Do partial home smoking bans signal progress toward a smoke-free home?

Abstract

Understanding who establishes partial home smoking bans, what these bans cover, and whether they are an intermediate step in going smoke-free would help to inform smoke-free home interventions. Participants were recruited from United Way of Greater Atlanta’s 2-1-1 contact center. Data were collected at baseline, 3 and 6 months via telephone interview. Participants (n = 375) were mostly African American (84.2%) and female (84.3%). The majority (58.5%) had annual household incomes <$10 000. At baseline, 61.3% reported a partial smoking ban and 38.7% reported no ban. Existence of a partial ban as compared with no ban was associated with being female, having more than a high school education, being married and younger age. Partial bans most often meant smoking was allowed only in designated rooms (52.6%). Other common rules included: no smoking in the presence of children (18.4%) and smoking allowed only in combination with actions such as opening a window or running a fan (9.8%). A higher percentage of households with partial bans at baseline were smoke-free at 6 months (36.5%) compared with households with no bans at baseline (22.1%). Households with partial smoking bans may have a higher level of readiness to go smoke-free than households with no restrictions.

Introduction

For children and non-smokers, particularly those who live with a smoker, the home is often the primary source of second-hand smoke exposure (SHS) [1–5]. SHS exposure causes a range of adverse health effects in both children and adults [3, 6, 7]. Among adults, it causes strokes, lung cancer and coronary heart disease; in children it exacerbates asthma and causes impaired lung function, middle ear disease, respiratory illness and Sudden Infant Death Syndrome [3, 6–9].

Household smoking bans have been shown to decrease exposure to SHS and, recently, third-hand smoke [2, 10–12]. The latter refers to contamination in the form of nicotine and other chemicals deposited on indoor surfaces by tobacco smoke [13, 14]. Household smoking bans also support smoking cessation and reduced cigarette consumption [15, 16], and may help to de-normalize smoking for youth and prevent adolescents from starting to smoke [17, 18].

Although prevalence of home smoking bans has increased markedly over the past two decades, more than half of households with a smoker still allow smoking in the home [5, 19, 20]. In 2010–2011, 46.1% of adults in households with a smoker reported a smoke-free home in contrast to 91.4% of households with no smokers [21]. Smoke-free households are more likely when no smokers live in the home, when young children are present, in higher socioeconomic households, and when a smaller proportion of friends and family members smoke [5, 19, 22, 23].

Because no level of SHS is considered safe, tobacco control efforts and related research tends to focus on smoke-free homes, which are characterized by no smoking allowed anywhere at any time [7, 19]. A significant proportion of households that still allow smoking have some restrictions either in terms of when smoking is allowed or where it is allowed [10, 24–27]. Whether these rules cover other types of smoke and vapor, such as from little cigars, e-cigarettes and marijuana is unknown. Partial bans, defined as allowing smoking at some times or in some places, are often grouped with no bans in analyses [19]. Yet, studies suggest partial bans can result in less exposure and are fairly common among households with a smoker [10, 12, 23].

Households with partial bans may represent settings with higher levels of readiness to go smoke-free than households that have not attempted to reduce exposure to SHS in any concrete way [27, 28]. Alternatively, these households may not be any closer to establishing a smoke-free policy and the difference between a partial ban and no ban may be a matter of interpretation on the part of an individual household member. A qualitative study in the United Kingdom documented that mothers who smoked ‘everywhere’ in their home agreed that smoking in a child’s bedroom would never occur [29]. Others who reported non-smoking homes described situations (e.g. visitors) when smoking occurred inside. Several US studies have noted discrepant reporting of household smoking restrictions by members of the same household, although these discrepancies have decreased in recent years as smoke-free environments have become the norm [5, 30].

To the best of our knowledge, no prior studies have carefully described partial smoking bans and examined whether they indicate a household with increased readiness to implement a full ban. This study uses longitudinal data from an intervention trial to understand similarities and differences in households that allow smoking anywhere inside the home (no ban) and those that allow it in some places or at some times (partial bans). In addition to describing partial bans and their correlates, this article also examines whether households with partial bans are more likely to progress to a total ban over time than are households with no smoking restrictions.

