Home environment and cigarette quitting behaviors among rural Black/African American women caregivers

Abstract

Objective:

This cross-sectional study described the home tobacco environment and its association with quitting behaviors among Black/African American women caregivers who smoke cigarettes and live in rural, low-resourced areas.

Methods:

A baseline survey was administered to caregivers enrolled in a randomized trial from 2020–2022 (n=147). Logistic regressions identified the associations between the independent variables (home cigarette smoking bans, caregiver restrictions on child cigarette access, number of people in the home who smoked around the caregiver during the past week, and who smoked in the caregiver’s home) and three outcome variables: lifetime quit attempt, past year quit attempt and use of evidence-based cessation strategies during a last quit attempt.

Results:

Caregivers have multiple generations of family smoking in their home, including caregivers’ children/nieces/nephews (21%) and their parents (36%). Young family members smoking in the home was related to the caregiver’s parents (p = .046) and grandparents (p = .03) smoking in the home. The number of people smoking around the caregiver was associated with lower odds of a lifetime quit attempt (OR=0.63, CI: 0.47, 0.85; AOR=0.61, CI: 0.45, 0.84). No independent variables were significantly related to past year quit attempts in unadjusted or adjusted models. Caregivers with young family members smoking in the home were more likely to have used evidence-based cessation strategies versus those without young family smoking (OR= 16.96, CI: 1.01, 283.68).

Conclusions.

Black/African women caregivers who smoke and live in rural, low-resourced areas are exposed to numerous family members smoking in their homes which may affect quitting.

Public Health Significance:

This research advances our understanding of how home environments impact cigarette quitting behaviors among Black/African American women caregivers who smoke and live in low-resource, rural areas. Our findings indicate that multiple generations are smoking within caregivers’ homes and that both the quantity and source of home-based caregiver tobacco exposure relates to caregiver quitting behaviors. Understanding Black/ African American rural caregivers’ home environment may improve the success of tobacco interventions.

Introduction

Black/ African American people suffer disproportionately from tobacco-caused diseases (American Association for Cancer Research, 2022; U.S. Department of Health and Human Services, 1998), and smoking may confer greater health risks for women (Allen et al., 2014; Mucha et al., 2006). Black/ African American women have the highest rate of cardiovascular disease mortality (Mehta et al., 2023) and the second highest rate of overall cancer mortality compared to women from other racialized groups (Islami et al., 2023). Cardiovascular and cancer disparities are also found among those with lower socioeconomic status and living in rural areas (Carnethon et al., 2017; Islami et al., 2023). Social determinants of health and smoking contribute to health disparities experienced by Black/ African American women living in rural, low resource areas (American Association for Cancer Research, 2022; Islami et al., 2023; Mehta et al., 2023; Mucha et al., 2006).

Parent or caregiver (henceforth referred to as caregiver) commercial tobacco smoking increases the risk for child environmental tobacco smoke exposure (ETSe), smoking initiation, and health conditions such as asthma, ear infections, wheezing and coughing, and acute respiratory illnesses in children (Leonardi-Bee et al., 2011; U.S. Department of Health and Human Services, 2014). Children who are Black/African American or live in poverty are more likely to be exposed to ETS in the home compared to other racialized groups and people who live at or above the poverty level (Tsai et al., 2018). Child ETSe is also higher in rural versus urban areas (Mantey et al., 2021; Mbulo et al., 2016).

To reduce family-level health disparities and child ETSe disparities in the homes of Black/African American women who live in rural and low resource communities, there is an urgent need to improve rates of smoking cessation among caregivers. Despite the intergenerational risk, little research has explored disparities in caregiver quitting behaviors. One study found that Black caregivers who presented with their child to an emergency room were more likely to make a past year quit attempt (75%) relative to white caregivers (65%); information on rurality was not presented (Mahabee-Gittens et al., 2019). Even fewer studies have used an intersectionality approach to examine how gender-race-socioeconomic-geographic factors influence caregiver cessation behaviors. Our prior study found that only 54% of Black/African American women caregivers who currently smoked and resided in rural, low-resource communities had a lifetime cigarette quit attempt and that only 55% of those with a lifetime quit attempt had a past year quit attempt (Jones et al., 2022). This work extends prior research documenting quitting-related disparities among Black adults, women, and those with lower socioeconomic status and living in rural areas: 1) Black adults are less likely to successfully quit smoking (5% vs. 7.7% for white adults) (Leventhal et al., 2022); 2) quit attempts and cessation are lower among those with less education and those living in rural areas, and lower income is associated with less cessation (Leventhal et al., 2022); 3) quit ratios (i.e., the percentage of ever smokers who quit) are lower among Black adults, those with lower SES, those living in the South and Midwest, and those who are uninsured (US Department of Health and Human Services, 2020); and 4) long-term abstinence may be less likely for women (Smith et al., 2016).

The lack of access to, quality of, and use of evidence-based strategies (EBS) for cessation may be contributing to cessation disparities. There is limited information on the use of EBS for quitting among Black/ African American women caregivers living in low resource, rural areas (US Department of Health and Human Services, 2020). However, a national study conducted among adults who smoke and made a past year quit attempt found that only 34% used any EBS, and Black adults (30.7%) were less likely to use EBS compared to white adults (36.9%) (Leventhal et al., 2022). The home environment may impact both child ETSe (e.g., number of people smoking in the home) and quitting behaviors among caregivers. Yet, little is known about the factors in the home environment that influence quitting among this group of women.

Multilevel influences (such as poverty, lack of cessation resources, and exposure to tobacco advertising) also perpetuate home environments with multiple generations of people who smoke, thereby reducing household members’ cessation success (Alvidrez et al., 2019; Chandola et al., 2004; Marbin et al., 2021; Talbot et al., 2019; Twyman et al., 2014). ETSe and living with people who smoke, which have been conceptualized as nicotine cues (Zhou et al., 2009), are associated with lower odds of cessation among the general population (Chandola et al., 2004; Lee & Kahende, 2007; Okoli & Kodet, 2015), women (Holahan et al., 2012), and caregivers (Nabi-Burza et al., 2021). Being around other people smoking is also associated with relapse among the general population (Zhou et al., 2009). Among children with caregivers who smoke, exposure to additional people smoking in their home further increases their risk for ETSe and smoking initiation (Leonardi-Bee et al., 2011).

On the other hand, adult quitting behaviors are also motivated by protecting children from tobacco exposure, especially among women (Vangeli & West, 2008). Findings from qualitative research suggest that the desire to prevent children from smoking is a motivator for quitting smoking and implementing home smoking bans among Black/ African American caregivers who smoke (Hoehn et al., 2016), including rural caregivers (Butler et al., 2009). In turn, home smoking bans are related to more quit attempts and cessation among the general population (Hyland et al., 2009; Kahende et al., 2011), socially disadvantaged individuals who called 211 (Haardörfer et al., 2018), and quitline callers (Yuan et al., 2019). Home smoking bans are also related to more quit attempts among Black/African American adults (Kahende et al., 2011) and caregivers (Nabi-Burza et al., 2021). Importantly, one study also found that home smoking bans were related to use of tobacco cessation medications among quitline callers (Yuan et al., 2019). Except for research on how partner’s smoking influences women’s smoking during and shortly after pregnancy (Scheffers-Van Schayck et al., 2019), there has been little attention paid to who is smoking around caregivers and how that affects quitting. Further, little is known about the familial relationships between the people who smoke in caregivers’ homes, information that could elucidate generational tobacco use clusters.

