Student Tobacco Use Behaviors on College Campuses by Strength of Tobacco Campus Policies

Abstract

Purpose:

To examine the association between strength of policy and self-reported tobacco use behavior, controlling for demographic characteristics, polytobacco use, knowledge of campus tobacco policy, and perceived policy compliance by others.

Design:

Cross-sectional, online survey.

Setting:

Ten participating State University of New York (SUNY) campuses; 5 with designated smoking/tobacco use areas and 5 with 100% tobacco-free policies.

Subjects:

Convenience sample of students from SUNY campuses: only tobacco users (N = 576 students) included for analysis.

Measures:

Items assessing tobacco use behaviors on campus, policy knowledge, and observation of others using tobacco on campus.

Analysis:

T tests and chi-square tests of association used to compare responses between tobacco users across campuses. Generalized estimating equations modeling used to evaluate predictors of tobacco use on campus; model estimated with students nested within campus.

Results:

Those on campuses without a comprehensive policy were more likely to report (in the past week) having seen others smoke on campus (98% vs 69%, P < .001), having personally smoked on campus (65% vs 36%, P < .001), and seeing others use tobacco products on campus (88% vs 67%, P < .001), compared to those on tobacco-free campuses.

Conclusion:

Tobacco-free campus policies provide numerous protective factors for tobacco users and nonusers. However, compliance strategies are imperative for intended policy success.

Introduction

Tobacco use continues to be the single greatest avoidable cause of disease and death in the United States, with 443 000 deaths annually.¹ Similarly, tobacco use on college campuses remains a concern, with more than 1 in 4 college students reporting tobacco use.² College students tend to switch more easily between daily and occasional smoking compared to older adults, indicating a key opportunity to intervene and reduce tobacco use among these at-risk young adults.³ Use of emerging tobacco products (eg, e-cigarettes) among college students is also on the rise,^2,4,5 reinforcing the need to target prevention and comprehensive tobacco control strategies on college campuses. Considering the associated increased health risks, health promotion efforts to combat secondhand smoke exposure and help tobacco users quit are critical to reduce the burden of tobacco use on society.^1,6

Although colleges and universities have made strides to reduce tobacco use on their campuses by adopting smoke- or tobacco-free policies, the majority of US college campuses still do not have comprehensive policies.⁷ The American College Health Association (ACHA) recognizes the dangers of tobacco initiation and secondhand smoke and encourages all colleges and universities to promote a tobacco-free environment.⁸ As of October 1, 2019, there were at least 2469 (100%) smoke- or tobacco-free campuses with comprehensive policies, meaning all campus grounds/property is covered.⁷ As there is impetus for comprehensive tobacco policies, there is a need to support campuses currently advocating for or implementing them. Similarly, campuses with existing policies need support for compliance and enforcement strategies, one of the most frequently cited barriers to policy implementation.⁹

Purpose

Research shows that comprehensive tobacco-free policies (ie, those prohibiting the use of all tobacco products) are more effective than those prohibiting cigarette use only or those with designated smoking/tobacco areas in reducing secondhand smoke exposure and intent to use tobacco.¹⁰ The purpose of this study was to examine the association between strength of policy and tobacco use behaviors, controlling for demographic characteristics, polytobacco use, knowledge of campus tobacco policy, and perceived on-campus tobacco use by others.

Methods

Design and Sample

The cross-sectional survey was distributed via e-mail to campus administrators of all 64 State University of New York (SUNY) campuses to share with students on their campus. In general, over 440 000 students are enrolled across all campuses and are made up of various racial and ethnic groups: 68% white, 12% Hispanic or Latino, 11% black, 6% Asian/Pacific Islander, 2% multiracial, and less than 1% Native American. The majority of students also tend to be in-state (from New York).¹¹

As part of this evaluation, 12 campuses decided to participate (from within the 64-school SUNY system) and invited students to complete the survey. Student responses from 2 participating campuses were not included in the analysis since fewer than 15 students replied from each. Among the 10 campuses retained for this analysis, the number of student responses per campus ranged from 82 to 557 students (mean = 252, standard deviation = 127). See Table 1 for a summary of campus characteristics.

