Article, Pediatrics

A Smoking Cessation Intervention for Low-Income Smokers in the ED

a b s t r a c t

Background: There is a high prevalence of smoking among caregivers who bring their children to the pediatric emergency department (PED) and even higher rates of tobacco smoke exposure (TSE) and related morbidity among their children. The PED visit presents an opportunity to intervene with caregivers, but it is unknown whether they are more likely to quit if their child has a TSE-related illness. We sought to examine a PED-based smoking cessation intervention and compare outcomes based on children’s TSE-related illness.

Methods: A single-arm, prospective trial, with baseline, 3, and 6 month assessments was used in this study. Care- givers whose child had either a TSE-related (n = 100) or non-TSE-related illness (n = 100) were given a brief intervention consisting of counseling, referral to the Quitline, and free nicotine replacement therapy.

Results: Participants were 91.5% female, 50.5% African American, 100% Medicaid recipients, 30.8 years old, child age mean of 5.5 years, 90% highly nicotine dependent, and 60.3% and 75.8% allowed smoking in the home and car, respectively. At follow-up (65% retention), 80% reported quit attempts at 3 months and 89% between 3 and 6 months. There were significant decreases in number of cigarettes smoked, time to first cigarette, and smoking in the home and car. Quit rates were 12.2% at 3 months, 14.6% at 6 months, and 7.3% at both time points (50% biochemically confirmed). There were no significant differences in outcomes based on children’s illness. Conclusions: A brief PED-based smoking cessation intervention resulted in quit attempts and successful quits. However, the presence of a TSE-related illness did not result in different cessation outcomes.

Introduction

An estimated 11 million low-income caregivers who smoke accom- pany their children to the pediatric emergency department (PED) annu- ally [1-3]. Children from low-income households are at higher risk for tobacco smoke exposure (TSE) and TSE-related morbidity compared with children from households above the poverty level [4-7]. Compared with higher income smokers, low-income smokers are at increased risk for tobacco-related disparities such as lower cessation rates, higher smoking-related illnesses, and lower access to cessation resources. The

? Funding source: This study was funded by the National Institutes of Health National Cancer Institute grant K22CA163747 (to Dr Mahabee-Gittens).

?? Financial disclosure: The authors have indicated that they have no financial rela-

tionships relevant to this article to disclose.

? Conflict of interest: The authors have indicated that they have no conflicts of inter-

est relevant to this article.

?? Clinical trial registration number: NCT01728038.

* Corresponding author. Division of Emergency Medicine, Cincinnati Children’s Hospital Medical Center, 3333 Burnet Avenue; Cincinnati, Ohio 45229-3039. Tel.: +1 513 636 7966; fax: +1 513 636 7967.

E-mail address: [email protected] (E.M. Mahabee-Gittens).

PED is a novel, yet largely underused setting in which to address these disparities. Much of the care delivered in the PED is nonemergent, and these visits are characterized by long wait times. This presents a unique opportunity to help caregivers quit smoking and, thus, reduce their chil- dren’s TSE.

Our previous cessation research indicates that caregivers who visit the PED are aware of the pediatric effects of TSE, are motivated to quit, and eager to receive cessation counseling [8-11]. The Health Belief Model [12] posits that one’s beliefs about health problems and the per- ceived benefits of action explain engagement in health-promoting be- havior, such as smoking cessation. In addition, the model posits that a stimulus or cue to action such as a child’s acute illness must also be pres- ent to trigger smoking cessation [13,14]. Prior research evaluated whether caregivers were more likely to quit smoking if a cessation in- tervention was presented as a way to improve their child’s health, with largely disappointing results [15]. Our prior research demonstrat- ed that caregivers who perceive that their child’s health is at risk be- cause of TSE are more motivated to quit, and those who believe that quitting will benefit their child’s health are more likely to quit [16,17]. However, it is not known if caregivers who smoke are more likely to quit if their child is brought to the PED for a TSE-related illness and

http://dx.doi.org/10.1016/j.ajem.2015.04.058

receive an intervention that includes information on the effects of TSE on their child’s health. We hypothesized that the delivery of a brief ces- sation intervention to caregivers during the PED visit would be more ef- fective for those whose child had a TSE-related illness vs those who did not. The primary aim of this study was to evaluate the feasibility and overall efficacy of a brief PED-based tobacco cessation intervention among low-income caregivers who brought their child to the PED and to compare efficacy in caregivers whose child presented with and with- out a TSE-related illness.