Methods

Participants and procedures

These data were obtained from a randomized-controlled trial (RCT) to promote the creation of a smoke free home [31]. Results from the RCT are reported elsewhere [31]. Participants were callers to the United Way of Greater Atlanta 2-1-1 contact center who were recruited to participate in the study. The mission of 2-1-1 contact centers is to connect community members to social services such as assistance with rent or utility bills [32]. Callers were eligible if they allowed any smoking in the home and if they were 18 years of age or older, were a smoker living with a non-smoker or a non-smoker living with a smoker, and if they spoke English. 2-1-1 line agents screened callers for eligibility, consented interested and eligible participants and conducted a baseline survey over the phone using a web-based tracking tool that provided scripts and allowed for online data entry. Line agents screened 4175 callers; 79.7% were ineligible and 8.4% declined to participate or did not complete the eligibility screening or consent process. The full sample of 498 participants was recruited from June 2012 to October 2012. Data were collected at baseline, 3 and 6 months post baseline. Those who completed data collection at all three time points (75.3%) are included in the analysis reported here (n = 375); 24.7% did not have complete data. Participants were compensated with a $25 gift card for completion of each telephone interview.

All participants were verbally consented to participate in the study. It was not feasible to obtain written consent due to the nature of the recruitment process and study procedures. All verbal consent was documented in the online tracking tool and included the date, time and name of agent conducting the informed consent process. Verbal informed consent and research protocol was approved by the Emory University Institutional Review Board. This study is registered with ClinicalTrials.gov (NCT01625468).

Although not the focus of this article per se, the 6-week intervention involved three mailings of educational materials supporting creation of a smoke-free home (e.g. Five Step Guide, Challenges and Solutions booklet, photonovella showing a family overcome barriers to going smoke-free), and a coaching call to help participants set goals related to establishing a smoke-free home [31, 33]. Briefly, prevalence of full smoking bans increased in the intervention group from 0% at baseline to 30.4% at 3 months and to 40.0% at 6 months [31]. Prevalence of full smoking bans also increased in the control group, likely due to a measurement effect, from 0% at baseline to 14.9% at 3 months to 25.4% at 6 months. Intent to treat analyses showed a significant intervention effect.

Measures

Because recruitment and baseline data were collected in the context of a busy call center, the baseline survey was short. Home ban status, smoking behaviors and selected demographic characteristics were collected at all three time points. More detailed questions regarding partial bans were included in the 3- and 6-month surveys. Actions to reduce SHS exposure and beliefs about SHS protection were collected at 3- and 6-month follow-up, respectively.

Household smoking bans

At all three time points, participants were asked: ‘Which statement best describes the rules about smoking inside your home?’ [34]. Response options were: smoking is not allowed anywhere inside your home; smoking is allowed in some places or at some times; smoking is allowed anywhere inside the home, or there are no rules about smoking inside the home.

At 6 month follow-up, those with partial bans were asked: ‘Can you describe the rule for me?’ They were also asked about the different kinds of smoke covered by the partial ban, including cigarette smoke, cigar or little cigar smoke, marijuana smoke, electronic cigarette smoke. The survey also included an open-ended question that asked participants to describe their reason for not having a full smoking ban.

Smoking behaviors

Current smokers were asked how long after waking they have their first cigarette, within 5, 6–30, 31–60 or >60 min [35].

Smoke-free home attempts

On the 3- and 6-month surveys, participants were asked, ‘In the last 3 months, has anyone tried to establish a smoke-free rule in your current home?’

Actions to reduce SHS exposure

At 3 months, participants were asked how often someone in their household did the following to reduce exposure to SHS: reminded smoker to go outside, opened the window, only smoked in certain rooms, smoked near a running fan, only smoked indoors when no one is home, only smoked indoors when the children were gone, used quit smoking medication, left the room to smoke, and left the room when smoker was smoking. They were also asked how often someone used air freshener to get rid of the smoke or smell [36, 37]. Responses ranged from 1 = never to 4 = very often

Beliefs about SHS protection

At 6 months, participants rated their beliefs about whether certain actions could protect non-smokers or children from SHS: Based on what you know or believe, how much each of the following actions protects non-smokers and children from SHS. The list of possible beliefs included: Banning smoking everywhere in your home except for one room, using a humidifier, smoking by an open window or door, smoking near a running fan and only smoking indoors when no one is home [36, 37]. Response options were 1 = not at all to 4 = a great deal.