Overall, there is limited data on how the home tobacco environment relates to cigarette quitting behaviors among Black/African American women caregivers who live in low-resource, rural areas. This is a critical gap given the greater exposure to home tobacco risk factors among families residing in rural communities (Talbot et al., 2019) and findings from our prior studies that offer insights into how the home tobacco environment influences quitting behaviors among this high priority group (Jones et al., 2021, 2022). Specifically, we found that 37% of caregivers lived with others who smoked in the home and only 26% had a comprehensive home smoking ban (Jones et al., 2022). Importantly, caregivers with comprehensive home smoking bans were more likely to intend to quit cigarette smoking in the next 30 days compared to those with partial or no bans (31% vs. 11.4%-13.3%); however, no association was found between home smoking ban status and lifetime or past year cigarette quit attempts.

Purpose of the Current Study

The present study describes the home tobacco environment and examines the relationships between the home tobacco environment and three cigarette quitting outcomes 1) lifetime quit history, 2) past year quit history, and 3) EBS use during a most recent quit attempt among Black/African American women caregivers living in rural, low-resource areas. First, we describe the home tobacco environment (i.e., the number of people in the home who smoke in caregiver’s presence, home smoking ban status, and caregiver restrictions on child’s access to cigarettes in the home), including who is smoking in caregivers’ homes (e.g., the familial relationship between the caregiver and others who smoke in the home). Second, we examined the familial relationships among others who smoke in the caregiver’s home to assess intergenerational clusters of tobacco use in the home, hypothesizing that there would be evidence of overall familial tobacco use clusters and specific intergenerational clusters of tobacco use. Lastly, we examined the relationships between the home tobacco environment and the three cigarette quitting outcomes. We hypothesized that the number of people in home who smoked in caregiver’s presence during the past week and having partners, parents, grandparents, aunts/uncles, siblings, and/or cousins smoke in the home are associated with lower odds of quit attempts and EBS use. We hypothesized that cigarette home smoking bans and caregiver restrictions on child access to cigarettes in the home are associated with higher odds of quit attempts and EBS use. This study will fill critical gaps in the literature by 1) describing who is smoking in caregivers homes, 2) elucidating whether the quantity of tobacco exposure in the home (e.g., number of people who smoke) and/or the source of tobacco exposure in the home lower the odds of quitting behaviors, and 3) investigating novel, potential facilitators, including caregiver restrictions on child access to cigarettes, for quitting behaviors among multiply marginalized caregivers.

Methods

This article includes descriptions of the sample, data exclusions and manipulations, and measures used in this study. The APA Reporting Standards for quantitative research were followed. The CONSORT diagram and analysis code are provided in the Online Supplement (A–B). Surveys are available by emailing the corresponding author. This trial was pre-registered on clinicaltrials.gov (NCT03476837).

Participants and Procedures

Participants who enrolled in Families Rising to Enforce Smokefree Homes (F.R.E.S.H.) study between 2020–2022 completed informed consent and the baseline survey prior to randomization. This randomized trial compared the influence of motivational counseling and culturally-relevant self-help materials on the implementation of comprehensive smoke free policies in the homes of Black/African American women caregivers who smoke cigarettes and/or little cigars/cigarillos and reside in Lee and Philips counties in rural Arkansas (N=188) (Jones et al., 2021). Lee and Phillips counties have the poorest health outcomes in the state and high rates of smoking (Lee: 25%; Phillips: 26%). A large proportion of children live in poverty (Lee: 45%; Phillips: 37%), and life expectancy is among the lowest in the state (Lee: 71.6 years; Phillips: 69.5 years) (University of Wisconsin Population Health Institute, 2022b, 2022a). This study was approved by the University of Arkansas for Medical Sciences Institutional Review Board (#207306, approved on 11/22/17).

Non-probability sampling techniques for recruiting historically marginalized populations were utilized (Bonevski et al., 2014). Research staff and community health workers collaborated with community organizations to host recruitment events and advertise the study (word of mouth; flyers; media). Inclusion criteria were: self-identified as a Black/African American woman; was 18–50 years old; resided in Lee or Phillips counties; spoke English; provided consent; had a working phone, home address, and email; was the primary caregiver (birth parent/ legal guardian) to a child aged 6 months-14 years old; was the primary decision maker in the home; endorsed ≥ one low income indicator (see Jones et al. 2021); and smoked cigarettes and/or little cigars/cigarillos for ≥1 year and in the past 30 days (Jones et al., 2021, 2022). Women with carbon monoxide < 5 ppm were excluded. Additional details on the sample and procedures have been published (Jones et al., 2021, 2022).

Measures

Dependent Variables

Lifetime Quit Attempt History.

Caregivers reported whether they had ever tried to quit smoking cigarettes completely (yes/no). This question is used in the Tobacco Use Supplement to the Current Population Survey (National Cancer Institute & Food and Drug Administration, 2023).

Past Year Quit Attempt History.

Caregivers who reported a lifetime cigarette smoking quit attempt were asked, “During the past 12 months, have you stopped cigarette smoking for one day or longer because you were trying to quit smoking?” (yes/no). This question is used in the Population Assessment of Tobacco and Health Study (Hyland et al., 2017; United States Department of Health and Human Services, 2022).

Use of EBS During Last Past Year Quit Attempt.

Caregivers who reported a past year cigarette smoking quit attempt were also asked whether they used different quit strategies in their most recent past year quit attempt (Online Supplement C details the quit strategies assessed). The questions assessing quit strategies were based on off questions used in the Population Assessment of Tobacco and Health Study and Tobacco Use Supplement to the Current Population Survey (National Cancer Institute & Food and Drug Administration, 2023; United States Department of Health and Human Services, 2022). We used the 2020 Surgeon General Report on Smoking Cessation (US Department of Health and Human Services, 2020), tobacco dependence clinical practice guidelines (Fiore MC et al., 2008), and the extant literature to identify the EBS quit strategies (Hooper et al., 2014; Livingstone-Banks et al., 2019; Orleans et al., 1998; Robinson et al., 2003; Webb Hooper et al., 2018), including: nicotine replacement therapy, prescription medication, quitline, cessation internet program, individual counseling, cessation clinic/class/support group, friend/family support, cessation books/pamphlets/videos, and gradually cutting down. Gradually cutting down on cigarette/nicotine consumption is recommended for those not ready to quit (US Department of Health and Human Services, 2020). Though there is heterogeneity across research examining varying types of social support, social support for quitting from family/friends was conceptualized as an EBS because 1) rural women with low SES prefer social support from their social network for cessation support (Mitchell et al., 2016); 2) social support for quitting from family, friends, and peers is associated with increased odds of smoking cessation among Black adults (U.S. National Cancer Institute, 2017), women (Westmaas et al., 2022), and caregivers (Tooley et al., 2015); and 3) social support is encouraged in public health initiatives (e.g., Smokefree.gov). We created a composite EBS quit strategy variable by summing the number of EBS participants reported using (range: 0–3). Due to the variable distribution, we dichotomized the EBS quit strategy variable such to use of ≥1 EBS vs. use of 0 EBS.

Independent Variables: Home Tobacco Environment

Number of People who Smoked around Caregiver.

The survey measured the number of people in the home that smoked in the presence of a caregiver during the past week (range: 0–12).

Family Members who Smoke in the Home.

Caregivers were asked if the following people currently smoked cigarettes or cigars in their home: Partner (spouse/partner/boyfriend/girlfriend); parent/stepparent; grandparent; sibling; aunt/uncle; cousin; and child/niece/nephew/grandchild (yes/no). We created a composite variable based on the number of categories that were endorsed by participants (range: 0–7).

Cigarette Home Smoking Ban.

