Health Education Research, Vol. 19, No. 3, 250-260,
June 1, 2004
© 2004 Oxford University Press
Changing channels for tobacco control with youth: developing an intervention for working teens
1 Center for Community-Based Research, Dana-Farber Cancer Institute, Boston, MA 02115, 2 Department of Health and Social Behavior, School of Public Health, Harvard University, Boston, MA 02115, 3 Tobacco Control Research Branch National Cancer Institute, Bethesda, MD 20852, 4 School of Public Health and Health Sciences University of Massachusetts, Amherst, MA 01003 and 5 Channing Laboratory, Brigham and Women’s Hospital, Harvard Medical School and Pediatric Oncology, Dana-Farber Cancer Institute, Boston, MA 02115, USA 6 Communication to: G. Sorensen, Center for Community-Based Research, Dana-Farber Cancer Institute, 44 Binney Street, Boston MA 02115, USA. e-mail: Glorian_Sorensen{at}dfci.harvard.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Worksites represent an untapped resource for reaching teens with tobacco control messages, given that 80% of teens have held at least one job by the time they graduate from high school. This paper presents formative research findings from a methods development study aimed at designing and testing a tobacco control intervention targeting working teens. Formative research included qualitative methods as well as quantitative data from a cross-sectional survey of teens employed in 10 participating grocery stores. Contrary to our a priori hypothesis, smoking rates among employed youth in this study were not higher than statewide averages and most of the teen workers were still in school, indicating that worksite interventions, at least in this setting, represent an alternative or adjunct to school-based programs, but do not necessarily capture a unique population. Employed teen tobacco use patterns and work characteristics that emerged from our formative research are presented in this paper, and may be useful in planning future worksite interventions for employed teens.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Approximately 80% of adult smokers initiated tobacco use before the age of 18 (Centers for Disease Control and Prevention, 2001
). Although schools have been the primary location for efforts to reduce tobacco use among youth (US Department of Health and Human Services, 1994), school-based interventions alone may not be sufficient to reduce smoking among this age group (Lynch and Bonnie, 1994; US Department of Health and Human Services, 1994; Lichtenstein, 1995; Peterson et al., 2000). One channel as yet unexplored for behavioral interventions targeting tobacco control for youth is the worksite. Being employed and working long hours are associated with increased risk of smoking among adolescents (Greenberger et al., 1981; Stanton et al., 1994; Bachman and Schulenberg, 1993). Fifty-seven percent of youth begin some type of work activity by age 14 and this percentage increases with age (US Department of Labor, 1999). By the time they have graduated from high school, 80% of teens have held at least one job. Fifty-two percent of employed youth work in retail settings, which includes restaurants, department stores and grocery stores (National Research Council, 1998). Innovative approaches in the worksite setting may increase the effectiveness of tobacco use prevention and cessation efforts with youth. This paper describes the SMART Teens Against the Risks of Tobacco study, which was designed to develop and conduct a preliminary test of a worksite-based tobacco control intervention for employed youth. The purposes of this manuscript are to: (1) describe the formative research process that was used to develop a tobacco control intervention for working teens, and (2) present the results of formative research and discuss their implications for intervention development. Specifically, this paper reports on the focus group and cross-sectional survey data collected prior to intervention development and delivery in the grocery store setting.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Study design
This phase II methods development study was based on the National Cancer Institute’s methodological framework for cancer control research (Greenwald and Cullen, 1984
; Flay, 1986). Following guidelines for phase II studies, we used formative research to develop and explore new approaches to health promotion, systematically assess the needs of working adolescents, identify intervention objectives to support behavior change, and estimate the efficacy and feasibility of a worksite-based tobacco control intervention for working teens. If preliminary findings suggest that intervention methods are feasible in the worksite setting, the lessons learned may inform a full- scale phase III efficacy assessment of a worksite-based tobacco control intervention for employed adolescents.
Study setting
SMART was implemented in ten grocery stores from a single corporation in the Boston, Massachusetts metropolitan area. We selected grocery stores as the setting for this study because they are one of the largest employers of teens (US Department of Labor, 2000). To be eligible for the study, stores were required to employ at least 40 teens.