Friends/relatives that smoke

Participants were asked about the number of smokers living in the home and the proportion of relatives and friends who smoked [38].

Sociodemographics

Participants reported their age, gender, race/ethnicity, employment status, educational level, type of housing and number of children in the home.

Statistical analyses

2-1-1 line agents entered baseline data into a web-based data collection system and research staff entered follow-up data into the same system. We ran univariate and bivariate descriptives for all variables and binary logistic regression analysis with ban status at baseline as the outcome variable and demographic and contextual variables as exposure variables. Key analyses were run for control group and intervention group participants separately and for the full sample. SAS 9.3 was used for all data analysis.

Results

Description of study participants

Of the 375 participants who provided data both at 3 and 6 months follow-up, most were female (84.3%), African American (84.2%), not employed (78.7%), and reported a household income of $10 000 or less (58.5%) (Table I). Most households had children in the home (77.3%). The large majority of participants were smokers (80.3%) and most reported not smoking within 60 min of waking up in the morning (76.3%).

Table I.

Baseline demographic and descriptive characteristics of study participants by ban status

	Total n = 375		Partial ban n = 230		No ban n = 145		P-valuea
	Mean	SD	Mean	SD	Mean	SD
Age	42.7	11.18	40.4	11.16	43.9	10.90	0.003
Gender	n	(%)	n	(%)	n	(%)
Male	59	15.7	29	49.2	30	50.9	0.90
Female	316	84.3	201	63.6	115	36.4	<0.0001
Race/Ethnicity
African American/Black	315	84.2	194	61.6	121	38.4	<0.0001
White	42	11.2	27	64.3	15	35.7	0.06
Other	17	4.6	8	47.1	9	52.9	0.81
Employment
Employed	80	21.3	50	62.5	30	37.5	0.03
Not employed	295	78.7	180	61.0	115	39.0	0.0002
Income
$10,000 or less	216	58.5	129	59.7	87	40.3	0.004
$10 001–$20 000	100	27.1	63	63.0	37	37.0	0.01
>$20 000	53	14.4	35	66.0	18	34.0	0.02
Education
Less than/some high school	94	25.1	49	52.1	45	47.9	0.68
High school graduate/GED	144	38.4	86	59.7	58	40.3	0.02
Higher than High School/GED	137	36.5	95	69.3	42	30.7	<0.0001
Marital status
Single	210	56.0	124	62.6	86	37.4	0.01
Married	66	17.6	44	66.7	22	33.3	0.01
Not married, living w/ partner	99	26.4	62	62.6	37	37.4	0.01
Smoking status
Smoker	301	80.3	180	59.8	121	40.2	0.001
Non-smoker	74	19.7	50	67.6	24	32.4	0.003
Time to first smoke after waking up (Smokers only; n = 300)
30 min or less	23	7.7	16	69.6	7	30.4	0.06
31–60 min	48	16.0	31	64.6	17	35.4	0.04
>60 min	229	76.3	132	57.6	97	42.4	0.02
Number of smokers in the home
1	187	50.0	111	59.4	76	40.6	0.01
2	137	36.6	82	59.9	55	40.2	0.02
3 or more	50	13.4	37	74.0	13	26.0	0.001
Children in the home
Children under 5 in the home	143	38.1	84	58.7	59	41.3	0.04
Children between 5 and 18 in the home	147	39.2	96	65.3	51	34.7	0.0002
No children in the home	85	22.7	50	58.8	35	41.2	0.10
Number of relatives and friends who smoke
Fewer than half	129	34.4	80	62.0	49	38.0	0.01
Half	123	32.8	80	65.0	43	35.0	0.001
More than half	123	32.8	70	56.9	53	43.1	0.13
Reasons for not having full smoking ban
Participant smokes	163	43.5	93	57.1	70	42.9	0.07
Participant lives with smoker	52	13.9	36	69.2	16	30.8	0.01
Didn't think of it	37	9.9	17	46.0	20	54.1	0.62
Smoker doesn't want a total ban	28	7.5	19	67.9	9	32.1	0.06
Too hard to comply with a rule	21	5.6	17	81.0	4	19.1	0.005
Unsafe to smoke outside	10	2.7	6	60.0	4	40.0	—
It’s not participant's home/not head of household	8	2.1	2	25.0	6	75.0	—
Smoking relieves stress	5	1.3	5	100.0	0	0.0	—
No good place to smoke outside	2	0.5	1	50.0	1	50.0	—
Can’t leave children alone to smoke outside	2	0.5	1	50.0	1	50.0	—
Someone else pays bills	1	0.3	1	100.0	0	0.0	—
Not sure/no reason	57	15.2	41	71.9	16	28.1	0.001