Methods

Participant, screening, and recruitment

Participants were recruited between March 2012 and June 2013 from the PED of Cincinnati Children’s Hospital Medical Center (CCHMC). A clinical research coordinator (CRC) accessed the electronic health system to screen for eligible caregivers who presented to the PED with a child younger than 18 years who was triaged in the nonurgent or urgent category and were recipients of Medicaid (used as a proxy for low-income status). These caregivers completed a screening question- naire that assessed demographics and smoking status. Those answering yes to “Have you smoked at least one cigarette in the past week?” were considered current smokers. Caregivers were excluded if they were en- rolled in a cessation program or using nicotine replacement therapy (NRT) or other pharmacologic cessation treatment, were non-English or non-Spanish speaking, had no working phone number or mailing ad- dress, or had plans to move within 8 months. To determine if there were differences in outcomes in caregivers whose child did or did not have a TSE-related illness, we recruited an equal number (n = 100) of care- givers with a child with a potentially TSE-related chief complaint as outlined by the US Surgeon General (eg, colds, ear pain, wheezing)

[18] and a non-TSE-related illness (n = 100). Eligible caregivers provid- ed informed consent. This study was approved by the institutional re- view board at CCHMC.

Assessments

Demographics were collected at baseline: caregiver and child age, sex, race/ethnicity, and education level. We also assessed caregivers’ smoking behavior, nicotine dependence, and prior quit attempts. To de- termine nicotine dependence, we used the Heavy Smoking Index (HSI), which is a validated, 2-item self-report measure of nicotine dependence (range, 0-8) derived from the Fagerstrom Test for Nicotine Dependence [19-21]. Caregivers were assessed for level of readiness to quit using the Contemplation Ladder (range, 0-10) [22]. Caregivers also reported smoking at home and in the car. To assess caregivers’ perception of their child’s smoking-related health risks, we used measures adapted from Wagener et al [23], which assess perceived child health risk be- cause of caregiver smoking and perceived child health benefit to caregiver cessation (range, 1-7). Caregivers were asked if they believed that their child’s PED visit was because of a TSE-related illness (yes/no). Finally, caregivers were queried regarding their acceptability of the ces- sation intervention (5-point Likert scale with responses ranging from “strongly disagree” to “strongly agree”).

The CRC conducted phone follow-up at 3 and 6 months after base- line to assess participants’ smoking behavior, nicotine dependence, quit attempts, readiness to quit, smoking in the home or car, and smoking abstinence (caregivers were asked to choose the response that best represented the amount they smoked “During the last 7 days.” Options were “I smoked every day,” “I smoked once in a while,” and “I haven’t smoked at all, not even a puff.” Only those who selected the latter category were considered abstinent). The primary outcome measures were repeated point prevalence of smoking at 3 and 6 months and prolonged abstinence at 6 months. Secondary outcomes included number of quit attempts, level of readiness to quit, participant retention,

and use of cessation resources. Self-reported abstinence was verified during home visits with exhaled carbon monoxide testing using a Bedfont piCO^+TM Smokerlyzer(R) carbon monoxide monitor (CoVita, LLC; Haddonfield, NJ) with more than 6 ppm as the cutoff indicating a positive smoking result [24].

Smoking cessation intervention

Caregivers received a brief (10-15 minutes) counseling session by the CRC while their child was waiting to be evaluated by the physician. This dedicated CRC was trained by the Principal Investigator (M.M.G.) before the study and observed delivering the counseling for the first 10 partici- pants. The counseling was based on the 5 As of the Public Health Service Clinical Practice Guideline for Treatment of Tobacco Dependence [25], highlighting the effects of TSE on children’s health. Participants responding positively to “Do you want to quit in the next six months?” were given a description of the Ohio Quitline (QL) and then asked about their interest in being referred. For those participants responding nega- tively, the CRC used the 5 Rs counseling technique to increase readiness to quit [25] and then gave an expanded description of the QL services. All participants were offered the opportunity to have a direct phone or fax referral to the QL during the PED visit. Those not wishing to be contacted by the QL were offered written QL cessation brochures. Counselors at the QL attempted to contact referred participants up to 3 times over a 7-day period. Caregivers contacted by the QL were provided stage-appropriate information and/or counseling. In the PED, participants who expressed an interest in receiving NRT were screened for eligibility. Eligible participants were given vouchers to be redeemed at CCHMC’s pharmacy for a free, 2-week supply of NRT (choice of gum, patch, or lozenge) with directions on proper use delivered by the CRC. Participants received gift cards of $10, $20, and $30 at baseline, 3 months, and 6-month follow-up, respectively; a bonus $10 gift card was received for completion of both follow-ups.