Table 1.

Campus Characteristics.

Campus	Campus Type	Campus Size (Acres)	Enrollment	Student Diversity (% White)	Policy Type at Time of Survey	Total Participants (Users and Nonusers)	Total % Tobacco Users of Study Participants (n)
A	Community College	50	4501	84.7%	Tobacco-free	82	24.4% (20)
B	Community College	191	6926	75%	Tobacco-free	225	13.3% (30)
C	Community College	130	9920	65%	Tobacco-free	228	26.3% (60)
D	University	220	5300	75.9%	Designated areas	704	19.0% (134)
E	Community College	120	11 888	75.6%	Tobacco-free	226	21.2% (48)
F	Community College	107	3038	80.2%	Tobacco-free	145	26.9% (39)
G	University	600	4504	32.0%	Designated areas	242	16.9% (41)
H	University	300	5719	62.9%	Designated areas	313	23.3% (73)
I	University	500	4206	53.4%	Designated areas	244	31.6% (77)
J	Technical college	850	2917	79.1%	Designated areas	231	23.4% (54)

Although a total of 2640 students completed at least one item in the online survey, only 2225 completed the portion of the survey pertaining to tobacco use (84% of respondents). Since a key outcome in this study is tobacco use on campus, we further limited the analysis to those of the 2225 who currently used at least one type of tobacco product (N = 576 or 26% of the eligible respondents). No data were available to estimate total tobacco use across all of the SUNY campuses.

Procedures

The student survey was distributed in mid-September 2016 and responses were collected until the end of March 2017. Student respondents were not excluded on the basis of sex, race/ethnicity, or student status. Survey participation was voluntary, and responses were anonymous and reported in aggregate for each campus. Per the request of SUNY Administration, the student survey was initially distributed via e-mail in September 2016 to the vice president of Student Affairs from each campus. The campus liaisons were then responsible for distributing the survey to the collective student body of their respective campuses. Campuses were not required to participate; as mentioned earlier, a total of 13 campuses sent the survey out to the student body. A reminder e-mail was sent by SUNY administration to all vice presidents of Student Affairs 1 week following the initial recruitment e-mail, again asking them to forward the survey request. In March 2017, 2 additional follow-up emails went out to campuses with zero participation to increase the response rate. However, no additional surveys were completed with these follow-up attempts. The survey was administered using Qualtrics, a secure survey administration and database software package. Only individuals who clicked on the anonymous link embedded in an email were able to complete the survey.

Measures

The electronic survey contained demographic items for age (in years), sex (male/female), and race/ethnicity (the response options were combined into the categories of “white, non-Hispanic” and “other racial/ethnic group”). Current tobacco use was assessed with the question “Within the last 30 days, on how many days did you use the following tobacco products?” Response options included “Never used,” “Have used (even tried once), but not within the last 30 days,” “1–2 days,” “3–5 days,” “6–9 days,” “10–19 days,” “20–29 days,” and “Used daily.” There was a separate row to record the frequency of use for each tobacco product, including “cigarettes,” “tobacco from a water pipe (hookah),” “cigars,” “little cigars,” “cigarillos,” “smokeless tobacco,” and “e-cigarettes.” The participant was coded as “yes” for current use of that product if they had used any number of days in the last 30 days. “Current cigarette use” was one indicator. “Current use of any tobacco product” was coded as “yes” if they had used any of the 7 listed items, and the “Current use of other tobacco products” indicator was coded as “yes” if they had used one or more tobacco products aside from cigarettes in the past 30 days. Polytobacco use was determined with these 7 items; those that indicated they had used at least 2 products in the past 30 days were coded as “yes” for polytobacco.