Rules about cigarette smoking inside caregivers’ homes were assessed (excluding porches, decks, and garages); responses included, “No one is allowed to smoke cigarettes anywhere inside the home” (full ban), “Cigarette smoking is allowed in some places or at some times” (partial ban), and “Cigarette smoking is permitted anywhere inside my home” (no ban). We dichotomized response options to full cigarette home smoking ban vs. partial/no ban.

Caregiver Restrictions on Child Access to Cigarettes in the Home.

Caregivers were asked if they kept cigarettes in places where children cannot reach and responses included “Yes”, “No”, and “I do not have cigarettes in my home.” We dichotomized the responses to: cigarettes are not in the home or children do not have access to them vs. cigarettes are in the home and children have access to them.

Covariates

Individual and Family Sociodemographic Characteristics.

Sociodemographic measures included age, race (Black/African American only or biracial/multiracial), highest level of education, annual household income, perceived financial situation, employment status, sexual orientation, marital status, ages of children living in the home, and insurance status.

Health Care.

We measured access to a regular primary care provider (yes/no) and if a healthcare provider had talked with them about their smoking during the past year (yes/no).

Caregiver Tobacco Use.

Separate questions assessed how often caregivers now used the following tobacco products: cigarettes, regular/large cigars, cigarillos/little cigars, electronic cigarettes (e-cigarettes), IQOS, hookah, pipes, and smokeless tobacco and response options were “not at all,” “some days,” or “every day.” For cigarettes, regular/large cigars, cigarillos/little cigars, and e-cigarettes, this question was only asked of those who endorsed a lifetime history of using the product. We created a composite “other tobacco use” variable identifying those who currently used IQOS, hookah, pipes, or smokeless tobacco on some days/every day vs. those who endorsed “not at all” for all products. Current cigarette smoking was defined as those who had smoked at least 100 cigarettes in their lifetime and currently smoked on “some days” or “every day.”

Past month use of mentholated cigarettes was measured using the responses: only smoked menthol; mostly menthol; half and half; mostly non-menthol; and only non-menthol. Due to variable distribution, a binary variable was created (only smoked menthol cigarettes (1) vs. no use/ some use of menthol cigarettes (0)). The Fagerstrom Test for Nicotine Dependence 6-item scale was used to measure nicotine dependence with greater scores reflecting greater cigarette dependence (scores 1 to 10) (Heatherton et al., 1991).

Data Analysis

The analytic sample for this study included 147 caregivers who currently smoked cigarettes (see the CONSORT diagram in Online Supplement A for more information on the larger sample). Sample sizes varied across quitting outcomes based on the subsamples who completed the relevant questions and missing data (see details in Online Supplement C; n=145 for lifetime quit history, n=76 for past year quit history, n=42 for use of EBS during last past year quit attempt). Given the small sample size for EBS use, analyses examining this outcome were exploratory and conducted to provide preliminary data to inform future research. All data analyses were performed in SPSS v29.

Preliminary Analyses and Model Building Process

First, descriptive and bivariate (ANOVA, Pearson correlation, chi-square, Fisher’s exact, and Fisher-Freeman-Haltman tests) analyses were used to identify covariates for the subsequent adjusted models and to identify the bivariate associations between home tobacco environment factors and the three quitting outcomes (lifetime and past year cigarette quit attempts and use of EBS during their most recent past year quit attempt). Bivariate analyses and prior literature were used to identify covariates (Hosmer & Lemeshow, 2000). A p value of 0.1 rather than .05 was used for covariate selection to allow for the inclusion of potentially important covariates; a p value of .25 was not used to avoid model overfitting, especially given our sample size (Hosmer & Lemeshow, 2000). Based on prior literature, all adjusted models controlled for cigarette dependence (FTND) and marital status. Table 1 describes the bivariate associations between sociodemographic, healthcare, and caregiver tobacco use characteristics by quitting outcomes. Online Supplements D describes how these results informed the model building process. E-cigarette use was not included as a covariate in adjusted models because of low n’s and cell counts of zero. Because the ordinal perceived financial situation had a cell with a count of zero when cross-tabulated with past year quit attempt history (see Table 1), a binary perceived financial situation variable (just met basic expenses/do not meet basic expenses vs. live comfortably/meets needs with some left over) was created and used in all adjusted models. Based on these analyses, the adjusted models examining lifetime quit attempts and past year quit attempts controlled for FTND, marital status, caregiver age, cigarette use frequency, past month menthol use, binary perceived financial situation, caregiver past year discussion with healthcare provider about tobacco, and ages of children in the home. A p value of 0.05 was used to determine statistical significance for all other analyses.

Table 1.

Sociodemographic, healthcare, and tobacco use characteristics by quit attempt histories and use of evidence based smoking cessation strategies.

