different groups.

Results

Compared to heterosexual women, lesbian and bisexual women were more likely to be current smokers (OR 2.9(1.8,4.5) and OR 3.6(2.4, 5.4) respectively). Employment, income and psychological distress were significant factors associated with smoking for lesbian women. Recent illicit drug use was the only significant factor associated with smoking for bisexual women.

Conclusions

We need to better understand the psychological, social and cultural factors that influence initiation, and sustain smoking among lesbian and bisexual women. Our findings demonstrate that sexual minority women in Australia warrant specific policy attention in a national framework.

Introduction

Smoking related disease continues to be a substantial burden globally, and one of the leading risk factors for early, and preventable, mortality (Reitsma et al., 2015). The prevalence of smoking among Australia’s general population has halved from 24% in 1991 to 12% in 2016 (Australian Institute of Health & Welfare, 2017). Despite this, minority populations such as Aboriginal and Torres Strait Islander people, prison populations, people with mental health problems and members of the lesbian, gay and bisexual communities continue to smoke at higher rates than the general community (Australian Institute of Health & Welfare, 2011, 2014; Johnston, Shahid, Schlumpp, & Wilkins, 2015; Lawrence, Mitrou, & Zubrick, 2009). These high rates of smoking have prompted a significant public health focus on reducing tobacco use, with Australia’s National Tobacco Strategy 2012–2018 making smoking among some minority groups a priority (Intergovernmental Committee on Drugs, 2012). What is currently missing from this national policy is consideration of the higher prevalence of smoking in sexual minorities and, in particular, sexual minority women (Intergovernmental Committee on Drugs, 2012).

International studies from industrialised countries show sexual minority women smoke at substantially higher rates than heterosexual women (Lee, Griffin, & Melvin, 2009; Lunn et al., 2017; Meads, Buckley, & Sanderson, 2007; Steele, Ross, Dobinson, Veldhuizen, & Tinmouth, 2009). A recent representative study from the United States shows lesbian and bisexual women have twice the odds of tobacco use than heterosexual women (McCabe et al., 2018). Another United States study found that lesbian and bisexual women have higher rates of smoking than heterosexual men and women, and sexual minority men (Wheldon, Kaufman, Kasza, & Moser, 2018). In Australia, there are few studies using representative samples of lesbian and bisexual women to examine smoking prevalence. One analysis, using National Drug Strategy Household Survey (NDSHS) data, reported lesbian/bisexual women had increased rates of smoking compared to heterosexual women (23.7% and 10.6% respectively; Roxburgh, Lea, de Wit, & Degenhardt, 2016). A more recent Australian study, using data from the Household, Income and Labour Dynamics in Australia (HILDA) survey, reported lesbian and bisexual women were significantly more likely to use tobacco than heterosexual women (Sabia, Wooden, & Nguyen, 2018).

There is a lack of consensus in the evidence from the United States, Canada and Australia, around differences in smoking rates between bisexual and lesbian women. Representative evidence from Australia is slim: two Australian representative studies investigating substance use did not separate lesbian and bisexual women (Hillier, De Visser, Kavanagh, & McNair, 2003; Roxburgh et al., 2016). A more recent study (using HILDA data) adjusted for demographic, family and personality controls, found only bisexual women were significantly more likely to smoke than heterosexual women, and results among lesbian women were non-significant (Sabia et al., 2018). Non-representative studies from the United States and Canada report bisexual women have higher odds of current smoking than lesbian women (Fallin, Goodin, Lee, & Bennett, 2015; Matthews, Steffen, Hughes, Aranda, & Martin, 2017).

Bisexual women are more likely to have higher perceived stress, lower mental health scores and less social support than both lesbian and heterosexual women (Hughes, Szalacha, & McNair, 2010). Bisexual women may also experience increased rates of poor general health than lesbian women (Fredriksen-Goldsen, Kim, Barkan, Balsam, & Mincer, 2010). Evidence from a recent review indicates bisexual women are at risk of increased rates of alcohol and illicit drug use (Green & Feinstein, 2012). These increased health risks may be a result of negative attitudes and prejudice from both the heterosexual and lesbian population (Mohr, 2004). Further investigation using Australian population level data, separating lesbian and bisexual women is needed.

Internationally, large representative studies report that smoking, risky or binge drinking, and illicit drugs are more highly correlated among sexual minority women than among the general population (Blosnich, Farmer, Lee, Silenzio, & Bowen, 2014; Kerr, Ding, Burke, & Ott-Walter, 2015; McCabe, Hughes, Bostwick, West, & Boyd, 2009). One longitudinal study found risky drinking significantly associated with tobacco use and associated with continued tobacco use over time among sexual minority women (Matthews et al., 2014). An Australian periodic survey of the health of lesbian, bisexual and queer women engaged with the Sydney’s lesbian, gay, bisexual and transgender (LGBT) communities (known as SWASH) found current smokers were significantly more likely to report illicit drug use than ex-smokers and significantly more likely to report regular binge drinking and illicit drug use than never smokers (Deacon & Mooney-Somers, 2017). One explanation for higher rates of smoking, drinking and illicit drug use is the attendance at LGBT bars and clubs. Socialising in bars may expose women to pro-smoking norms among the LGBT community, with the qualitative literature suggesting that for sexual minority women, smoking is both a way to connect with peers and a way of projecting an identity in these social spaces (Green & Feinstein, 2012; Gruskin, Byrne, Altschuler, & Dibble, 2008; Youatt, Johns, Pingel, Soler, & Bauermeister, 2015). We will examine the relationship between alcohol, illicit drug use and smoking among sexual minority women to confirm if these are significant factors for lesbian and bisexual women in the Australian population.

Meyer’s minority stress model (Meyer, 2003) is often used to explain high rates of smoking, risky drinking and illicit drug use (see for example, Lehavot & Simoni, 2011; Bariola, Lyons, & Leonard, 2016). This theory suggests that stressors such as internalised homophobia, anticipated stigma or stress and sexuality-related victimisation explain the higher rates of mental health diagnoses among sexual minorities and may have an effect on substance use in this population (Meyer, 2003).

Another approach to understanding factors associated with smoking is as a coping mechanism for psychological distress, potentially an outcome of minority stress experiences (Lea, de Wit, & Reynolds, 2014). International research shows a significant association between psychological distress (including depression, anxiety and worry, measured for example by the Kessler psychological distress (K10 scale; Kessler et al., 2002) and sexual minority status (Cochran, Greer Sullivan, & Mays, 2003; Platt & Scheitle, 2018). There has been limited work in Australia on the relationship between smoking and psychological distress among sexual minority women. In Australia, the SWASH survey found high rates of psychological distress among their sample of sexual minority women (42% reported medium to high psychological distress using the K6), however no significant association was found with smoking (Deacon & Mooney-Somers, 2017). The authors speculate that the high rates of psychological distress in their sample created a ceiling effect for the association between psychological distress and smoking (Deacon & Mooney-Somers, 2017). An Australian-based study found no clear link between higher rates of psychological distress and alcohol or illicit drug use; they did not examine smoking (Lea et al., 2014). Given these uncertain results, further investigation of the association between psychological distress and smoking in Australia is needed.