Formative research data were collected from interviews, workshops, advisory boards, a review of existing literature and health education materials, focus groups, and a cross-sectional survey. Focus groups and cross-sectional data were the primary data sources used to develop the intervention objectives; data from other sources were used to verify, support and refine intervention and evaluation design, intervention messages, and the development of peer leader training curricula. Based on all methods, we identified and adapted intervention methods that were suitable for employed teens. This paper reports on data from focus groups and the survey.
Focus groups
Data were collected from focus groups to define constructs for inclusion in the cross-sectional survey and to help develop intervention objectives and messages. We conducted eight focus groups with 41 employed teens, including two groups with teens who had never smoked, two groups with teens experimenting with smoking, one group with smokers and three groups of mixed smoking status. Data were collected on the beliefs and attitudes about smoking, the role of smoking, social influences to smoke and quit, employment experiences, sources of stress, coping with stress, quitting smoking, sources of social support, and logistics on survey implementation. Audiotapes of the focus groups were transcribed and analyzed to identify generative themes.
Employed Youth Survey (EYS)
We conducted the EYS among teens to assess their smoking patterns and work characteristics.
Data collection
The EYS was administered between October and December 1998. Teens aged 15–18 who worked an average of at least 5 hours per week in one of the 10 study sites were eligible to participate in the survey. We trained research staff and store contacts (i.e. store managers and/or adult employees) to administer the surveys following standard procedures. During the 2-month administration period, 456 eligible teens received the survey and 83% completed it (n = 379; range 50–98% by store).
Measures
Items were derived from existing youth surveys, suggestions from other researchers and findings from our qualitative research. We used five measures of cigarette smoking that were comparable with data from the Youth Risk Behavior Survey (YRBS) (Kann et al., 2000), including: lifetime cigarette use (ever tried cigarette smoking, even one or two puffs), lifetime daily cigarette use (ever smoked one or more cigarettes per day for a whole month), current cigarette use (smoked cigarettes on 1 day or more of the 30 days preceding the survey), frequent cigarette use (smoked cigarettes on 20 days or more of the 30 days preceding the survey) and smoking more than 10 cigarettes per day.
Sociodemographic and employment measures included age, race/ethnicity, school attendance, average grade in school, future educational plans, educational status of parents, hours worked per week, time spent on the job, work shifts, expected duration of employment and main reason for working. We assessed the work experiences of teens in order to identify and examine work-related predictors of smoking behavior.
Analysis
We selected the dichotomous measure, whether the teen had smoked one or more cigarettes in the last 30 days, as our primary measure of smoking status and explored associations of this measure with the other teen characteristics. For bivariate associations we used cross-classification and the 
2-test of homogeneity. For multivariable associations we used multiple logistic regression analysis. Variables were included in the multiple regression analysis if they were statistically significantly associated with smoking status at the 10% significance level. Measures no longer significantly associated with smoking status (P < 0.05) were removed from the analysis. Interaction effects of the remaining measures were then tested for statistical significance. All analyses were carried out using the personal computer version of SAS (release 8.01; Cary, NC) statistical software.
Because teens were recruited for the survey in 10 participating stores, we tested the assumption that there was no significant intra-class correlation of the smoking measures with store (data not shown). Although we found significant differences by store in gender and race distributions, there was no significant intra-class correlation of the association of these characteristics and smoking by store. Therefore, for the purpose of these analyses we ignored the clustering of teens in stores. This approach meant that each teen had equal weight in the analysis, as compared to each store having equal weight. We felt that this decision reflected the appropriate weighting for the analysis of factors associated with smoking of individual teens.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Focus group results
We identified the following major themes in our focus group analysis: teen employment experiences, sources of stress, the functional meaning of smoking, beliefs and attitudes, and social influences at work. Adolescents reported frustrations with inadequate break time and scheduling of inconsistent and minimal hours. Teens indicated that on their 15-min breaks they eat and socialize with other co-workers; some teens reported smoking on their breaks. Adolescents indicated that parents were the most consistent source of general stress and customers were the most consistent source of work-related stress. Teens in general reported positive relationships with co-workers who were not their supervisors.