Caregivers who Smoke Cigarettes on Some Days/Every Day (n= 147)	Overall	Lifetime Quit Attempt (n =145) ‡		Past Year Quit Attempt (n=76) ‡		Used Evidence-Based Cessation Strategy a (n=42) ‡
		No	Yes	No	Yes	No	Yes
Individual and Family Sociodemographic Variables
Age in years (Mean ± SD)	34.6 (8.7)	32.5 (8.0) **	36.4 (9.0)	38.2 (8.3) †	34.6 (9.3)	34.8 (9.4)	34.3 (9.4)
Race
Black/ African American only	139 (95.9)	61 (44.2) b, †	77 (55.8)	33 (44.0) b	42 (56.0)	23 (54.8) c	19 (45.2)
Biracial/ Multiracial	6 (4.1)	5 (83.3)	1 (16.7)	1 (100.0)	0(0)	0 (0)	(0)
Highest level of education
Less than high school	44 (30.1)	21 (47.7)	23 (52.3)	8 (34.8)	15 (65.2)	8 (53.3)	7 (46.7)
High school or greater	102 (69.9)	46 (45.5)	55 (54.5)	26 (49.1)	27 (50.9)	15 (55.6)	12 (44.4)
Annual household income
<$10,000	96 (65.8)	47 (49.5)	48 (50.5)	20 (43.5)	26 (56.5)	13 (50.0)	13 (50.0)
≥$10,000	50 (34.2)	20 (40.0)	30 (60.0)	14 (46.7)	16 (53.3)	10 (62.5)	6 (37.5)
Perceived financial situation
Finances do not meet basic expenses	19 (13.0)	9 (50.0)	9 (50.0)	0 (0.0) b, *	9 (100.0)	5 (55.6) b	4 (44.4)
Finances just meet basic expenses	37 (25.3)	19 (51.4)	18 (48.6)	10 (55.6)	8 (44.4)	3 (37.5)	5 (62.5)
Finances meet needs with a little left over	37 (25.3)	13 (35.1)	24 (64.9)	12 (52.2)	11 (47.8)	7 (63.6)	4 (36.4)
Live comfortably	53 (36.3)	26 (49.1)	27 (50.9)	12 (46.2)	14 (53.8)	8 (57.1)	6 (42.9)
Employment Status
Not currently working for pay	96 (65.8)	40 (42.1)	55 (57.9)	25 (47.2) b	28 (52.8)	14 (50.0) b	14 (50.0)
Part-time	22 (15.1)	14 (63.6)	8 (36.4)	2 (25.0)	6 (75.0)	5 (83.3)	1 (16.7)
Full-time	28 (19.2)	13 (46.4)	15 (53.6)	7 (46.7)	8 (53.3)	4 (50.0)	4 (50.0)
Sexual Orientation
Heterosexual/ straight	107 (77.5)	44 (41.1)	63 (58.9)	30 (48.4)	32 (51.6)	19 (59.4) b	13 (40.6)
Gay/Lesbian/Bisexual/Other	31 (22.5)	17 (54.8)	14 (45.2)	4 (30.8)	9 (69.2)	3 (33.3)	6 (66.7)
Marital Status
Not married or living with partner	84 (57.9)	40 (48.2)	43 (51.8)	20 (47.6)	22 (52.4)	11 (50.0)	11 (50.0)
Married or living with partner	61 (42.1)	26 (42.6)	35 (57.4)	14 (41.2)	20 (58.8)	12 (60.0)	8 (40.0)
Ages of children in the home (Checked Yes)
0–5 yrs.	76 (51.7)	35 (46.1)	41 (53.9)	14 (35.0) †	26 (65.0)	14 (53.8)	12 (46.2)
6–14 yrs.	102 (69.4)	47 (46.5)	54 (53.5)	27 (50.9) †	26 (49.1)	14 (53.8)	12 (46.2)
15+ yrs.	14 (9.5)	6 (42.9)	8 (57.1)	3 (37.5) b	5 (62.5)	5 (100.0) b, †	0 (0)
Healthcare Variables
Insurance Status
Uninsured	31 (21.5)	15 (50.0)	15 (50.0)	6 (42.9)	8 (57.1)	5 (62.5) b	3 (37.5)
Insured	113 (78.5)	51 (45.1)	62 (54.9)	28 (45.9)	33 (54.1)	18 (54.5)	15 (45.5)
Regular Primary Care Provider
No	21 (14.4)	11 (55.5)	9 (45.0)	5 (55.6) b	4 (44.4)	2 (50.0) b	2 (50.0)
Yes	125 (85.6)	56 (44.8)	69 (55.2)	29 (43.3)	38 (56.7)	21 (55.3)	17 (44.7)
Healthcare provider has talked with participant about smoking in the past year
No	97 (66.4)	43 (44.8)	53 (55.2)	23 (45.1)	28 (54.9)	18 (64.3) †	10 (35.7)
Yes	49 (33.6)	24 (49.0)	25 (51.0)	11 (44.0)	14 (56.0)	5 (35.7)	9 (64.3)
Caregivers Tobacco Use Variables	Overall	Lifetime Quit Attempt (n =145) ‡		Past Year Quit Attempt (n=76) ‡		Used Evidence-Based Cessation Strategy a (n=42) ‡
		No	Yes	No	Yes	No	Yes
Cigarette Use
Some days	32 (21.8)	12 (37.5)	20 (62.5)	4 (21.1) *	15 (78.9)	9 (60.0)	6 (40.0)
Every day	115 (78.2)	55 (48.7)	58 (51.3)	30 (52.6)	27 (47.4)	14 (51.9)	13 (48.1)
Regular Cigar Use d
Not at all	12 (32.4)	5 (41.7) b	7 (58.3)	3 (50.0) b	3 (50.0)	0 (0.0) b	3 (100.0)
Some days	19 (51.4)	10 (52.6)	9 (47.4)	4 (44.4)	5 (55.6)	2 (40.0)	3 (60.0)
Every day	6 (16.2)	2 (33.3)	4 (66.7)	1 (25.0)	3 (75.0)	1 (33.3)	2 (66.7)
Cigarillo/Little Cigar Use d
Not at all	21 (36.8)	8 (38.1) b	13 (61.9)	5 (38.5) b	8 (61.5)	1 (12.5) b	7 (87.5)
Some days	28 (49.1)	16 (57.1)	12 (42.9)	5 (50.0)	5 (50.0)	2 (40.0)	3 (60.0)
Every day	8 (14.0)	2 (25.0)	6 (75.0)	2 (33.3)	4 (66.7)	3 (75.0)	1 (25.0)
E-Cigarette Use d
Not at all	10 (52.6)	8 (80.0) b, *	2 (20.0)	1 (50.0) b	1 (50.0)	0 (0.0) b	1 (100.0)
Some days	8 (42.1)	2 (25.0)	6 (75.0)	2 (40.0)	3 (60.0)	2 (66.7)	1 (33.3)
Every day	1 (5.3)	0 (0.0)	1 (100.0)	0 (0.0)	1 (100.0)	1 (100.0)	0 (0.0)
Use of Other Tobacco Product on Some or Every Day e
No	138 (94.5)	63 (46.0) b	74 (54.0)	33 (45.8) b	39 (54.2)	22 (56.4) b	17 (43.6)
Yes	8 (5.5)	4 (50.0)	4 (50.0)	1 (25.0)	3 (75.0)	1 (33.3)	2 (66.7)
Fagerstrom Test for Nicotine Dependence (M (SD))	4.6 (2.1)	4.9 (1.8)	4.4 (2.4)	4.9 (2.1)	4.1 (2.6)	3.9 (2.5)	4.4 (2.9)
Past month menthol cigarette use
No use or some use of menthol cigarettes	29 (19.7)	9 (31.0)	20 (69.0)	9 (47.4)	10 (52.6)	4 (40.0) b	6 (60.0)
Only smoked menthol cigarettes	118 (80.3)	58 (50.0)	58 (50.0)	25 (43.9)	32 (56.1)	19 (59.4)	13 (40.6)

Notes: Unless otherwise noted, the numbers represent “n (%)”.

^*p < .05

^**p < .01

^***p < .001.

^†p < .10.

^‡Valid percents presented.

Boldface indicates a statistically significant association. Sample sizes varied across analyses due to missing data. One-way ANOVAs were used for age and Fagerstrom Test for Nicotine Dependence (M (SD)); Chi-square tests were used for other variables unless noted otherwise.

^aComposite variable identifying use of evidence-based cessation methods that included nicotine replacement therapy, medication (Chantix/varenicline, Zyban/bupropion/wellbutrin, other), cessation quitline, cessation internet program, individual counseling, cessation clinic/class/support group, friend/family support, cessation books, pamphlets, videos, gradually cutting down.

^bFisher’s or Fisher-Freeman-Haltman exact test; 2-sided exact p-value used.

^cNo test of association was computed.

^dOnly people who a lifetime history of use answered this question.

^eCurrent use of IQOS, hookah, pipes, or smokeless tobacco; all participants answered these questions.

Multivariable Analyses

Next, we conducted separate unadjusted and multivariable logistic regression models to examine the relationships between the independent variables (number of people in home who smoke in caregiver’s presence, who smokes in the home, cigarette home smoking ban status, and caregiver restrictions on child’s access to cigarettes in the home) and each quitting outcome: lifetime quit attempt (Model Set 1), past year quit attempt (Model Set 2), and EBS use (Model 3- Unadjusted Model Only). We did not report the adjusted model results examining EBS use due to the small sample size (n=40) and problems with model convergence with the fully adjusted model. As previously noted, EBS findings were exploratory and conducted to provide preliminary data. The SWEEP algorithm, a collinearity diagnostic, was used to identify collinearity between the independent variables; based on this, the composite sum of the number of family member types who smoked in the home was removed from models.

Results

Sociodemographic Characteristics

The average caregiver age was 35 years and most identified as Black/African American only (96%). Most caregivers reported an annual household income < $10,000 (66%), most identified as heterosexual/straight (78%), and more than half were not married or living with a partner (58%; Table 1). Almost 80% smoked cigarettes daily and 80% only smoked menthol cigarettes during the past month.

Caregiver Tobacco Use

Overall, 54% of caregivers had a lifetime history of trying to quit smoking (Online Supplement C). Of those who ever tried to quit, 55% had tried to quit in the past year. Among those with a past year quit attempt, 42.9% had tried to quit once during the past year, 50.0% had tried to quit 2–3 times, and 7.1% had tried to quit ≥4 times. Among those with a past year quit attempt, only 45% used an EBS during their last quit attempt, while 71% tried to quit smoking all at once. Of those who tried to quit all at once, 40% also used an EBS during their last quit attempt, while 60% only tried to quit all at once. See Online Supplement C for details on the individual EBS used.