The item to assess knowledge of campus tobacco policy was “My university’s/college’s current policy on smoking or tobacco use is:” “100% tobacco-free: Use of all tobacco products are not allowed in all areas on campus inside and out,” “100% smoke-free: Use of cigarettes and other smoked tobacco products (eg, cigars, hookah) are not allowed in all areas on campus inside and out,” “Designated smoking areas (cigarettes, cigars, etc are allowed in those areas only),” “Designated tobacco use areas (all tobacco products are allowed in those areas only),” and “Not sure.” Students who either chose a policy that was not the same as their campus” actual tobacco policy or who indicated they were unsure were combined to form the “Incorrect/Unsure” group for this coded item. Those who indicated the policy that aligned with that of their campus were in the “Correct” group. The actual campus policies were combined into 2 policy groups: “Tobacco-free” and “Non-comprehensive.”

Exposure to on-campus tobacco use was assessed with the items “In the past 7 days, I have seen someone (not including myself) smoke cigarettes on campus” and “In the past 7 days, I have seen someone (not including myself) using another type of tobacco product (eg, smokeless tobacco, cigarillo, hookah, e-cigarettes) on campus,” each with yes/no responses. Self-report of personal on-campus use was measured with “In the past 7 days, I have smoked cigarettes on campus” and “In the past 7 days, I have used another type of tobacco product (e.g., smokeless tobacco, cigarillo, hookah, e-cigarettes) on campus;” the response options were “yes,” “no,” and either “I do not smoke cigarettes” or “I do not use any tobacco products,” respectively.

Data Analysis

Descriptive analysis, including means and standard deviations or frequency distributions, was used to summarize the survey data. Comparisons of responses between tobacco users from tobacco-free campuses and those from campuses without comprehensive tobacco policies were compared using 2-sample t tests and chi-square tests of association. Generalized estimating equations modeling was used to evaluate predictors of cigarette use on campus (among current smokers) and other tobacco use on campus (among those who indicated they were current users of other tobacco products); each model was estimated with students nested within campus. Data analysis was done using SAS, version 9.4; an alpha level of .05 was used throughout.

Results

Among the 576 students who responded to the survey and indicated they had used at least one type of tobacco product in the last 30 days, 197 were from campuses with a tobacco-free policy (34.0%) and the remaining 379 were from a campus without a comprehensive policy (66.0%). Of the 5 tobacco-free campuses, 4 explicitly prohibited e-cigarette use while the remaining one did not. Of the 5 campuses without a comprehensive policy, one had a current policy of designated smoking areas, with implementation of a tobacco-free policy to begin shortly after the survey was conducted, and the other 4 had designated smoking areas, with no pending additional policy implementation.

Among all tobacco users, the average age was 23 years, and most participants were male (56%) and white, non-Hispanic (79%). The prevalence of current cigarette use in this group is 69%, and 42% had used at least 2 tobacco products in the last month. More than four-fifths correctly identified their campus tobacco policy (83%); the remainder incorrectly identified the policy or were unsure of it. Most had seen others smoke cigarettes on campus in the past week (88%), and over half of current smokers had smoked on campus in the same time period (55%). The majority had seen others use tobacco products other than cigarettes on campus (81%), and a smaller percentage of other tobacco users had used them on campus in the same time interval (39%).

Compared to students from campuses with a noncomprehensive policy, those from tobacco-free campuses tended to be older (21 vs 26, P < .001) and more likely to be female (39% vs 52%, P = .005; see Table 2). Those on campuses with noncomprehensive policies were more likely to incorrectly identify or be unsure of their campus policy, compared to students on tobacco-free campuses (14% vs 8%, P = .023). Those on campuses without a comprehensive policy were more likely to report having seen others smoke on campus in the past week (98% vs 69%, P < .001) and having personally smoked on campus in the past week (65% vs 36%, P < .001), compared to those on tobacco-free campuses. Similarly, those on campuses without comprehensive tobacco policy were more likely to indicate they had seen others use tobacco products on campus in the past week (88% vs 67%, P < .001). Having personally used other tobacco products on campus in the past week approached significance, with students on campuses with noncomprehensive policies more likely to report use (42% vs 34%, P = .056) compared to those on tobacco-free campuses. There were no differences between students on campuses with noncomprehensive policies and those on tobacco-free campuses in racial/ethnic distribution, current cigarette use, and current polytobacco status.