Many teens reported that their first experiences of smoking were related to peer pressure and, after initiating smoking, stress was cited as a reason for currently smoking. Non-smokers mentioned that they felt ‘weird’ in a setting with smokers, while experimenters reported that ‘it was not an issue’ to be the only one not smoking. Teens stated that they liked the ‘rush’ they felt after smoking. Most teens felt that they had to quit smoking at some point in the future because of the health hazards and that they could quit on their own using cold turkey methods or gradually cutting back until they quit.
Teens reported that a high proportion of their adult co-workers smoked. When asked if most people smoke at work, one female respondent stated: ‘It’s mostly older people [our] parents age...people who work in the daytime, who’ve been here forever’. Teens mentioned that adults at work do not reprimand them for smoking.
Results from focus groups helped confirm the need to incorporate measures of work-related variables, social influences, social support from co-workers, worksite social norms and stress variables into the EYS. For examples of objectives, see Table V.
|
EYS results
Demographic characteristics of adolescent workers
Table I summarizes the demographic and related characteristics of respondents.
|
Smoking habits of teens
We compared the smoking patterns of teens in this sample with Massachusetts high school students responding to the 1999 YRBS (Kann et al., 2000
), administered between February and June 1999, in randomly selected public high schools around the state. The survey was completed with 4415 students, representing 79% of students enrolled in the classes originally selected. As shown in Table II, results from the two surveys indicate that the employed teens in this study have patterns of cigarette use similar to Massachusetts teens overall.
|
Employment and demographic characteristics by smoking status
We examined respondent characteristics by current cigarette use (smoked cigarettes on at least one day in the last 30 days), as shown in Table III. Girls were more likely to be current smokers than boys. Smoking frequency increased as grades decreased. Teens who reported that they spent most of their free time with school friends were least likely to have smoked in the last month, whereas those who reported spending time with friends who were neither school nor work friends were most likely to have smoked.
|
To explore the independent association of the measures with current smoking, we computed a logistic regression analysis including gender, average grade in school, who teens spend time with, race/ethnicity, expected duration of employment and reason for working. When all six measures were included in the analysis, main reason for working was no longer associated with smoking. None of the interaction effects among the explanatory variables was associated with current smoking. The final model (Table IV) indicates that girls and teens of white, non-Hispanic ethnicity were more likely to smoke, while teens with higher grades and those who hang out with school or work friends were less likely to smoke. Finally, teens who expect to remain employed in the grocery store less than 1 year were also more likely to smoke.
|
Incorporation of survey results into intervention activities
To incorporate survey results into the design of our intervention activities we used the intervention planning process suggested by Perry et al. (Perry et al., 1999
) to (1) identify factors predictive of teen smoking, (2) formulate intervention objectives that describe how the intervention would change the predictive factors, and (3) plan intervention activities that have been shown to be associated with behavior change in teens, attract teens to participate and are feasible in the grocery store environment. Table V summarizes this planning process and gives examples of ways in which we incorporated EYS data into the design of intervention activities. For example, because 44% of teen respondents over-estimated the percentage of youth their age who smoke, we created ‘weird facts table tents’ for store breakrooms that reported, among other facts suggested by teens, the actual percentage of employed teens and Massachusetts teens who smoke as 16%. Teens reported support from their friends for non-smoking; 70% of teens report that their best friends would disapprove of their smoking and 44% said that their friends would help them quit. We created bulletin board displays on ways teens could support their friends in their quit smoking attempts. To create the awareness that teen smokers, although they have not been smoking as long as many adults, can become addicted to nicotine, we included facts about signs of nicotine addiction on table tents, bulletin boards and discussed addiction at teen advisory board meetings. See Table V for additional examples of ways in which we incorporated formative data into intervention design. |
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This phase II methods development study used formative research to develop a tobacco control intervention for working teens. Results from the focus groups confirmed information from the literature on teen attitudes and beliefs about smoking, the relationship between stress and smoking, and social influences for smoking. These formative results also generated new information about teen worksite experiences, worksite sources of stress and how to structure interventions so that employed teens are exposed to tobacco control messages.