Table 1 displays the bivariate associations between sociodemographic, healthcare, and caregiver tobacco use characteristics by quitting outcomes; Online Supplements D describes how these results informed the model building process. Results showed that being older (p=.007) and using e-cigarettes (p =.03) were related lifetime quit history. Smoking cigarettes on some days (versus daily; p =.02) and having finances that did not meet basic expenses (versus those with better financial situations, p=.02) were related having a past year quit attempt. There were no significant associations between sociodemographic, healthcare, and caregiver tobacco use characteristics by EBS use.

Characterization of the Home Tobacco Environment

Overall, caregivers reported that on average about two people in their home had smoked around them during the past week (M = 1.8 (SD = 2.1), range: 0–12; Table 2) and 21% reported that a young family member (e.g., child, grandchild, niece, or nephew) smoked cigarettes or cigars in the home. Additionally, about 54% reported that their partners smoked in the home. Caregivers reported that other family members also smoked in their home, including their parents (36%), grandparents (15%), siblings (50%), aunts/uncles (53%), and cousins (61%). There was a positive correlation between the number of people who smoked around the caregiver in the past week and the number of family member types who currently smoked in the caregiver’s home (r=.28, p=.001, [95% CI: .12, .43]).

Table 2.

Factors associated with home tobacco environment by caregiver quit attempt histories and use of evidence based smoking cessation strategies.

Caregivers who Smoke Cigarettes on Some Days/Every Day (n= 147)		Lifetime Quit Attempt (n =145) ‡		Past Year Quit Attempt (n=76) ‡		Used Evidence-Based Cessation Strategy a (n=42) ‡
	Overall	No	Yes	No	Yes	No	Yes
Number of people in home who smoked in caregiver’s presence in past week (M (SD))	1.8 (2.1)	2.5 (2.6) ***	1.2 (1.2)	1.2 (1.3)	1.2 (1.2)	1.1 (1.0)	1.3 (1.4)
Partner smokes cigarettes or cigars in home b
No	66 (45.8)	31 (47.7)	34 (52.3)	17 (51.5)	16 (48.5)	11 (68.8)	5 (31.3)
Yes	78 (54.2)	34 (43.6)	44 (56.4)	17 (39.5)	26 (60.5)	12 (46.2)	14 (53.8)
Parent smokes cigarettes or cigars in home c
No	90 (63.8)	36 (40.0)	54 (60.0)	23 (43.4)	30 (56.6)	17 (56.7)	13 (43.3)
Yes	51 (36.2)	26 (52.0)	24 (48.0)	11 (47.8)	12 (52.2)	6 (50.0)	6 (50.0)
Grandparent smokes cigarettes or cigars in home
No	116 (84.7)	50 (43.5)	65 (56.5)	29 (46.0) e	34 (54.0)	19 (55.9) e	15 (44.1)
Yes	21 (15.3)	12 (57.1)	9 (42.9)	4 (44.4)	5 (55.6)	3 (60.0)	2 (40.0)
Aunt/uncle smokes cigarettes or cigars in home
No	67 (47.5)	28 (41.8)	39 (58.2)	16 (42.1)	22 (57.9)	12 (54.5)	10 (45.5)
Yes	74 (52.5)	36 (49.3)	37 (50.7)	18 (50.0)	18 (50.0)	11 (61.1)	7 (38.9)
Sibling smokes cigarettes or cigars in home
No	72 (50.0)	28 (38.9)	44 (61.1)	16 (38.1)	26 (61.9)	17 (65.4) †	9 (34.6)
Yes	72 (50.0)	37 (52.1)	34 (47.9)	18 (52.9)	16 (47.1)	6 (37.5)	10 (62.5)
Cousin smokes cigarettes or cigars in home
No	56 (39.2)	22 (39.3)	34 (60.7)	13 (39.4)	20 (60.6)	12 (60.0)	8 (40.0)
Yes	87 (60.8)	42 (48.8)	44 (51.2)	21 (48.8)	22 (51.2)	11 (50.0)	11 (50.0)
Young family member smokes cigarettes or cigars in home d
No	109 (79.0)	47 (43.5)	61 (56.5)	25 (42.4)	34 (57.6)	22 (64.7) e, *	12 (35.3)
Yes	29 (21.0)	16 (55.2)	13 (44.8)	6 (46.2)	7 (53.8)	1 (14.3)	6 (85.7)
Sum of types of family members (i.e., relationship to caregiver) who smoke in caregiver’s home	2.8 (2.1)	3.1 (2.2)	2.6 (1.9)	2.7 (1.9)	2.5 (1.9)	2.3 (2.1)	2.9 (1.8)
Cigarette Home Smoking Ban
Partial ban/No Ban	100 (68.5)	43 (43.4)	56 (56.6)	28 (51.9) †	26 (48.1)	12 (46.2)	14 (53.8)
Full ban	46 (31.5)	24 (52.2)	22 (47.8)	6 (27.3)	16 (72.7)	11 (68.8)	5 (31.3)
Caregiver Restrictions on Child Access to Cigarettes in the Home
Cigarettes are kept in the home and child has access to them	31 (21.2)	16 (53.3)	14 (46.7)	8 (61.5)	5 (38.5)	3 (60.0) e	2 (40.0)
Cigarettes are not kept in the home or child does not have access to them	115 (78.8)	51 (44.3)	64 (55.7)	26 (41.3)	37 (58.7)	20 (54.1)	17 (45.9)

^*p < .05.

^**p < .01.

^***p < .001.

^†p < .10.

^‡Valid percents presented. Boldface indicates statistically significant difference. Sample sizes varied across analyses due to missing data. One-way ANOVAs were used for number of people in home who smoked in caregiver’s presence in past week (range: 0–12) and number of variables endorsed for which family members smoke in home (range: 0–7); Chi-square tests were used for other variables unless noted otherwise.

^aComposite variable identifying use of evidence-based cessation methods that included nicotine replacement therapy, medication, cessation quitline, cessation internet program, individual counseling, cessation clinic/class/support group, friend/family support, cessation books, pamphlets, videos, gradually cutting down.

^bSpouse/boyfriend/girlfriend.

^cParent/step-parent.

^dChild/ grandchild/niece/nephew.

^eFisher’s or Fisher-Freeman-Haltman exact test; 2-sided exact p-value used.

Tobacco-related protections in the home were also reported (Table 2): 32% of caregivers had a full cigarette smoking ban in the home and 79% had restrictions in place to keep cigarettes inaccessible to children in the home (i.e., did not keep cigarettes in the home or if they did, they kept inaccessible to children in the home). Having a home cigarette smoking ban and having child cigarette access restrictions were not significantly related (x ²(1, n=146)= .11, p = .74).

Home Tobacco Environment Factors by Lifetime Quit Attempts, Past Year Quit Attempts, and EBS Use

Bivariate associations of caregivers’ home tobacco environment by quitting outcomes are presented in Table 2. There was a significant difference in the number of people in the home who smoked in the caregiver’s presence in the past week by lifetime quit attempts. On average, caregivers with a lifetime history of a quit attempt reported fewer people in their home who smoked in their presence in the past week than caregiver’s without a lifetime history of a quit attempt (M = 1.2 (SD = 1.2) vs. M = 2.5 (SD = 2.6), F(2, 144)= 14.5, p < .001). No significant associations were found with other measures of the home tobacco environment by lifetime or past year quit attempts. Although not significantly different, caregivers with no or partial home cigarette smoking bans were less likely (48.1% vs.72.7%, p = .051) to have had made a past year quit attempt compared to caregivers with full cigarette smoking bans. There was a significant difference in the presence of young family members who smoked in the home by caregiver EBS use. Caregivers who reported that a young family member smoked cigarettes or cigars in the home were more likely (85.7% vs. 35.3%, p = .03) to have used an EBS during their most recent quit attempt compared to caregivers who did not report a young family member smoked in the home. No associations were found with other home environment factors by EBS use.