Table 2.

Descriptive Summary and Campus Tobacco Policy Group Comparisons of Study Variables.^a

Variable	Total Sample Mean (SD) or n (%)	Campus Policy		P
		Tobacco-free (n = 197 From 5 Campuses)	Noncomprehensive (n = 379 From 5 Campuses)
Age	22.6 (7.1)	25.2 (9.7)	21.2 (4.6)	<.001
Sex
Male	320 (56.3%)	94 (48.2%)	226 (60.6%)	.005
Female	248 (43.7%)	101 (51.8%)	147 (39.4)
Race/ethnicity
White, non-Hispanic	448 (79.0%)	151 (77.8%)	297 (79.6%)	.62
Hispanic or other than White race	119 (21.0%)	43 (22.2%)	76 (20.4%)
Cigarette use past month
Yes	392 (68.8%)	136 (70.1%)	256 (68.1%)	.62
No	178 (31.2%)	58 (29.9%)	120 (31.9%)
Polytobacco use past month
Yes	241 (41.8%)	88 (44.7%)	153 (40.4%)	.32
No	335 (58.2%)	109 (55.3%)	226 (59.6%)
Knowledge of their campus policy
Incorrect/unsure	95 (16.6%)	15 (7.6%)	53 (14.1%)	.023
Correct	477 (83.4%)	181 (92.4%)	323 (85.9%)
Seen others smoke on campus (past week)
Yes	505 (87.8%)	136 (69.0%)	369 (97.6%)	<.001
No	70 (12.2%)	61 (31.0%)	9 (2.4%)
Personally smoked on campus (past week)b
Yes	252 (55.0%)	56 (36.4%)	196 (64.5%)	<.001
No	206 (45.0%)	98 (63.6%)	108 (35.5%)
Seen others use other tobacco products on campus (past week)
Yes	467 (81.1%)	132 (67.0%)	335 (88.4%)	<.001
No	109 (18.9%)	65 (33.0%)	44 (11.6%)
Personally used other tobacco products on campus (past week)
Yes	227 (39.4%)	67 (34.0%)	160 (42.2%)	.056
No	349 (60.6%)	130 (66.0%)	219 (57.8%)

Abbreviation: SD, standard deviation.

^aN = 576 students from 10 campuses.

^bAmong smokers only; this includes n = 458 students.

As shown in Table 3, the significant predictors of on-campus cigarette smoking (among current smokers only) included age, polytobacco use status, having seen others smoking on campus in the past week, and tobacco policy type. For each additional year in age, a current smoker was 20% more likely to use cigarettes on campus (odds ratio [OR] = 1.20, P = .015). Current smokers who also used at least one other type of tobacco were 85% more likely than cigarette-only users to smoke cigarettes on campus (OR = 2.85, P < .001). Current smokers who had seen others smoke on campus in the last week were 10 times more likely to report having smoked on campus themselves (OR = 10.67, P < .001). Finally, smokers who attended school on a tobacco-free campus were 58% less likely to indicate they had smoked on campus themselves (OR = 0.42, P = .004). Self-report of smoking on campus was not associated with gender, race/ethnicity, or policy knowledge.

Table 3.

Generalized Estimating Equations Modeling With Students Nested Within Campuses: Predictors of On-Campus Use of Cigarettes Among Smokers and On-Campus Use of Other Tobacco Use Products Among Those Who Use at Least One of These Products.