Analyses

Univariate analyses identified potential outliers and examined dis- tributional properties of the dependent and independent variables. Bi- variate analysis examined the relationships between dependent and independent measures. As we had measurements recorded over time, the statistical analysis used a method to account for these multiple mea- surements and also for any missing time points. The technique used is called a mixed model and uses generalized estimating equations, allowing examination of changes over time. The subject identifier is used in the regression model and is included as a random effect, as it is unique to each participant. Based on the literature, 6 predetermined covariates (caregiver age, race and education, HSI, child age, and TSE- related illness) [26-28] were included in the multivariable analyses, re- gardless of statistical significance. Adjusted odds ratios (AOR) and 95% confidence intervals (CIs) are reported from the final models. All analy- ses were conducted using SAS version 9.3 (SAS Institute, Cary, NC); PROC GENMOD was used for the repeated measures analysis. An a priori level of P b .05 was considered statistically significant.

Results

Participant characteristics

A total of 1809 caregivers were approached; 865 (47.8%) screened positive for current smoking, but 355 (41%) refused further study screen- ing. Of the remaining 510, 399 (78.2%) were current smokers with Medicaid. Of these, 252 (63.1%) were eligible, 212 (84.1%) consented, and 12 were duplicates and were removed; thus, 200 (79.4%) received the intervention and were included in the analysis (Figure). The sociodemographics, smoking behavior, and TSE-related characteristics of the study population at baseline are presented in Table 1.

Analyzed at 3 months (n = 131)

Analyzed at 6 months (n = 130)

Figure. Flow of participants in the study.

-Non-Smoker (n = 944)

-Smoker (n = 865)

-Refused further screening (n = 355)

-Smokers who were screened (n = 510)

-Ineligible (n = 258)

-Non-Medicaid (n = 111)

-Already on NRT (n = 27)

-Plans to move in 8 months (n = 84)

-No telephone (n = 14)

-No permanent residence (n = 5)

-Discharged before study completed (n = 4)

-No parent/legal guardian (n = 12)

-Non-Spanish Interpreter needed (n = 1)

-Refused to Consent (n = 40)

-Duplicates (n = 12)

Assessed for smoking (n = 1809)

Met Study Criteria (n = 200)

Lost to follow up at 3 months (n = 35)

Lost to follow up at 6 months (n = 36)

Caregivers of Children with SHSe Illnesses

(n = 100)

Lost to follow up at 3 months (n = 34)

Lost to follow up at 6 months (n = 34)

Caregivers of Children with non-SHSe Illnesses

(n = 100)

Analysis

Follow up

TSE-related characteristics

Allocation

Enrollment

Of the total sample, 43.5% believed that their child’s health would improve if they stopped smoking, but only 27.6% believed that their child would develop a Serious disease if they continued smoking, and there were no differences in these beliefs if the child presented with TSE-related illness. When caregivers whose children presented with TSE-related complaints were compared with those whose children pre- sented with a non-TSE-related complaint, they were more likely to be younger, have a younger child, not allow smoking inside the home, and believe that their child’s visit was because of TSE. There were no other statistically significant differences in demographic or smoking- related measures related to TSE (Table 1).

Participant retention

The CRC obtained questionnaires from 131 (65.5%) participants at 3 months and 130 (65%) at 6 months; 109 (54.5%) were completed at both 3 and 6 months; 48 (24%) were lost to follow-up at both time points. The primary reasons why we were unable to reach participants were no answer or the wrong telephone number given. Those who completed both 3 and 6 months were more likely to have older children (mean, SD: 6.4, 5.2 years, vs 4.5, 4.6 years; P = .008) and to be older than 30 years (62.6% vs 93.4%, P b .001).

Changes in caregiver attitudes about TSE

There were no differences at 3 or 6 months in caregivers’ belief that their child’s health would worsen if they continued smoking or that their child’s health would improve if they quit, comparing those whose child did and did not present with a TSE-related illness.