Familial Relationships among Others who Smoke in Caregivers’ Homes

Table 3 shows the associations between who smokes in the caregiver’s home. There were numerous significant associations between which family members smoked in the home, highlighting intergenerational and intragenerational smoking clusters. Intergenerational clusters were conceptualized as associations between smoking among family members from an older generation and a younger generation (e.g., associations between caregiver smoking and their parents smoking). Intragenerational clusters were conceptualized as associations between smoking among family members from similar generations (e.g., associations between smoking among siblings). For example, young family members smoking in the caregiver’s home was associated with older generations smoking in the home including, the caregiver’s parents (p = .046), grandparents (p = .03), and siblings smoking in the home (p = .01). Caregivers whose parents smoked in the home were more likely to have grandparents (p <.001), aunts/uncles (p <.001), siblings (p <.001), and cousins (p <.001) who also smoked in the home (Table 3).

Table 3.

Associations between caregivers’ family members smoking in the caregiver’s home.

Caregivers who Smoke Cigarettes on Some Days/Every Day (n= 147)
	Who Smokes Cigarettes or Cigars in Caregiver’s Home
Who Smokes Cigarettes or Cigars in Caregiver’s Home	Partner Smokes in Home a		Parent Smokes in Home b		Grandparent Smokes in Home		Aunt/uncle Smokes in Home		Sibling Smokes in Home		Cousin Smokes in Home		Younger Family Member Smokes in Home c
	No	Yes	No	Yes	No	Yes	No	Yes	No	Yes	No	Yes	No	Yes
Partner a
No	-	-	46 (71.9) †	18 (28.1)	62 (95.4) ***	3 (4.6)	38 (57.6) *	28 (42.4)	41 (62.1) *	25 (37.9)	35 (53.0) **	31 (47.0)	54 (83.1)	11 (16.9)
Yes	-	-	44 (57.1)	33 (42.9)	54 (75.0)	18 (25.0)	29 (38.7)	46 (61.3)	31 (40.3)	46 (59.7)	21 (27.3)	56 (72.7)	55 (76.4)	17 (23.6)
Parent b
No	46 (51.1) †	44 (48.9)	-	-	84 (95.5) ***	4 (4.5)	59 (65.6) ***	31 (34.4)	61 (67.8) ***	29 (32.2)	50 (55.6) ***	40 (44.4)	75 (84.3) *	14 (15.7)
Yes	18 (35.3)	33 (64.7)	-	-	31 (66.0)	16 (34.0)	7 (14.6)	41 (85.4)	9 (17.6)	42 (82.4)	5 (10.0)	45 (90.0)	32 (69.6)	14 (30.4)
Grandparent
No	62 (53.4) ***	54 (46.6)	84 (73.0) ***	31 (27.0)	-	-	63 (54.8) ***	52 (45.2)	67 (57.8) ***	49 (42.2)	53 (45.7) ***	63 (54.3)	95 (83.3) d, *	19 (16.7)
Yes	3 (14.3)	18 (85.7)	4 (20.0)	16 (80.0)	-	-	2 (9.5)	19 (90.5)	3 (14.3)	18 (85.7)	1 (4.8)	20 (95.2)	12 (60.0)	8 (40.0)
Aunt/uncle
No	38 (56.7) *	29 (43.3)	59 (89.4) ***	7 (10.6)	63 (96.9) ***	2 (3.1)	-	-	53 (79.1) ***	14 (20.9)	52 (77.6) ***	15 (22.4)	56 (86.2)	9 (13.8)
Yes	28 (37.8)	46 (62.2)	31 (43.1)	41 (56.9)	52 (73.2)	19 (26.8)	-	-	19 (26.0)	54 (74.0)	4 (5.5)	69 (94.5)	53 (75.7)	17 (24.3)
Sibling
No	41 (56.9) *	31 (43.1)	61 (87.1) ***	9 (12.9)	67 (95.7) ***	3 (4.3)	53 (73.6) ***	19 (26.4)	-	-	47 (65.3) ***	25 (34.7)	63 (88.7)	8 (11.3)
Yes	25 (35.2)	46 (64.8)	29 (40.8)	42 (59.2)	49 (73.1)	18 (26.9)	14 (20.6)	54 (79.4)	-	-	9 (12.9)	61 (87.1)	46 (69.7) **	20 (30.3)
Cousin
No	35 (62.5) **	21 (37.5)	50 (90.9) ***	5 (9.1)	53 (98.1) ***	1 (1.9)	52 (92.9) ***	4 (7.1)	47 (83.9) ***	9 (16.1)	-	-	48 (87.3)	7 (12.7)
Yes	31 (35.6)	56 (64.4)	40 (47.1)	45 (52.9)	63 (75.9)	20 (24.1)	15 (17.9)	69 (82.1)	25 (29.1)	61 (70.9)	-	-	61 (74.4)†	21 (25.6)
Younger family member c
No	54 (49.5)	55 (50.5)	75 (70.1) *	32 (29.9)	95 (88.8) d, *	12 (11.2)	56 (51.4)	53 (48.6)	63 (57.8) **	46 (42.2)	48 (44.0) †	61 (56.0)	-	-
Yes	11 (39.3)	17 (60.7)	14 (50.0)	14 (50.0)	19 (70.4)	8 (29.6)	9 (34.6)	17 (65.4)	8 (28.6)	20 (71.4)	7 (25.0)	21 (75.0)	-	-

Notes: The numbers represent “n (%)”.

^*p < .05.

^**p < .01.

^***p < .001.

^†p < .10.

Valid percents presented. Chi-square tests were used unless noted otherwise.

^aSpouse/boyfriend/girlfriend.

^bCaregivers’ parent/step-parent.

^cCaregivers’ child/grandchild/niece/nephew.

^dFisher’s or Fisher-Freeman-Haltman exact test; 2-sided exact p-value used.

Logistic Regression Examining Home Tobacco Environment Factors and Outcomes of Lifetime and Past Year Quit Attempts

Unadjusted and adjusted multivariable logistic regression models identified associations between home tobacco environment factors and the quitting outcomes (Tables 4; Model Sets 1–2). In unadjusted analyses, there was a 37% decrease in the odds of having a lifetime quit attempt for every 1 unit increase in the number of people in the home who smoked in the caregiver’s presence in the past week (OR: 0.63, [95% CI: 0.47, 0.85]). This association remained statistically significant (AOR: 0.61, [95% CI: 0.45, 0.84]) following adjustment for select caregiver sociodemographic, healthcare, and tobacco use characteristics. We did not observe any significant associations with past year quit attempts in unadjusted or adjusted models. Although not statistically significant (p = .06), there was an association between caregiver restrictions on child access to cigarettes in the home and past year quit attempt history (AOR=7.72, CI: 0.92, 64.84).

Table 4.

Multivariable models examining associations between home tobacco environment by caregiver quit attempt histories and use of evidence based smoking cessation strategies.