	Cigarette Use on Campus (n = 446 Smokers From 10 Campuses)			Other Tobacco Use on Campus (n = 562 Other Tobacco Users From 10 Campuses)
	OR	95% CI for OR	P	OR	95% CI for OR	P
Age	1.04	1.01–1.07	.015	0.96	0.91–1.03	.24
Gender
Male	0.81	0.51–1.27	.36	2.11	1.60–2.79	<.001
Female	1.00			1.00
Race/ethnicity
White, non-Hispanic	0.73	0.48–1.12	.15	0.89	0.48–1.64	.70
Hispanic or other than white race	1.00			1.00
Polytobacco use
Yes	1.85	1.34–2.54	<.001	3.93	3.04–5.08	<.001
No	1.00			1.00
Knowledge of policy
Incorrect/unsure	1.02	0.53–1.95	.95	1.85	1.03–3.31	.038
Correct	1.00			1.00
Seen others smoke/use tobacco on campus
Yes	10.67	3.31–34.36	<.001	5.86	2.74–12.55	<.001
No	1.00			1.00
Policy type
Tobacco free	0.42	0.23–0.76	.004	1.06	0.62–1.82	.84
Noncomprehensive	1.00			1.00

Abbreviations: CI, confidence interval; OR, odds ratio.

Use of other tobacco products on campus was associated with gender, polytobacco use status, having seen other using other tobacco products, and policy knowledge (Table 2). Among those who used other forms of tobacco (not including cigarettes), males were 111% more likely than females to report they had done so on campus within the past week (OR = 2.11, P < .001). Among other tobacco users, those who used more than one type of tobacco product were 293% more likely to have used a tobacco product other than cigarettes on campus (OR = 3.93, P < .001). Among other tobacco users, those who had seen others use tobacco products other than cigarettes were 486% more likely to report the same of themselves (OR = 5.86, P < .001). In addition, those who had incorrect or unsure knowledge of their campus tobacco policy were 85% more likely to use other tobacco products on campus (OR = 1.85, P = .038). Use of other tobacco products on campus by current users of other tobacco products was not related to age, race/ethnicity, or type of campus tobacco policy.

Discussion

Although there is a push to implement comprehensive tobacco-free campus policies,⁸ there is limited research on the impact of¹⁰ and compliance with these policies.¹² Although Fallin and colleagues reinforced that comprehensive tobacco-free policies are more effective in reducing self-reported exposure to secondhand smoke and intent to use tobacco among students on college campuses compared to noncomprehensive policies,¹⁰ there is also a need to understand factors that may affect tobacco use on campuses with varying policies. The findings of this study reinforce the benefits of comprehensive tobacco-free campus policies, including greater policy awareness, changing protobacco social norms (ie, seeing less people use tobacco while on campus), and less tobacco use on campus. Given these benefits, health promotion professionals should continue to reinforce the importance of strength of policy and evidence-based practices.

Those on campuses with tobacco-free policies were more likely to correctly identify their campus policy compared to students on campuses with noncomprehensive policies. For tobacco use on campus, those who had incorrect or unsure knowledge of their campus tobacco policy were 85% more likely to use other tobacco products on campus, further reinforcing the importance of policy communication. Campus policies only covering combustible products and/or permitting use on some areas of campus may create confusion among the campus community, reinforcing the importance of comprehensive policies. As prevalence of emerging tobacco products such as e-cigarettes increases among college students,² there is also a need to determine effective policy messaging with targeted groups on campus.¹³ It may be beneficial for health promotion and health communication professionals to collaborate to develop and test tailored messaging.

Students on tobacco-free campuses were less likely to report having seen others smoke or use tobacco on campus in the past week compared to those on campuses with noncomprehensive policies. Similarly, those on tobacco-free campuses were less likely to report personal smoking or tobacco use on campus in the past week. These findings are consistent with previous research on California campuses with varying strength of policies¹⁰ and also help support campuses working to adopt comprehensive tobacco-free policies, as they reinforce the resulting benefits of policy implementation. As stated by ACHA, tobacco-free campus policies provide an opportunity to decrease tobacco use, decrease exposure to secondhand smoke, and to shift the cultural norm.⁸ Given what we know about tobacco-related health and economic consequences,¹ health promotion professionals should work with campuses to advocate for comprehensive policies. Future longitudinal research is needed to determine how these outcomes shift over time with policy adoption, implementation, enforcement, and sustainability and how they may affect students and employees differently.