Note: Percentages might not add up to 100% due to rounding or refusal to answer.—No tests were performed due to small numbers. ^aChi-square tests for categorical variables; independent t-tests for continuous variables.

Socio-demographic differences by ban status

Overall, more participants reported partial bans (61.3%, n = 230) than no ban (38.7%, n = 145) at baseline (p < 0.0001). For most demographic categories, this pattern held with partial bans more common than no bans (Table I). Gender was an exception, with men about equally likely to report a partial ban (49.2%) or no ban (50.9%), in contrast to women who were more likely to report a partial ban. Both smokers and non-smokers were more likely to have partial bans than no bans. In additional, more partial smoking bans were reported across the entire sample independent of the number of smokers in the home.

In terms of household composition, those with children under 5 years of age in the home reported higher rates of having a partial ban (58.7%) than having no ban (41.3%). Those with children ages 5–18 had even higher rates of partial bans (65.3%). However, households without children were equally likely to have a partial ban or no ban. Those for whom half or fewer relatives and friends smoked were more likely to report a partial ban than no ban at all.

Reasons for not having a full ban

When asked why they did not having a full smoking ban, most participants (43.5%) reported it was because they smoked themselves (Table I). Other common reasons were living with a smoker (13.9%), did not think of it (9.9%), smoker does not want a ban (7.5%) and no reason (15.2%). Less common reasons were safety concerns about going outside to smoke, the participant was not the head of household, smoking relieves stress, cannot leave children alone and no good place to smoke outside. Differences in reasons by ban status were observed for living with a smoker, too hard to comply and no reason, with all three more commonly reported by those with partial bans.

Description of partial bans

During the 6-month follow-up survey, participants were asked to describe their partial bans. The majority with a partial ban had set up a rule to allow smoking only in designated rooms (52.6%). Commonly mentioned designated rooms included: the bedroom, bathroom and living/common areas. Some of the partial bans were structured around children rather than designated rooms. Approximately 18.4% did not allow smoking in the presence of children. A few (5.6%) described that smoking was not allowed anywhere in the home when children were home or in the children’s room. Others allowed smoking in the home in conjunction with actions such as running a fan or opening a window or door (9.8%). Other rules focused on smoking being permitted in bad weather, for certain guests and at parties or celebrations.

Participants reported that the partial ban covers smoke from cigarettes (96.5%), cigars or little cigars (75.3%) and marijuana (63.5%). More than half (54.7%) included electronic cigarettes in their partial ban.

Multivariate predictors of partial bans

Logistic regression analysis (Table II) investigating predictors of partial smoking bans in contrast to no ban indicate that women were almost twice as likely as men to report a partial ban (OR = 1.93; CI = 1.13, 3.27). Those with educational attainment beyond high school were more likely to have a partial ban (OR = 2.3; CI = 1.36, 3.92) as were those who were married (1.83; CI = 1.01, 3.30), but older participants were less likely to have a partial ban. Participants classified as other race (e.g. not African American or white or mixed race) were less likely to have a partial ban (OR = 0.41; CI = 0.17, 0.99) than African Americans. There were no significant differences due to smoking status, level of addiction, household composition or smoking habits of friends and relatives.