Changes in smoking behaviors

Of the 131 caregivers that we assessed at 3 months, 16 caregivers (12.2%) reported that they had quit at 3 months (9/16 [56%] completed the CO test, 7/16 [44%] biochemically confirmed). Of the 130 caregivers that we assessed at 6 months, 19 (14.6%) reported that they had quit at 6 months (8/19 [42%] completed the CO test, 6/19 [32%] biochemically confirmed). Of the 109 caregivers that we assessed at both 3 and 6 months, 8 (7.3%) reported that they had quit (5/8 [62.5%] completed both CO tests, 4/8 [50%] biochemically confirmed). The 8 reporting quit- ting at both time points are included in the 16 and 19 that reported quit- ting at 3 months and 6 months, respectively. In addition, 80% of caregivers made a quit attempt between baseline and 3 months, and 89.2% made a quit attempt between 3 and 6 months. Using an intent to treat analysis, 16 caregivers (8%) reported that they had quit at 3 months, 19 (9.5%) reported that they had quit at 6 months, and 8 (4%) reported that they had quit at both 3- and 6-month follow-ups.

Table 1 Population demographics, caregiver smoking belief, and caregiver’s beliefs
	Total sample	TSE-related illness	Non-TSE-related illness	P
	(n = 200), n (%)	(n = 100), n (%)	(n = 100), n (%)
Parent age (mean, SD)	30.8, 8.0	29.3, 7.2	32.3, 8.5
b30 y	95 (48.2)	56 (57.1)	39 (39.4)	.01
>= 30 y	102 (51.8)	42 (42.9)	60 (60.6)	.01
Child age (mean, SD)	5.5, 5.0	4.0, 4.4	7.0, 5.3	b.0001
Parent gender, female	183 (91.5)	90 (90.0)	93 (93.0)	NS
Parent race				NS
White	87 (43.7)	43 (43.0)	44 (44.4)
African American	101 (50.8)	52 (52.0)	49 (49.5)
Asian, American Indian, unspecified	11 (5.5)	5 (5.0)	6 (6.1)
Parent education				NS
Less than college	124 (62.3)	61 (61.0)	63 (63.6)
Some college and above	75 (37.7)	39 (39.0)	36 (36.4)
Number of cigarettes/day				NS
1-10	100 (50.0)	54 (54.0)	46 (46.0)
N 10	100 (50.0)	46 (46.0)	54 (54.0)
Minutes to first cigarette upon awakening				NS
1-5	100 (50.0)	44 (44.0)	56 (56.0)
>= 6	100 (50.0)	56 (56.0)	44 (44.0)
HSI >=4^a	179 (90.0)	89 (89.0)	90 (90.0)	NS
motivation to quit (range, 0-10; mean, SD)	7.1, 2.3	7.1, 2.1	7.0, 2.5	NS
Smoking not allowed inside the home	79 (39.7)	47 (47.0)	32 (32.3)	.03
Smoking not allowed inside the car	29 (24.2)	13 (23.6)	16 (24.6)	NS
Caregiver’s belief of child disease >=50% likely if they do not stop smoking	55 (27.6)	28 (28.0)	27 (27.3)	NS
Caregiver’s belief that child’s health would get “much better” or “completely	87 (43.5)	42 (42.0)	45 (45.0)	NS
better” if they quit smoking
Caregiver’s belief that child’s PED visit is because of TSE	14 (7.0)	12 (12.0)	2 (2.0)	b.006

^a Heavy Smoking Index is a validated, 2-item self-report measure of nicotine dependence derived from the Fagerstrom Test for Nicotine Dependence [20-22]; please see text.

As shown in Table 2, there were statistically significant changes from baseline to both 3 and 6 months in cigarettes smoked, time to first cig- arette, and HSI. Whites had higher odds of smoking less than 10 ciga- rettes per day than African Americans (AOR, 0.11; 95% CI, 0.06-0.2) but also higher odds of smoking their first cigarette within 5 minutes of waking compared with African Americans (AOR, 1.9; 95% CI, 1.1- 3.3). In addition, whites had higher odds of having HSI of 4 or more com- pared with African Americans (AOR, 3.4; 95% CI, 1.9-6.2), and caregivers older than 30 years had higher odds of an increased HSI compared with younger caregivers (AOR, 2.2; 95% CI, 1.1-4.2). There were significant in- creases in the number of participants who did not allow smoking in the home over time from baseline (39.7%) to 3 months (60.3%) (P b .01) but not from baseline (39.7%) to 6 months (53.9%); whites had higher odds