The worksite’s social milieu is relevant to an increasing number of teens; the workforce of youth ages 16–19 is expected to increase by 14% between 1996 and 2006 (US Department of Labor, 1999). Data collected from teens employed in the 10 grocery stores participating in the EYS indicate that the smoking patterns of these teens are fairly comparable to Massachusetts high school students, using comparisons with 1999 YRBS data (Kann et al., 2000). Contrary to our expectations, these smoking rates were not higher than Massachusetts high school students overall. Rather than reaching teens who had dropped out of school, 98% of our sample was still enrolled in school. Worksites may provide an important supplemental intervention channel to complement school- and community-based tobacco control efforts with youth.
An important step in phase II studies is the assessment of the feasibility and acceptance of the intervention. This is illustrated by the iterative process we used to develop intervention activities that were salient to teens. For example, we found that the peer leader model was difficult to implement in the worksite setting for several reasons, including the fact that most teens’ close friendship groups which are conducive to peer influence were with school friends and not co-workers and because the small number of peer leaders that was feasible in the each store was not enough to foster group cohesion. Therefore, we changed our approach and formed teen advisory boards with seven to 10 teens and focused on their primary roles as advisers for those who were uncomfortable delivering intervention messages one-to-one. In addition, when we observed that teens were not taking educational materials from the wall pockets in the break rooms, we switched to using plastic table tents containing information in an interview format and positioned as ‘weird facts’ that we placed on tables throughout the break rooms.
Our experiences in the grocery store setting provide information about potential challenges to as well as some of the advantages of delivering tobacco control interventions with teens in worksites. The majority of teens have a relatively short tenure with an individual employer. Among 16 and 17 year olds, 78% worked 12 months or less for a particular employer, and 10% worked between 13 and 23 months (US Department of Labor, 1998).
Because of their high turnover rates, management is likely to have a lower level of commitment to teen workers than to more stable adult workers. Recruitment of worksites to this study was difficult because employers were apparently reluctant to devote resources to this relatively unstable group of employees. This observation suggests that teen workers might benefit from worksite programs addressing tobacco control for adult as well as teen employees and that future programs include interventions with management to increase their level of support.
In addition, most adolescents work part-time. In 1996–1998, 15–17 year olds worked an average of 16.5 hours per week during the school months and 23 hours per week during the summer months (US Department of Labor, 2000). Part-time workers are likely to have less break time than full-time workers and as a result have fewer opportunities for participating in the intervention. Also, teens in these settings often worked varying shifts. Among 16–19 year old full-time workers, 8.8% work irregular shifts, compared to 3.8% of full-time workers 20 years of age or older (US Department of Labor, 1997). Their inconsistent shifts meant that teens did not form a highly cohesive group that may have enhanced the effectiveness of the peer model and offered a higher level of group support for participating in project activities. Despite the instability of teen workers, the influx of new employees offered an opportunity to provide at least minimal exposure to the intervention to a large number of teens over the course of 12 months.
Several limitations suggest caution in the interpretation of these results. These data were based on self-report of smoking status. Also, the work experiences of teens employed in grocery stores may not be representative of all working teens, who may be employed in a range of other settings.
Contrary to our a priori hypothesis, smoking rates among employed youth in this study were not higher than statewide averages and most of the teen workers were still in school, indicating that worksite interventions, at least in this setting, represent an alternative or adjunct to school-based programs, but do not necessarily capture a unique population. Employed teen tobacco use patterns and work characteristics that emerged from our formative research are presented in this paper, and may be useful in planning future worksite interventions for employed teens.