Table II.

Logistic regression model predicting partial home smoking bans

Variable	OR	(95% CI)		P-value
Gender
Male	1.00
Female	1.93	1.13	3.27	0.02
Race/Ethnicity
African American/Black	1.00
White	0.86	0.46	1.62	0.64
Other	0.41	0.17	0.99	0.048
Employment
Employed	1.00
Not employed	0.84	0.52	1.36	0.48
Income
$10 000 or less	1.00
$10 001–$20 000	1.18	0.74	1.89	0.49
>$20 000	1.33	0.72	2.47	0.37
Education
Less than/some high school	1.00
High school graduate/GED	1.53	0.94	2.51	0.09
Higher than high school/GED	2.30	1.36	3.92	0.002
Marital status
Single	1.00
Married	1.83	1.01	3.30	0.047
Not married, living w/ partner	1.10	0.69	1.75	0.70
Age	0.97	0.95	0.99	0.004
Smoking status
Smoker	1.00
Non-smoker	0.82	0.33	2.08	0.68
Time to first smoke after waking up
<30 min	1.00
30–60 min	0.87	0.33	2.32	0.78
>60 min	0.61	0.26	1.44	0.26
Number of smokers in the home
1	1.00
2	1.00	0.64	1.55	0.99
3 or more	1.22	0.67	2.22	0.52
Children in the home
Children under 5 in the home	1.00
Children between 5 and 18 in the home	1.24	0.79	1.93	0.35
No children in the home	0.98	0.54	1.77	0.94
Number of relatives and friends who smoke
Fewer than half	1.00
Half	1.16	0.71	1.89	0.56
More than half	0.81	0.50	1.32	0.40

Actions and beliefs about reducing SHS exposure

Using air fresheners to get rid of the smoke or smell, only smoking when no one else is home, and only smoking in certain rooms were the most common actions to address SHS. Such actions were generally more common in households with partial bans, with one exception. Participants with no ban reported higher frequency of smoking indoors only when no one else was home.

When asked about their beliefs on the effectiveness of potential actions to reduce exposure, the only difference was that those with a partial ban had stronger beliefs that banning smoking in the home except for one room would protect non-smokers and children from SHS than did those with no ban. It should be noted that average scores were not high, suggesting beliefs in the protective effects of these actions were not strong (Table III).

Table III.

Actions to reduce SHS exposure at 3 months post-baseline and beliefs about SHS protection at 6 months-post baseline by ban status at time of data collection

Home smoking ban status at follow-up	Combined		Partial ban		No ban		P-valuec
Actions to reduce SHS exposure at 3 monthsa	n = 293		n = 187		n = 106
	Mean	SD	Mean	SD	Mean	SD
Reminded smoker to go outside	2.60	1.11	2.86	1.05	2.13	1.07	<0.0001
Opened the window	3.02	1.08	3.22	1.00	2.66	1.14	<0.0001
Only smoked in certain rooms	3.11	1.05	3.40	0.83	2.58	1.18	<0.0001
Smoked near running fan	2.41	1.16	2.40	1.18	2.42	1.13	0.86
Only smoked indoors when no one is home	3.13	1.15	3.03	1.02	3.31	1.04	0.02
Only smoked indoors when the children were gone	3.02	1.08	2.98	1.06	3.09	1.11	0.41
Used quit smoking medication	1.46	0.81	1.49	0.84	1.42	0.72	0.46
Left the room to smoke	2.88	1.13	3.18	0.98	2.32	1.08	<0.0001
Left the room when smoker was smoking	2.47	1.21	2.75	1.17	2.34	1.12	0.004
Used air freshener	3.36	0.91	3.45	0.88	3.19	0.94	0.02
	n = 259		n = 170		n = 89
Beliefs about SHS protection at 6 monthsb	Mean	SD	Mean	SD	Mean	SD	P-value
Banning smoking everywhere in your home except for one room	2.49	1.19	2.61	1.18	2.27	1.18	0.03
Using a humidifier	2.25	1.19	2.33	1.19	2.10	1.18	0.14
Smoking by an open window or door	2.56	1.03	2.64	1.02	2.43	1.05	0.12
Smoking near a running fan	1.78	1.00	1.73	1.01	1.87	0.99	0.30
Only smoking indoors when no one is home	2.09	1.18	2.04	1.13	2.19	1.27	0.33

^aResponse options: 1 = never to 4 = very often. ^bResponse options: 1 = not at all, 2 = a little, 3 = somewhat or 4 = a great deal. ^cIndependent t-tests.