This study also looked at the significant predictors of on-campus cigarette smoking and use of other tobacco products on campus separately. Similarities included polytobacco use status and having seen others smoke or use tobacco on campus in the past week. Current smokers who also used at least one other type of tobacco were 85% more likely than cigarette-only users to smoke cigarettes on campus and among other tobacco users, those who used more than one type of tobacco product were 293% more likely to have used a tobacco product other than cigarettes on campus. These findings underscore the increased risk for dependence among college polytobacco users¹⁴ and the need for targeted tobacco treatment on college campuses.

Current smokers who had seen others smoke on campus in the last week were 10 times more likely to report having smoked on campus themselves and among other tobacco users, those who had seen others use tobacco products other than cigarettes were more than 5 times more likely to self-report tobacco use on campus. Although these findings are not surprising, they reinforce the continued social influence around tobacco use behaviors,^3,15,16 specifically among students on college campuses. Very limited research exists in relation to campus policies and social norms, but perceived social reward among tobacco users on campuses with noncomprehensive policies (ie, those with designated areas) may lead to increased frequency of tobacco use.¹⁷ There is a need for future research on college campuses assessing the impact observation of smokers and tobacco users may have on tobacco initiation and continued use.

Interestingly, use of other tobacco products on campus was not related to type of campus tobacco policy. Although our findings support stronger campus policies, over a third of students on campuses with tobacco-free policies reported smoking and/or using tobacco while on campus. Policy awareness is important, but even with visible signage and appropriate policy communication, students tend to disregard the policy if a plan for compliance or enforcement is lacking.¹⁸ Implementing evidence-based and sustainable compliance and enforcement strategies is critical to providing support for tobacco-free campus policies.^12,19 There is a need to not only evaluate strength of policy based on policy language but also to assess the association of implementation strategies, including compliance and enforcement efforts, on policy effectiveness.

Strengths and Weaknesses

This study is based on a relatively large sample of tobacco-using students from one of 10 of the 64 campuses that make up the SUNY system. The survey measured not only current use of cigarettes and other tobacco products but also use of tobacco products on campus (including both personal use and witnessing others’ use), as well as knowledge of campus tobacco policy. The primary limitation is that not all campus administrators invited their students to participate, which may have limited not only the sample size but the representativeness of the sample. We also were not able to estimate the full response rate given we do not know how many students who were current tobacco users received the invitation to be part of this study.

Conclusion

The findings of this study reinforce the benefits of comprehensive tobacco-free campus policies, including greater policy awareness, changing protobacco social norms (ie, seeing less people use tobacco while on campus), and less tobacco use on campus. There is a need for continued research on campuses with varying strength of policies to support efforts to advocate for comprehensive tobacco-free campus policies.

So What? (Implications for Health Promotion Practitioners and Researchers).

What is already known on this topic?

Tobacco-free campus policies are more effective in reducing self-reported exposure to secondhand smoke and intent to use tobacco among students compared to noncomprehensive policies.

What does this article add?

The findings of this study reinforce the benefits of comprehensive tobacco-free campus policies, including greater policy awareness, changing pro-tobacco social norms, and less tobacco use on campus. Those who used more than one type of tobacco product were more likely to have used a tobacco product on campus.

What are the implications for health promotion practice or research?

Given the benefits of tobacco-free campus policies, health promotion professionals should continue to advocate for comprehensive policies. Further research is needed to assess implementation and compliance factors as related to policy effectiveness. These findings also underscore the increased risk for dependence among college polytobacco users and the need for targeted tobacco treatment on college campuses.