Distinct patterns of cigarette smoking intensity and other substance use among women who experience housing instability

Abstract

Background.

Recent insights into substance use cessation suggest that outcomes short of long-term abstinence are clinically meaningful and may offer more realistic incremental goals, particularly for highly vulnerable individuals. With the goal of informing tobacco treatment programs, we examined distinct patterns of cigarette smoking and their association with the ongoing use of other substances in women who experience housing instability.

Methods.

We recruited participants from a longitudinal study of women experiencing housing instability. Between June 2017 and January 2019, participants completed six monthly survey interviews regarding social conditions and the use of multiple substances. We examined associations between cigarette smoking intensity, including number of cigarettes smoked per day, heavy smoking, and an increase in number of cigarettes smoked from the previous 30-days, and other substance use in the past 7-days.

Results.

Of the 243 participants, 69% were current smokers and 58% were daily smokers. Number of cigarettes smoked per day (Adjusted odds ratio [AOR] 1.02, 95% CI 1.00–1.03), heavy cigarette smoking, compared to none or light smoking (AOR 2.02, 95% CI 1.46–2.79), and an increase in number of cigarettes smoked from the previous 30-days (AOR 1.06, 95% CI 1.01–1.12) were all significantly associated with methamphetamine use in the past 7-days. Associations with other substance use were not as strong.

Conclusions.

In a sample of unstably housed women, where almost half used multiple substances, methamphetamine use was associated with higher cigarette smoking intensity. Our findings highlight a potential role for integrating tobacco and methamphetamine use treatment to reduce tobacco use among unstably housed women.

1. Introduction

Cigarette smoking among women experiencing housing instability is common, with an estimated prevalence of 50% to 70% compared with 13% in the general population (Kim et al., 2017; Vijayaraghavan et al., 2012). Smoking-related chronic diseases, including cancers of the lung, trachea, and bronchus, are the leading causes of morbidity and mortality among individuals experiencing homelessness (Baggett et al., 2015). Substance use is another leading contributor to morbidity and mortality in this population, and tobacco use accounts for more than half of those substance-use related deaths (Baggett et al., 2015).

A recent study of women who experienced homelessness or unstable housing showed that current tobacco use was significantly higher among those who used alcohol, cannabis, opioids, or stimulants, with the odds of tobacco use more than doubling for every additional substance used (Riley et al., 2022). These findings are consistent with prior reports indicating high rates of polysubstance use in vulnerable populations (Prochaska et al., 2017). Some treatment programs have responded by simultaneously treating multiple addictions, and research shows that integrating tobacco and other substance use treatment results in a 25% increase in the likelihood of long-term abstinence from all substances when both are treated together (Prochaska et al., 2004).

In addition, a study among people experiencing homelessness showed that treating tobacco use did not interfere with substance use treatment and may even reduce number of drinking days (Reitzel et al., 2014), and that co-treatment effects are robust, even in population-level analysis among high-risk adults with major depression, substance use disorders or both (Han et al., 2022). Despite the increased momentum for co-treatment of tobacco and substance use in the general population, there exists gaps in offering integrated treatment for cigarette smoking and other substance use among especially high-risk populations like unstably housed women.

Much of the existing evidence informing tobacco and substance use treatment is based on abstinence (i.e., any use versus no use), but there is evidence that reduction in use can support cessation. In particular, over half of people experiencing homelessness who participated in a motivational interviewing randomized trial reduced smoking by at least 50%, indicating that reduction is feasible among homeless smokers (Battacharya et al., 2022). Adopting reductions in use for substance use interventions in general may also offer more realistic incremental goals, particularly for highly vulnerable individuals (Volkow et al., 2018), including adults experiencing homelessness who smoke who have more barriers to quitting (Baggett et al., 2018).

In this study we build on prior research, which indicated that the odds of any tobacco use increased with every additional substance used among women who experience homelessness or housing instability (Riley et al., 2022). Here, we extend these findings by examining the association between other substances used and intensity of cigarette smoking, defined as a spectrum rather than a binary factor of any smoking vs. not. We hypothesized that, in addition to associations with abstinence, the use of other substances is associated with higher incremental levels of cigarette smoking. Specifically, compared to lower levels of use, an increasing number of cigarettes smoked per day, heavy smoking, and an increase in cigarette consumption from the previous 30 days would all be associated with the ongoing use of other substances. Findings from this study, which are intended to elucidate more complex relationships between tobacco use and other factors, could inform the development of integrated tobacco and substance use interventions scaled to service settings for especially high-risk populations such as women experiencing housing instability.

