HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Carroll, S. L. Articles by Huang, T. T.-K. Search for Related Content PubMed PubMed Citation Articles by Carroll, S. L. Articles by Huang, T. T.-K.

Smoking, Weight Loss Intention and Obesity-Promoting Behaviors in College Students

Watkins Memorial Health Center, University of Kansas, Lawrence, Kansas (S.L.C., M.L.S.)
Department of Health and Human Performance, University of Houston, Houston, Texas (R.E.L.)
Departments of University of Minnesota School of Medicine, Minneapolis, Minnesota (H.K.)
Department of Psychology, University of Montana, Missoula, Montana (K.J.H.)
Friedman School of Nutrition Science and Policy, Tufts University, Boston, Massachusetts (T.T.-K.H.)

Address reprint requests to: Dr. Terry T.-K. Huang, Health Scientist Administrator and Program Director of Pediatric Obesity and Metabolic Syndrome, Endocrinology, Nutrition and Growth Branch, National Institute of Child Health and Human Development, 6100 Executive Boulevard, 4B11, MSC 7510, Rockville, MD 20852. E-mail: huangter{at}mail.nih.gov

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS CONCLUSION ACKNOWLEDGMENTS REFERENCES

 
Objective: To examine whether college smoking was associated with trying to lose weight and other weight-related behaviors.

Methods: We surveyed 300 students at the University of Kansas about smoking (ever, current, and amount), weight loss intention (y/n), weight-related attitudes, and eating and exercise behavior. Weight, height, and body fat were measured.

Results: About half the students (49%) self-identified as having ever smoked while 53 (17.6%) self-identified as current smokers. After controlling for sex, age, and ethnicity, ever smoking was not related to weight loss intention but was associated with greater pressure to maintain a healthy weight (p = 0.05), and having engaged in mild exercise on more days in the previous year (p = 0.05). Compared to nonsmokers, current smokers ate more at restaurants serving high calorie foods (p < 0.05) and ate more frequently in front of the TV (p < 0.01). Amount smoked was related to diminished use of exercise facilities (p = 0.03) and more frequent eating at restaurants serving high calorie foods (p < 0.05) and in front of the TV (p = 0.01).

Conclusions: Current smoking among college students was related to weight loss intention. Despite wanting to lose weight, current smoking was concomitant with obesity-promoting behaviors such as eating higher calorie foods and eating in front of the TV. College-based interventions to prevent smoking initiation or promote smoking cessation should include a focus on healthy eating, exercise and healthful ways to lose or maintain weight.

Key words: smoking, diet, weight loss, physical activity, college health

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS CONCLUSION ACKNOWLEDGMENTS REFERENCES

 
Cigarette smoking is a major health problem among college students, with 28.5% of students identifying themselves as smokers [1]. As many as half of self-reported smokers indicate they initiated smoking during their college years [2,3]. Studies of smokers in the general population indicate that smoking is concomitant with high rates of additional unhealthy behaviors [4,5], but studies examining these relationships in young adults are scarce and often show contradictory results [6]. For example, while some studies have shown associations of smoking with other unhealthy behaviors such as poor eating and physical activity habits [7] or negative cardiovascular outcomes [8], others have not shown similar findings [9,10]. Gender differences might also exist in these associations [11].

It is unclear whether dieting is a significant risk factor for smoking among college students [2,10]. Among adolescents, studies have shown that smoking initiation is higher in girls who diet [12], express anxiety about their weight [13,14], or are overweight [15]. Adolescent girls also report using cigarettes to control binge eating and believe that smoking helps limit weight gain [16,17]. Adolescent and young adult female smokers are more likely to report tobacco use as a method of weight control as they age. This may be because young adults are able to purchase tobacco legally and smoke on a regular basis, which is necessary if smoking is used to replace eating [18].