Trajectories to full bans

We examined whether households with partial bans at baseline (n = 230) were more likely to have full bans at 3 and 6 months relative to those with no ban at baseline (Table IV). Of the participants who started with a partial ban, 24.0% reported a full ban at 3 months and 36.5% reported a full ban at 6 months. In contrast, of those with no ban at baseline (n = 145), 17.2% reported a full ban at 3 months and 22.1% did so at 6 months. Among control group participants only, of those who started with a partial ban, 17.4% reported a full ban at 3 months and 30.6% reported a full ban at 6 months. Of control participants with no ban at baseline, 7.6% reported a full ban at 3 months and 12.7% did so at 6 months. Overall, at both follow-up time-points, rates for increasingly strict bans were significantly higher for those who had a partial ban at baseline compared with those who had no ban (P < 0.0001).

Table IV.

Home smoking ban status at baseline, 3 and 6 months by baseline ban status

	Baseline home smoking ban status (all participants)					Baseline home smoking ban status (control group participants)
Home smoking ban status	Partial ban		No ban		P-value	Partial Ban		No Ban		P-value
	n	%	n	%		n	%	n	%
Baseline	230	61.3	145	38.7		121	60.5	79	39.5
3 months
Full ban	55	24.0	25	17.2		21	17.4	6	7.6
Partial ban	148	64.6	40	27.6		81	66.9	17	21.5
No ban	26	11.4	80	55.2	<0.0001	19	15.7	56	70.9	<0.0001
6 months
Full ban	84	36.5	32	22.1		37	30.6	10	12.7
Partial ban	118	51.3	52	35.9		64	52.9	26	32.9
No ban	28	12.2	61	42.1	<0.0001	20	16.5	43	54.4	<0.001

Note: Chi-squared tests

Figure 1 shows trajectories of control group participants who succeeded in making their home smoke-free by 6 months (no participants had a full ban at baseline). Of those with no ban at baseline who reported a full ban at 6 months, half (50.0%) had implemented a full ban already at 3 months, while some implemented a partial ban first (30.2%), and a few still had no restrictions at 3 months. Of those with a partial ban at baseline, 41.6% had a full ban by 6 months, with 58.4% of those already implementing the full ban at 3 months.

Fig. 1.

Trajectories from no or partial ban to full household smoking bans among control group participants.

Discussion

Households with partial bans may be more ready to go smoke-free and/or may signify some level of negotiation among family members to reduce SHS exposure. Our results show that individuals with existing partial bans were more likely than those with no bans to implement full bans over time. These data indicate that those with partial bans may have a greater level of readiness to adopt full bans.

Our data on the trajectories of participants, who succeeded in making their homes smoke free; however do not support the notion that families typically take a stepped approach to implementing full bans. Half of participants who had no ban at baseline and a full ban at 6 months, also had a full ban at 3 months, while a smaller proportion went to partial bans first (at 3 months) prior to implementing full bans at 6 months. These data seem to show that while individuals with partial bans may be more ready to implement full bans than those with no bans, those who are successful at making their home smoke free often go straight from no ban to a full ban.

In the literature, it is uncommon to report progress towards the adoption of full household smoking bans [39]. Often times, households that report partial bans are grouped with those that report no ban in data analyses [5, 40] given the modest effectiveness of partial versus full smoking bans in reducing exposure to SHS and that no level of exposure is considered safe [3, 12, 41]. Unfortunately, this analytical approach makes it difficult to tease out the real-world transitions that households move through on their way to becoming a smoke-free home. Even if lines between no bans and partial bans are blurry when operationalized, those that describe themselves as having a partial ban may be more amenable to a full ban than those saying they allow smoking anywhere at any time. Our study is one of the first to attempt to examine these trajectories.