2. Material and Methods

2.1. Study procedures

The present study was a sub-study within the “Polysubstance Use and Health Outcomes Evaluation” (PULSE), an investigation of relationships between substance use and cardiac dysfunction in homeless and unstably housed women (Riley et al., 2020, 2022). Data collection took place between June 2017– January 2019. Study participants were recruited in one of three ways: (1) random sampling from shelters and free meal programs; (2) area sampling of street encampments, a technique used when no complete frame of reference is available, and maps rather than lists serve as the sampling frame; and (3) sampling with probability proportionate to the number of women housed (Riley et al., 2020, 2022) from San Francisco’s low-income single room occupancy hotels. Inclusion criteria included being born as female, age ≥ 18 years or older, and having a lifetime history of housing instability (i.e., slept in public or a homeless shelter, or stayed with a series of associates because there was no other place to sleep [“couch-surfed”]). Participants completed monthly study visits, including an interview, blood draw and urine collection, for six months and were reimbursed $40 for each interview. Participants had a 91% response rate, with a median of 5 completed study visits each, and a total of 1051 study visits (Riley et al., 2022). The University of California, San Francisco Institution Board approved study procedures (IRB # 18-26797).

2.2. Data collection

The goal of the current study was to build on our prior work, which indicated an increase in the odds of tobacco use with every additional substance used, with a more detailed investigation of smoking across a spectrum of cigarette use levels. We categorized cigarette smoking intensity as: (1) number of cigarettes consumed per day on average, (2) heavy smoking (≥10 cigarettes/day versus < 10 cigarettes per day) (Pierce et al., 2011) and (3) increase in cigarettes smoked from the previous 30-days among those who had at least two consecutive visits for which there was a documented smoking status (≥10 cigarette increase from the previous 30-days vs. not). This measure was only calculated for visit two onward. We assessed other substances used in the past 7-days as any self-reported use of alcohol, cannabis, crack/cocaine, heroin, or methamphetamines. Among people who used alcohol in the past 7-days, we assessed heavy use of alcohol (>7 drinks/day) (National Institute on Alcohol Abuse and Alcoholism).

2.3. Statistical Analysis

We described sample statistics using median and interquartile range (IQR) for continuous variables and proportions for categorical variables. Using generalized estimating equations (GEE), accounting for repeated measurements and within-cluster correlations, we examined the associations between the three measures of cigarette smoking intensity, including number of cigarettes smoked per day, heavy smoking, and an increase in cigarettes smoked from the previous 30-days, and other substance use (i.e., alcohol, cannabis, crack/cocaine, methamphetamines, and heroin) in the past 7-days. We specified an exchangeable working correlation structure to account for the higher level of clustering (i.e., potential nonindependence of repeat observations within individuals), as this has been previously demonstrated to yield improved coverage of resultant 95% confidence intervals (CIs) in simulation studies (Bottomley et al., 2016). We ran separate models for each of the tobacco use measures and each of substance use variables. Models were adjusted for age, education, and race/ethnicity. The analytic sample for the models was 240 after excluding 3 individuals who were missing substance use data. All statistical analyses were conducted using R statistical software version 3.5.6 (R Foundation for Statistical Computing, Vienna, Austria).

3. Results

Of the 243 participants, 32.1% were between 45–54 years, 27.2% were White, 39.9% were Black/African American, 15.2% were Latina, and 19.3% were multiracial (Table 1, Supplementary table). Most participants were current smokers (69%), over half were daily smokers (58%). Average number of cigarettes smoked per day was 6.0, 19.0% reported heavy cigarette smoking (≥10 cigarettes/day), and 20.2% had an increase of 10 cigarettes smoked from the previous 30-days. Substances used at baseline included alcohol (47.1%), cannabis (49.6%), crack/cocaine (42.6%), heroin (11.2%), and methamphetamines (21.9%). Among alcohol users in the past 7-days, 54% reported heavy alcohol use. Among current tobacco users, 53% used alcohol, 54% used cannabis, 55% used cocaine, 15% used heroin, and 28% used methamphetamines in the past 7-days.

Table 1:

Participant characteristics at baseline (N=243)

Characteristics	N (% or 95% CI)
Age
< 45 years	58 (23.9)
45–54 years	78 (32.1)
55–64 years	89 (36.6)
>65	18 (7.4)
Education
Less than high school	72 (29.6)
High school or GED	138 (56.8)
Some college	33 (13.6)
Race
Alaskan Native or Native American	16 (6.6)
Asian and Pacific Islander	8 (3.3)
Black	97 (39.9)
White	66 (27.2)
Other	9 (3.7)
Multiracial	47 (19.3)
Latina Ethnicity	37 (15.2)
Tobacco use characteristics
Cigarettes smoked per day	6.0 (4.7 – 7.3)
Heavy tobacco use a	46 (19.0)
Increase in 10 cigarettes smoked from the previous 30-days b	49 (20.2)
Substance use in the past 7 days c
Alcohol use	114 (47.1)
Cannabis use	120 (49.6)
Crack/cocaine use	103 (42.6)
Opioid use	27 (11.2)
Methamphetamine	53 (21.9)

^aDefined as smoking ≥ 10 cigarettes per day versus not

^bProportion who had ≥10 cigarette increases from the previous 30-days among those who had 2 or more consecutive visits after the baseline visit

^cSelf-reported substance use in the past 7-days

In adjusted models, a 5 unit increase in cigarettes smoked per day (Adjusted odds ratio [AOR] 1.02, 95% CI 1.00–1.03), heavy cigarette smoking (AOR 2.02, 95% CI 1.46–2.79), and a 10 unit increase in the cigarettes smoked from the previous 30-days (AOR 1.06, 95% CI 1.01–1.12) were significantly associated with methamphetamine use in the past 7-days. Associations with past 7-days use of alcohol, cannabis, crack/cocaine, and heroin did not reach levels of significance for any of the smoking intensity variables (Table 2).