Caregivers who Smoke Cigarettes on Some Days/Every Day (n= 147)	Lifetime History of Quit Attempt				Past Year Quit Attempt				Used Evidence-Based Cessation Strategy a
	Unadjusted Model		Adjusted Model		Unadjusted Model		Adjusted Model		Unadjusted Model
	Model (n=130): x2(10)= 21.59, p = .02		Model (n=126): x2(20)= 32.80, p = .04		Model (n=68): x2(10)= 7.16, p = .71		Model (n=67): x2(20)= 26.21, p = .16		Model (n=37): x2(10)= 12.45, p = .26
	B (SE)	OR (95% CI)	B (SE)	AOR ( 95% CI)	B (SE)	OR (95% CI)	B (SE)	AOR (95% CI)	B (SE)	OR ( 95% CI)
Number of people in home who smoked in caregiver’s presence in past week (M (SD))	−0.46 (0.15) **	0.63 (0.47, 0.85)	−0.49 (0.16) **	0.61 (0.45, 0.84)	0.07 (0.22)	1.08 (0.70, 1.66)	0.17 (0.28)	1.19 (0.69, 2.04)	−0.004 (0.36)	1.00 (0.49, 2.02)
Partner smokes cigarettes or cigars in home b
No (Ref)
Yes	0.57 (0.42)	1.77 (0.77, 4.04)	0.72 (0.48)	2.06 (0.80, 5.33)	0.48 (0.57)	1.62 (0.53, 4.92)	0.24 (0.72)	1.27 (0.31, 5.20)	0.14 (0.91)	1.15 (0.19, 6.89)
Parent smokes cigarettes or cigars in home c
No (Ref)
Yes	−0.45 (0.50)	0.64 (0.24, 1.68)	−0.32 (0.59)	0.73 (0.23, 2.31)	0.28 (0.73)	1.32 (0.32, 5.53)	−0.25 (1.11)	0.78 (0.09, 6.86)	−1.23 (1.22)	0.29 (0.03, 3.20)
Grandparent smokes cigarettes or cigars in home
No (Ref)
Yes	0.02 (0.65)	1.02 (0.29, 3.66)	0.14 (0.72)	1.15 (0.28, 4.71)	0.54 (0.98)	1.71 (0.25, 11.61)	0.52 (1.42)	1.67 (0.10, 27.12)	−2.06 (1.53)	0.13 (0.01, 3.20)
Aunt/uncle smokes cigarettes or cigars in home
No (Ref)
Yes	0.62 (0.59)	1.86 (0.59, 5.91)	0.63 (0.66)	1.87 (0.51, 6.79)	0.27 (0.76)	1.31 (0.30, 5.81)	0.62 (0.96)	1.85 (0.28, 12.20)	−1.76 (1.93)	0.17 (0.004, 7.47)
Sibling smokes cigarettes or cigars in home
No (Ref)
Yes	−0.60 (0.48)	0.55 (0.21, 1.41)	−0.73 (0.54)	0.48 (0.17, 1.38)	−0.55 (0.65)	0.58 (0.16, 2.08)	−0.65 (0.84)	0.52 (0.10, 2.71)	2.33 (1.19)†	10.24 (1.00, 105.13)
Cousin smokes cigarettes or cigars in home
No (Ref)
Yes	−0.17 (0.60)	0.85 (0.26, 2.75)	0 (0.66)	1.00 (0.29, 3.62)	−0.46 (0.80)	0.63 (0.13, 3.02)	−0.48 (0.96)	0.62 (0.09, 4.06)	1.45 (1.97)	4.27 (0.09, 201.97)
Younger family member smokes cigarettes or cigars in home d
No (Ref)
Yes	−0.19 (0.51)	0.83 (0.30, 2.24)	−0.45 (0.57)	0.64 (0.21, 1.94)	−0.35 (0.74)	0.70 (0.16, 3.02)	−0.88 (1.05)	0.42 (0.05, 3.27)	2.83 (1.44) *	16.96 (1.01, 283.68)
Cigarette Home Smoking Ban
Partial ban/No Ban (Ref)
Full ban	−0.53 (0.42)	0.59 (0.26, 1.33)	−0.63 (0.47)	0.53 (0.21, 1.33)	0.97 (0.60)	2.62 (0.81, 8.54)	0.96 (0.76)	2.60 (0.58, 11.59)	−0.95 (0.91)	0.39 (0.07, 2.30)
Caregiver Restrictions on Child Access to Cigarettes in the Home
Cigarettes are kept in the home and child has access to them (Ref)
Cigarettes are not kept in the home or child does not have access to them	0.32 (0.47)	1.38 (0.55, 3.44)	0.48 (0.52)	1.61 (0.58, 4.44)	0.49 (0.69)	1.63 (0.42, 6.37)	2.04 (1.09) ††	7.72 (0.92, 64.84)	−0.33 (120)	0.72 (0.07, 7.58)

Notes:

^*p < .05.

^**p < .01.

^***p < .001.

^†p = .05.

^††p = .06.

Boldface indicates a statistically significant difference. Unadjusted models had only the independent variables entered into the models.

^aComposite variable identifying use of evidence-based cessation methods that included nicotine replacement therapy, medication, cessation quitline, cessation internet program, individual counseling, cessation clinic/class/support group, friend/family support, cessation books, pamphlets, videos, gradually cutting down.

^bSpouse/boyfriend/girlfriend.

^cParent/step-parent.

^dChild/ grandchild/niece/nephew. Adjusted multivariable models controlled for: caregiver age, FTND, marital status, cigarette use frequency, past month menthol use, binary perceived financial situation, caregiver past year discussion with healthcare provider about tobacco use, and ages of children in the home.

Logistic Regression Examining Home Tobacco Environment Factors and Outcome of EBS Use

In unadjusted analyses (Table 4; Model 3), caregivers with young family members (i.e., child/ grandchild/ niece/ nephew) who smoked in the home had 16 times the odds of EBS use during their past year quit attempt (OR: 16.96 [95% CI: 1.01, 283.68]) compared to caregivers who did not have a young family member who smoked in their home. Although not statistically significant, it is worth noting that caregivers with siblings who smoked in their home were more likely (OR: 10.24 [95% CI: 1.00, 105.13], p = .05) to have used an EBS during their recent quit attempt compared to caregivers who did not have siblings who smoked in the home.

Discussion

This study fills a critical gap in the literature by describing the home tobacco environment, including the associations between who is smoking in caregivers’ homes, and the relationship between home tobacco environment factors and three quitting outcomes (lifetime and past year cigarette quit attempts and EBS use during a most recent quit attempt) among Black/African American women caregivers who smoke and live in low-resource, rural areas. Our hypotheses were partially met. First, building upon our prior work (Jones et al. 2021, 2022), we found numerous associations with who smoked in caregivers’ homes overall and by the cessation outcomes, highlighting intergenerational and overall familial clusters of tobacco use (e.g., caregivers with young people who smoked in their homes were more likely to also have older generations who smoked in their homes). Second, as hypothesized, caregivers with greater past week home ETSe due to people smoking around them were less likely to have a lifetime quit attempt. Contrary to our hypotheses, who was smoking in the home was unrelated to caregiver lifetime or past year quit history; however, caregivers with young family members smoking in the home were more likely to have used EBS. Our hypotheses regarding cigarette home smoking bans and caregiver restrictions on child access to cigarettes being facilitators to quitting were largely unsupported.

The present study is a novel contribution to the literature that provides important insights for future studies. The study population is underrepresented in cessation research, despite greater exposure to multilevel drivers of tobacco-caused inequities commonly faced by socially-disadvantaged people who smoke. Our current findings build upon our prior work (Jones et al. 2021, 2022) and may inform interventions that better address home-based barriers and facilitators to caregiver cessation among Black/African American people and people residing in rural areas with limited resources.