In light of the obesity epidemic and the pressure for thinness, dieting is more common than ever before [6]. This is true even among people of normal weight [6]. In 2000, the American Dietetic Association found that 28% of Americans reported significantly changing their eating behavior to try to achieve a healthier, more nutritious diet. This was 2% higher than in 1997, and the highest percentage since the survey began in 1991 [19]. Whether smoking is effective for weight loss is not clear, but smoking cessation is often associated with weight gain. Levels of gain vary and mechanisms are not well understood [20,21], but associations among smoking cessation, increased food intake, and decreased metabolism have been shown [22].

The associations between smoking and weight control among adolescents, as well as the biological link between smoking cessation and weight gain, suggest that college students may also attempt to use smoking as a dieting strategy. Therefore, in this study, we aimed to examine associations among smoking, body composition, weight-related attitudes, weight loss intention, and eating and physical activity patterns among students from the University of Kansas, Lawrence.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS CONCLUSION ACKNOWLEDGMENTS REFERENCES

 
Data were collected from September 2003 to April 2005 as part of the Monitoring University Students Tackling Diabetes and Obesity (MUST-DO) Study which aimed to examine the relationships among diet, activity, psychosocial variables, body composition, and metabolic outcomes among college students. In this portion of the study, three hundred students (aged 18–24 y) were asked questions regarding their smoking status, weight attitudes, weight loss intention, eating habits, as well as their level of physical activity. Students were recruited by posting flyers advertising the study in heavily frequented areas such as campus lounges and by sending e-mail over club listserves. Due to the subject area of the study, there were more female recruits than males and emphasis was placed on recruiting males at the end of the study period. This was done by e-mailing fraternities and male-dominated organizations such as the engineering club. Interested females were not refused due to gender. One hundred two (34%) participants were male and 198 (66%) were female. Most students were white (82%, the white population at the University of Kansas is close to 90%). All students provided informed consent prior to testing. The study was approved by the Institutional Review Boards of the University of Kansas, Lawrence, and Tufts University.

Smoking behavior was assessed by three variables: ever smoking (y/n), current smoking (y/n in the past 6 months), and amount of cigarettes smoked (number of cigarettes per day in last 30 days) [23]. Basic body composition indices were measured by trained staff. Measured weight and height were used to calculate body mass index (BMI, kg/m²). Percent body fat was measured using leg-to-leg bio-impedance analysis (Tanita 300A, Arlington Heights, IL).

In addition, students were asked by questionnaire about their perception of their own weight and weight loss intention (how they described their current weight (–2 = "very underweight" to –2 = "very overweight") [24], whether they were trying to lose weight (y/n) [24], whether their diet was high, medium, or low in fat (0 = "low" to 2 = "high") [25], and if they felt pressured to maintain a healthy weight (0 = "not true" to 6 = "totally true"). Students were also asked whether they had received weight-related clinical counseling in the past (if a health care provider spoke to them regarding their risk for obesity, advised them to change their eating habits to lose weight, or to exercise or get more physical activity to lose weight) (y/n). Students were questioned regarding their level of physical activity—how often they engaged in strenuous, moderate, and mild exercise (number of days per year) [26] or used university exercise facilities (y/n). Finally students were asked about their food intake behavior (how many days per year they ate at restaurants serving fried chicken, burgers, pizza, Mexican, Chinese, fried fish, and fast food) [27]. Information was also gathered regarding how often students ate all-you-can-eat buffets, vegetarian meals, prepared meals (e.g., canned, frozen, or microwave entrees) or ate in front of the television (how many times per year).

Log transformations were performed with variables not normally distributed. T-tests and chi square analyses were used to compare means and proportions between the two genders. In separate models, general linear regression analyses were performed to model weight-related attitudes, weight loss intention, and eating and activity variables on each of the three smoking variables. All regression analyses controlled for age, gender, ethnicity (white vs. non-white), and the interaction between smoking and gender. Stratified analyses by gender were performed when there was a significant smoking by gender interaction. Analyses were conducted using MINITAB v. 13 (State College, PA) and SAS v. 9.1 (Cary, NC), with an alpha set at 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS CONCLUSION ACKNOWLEDGMENTS REFERENCES

 
Table 1 shows the descriptive characteristics in the total sample as well as by gender. Of the 300 participants, 49% (n = 153) self-identified as having ever smoked and 17% (n = 53) self-identified as current smokers (smoking in the past 30 days) (Table 1). Thirty seven percent were trying to lose weight, and 83% said they used university exercise facilities. The average participant had a BMI of 23.77 ± 4.91 (mean ± SD) kg/m². Male students had an average BMI of 25.60 ± 6.97 kg/m² while female students had an average BMI of 22.84 ± 3.01 kg/m². Overall, 29% of the subjects were overweight (BMI 25 kg/m²).