Table 2:

Association of number of cigarettes smoked per day, heavy cigarette smoking, and increase in cigarettes smoked from the previous 30 days with past 7-days use of alcohol, cannabis, stimulants, amphetamines, and heroin^e (N=240)

	Alcohol d (Adjusted odds ratio, 95% CI)	Cannabis d (Adjusted odds ratio, 95% CI)	Crack/cocaine d (Adjusted odds ratio, 95% CI)	Heroin d (Adjusted odds ratio, 95% CI)	Methamphetamine d (Adjusted odds ratio, 95% CI)
Number of cigarettes smoked per day a	1.00 (0.98,1.02)	0.99 (0.97,1.01)	1.00 (0.99,1.02)	1.02 (0.98,1.05)	1.02 (1.00,1.03) *
Heavy cigarette smoking b	0.97 (0.64,1.48)	0.81 (0.54,1.22)	0.89 (0.55,1.45)	1.44 (0.74,2.78)	2.02 (1.46,2.79) *
Increase in number of cigarettes from the previous month c	1.04 (0.96,1.12)	0.98 (0.91,1.04)	0.99 (0.93,1.05)	1.02 (0.99,1.06)	1.06 (1.01,1.12) *

^aOdds ratios correspond to an increase in 5 cigarettes per day

^bDefined as smoking ≥ 10 cigarettes per day versus not

^cProportion who had ≥10 cigarette increases from the previous 30-days among those who had 2 or more consecutive visits after the baseline visit

^dSelf-reported substance use in the past 7-days

^eModels were adjusted for age, education, and race/ethnicity

^*p<0.05

4. Discussion

Results reported here extend our previous findings by showing that, beyond associations between any use (Riley et al., 2022), number of cigarettes smoked per day, heavy (daily) tobacco use, and an increase in the number of cigarettes smoked from the previous 30-days are all associated with methamphetamine use among unstably housed women. These results argue against the idea that associations between cigarette smoking and other substance use are similar for unstably housed women who smoke at any level. Rather results show that the odds of methamphetamine use are stronger for every additional increase in cigarette smoking intensity, suggesting an especially strong association at the highest levels of use.

Prior research in the general population on approaches to smoking cessation, which may include reduction-to-quit or abrupt quitting, suggests that more research is needed to determine which of these may be the more optimal approach (Lindson et al., 2019; Klemperer et al., 2019). Studies have shown that integrated tobacco and substance use treatment does result in reductions in tobacco use and an increase in the use of tobacco treatment (McCuistian et al., 2022). Results presented here extend this dialog by suggesting that acknowledging the level of tobacco use in interventions for the co-treatment of tobacco and methamphetamine (e.g., content, intensity, or duration of treatment by level of tobacco use at baseline) may improve intervention results. Whether variations in intervention by baseline tobacco use intensity influence intervention effectiveness, and/or whether the effectiveness of co-treatment without variation is modified by baseline tobacco use intensity merits further investigation.

Current treatments for methamphetamine use disorder often rely on cognitive behavioral therapy and contingency management approaches, common also to the treatment of tobacco use (Lee et al., 2008; Rash et al., 2018; Vijayaraghavan et al., 2020). Pilot studies of cognitive behavioral therapy with contingency management to simultaneously address concurrent cigarette and methamphetamine use, with particular attention to individuals with high-intensity tobacco use, are warranted. For example, prior research suggests a trauma-informed approach may increase engagement in tobacco cessation treatment for women experiencing homelessness or housing instability (Rubin et al., 2021), a possibility that may also apply to co-treatment. Prior research also suggests that co-treatment of tobacco and substance use may be more effective than treating tobacco use alone among people experiencing homelessness who use other substances (Vijayaraghavan et al., 2020). Such findings motivated research that led to the recent emergence of evidence-based pharmacotherapies tested in the context of cigarette and stimulant co-use, such as bupropion (Gao et al., 2023), naltrexone (Santos et al., 2016), or bupropion plus naltrexone (Schmitz et al., 2023). In combination, the growing treatment-related evidence base continues to provide more options for tailored interventions for tobacco and substance use co-treatment.

4.1. Conclusions

Our study is limited by a small sample size. However, our findings present a particularly robust association between cigarette smoking intensity and methamphetamine use and highlights the potential utility of integrated interventions among women who experience housing instability.

Highlights.

• Among unstably housed women, 69 % were current smokers and 58 % were daily smokers.

• Cigarette smoking intensity was associated with past 7-days methamphetamine use.

• Integrating cigarette smoking and methamphetamine use treatment for high-frequency cigarette smokers is warranted.