One particularly novel aspect of this study is our characterization of who smoked in caregivers’ homes and how smoking in the home clustered among family members. Most caregivers were exposed to several family members smoking in their home, including their partners (54%), children/nieces/nephews (21%), parents (36%), grandparents (15%), siblings (50%), aunts/uncles (53%), and cousins (61%). Our focus on who smokes in the home is innovative because there is limited data on family influences on adult smoking, especially in the literature focused on priority populations (U.S. National Cancer Institute, 2017). We also identified numerous associations between which family members smoked in the home, including intergenerational clusters (i.e., younger family members were more likely to smoke in the caregiver’s home if older generations smoked in the home) and intragenerational clusters (e.g., caregivers’ parents were more likely to smoke in caregivers’ home if caregivers’ aunts/uncles smoked in the home). These findings expand upon the research documenting that grandparent, caregiver, and sibling smoking increases the risk of youth smoking onset (Leonardi-Bee et al., 2011; Vandewater et al., 2014); partner smoking increases the risk of continued smoking among pregnant and recently pregnant women (Scheffers-Van Schayck et al., 2019); and that smoking cessation among adults’ spouses, siblings, friends, and coworkers decreased the chance of adult smoking (Christakis & Fowler, 2008). Our findings underscore the importance of understanding who smokes in Black/ African American rural caregivers’ homes to improve the success of tobacco interventions focused on caregiver cessation, family ETSe reduction, and child tobacco use prevention (Butler et al., 2009; Hoehn et al., 2016; Marbin et al., 2021). Future research using longitudinal or multi-generational informant designs, or more advanced analytic techniques (e.g., mixture modeling), may help to further illuminate the associations between family members smoking in the homes of rural, Black/ African American women caregivers.

Our findings related to how home tobacco environment factors are related to quitting behaviors also dovetail with the available literature. We found that caregivers with more people who smoked around them in the past week were less likely to have a lifetime quit attempt. This finding is consistent with the extant literature showing that living with people who smoke and ETSe reduce the odds of cessation among the general population (Chandola et al., 2004; Okoli & Kodet, 2015), women (Holahan et al., 2012), and caregivers (Nabi-Burza et al., 2021).

Prior research in this area has largely focused on the quantity of household tobacco exposure (e.g., number of people who smoke in the home) rather than on how who is smoking around caregivers affects cessation. The notable exception is research focused on how partner smoking affects cessation among pregnant people (e.g., (Scheffers-Van Schayck et al., 2019)). Therefore, our study expanded the literature by examining how who was smoking in the home related to caregiver lifetime or past year quit history and EBS use, finding that caregivers with young family members smoking in the home were more likely to have used EBS during a most recent quit attempt. However, we did not find evidence supporting relationships between who was smoking in caregivers’ homes and quit attempts. The finding documenting how having young people smoking in the home increased the odds of EBS use is aligned with qualitative data highlighting how Black/ African American caregivers’ motivation to prevent their children from using tobacco is a facilitator for smoking cessation and home smoking ban implementation (Butler et al., 2009; Hoehn et al., 2016). Caregivers’ efforts to protect their family from tobacco were also seen in other ways: About 32% of caregivers had a full home cigarette smoking ban and 79% kept cigarettes inaccessible to children in the home. Although our study failed to identify statistically significant findings between these behaviors and quitting, our findings provide preliminary insights and support the need for future, larger scale research examining how having full home cigarette smoking bans and restrictions on child access to cigarettes in the home may facilitate caregiver quit attempts. This is especially warranted given a prior study that found that home smoking bans were related to use of tobacco cessation medications among quitline callers (Yuan et al., 2019). Collectively, our findings and the extant literature reify the need for systems-oriented cessation interventions that address the home environment.

Finally, our novel finding identifying that 40% of Black/African American women caregivers residing in rural, low-resource communities used an EBS during a past year quit attempt is somewhat consistent with the literature looking at EBS use rates from a non-intersectional lens. It also highlights a critical barrier to smoking cessation among this priority population. For example, in prior national studies, estimates of EBS use among women (37%) and Black adults (30.7%) is lower than what was found in the present study (Leventhal et al., 2022). However, prior studies did not stratify by race, gender, region, and/or income. Notably, our estimate of EBS use is higher than Leventhal et al.’s (2022) estimate among Black/African American people, which could be attributed to our intersectional approach and/or our broader/more inclusive EBS definition compared to the Leventhal et al. (2022), who only included nicotine replacement therapy, medication, quitline, counseling, and digital applications. Future research is needed to enhance our understanding of EBS use among these caregivers.

Generally, successful quitting often involves numerous quit attempts, with low long-term abstinence rates among individuals not receiving any cessation treatment (Chaiton et al., 2016; Hughes et al., 2004). Understanding how home environment factors relate to quit attempts and EBS use among women caregivers with these intersectional social experiences is critical because they are underrepresented in research, despite the poorer cessation outcomes among Black/ African American adults and those from socially disadvantaged and rural backgrounds (Leventhal et al., 2022; US Department of Health and Human Services, 2020) (and potentially women (Smith et al., 2016)). For these women, cessation interventions may benefit from addressing home-based factors to support cessation and reduce tobacco exposure across generations.

Limitations

There are several limitations. The generalizability of our findings may be limited to populations similar to the population investigated. However, as noted in recent published works (Potter et al., 2021; Tan et al., 2023), our purposeful focus on Black women caregivers living low resource, rural areas may provide critically needed data to help improve cessation outcomes and reduce tobacco-caused health inequities among these women. Caregivers self-reported on if they were a primary decision maker in the home, and this was not independently verified; future research should consider additional methods for assessing primary decision-making roles in the home as this could influence outcomes. Our sampling frame and sample size likely affected the precision of some estimates (e.g., the confidence intervals for some estimates in the multivariable EBS model were wide). Our covariate selection approach may have failed to include some relevant covariates (Hosmer & Lemeshow, 2000); future research with larger samples should explore other covariate selection techniques and examine if that impacts the covariates included and overall pattern of results (Chowdhury & Turin, 2020). Regarding identification of EBS use, information on which digital interventions or self-help materials caregivers used was not available, therefore, while the broad category was evidence-based, it is unknown if the specific cessation strategy they used was.

Future research may benefit from expanding this work to investigate how additional multilevel influences, including additional home (e.g., housing security), family (e.g., family functioning), and contextual influences (e.g., area deprivation, access to cessation resources), predict, and perhaps interact, to influence quitting behaviors among this population. The National Institute of Minority Health and Health Disparities Research Framework (Alvidrez et al., 2019) and socioecological approach for addressing tobacco disparities (U.S. National Cancer Institute, 2017) may be helpful for identifying multilevel influences. Larger scale research in this area may allow for examination of other important variables (e.g., monoracial/ multiracial identity). We were unable to assess cessation as an outcome because all data were from participants who currently smoked; future research would benefit from assessing the association between the home environment and cessation. Furthermore, conclusions about causality are limited by the cross-sectional nature of this study; longitudinal research on how the home environment predicts quit attempts and cessation is needed.

Conclusions

Black/African American women caregivers who smoke and live in low-resource, rural areas are exposed to numerous home environments factors that may impede cessation, including having numerous family members who smoke in their homes. There was significant clustering surrounding who smoked in the caregivers’ homes, with notable intergenerational and intragenerational smoking patterns. Greater past week exposure to other people smoking in the home was related to lower odds of a lifetime quit attempt and having young people smoking in the home was related to EBS use during a most recent quit attempt, highlighting the importance of understanding both the quantity and source of home-based tobacco exposure among caregivers. Understanding Black/ African American rural caregivers’ home environment may help to improve the success of caregiver cessation, family ETSe reduction, and child tobacco use prevention interventions.

Positionality statement:

Mindful that our identities can influence our approach to science (Roberts, et al. 2020), the authors wish to provide the reader with information about our backgrounds. With respect to gender, when the manuscript was drafted, 6 authors self-identified as women and 1 author as a man. With respect to race, 1 author self-identified as white and 6 as Black/Kenyan or African American.