Regression analyses were performed to model weight-related variables and eating and physical activity patterns on smoking indicators, adjusting for age, gender, and ethnicity. No significant differences were found between ever smokers and never smokers in terms of body composition. However, compared with never smokers, students who had ever smoked reported having experienced significantly greater pressure to maintain a healthy weight (adjusted means ± SE = 2.29 ± 0.18 vs. 1.84 ± 0.19, p = 0.05), having engaged in mild exercise on more days in the previous year (adjusted means ± SE = 208.10 ± 18.98 vs. 161.31 ± 19.55, p = 0.05), and having eaten more at restaurants serving Mexican food (adjusted logged means ± SE = 3.13 ± 0.08 vs. 2.80 ± 0.08, p = 0.008). There was no significant interaction between ever smoking and gender.

Comparisons between current smokers and non-smokers showed that current smokers had a significantly higher BMI (adjusted means ± SE = 25.88 ± 0.67 vs. 24.22 ± 0.39, p = 0.02). Students who were current smokers reported receiving counseling from a health care provider about their risk for obesity (20% vs. 6%, p = 0.003) and being advised by a health care provider to increase their physical activity to lose or control weight (21% vs. 10%, p = 0.05) more than nonsmokers. No differences were found between current smokers and non-smokers regarding time spent on mild, moderate or vigorous exercise, or on use of university exercise facilities. Instead, current smokers reported eating at restaurants serving burgers (adjusted logged means ± SE = 3.56 ± 0.13 vs. 3.27 ± 0.08, p = 0.04), Mexican food (adjusted logged means ± SE = 3.16 ± 0.12 vs. 2.20 ± 0.07, p = 0.05), Chinese food (adjusted logged means ± SE = 3.13 ± 0.12 vs. 2.86 ± 0.07, p = 0.04), and fried fish (adjusted logged means ± SE = 2.17 ± 0.11 vs. 1.89 ± 0.06, p = 0.01) more than nonsmokers. Current smokers also reported eating meals or snacks in front of the television more often than nonsmokers (adjusted logged means ± SE = 4.68 ± 0.19 vs. 4.08 ± 0.11, p = 0.00).

The interaction between current smoking and gender was significant for models with the following dependent outcomes: BMI, percent body fat, strenuous exercise, physician counsel to maintain a healthy weight, and eating at restaurants serving pizza. Stratified analyses were conducted for males and for females using these variables. Male current smokers had a higher BMI than male nonsmokers (adjusted mean ± SE = 29.26 ± 1.41 vs. 25.75 ± 0.92, p = 0.02). Male current smokers also had a higher percent body fat (adjusted mean ± SE = 20.35 ± 1.32 vs. 16.83 ± 0.86, p = 0.01). Compared to female nonsmokers, female current smokers ate more often at restaurants serving pizza (adjusted logged means ± SE = 3.79 ± 0.16 vs. 3.27 ± 0.07, p = 0.003) and were less likely to indicate their doctor believed they should maintain a healthy weight (adjusted mean ± SE = 2.80 ± 0.37 vs. 3.62 ± 0.18, p = 0.001). However, female current smokers indicated engaging in strenuous exercise more often than female nonsmokers (adjusted mean ± SE = 206.67 ± 28.87 vs. 111.21 ± 13.99, p = 0.001).