|
Acknowledgements |
|---|
The authors are grateful to the investigators and staff who participated on the project, including Gina Escamilla, Chris Grasso, Dana Jessup, Kerry Kokkinogenis, Hye-Seung Lee, Ruth Lederman, Stefania Maggi, Richard Martins, Rachel Noriscat, Anil Pillay, Steve Potter, Lois Rasmussen Norstrom, Prabhjyot Singh, Dana Spain, Evelyn Stein, Rosemary Thom, Travis Trammell, David Wilson and Kathleen Yaus. The authors would also like to thank the Scientific Advisory Board for their contributions to the development of the assessment tools and other study components, including Drs J. Allan Best, Graham A. Colditz, William DeJong, Steven L. Gortmaker, Nancy Rigotti, Michael Segal, Ellen Frank, Judy Foley and Jeanne M. Medas. This work would not have been possible without the collaborative efforts of the upper management, store managers, and teen employees of the worksites that participated in this study. This study was supported by the National Institute of Nursing Research and the National Cancer Institute, grant no. R01 NR04748. This study was reviewed and approved by the Institutional Review Board of the Dana-Farber Cancer Institute.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Bachman, J.G. and Schulenberg, J. (1993) How part-time work intensity relates to drug use, problem behavior, time use and satisfaction among high school seniors: are these consequences or merely correlates? Developmental Psychology, 29, 220–235.[CrossRef]
Centers for Disease Control and Prevention (2001) CDC Surveillance Summaries, Youth Tobacco Surveillance, United States, 2000. Morbidity and Mortality Weekly Report, 50, SS-4.
Flay, B.R. (1986) Efficacy and effectiveness trials (and other phases of research) in the development of health promotion programs. Preventive Medicine, 15, 451–474.[CrossRef][Web of Science][Medline]
Greenberger, A., Steinberg, L.D. and Vaux, A. (1981) Adolescents who work: health and behavioral consequences of job stress. Developmental Psychology, 17, 691–703.[CrossRef]
Greenwald, P. and Cullen, J.W. (1984) A scientific approach to cancer control. Cancer, 25, 236–244.
Kann, L., Kinchen, S.A., Williams, B.I., Ross, J.G., Lowry, R., Grunbaum, J.A. and Kolbe, L.J. (2000) Youth risk behavior surveillance—United States, 1999. Morbidity and Mortality Weekly Report, 49, 1–32.
Lichtenstein, E. (1995) Behavioral research contributions and needs in cancer prevention and control: tobacco use prevention and cessation. In NCI/National Cancer Advisory Board Workshop on Behavioral Research in Cancer Prevention and Control, Oregon Research Institute, Eugene, OR.
Lynch, B. and Bonnie, R. (1994) Growing Up Tobacco Free. National Academy Press, Washington, DC.
National Research Council (1998) Protecting Youth at Work: Health Safety and Development of Working Children and Adolescents in the United States. National Academy Press, Washington, DC.
Perry, C.L. (1999) Creating Health Behavior Change: How to Develop Community-wide Programs for Youth. Sage, Thousand Oaks, CA
Peterson, A.V.J., Kealey, K.A., Mann, S.L., Marek, P.M. and Sarason, I.G. (2000) Hutchinson Smoking Prevention Project: long-term randomized trial in school-based tobacco use prevention, results on smoking. Journal of the National Cancer Institute, 92, 1979–1991.
Stanton, W.R., Oei, T.P. and Silva, P.A. (1994) Sociodemographic characteristics of adolescent smokers. International Journal of Addictions, 29, 913–925.[Web of Science][Medline]
US Department of Health and Human Services (1994) Preventing Tobacco Use among Young People: A Report of the Surgeon General. US Department of Health and Human Services, Public Health Service, Centers for Disease Control and Preventive, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health, Atlanta, GA.
US Department of Labor (1997) Workers on Flexible and Shift Schedules in 1997. US Department of Labor, Bureau of Labor Statistics, Washington, DC.
US Department of Labor (1998) Employee tenure in 1998. US Department of Labor, Bureau of Labor Statistics, Washington, DC. Available at: http://stat/bls.gov/newsres. htm (accessed 23 February, 2004).
US Department of Labor (1999) Civilian Labor Force. US Department of Labor, Bureau of Labor Statistics, Washington, DC.
US Department of Labor (2000) Report on the Youth Labor Force. US Department of Labor, Washington, DC.
Received on May 22, 2001; accepted on February 20, 2002
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||