of banning smoking in the home (AOR, 1.8; 95% CI, 1.1-3.1; P b .05), compared with African Americans, as did non-heavy smokers com- pared with heavy smokers (AOR, 1.8; 95% CI, 1.0-3.1; P b .05) and care- givers who had Younger children (AOR, 1.08; 95% CI, 1.01-1.14; P b .05 for each year decrease in age). Similarly, there were decreases in the number of caregivers who smoked in their cars over time; caregivers with younger children had higher odds of not smoking in the car (AOR, 1.19; 95% CI, 1.08-1.30; P b .001 for each year decrease in age).

There were no statistically significant differences in caregiver age, sex, education, prior quit attempts, or readiness to quit in those who re- duced smoking at follow-up. However, a higher percentage of white caregivers reported smoking fewer cigarettes compared with all other races both at 3 months (68% vs 40%, P b .01) and at 6 months (69% vs

Table 2 Changes in smoking characteristics
	Baseline	3 mo	6 mo	P for unadjusted/adjusted^a	Specific difference in adjusted GEE was
	(n = 200)	(n = 131)	(n = 130)	GEE analysis over time	present in baseline to 3 mo, 3 to 6 mo, or
					baseline to 6 mo
No. of cigarettes/day 0-10	100 (50.0)	120 (91.6)	117 (90.0)	b.0001/b.0001	Base to 3 mo (P b .0001)
N 10	100 (50.0)	11 (8.4)	13 (10.0)		Base to 6 mo (P b .0001)
Time to first cigarette (min) <= 5	100 (50.0)	22 (16.8)	19 (14.6)	b.0001/b.0001	Base to 3 mo (P b .0001)
>= 6	100 (50.0)	109 (83.2)	111 (85.4)		Base to 6 mo (P b .0001)
HSI >=4^b	179 (90.0)	54 (41.2)	43 (33.1)	b.0001/b.0001	Base to 3 mo (P b .0001)
					Base to 6 mo (P b .0001)
Motivation to quit (range, 0-10; mean, SD)	7.1, 2.3	7.6, 1.8	7.6, 2.0	.01/.3
Motivation to quit >= 7 (range, 0-10)	110 (55.0)	89 (77.4)	87 (78.4)	b.0001/.001	Base to 3 mo (P = .001)
					Base to 6 mo (P = .004)
Smoking not allowed inside the home	79 (39.7)	79 (60.3)	70 (53.9)	b.0001/.023	Base to 3 mo (P = .008)
Smoking not allowed inside the car (n = 120)^c	29 (24.2)	28 (40.0)	32 (43.8)	.004/.009	Base to 3 mo (P = .01)
					Base to 6 mo (P = .002)

Abbreviation: GEE, generalized estimating equations.

^a Adjusted for secondhand smoking-related illness, heavy smoking, race, parent age group, parent education, and child age. However, heavy smoking was not used as a covariate in the model for number of cigarettes per day or time to first cigarette.

^b Heavy Smoking Index is a validated, 2-item self-report measure of nicotine dependence derived from the Fagerstrom Test for Nicotine Dependence [20-22]; please see text.

^c Eighty caregivers did not own cars.

40%, P b .01). There were no differences in quit rates or reduction in smoking when we compared those who did and did not bring their child in for a TSE-related illness.

Use of cessation resources

Four participants accepted direct and 103 accepted fax QL referral. Of those who accepted fax referral, 25% reported on follow-up assessments that they had received QL counseling. All who expressed an interest in receiving NRT (n = 184, 92%) were screened for NRT eligibility; 169 (92%) were eligible and received NRT vouchers. Of those given the voucher, 79 (47%) redeemed it; 41 (52%) received the patch, 29 (37%) the gum, and 9 (11%) the lozenge.

Consumer satisfaction

Of the 198 (99%) caregivers who completed the acceptability survey, most strongly agreed or agreed that they were “satisfied” with the inter- vention (n = 187, 94%) and that the intervention was “useful” (n = 187, 94%) and “interesting” (n = 189, 95%). In addition, most said they would “recommend this study to a friend or family member” (n = 190, 96%) and that “smoking advice should be given in the PED” (n = 182, 92%).