When amount smoked (i.e., number of cigarettes consumed in the past 30 days) was examined as the independent variable, significant results were found for the use of university exercise facilities (ß ± SE = –0.01 ± 0.01, p = 0.03), the number of days students engaged in mild exercise (ß ± SE = 7.70 ± 2.79, p = 0.005), the frequency at which students consumed food from restaurants serving burgers (ß ± SE = 0.04 ± 0.01, p < 0.001), pizza (ß ± SE = 0.03 ± 0.01, p = 0.005), Mexican food (ß ± SE = 0.03 ± 0.01, p = 0.03), or fried fish (ß ± SE = 0.02 ± 0.01, p = 0.02), and the frequency at which they ate vegetarian meals (ß ± SE = 0.06 ± 0.02, p = 0.01). Amount smoked was also associated with how often students ate in front of the television (ß ± SE = 0.05 ± 0.02, p = 0.01). Significant interaction between amount smoked and gender was found for only two dependent outcomes: a health care provider counseling on obesity was significant for males (ß ± SE = 0.02 ± 0.006, p < 0.001) but not for females (ß ± SE = –0.001 ± 0.004, p = 0.76), and body weight was significant for males (ß ± SE = 1.19 ± 0.54, p = 0.03) but not for females (ß ± SE = –0.05 ± 0.27, p = 0.84).

Many unhealthy behaviors followed a dose-response relationship with the amount smoked per day. The more students smoked, the more often they ate at restaurants serving burgers and ate in front of the TV (p < 0.05 for trend, Fig. 1). In addition, the more students smoked, the less frequently they used university exercise facilities (p = 0.03 for trend).

View larger version (22K): [in this window] [in a new window]   Fig. 1. There was a dose response relationship between consumption of food at restaurants serving burgers and consumption of food in front of the TV with amount smoked per day (p < 0.05). Consumption of food at restaurants serving burgers (d/year) by amount of smoking is shown in black bars. Food consumption in front of the TV (times/year) by amount of smoking is shown in gray bars. The frequency of activity reported on the y axis (either day or times per year as indicated above) is reported as adjusted means. Data were log transformed for analyses, but were exponentiated (back transformed) in Fig. 1.

	DISCUSSION

Studies that focus on smoking among college students tend to focus on current smokers [2,7–9]. We presented some unique data on ever smokers vs. never smokers. Results for students who ever smoked may indicate that some students tried smoking during adolescence because they believed it was an acceptable way to maintain or reduce weight, but quit smoking when they realized the negative health consequences associated with smoking. This would explain why feeling pressured to maintain a healthy weight was related to having ever smoked but not necessarily related to current smoking. It may also explain why ever smokers were significantly more likely to engage in mild exercise. Indeed, contrasting results from models of ever smoking vs. models of current smoking might suggest that quitting smoking is concomitant with a range of improvement in health behaviors.

Current smoking was associated with higher body mass and fat (particularly among males) and the intention to lose weight. Current smoking was also associated with a number of poor health behaviors including eating at restaurants serving high calorie foods, eating in front of the television, and decreased use of university exercise facilities. Many of these unhealthful behaviors increased with increasing amount of smoking. These seemingly paradoxical findings suggest that current smoking is a risk factor that clusters with many other unhealthy lifestyle choices. For these students, the intention to lose weight was not correlated well with healthy behaviors that might help them achieve that goal.

Some interesting gender differences were found. Male current smokers had a higher BMI and percent body fat compared to their nonsmoking counterparts. Current smoking status among females was not related to differences in BMI compared with non-smokers but to a lower frequency of receiving weight counseling from physicians and higher frequency of strenuous activity. Smoking among females may be related more to a perception of weight problems, especially since females experienced greater pressure to maintain weight and 40% of them reported trying to lose weight (Table 1). These gender differences may be important to consider in the design of intervention programs.

Based on this study, college health interventions should not view smoking as an independent health behavior. It is evident that smoking is one component of a cluster of unhealthy lifestyles. Comprehensive prevention or intervention programs are needed in the college setting to address the complexity of these unhealthy behaviors in this unique developmental stage.