It is interesting to observe characteristics associated with having or not having a partial smoking ban. In our sample, men and those who had not graduated from high school or a GED program were less likely to have a partial ban. This may be due to lower levels of knowledge about the harmfulness of SHS exposure [42]. Households with children and those with half or fewer friends and relatives who smoked were more likely to have a partial ban. It is likely that those who are exposed to normative smoking behavior infrequently by their peers are more likely to ban smoking in their homes at least at some times or in some parts of the home. Furthermore, parents are likely more concerned than those without children about exposure of the young non-smokers in their home to SHS. These findings are generally consistent with prior research examining correlates of full home bans [5, 19, 22, 23].

In examining multivariate predictors of partial smoking bans, it was most notable that smoking status, time to first cigarette, and friends’ smoking habits did not predict having a partial ban. The characteristics that predicted partial bans versus no bans were all demographic in nature. Participants who were women, married, younger and/or with education beyond high school were more likely to have a partial smoking ban. The findings indicate that people in these demographic categories may be more open to adopting smoke-free home policies. Future interventions should incorporate components that increase their appeal to and effectiveness with single, male and/or low education participants.

The specific nature of partial bans varied across participants. Most who reported partial bans allowed smoking in designated rooms, and about 10% allowed smoking in conjunction with purported SHS reduction behaviors like running a fan, implying that either they incorrectly assumed this would provide adequate protection from SHS or that they acknowledged that this does not provide a safe environment, but may provide other benefits such as reduced smell. Prior research has suggested that the smell of SHS can stimulate rules about smoking in the home [25]. This misinformation and/or harm reduction effort also extends to behaviors aimed specifically at reducing children’s SHS exposure, such as not allowing smoking in children’s presence or in children’s rooms. Further research into households with partial smoking bans should focus on understanding whether they are misinformed (which could be corrected with an informational intervention), or engaging in harm reduction, perhaps because there are barriers to adopting a full ban (in which case interventions could address reducing those barriers). It is of note that more than half of participants stated that e-cigarettes were included in their partial bans, implying that they understand and acknowledge that second-hand vapor may also be harmful.

Not surprisingly, homes with a partial ban status were more likely to report engaging in behaviors to reduce exposure such as reminding smokers to go outside to smoke, opening windows when smoking, and smoking only in certain rooms. Because use of these strategies are conflated with the definition of partial ban it is difficult to tease out whether these are seen as additional attempts to protect household members from SHS or if they represent implementation of the partial ban itself.

The belief that limiting smoking to one room would protect non-smokers and children (a practice carried out by the majority of those with a partial smoking ban) was most strongly held by those with a partial ban while all other beliefs about SHS protection behaviors were similar among those with and without partial bans. These results may indicate a desire on the part of individuals with partial smoking bans to argue that their efforts, while only ‘partial’, can be effective.

Our study had several limitations that should be considered in interpreting results. First our study population was mainly African American and low-income with a lower level of addiction than commonly reported for other groups of smokers. The majority also lived in the Atlanta metropolitan area. Thus, these results may not generalize to other racial/ethnic groups and/or regions and countries. Second, some social desirability bias may have been present given that participants were enrolled in an intervention study to create smoke-free homes. They may also have been more receptive to changing home smoking rules than those choosing not to enroll in the study, thus affecting generalizability. Last, sample size was relatively small for some of the analyses by ban status and study condition.

For public health researchers and practitioners, recognizing partial bans as an intermediate step may facilitate the development of interventions that are tailored to stage of ban adoption. Results from this study show households with partial bans had a greater likelihood of strengthening their ban status. Practitioners may want to target households with partial bans given their possible receptivity to smoke-free home interventions.

Funding

This work was supported by the National Cancer Institute’s State and Community Tobacco Control Research Initiative, Grant Number U01CA154282.

Conflict of interest statement

None declared.