Discussion

To our knowledge, this is the first study to test the effects of a brief cessation intervention including NRT in the PED setting. Although we hypothesized that caregivers who brought their children in for a TSE- related illness would have higher cessation rates, our results did not support this hypothesis. Moreover, we did not see differences in beliefs about the effects of smoking or the benefits of quitting on their child. Similar to our previous studies, only 28% of caregivers in our study whose child had a TSE-related illness believed that their smoking was putting their child’s health at risk, and 44% of caregivers believed that quitting would benefit their child [17]. Our findings support previous primary care or hospital-based research that found no clear evidence of cessation success with caregivers whose children have a TSE- related or even a general “ill child” visit [29-32]. However, a recent meta-analysis indicates that interventions designed to achieve cessa- tion among caregivers for the sake of their children appear to increase quit rates, especially when interventions include the use of medications [15]. Our lack of findings may be caused by nonstandardized informa- tion about how TSE relates to their child’s illness included in our inter- vention. In addition, our lack of findings may be caused by our broad definition of a TSE-related illness, which included illnesses that are not as commonly recognized by caregivers as being associated with TSE, such as colds and ear infections. There were encouraging results in ces- sation, number of cigarettes smoked, nicotine dependency, readiness to quit, and reported TSE in the home and car. Moreover, similar to previ- ous research, we found that caregivers who had younger children were more likely to have imposed smoking bans and may be more likely to perceive harm from smoking [23] and to quit after an intervention [16]. The current study is important for several reasons. First, it docu- ments the feasibility and acceptability of a PED-based smoking cessation intervention for caregivers who are not the patient and are not expecting to be targeted in an intervention for themselves. We achieved high recruitment rates (84%), and most (94%) found the intervention acceptable. Second, this study highlights the need for interventions that provide education for caregivers about the effects of smoke expo- sure on children. Caregivers did not understand the effects of TSE on their children or the benefits of quitting on their children’s health. Third, our results suggest that this intervention may be effective in re- ducing smoking and increasing cessation among caregivers identified in the PED. Finally, this study highlights methodological issues related to evaluating a cessation intervention targeting caregivers recruited

from the PED setting. Although most caregivers welcomed the vouchers for NRT, only 47% actually picked up the NRT from the on-site pharmacy (which was just down the hall from the PED). It may have been better to provide caregivers with NRT during the PED visit. In addition, we had difficulty with participant retention despite multiple contacts, generous incentives, and the option of home visits to caregivers who reported ab- stinence. Such difficulties with retention are common in low-income populations [33].

A number of limitations should be considered when interpreting these results. First, this was a small sample, limiting the overall Power of the study. However, despite the sample size, a number of significant and important findings were detected related to cessation in this popu- lation that has not been extensively studied. In addition, the sample was drawn exclusively from a population of low-income smokers who pre- sented to a Midwestern, tertiary care PED at 1 children’s hospital, which limits generalizability. On a related note, our sample was largely female, which is because female caregivers are much more likely to bring their children to the PED than males [34]. Greater efforts are need- ed to recruit and enroll male caregivers who smoke into PED-based in- terventions. In addition, because of the low socioeconomic status of our PED population, we experienced a high attrition rate at follow-up. Final- ly, no control condition was included to determine whether changes were because of the intervention or simply to the changing smoking patterns of caregivers; however, based on our previous research in this setting, we know that it is unlikely that these caregivers would have been assisted in quitting without this intervention [28,35,36]. Fu- ture PED-based efficacy trials should include a control arm and a larger sample of caregivers. Despite these limitations, results from our re- search may guide future research on conducting cessation interventions for low-income caregivers in the PED and other Acute care settings.

Conclusions

The results of our pilot study are promising and suggest the need for further research in this area. The intervention model was viable and ac- ceptable to caregivers, and there was preliminary evidence of efficacy. However, it is necessary to conduct full-scale randomized control effec- tiveness trials. In addition, future studies need to improve retention rates in this transient, low-income population. Our findings did not show differences in cessation between caregivers whose children have/did not have TSE-related illnesses. Future research should focus on improving and testing the TSE intervention component. Encourag- ingly, our brief intervention prompted a substantial number of quit at- tempts, reduced cigarette consumption, increased smoking bans, and reduced smoking prevalence among this underserved population.

References

AJEM