This study was limited by its cross sectional design and its reliance on self reports. Because analysis of smoking was not the main focus of the original study, limited questions on smoking behavior were available. For the same reason, we did not over-sample smokers in the study in order to provide a larger smoking sample size. Finally, because of the population demographics in Kansas, our study included mostly white students; thus, generalization of findings to other ethnic minorities may not be appropriate.

	CONCLUSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS CONCLUSION ACKNOWLEDGMENTS REFERENCES

 
During the 1990s, smoking dramatically increased among young adults in spite of progress made among younger adolescents [28,29]. Tobacco companies target college-age consumers because of laws that prohibit marketing to younger groups and because this period is considered transitional for experimental smokers who either quit or become regular tobacco users [30,9]. Therefore, public health professionals ought to also address this important age group to ensure that experimental smokers quit and nonsmokers do not start smoking during these impressionable years. This study shows that smoking among college students is complex and related to a cluster of poor health behaviors, despite the fact that smokers are more likely to indicate weight loss intention. This study also demonstrates important differences in behaviors between ever smokers and current smokers as well as between male and female smokers. University interventions may benefit from comprehensive health models that incorporate not just anti-smoking messages but also other health behaviors, including healthy eating and exercise. Strategies should be developed to improve overall health. Finally, the etiology of this cluster of unhealthy behaviors in this developmental stage warrants much further study.

	ACKNOWLEDGMENTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS CONCLUSION ACKNOWLEDGMENTS REFERENCES

 
This study was funded by American Heart Association grant #0365447Z (PI: TTK Huang). Dr. Huang conducted this research when he was at the Friedman School of Nutrition Science and Policy at Tufts University.

Received September 12, 2005. Accepted March 16, 2006.

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS RESULTS CONCLUSION ACKNOWLEDGMENTS REFERENCES

 

1. Wechsler H, Rigotti N, Gledhill-Hoyt J, Lee H: Increased level of cigarette use among college students: a cause for national concern. JAMA280 :1673 –1678,1998 .[Abstract/Free Full Text]
2. Saules K, Pomerleau C, Snedecor S, Mehringer A, Shadlea M, Kurth C, Krahn D: Relationship of onset of cigarette smoking during college to alcohol use, dieting concerns, and depressed mood: results from the young women’s health survey. Addict Behav29 :893 –899,2004 .[Medline]
3. DeBernardo R, Aldinger C, Dawood O, Hanson R, Lee S, Rinaldi S: An e-mail assessment of undergraduates’ attitudes toward smoking. J Am Coll Health48 :61 –66,1999 .[Medline]
4. Serdula M, Byers T, Mokdad A, Simoes E, Mendlein J, Coates R: The association between fruit and vegetable intake and chronic disease risk factors. Epidemiology7 :161 –165,1996 .[Medline]
5. Rosal M, Ockene J, Yunsheng M, Hebert J, Merriam P, Matthews C, Ockene I: Behavioral risk factors among members of a health maintenance organization. Prev Med33 :586 –594,2001 .[Medline]
6. Saarni S, Silventoinen K, Rissanen A, Sarlio-La‘hteenkorva S, Kaprio J: Intentional weight loss and smoking in young adults. Int J Obes28 :796 –802,2004 .[Medline]
7. Oleckno W, Blacconiere M: A multiple discriminate analysis of smoking status and health related attitudes and behaviors. Am J Prev Med6 :323 –329,1990 .[Medline]
8. Spencer L: Results of a heart disease risk factor screening among traditional college students. J Am Coll Health50 :291 –296,2002 .[Medline]
9. Kenford S, Wetter D, Welsch S, Smith S, Fiore M, Baker T: Progression of college-age cigarette samplers: what influences outcome. Addict Behav30 :285 –294,2005 .[Medline]
10. Lenz B: Tobacco, depression, and lifestyle choices in the pivotal early college years. J Am Coll Health52 :213 –220,2004 .[Medline]
11. Psujek J, Martz D, Curtin L, Michael K, Aeschleman S: Gender differences in the association among nicotine dependence, body image, depression, and anxiety within a college population. Addict Behav29 :375 –380,2004 .[Medline]
12. French S, Perry C, Leon G, Fulkerson J: Weight concerns, dieting behaviors, and smoking initiation in adolescents: a prospective epidemiological study. Am J Public Health84 :1818 –1820,1994 .[Abstract/Free Full Text]
13. Honjo K, Siegel M: Perceived importance of being thin and smoking initiation among young girls. Tob Control12 :289 –295,2003 .[Abstract/Free Full Text]
14. French S, Perry C: Smoking among adolescent girls: prevalence and etiology. J Am Med Womens Assoc51 :25 –28,1996 .
15. Camp D, Klesges R, Relyea G: The relationship between body weight concerns and adolescent smoking. Health Psychol12 :24 –32,1993 .[Medline]
16. Crisp A, Sedgwick P, Halek C, Joughin N, Humphrey H: Why may teenage girls persist in smoking? J Adolesc22 :657 –672,1999 .[Medline]
17. Crisp AH, Stavrakaki C, Halek C, Williams E, Sedgwick P, Kiossis I: Smoking and pursuit of thinness in schoolgirls in London and Ottawa. Postgrad Med J74 :473 –479,1998 .[Abstract]
18. Nichter M, Nichter M, Vuckovic N, Tesler L, Adrian S, Ritenbaugh C: Smoking as a weight-control strategy among adolescent girls and young women: a reconsideration. Med Anthropol Q18 :305 –324,2004 .[Medline]
19. American Dietetic Association: American’s Food and Nutrition Attitudes and Behaviors—American Dietetic Association’s Nutrition and You: Trends 2000. Available on-line May 25,2005 : http://www.eatright.org/Public/Media/PublicMedia_10333.cfm.
20. Klesges R, Winders S, Meyers A, Eck L, Ward K, Hultquist C, Ray J, Sadish W: How much weight gain occurs following smoking cessation? a comparison of weight gain using both continuous and point prevalence abstinence. J Consult Clin Psychol65 :286 –291,1997 .[Medline]
21. Swan G, Carmelli D: Characteristics associated with excessive weight gain after smoking cessation in men. Am J Public Health85 :73 –77,1995 .[Abstract/Free Full Text]
22. Filozof C, Fernández Pinilla M, Fernández-Cruz A: Smoking cessation and weight gain. Obesity5 :95 –103,2004 .
23. Centers for Disease Control and Prevention (CDC): Behavioral Risk Factor Surveillance System Survey Questionnaire. Atlanta, Georgia: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention,1999 .
24. Centers for Disease Control and Prevention (CDC): Behavioral Risk Factor Surveillance System. Centers for Disease Control and Prevention. Available on-line 6/30/05: http://apps.nccd.cdc.gov/BRFSSQuest/
25. National Cancer Institute (NCI): Percent Energy from Fat Screener. National Institutes of Health. Available on-line: 6/30/05: http://riskfactor.cancer.gov/diet/screeners/fat/
26. Sallis J, Buono M, Roby J, Micale F, Nelson J: Seven day recall and other physical activity reports in children and adolescents. Med Sci in Sports Exerc1 :99 –108,1993 .
27. McCrory M, Fuss P, Hays N, Vinken A, Greenberg A, Roberts S: Overeating in America: association between food consumption and body fatness in healthy adult men and women ages 19 to 80. Obes Res7 :564 –571,1999 .[Medline]
28. Kear M: Psychosocial determinants of cigarette smoking among college students. J Community Health Nurs19 :245 –257,2002 .[Medline]
29. Wechsler H, Kelley K, Seibring M, Kuo M, Rigotti N: College smoking policies and smoking cessation programs: results of a survey of college health center directors. J Am Coll Health49 :1 –8,2001 .
30. Rigotti N, Moran S, Wechsler H: US college students’ exposure to tobacco promotions: prevalence and association with tobacco use. Am J Public Health95 :138 –144,2005 .[Abstract/Free Full Text]

This Article Abstract Figures Only Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Carroll, S. L. Articles by Huang, T. T.-K. Search for Related Content PubMed PubMed Citation Articles by Carroll, S. L. Articles by Huang, T. T.-K.