Development of a Tailored, Internet-based Smoking Cessation Intervention for Adolescents


  • Tracy Meis M.S.,

    Corresponding author
    1. Senior Research Specialist for the Center for Health Systems Research and Analysis at the University of Wisconsin-Madison.
      Address: CHESS Project, 1130 WARF Building, 610 Walnut St., Madison, WI 53705-2397.
    Search for more papers by this author
  • Martha Gaie M.S., R.N.,

    Corresponding author
    1. Doctoral student at the University of Wisconsin-Madison and a Research Assistant for the Center for Health Systems Research and Analysis.
      Address: CHESS Project, 1131 WARF Building, 610 Walnut St., Madison, WI 53705-2397.
    Search for more papers by this author
  • Suzanne Pingree Ph.D.,

    Corresponding author
    1. Professor and Chair of Family and Consumer Communication at the University of Wisconsin-Madison.
      Address: 440 Henry Mall, University of Wisconsin, Madison, WI 53705-2397.
    Search for more papers by this author
  • Eric W. Boberg Ph.D.,

    Corresponding author
    1. Associate Researcher and the Project Director of CHESS at the Center for Health Systems Research and Analysis at the University of Wisconsin-Madison.
      Address: CHESS Project, 1122 WARF Building, 610 Walnut St., Madison, WI 53705-2397.
    Search for more papers by this author
  • Christi A. Patten Ph.D.,

    Corresponding author
    1. Assistant Professor of Psychology at the Mayo Medical School in Rochester, Minnesota.
      Address: Nicotine Research Center, Mayo Clinic, 200 First St. SW, Rochester, MN 55905.
    Search for more papers by this author
  • Kenneth P. Offord M.S.,

    Corresponding author
    1. Co-Director for Research at the Nicotine Dependence Center at Mayo Clinic in Rochester, Minnesota.
      Address: Nicotine Research Center, Mayo Clinic, 200 First St. SW, Rochester, MN 55905.
    Search for more papers by this author
  • Kelly L. Berry B.A.,

    Corresponding author
    1. Lead graphic designer for the CHESS project at the Center for Health Systems Research and Analysis at the University of Wisconsin-Madison.
      Address: CHESS Project, 5 WARF Building, 610 Walnut St., Madison, WI 53705-2397.
    Search for more papers by this author
  • David H. Gustafson Ph.D.

    Corresponding author
    1. Professor in the Department of Industrial Engineering and the Department of Preventive Medicine at the University of Wisconsin-Madison. He is the founder of the Center for Health Systems Research and Analysis.
      Address: CHESS Project, 1119 WARF Building, 610 Walnut St., Madison, WI 53705-2397.
    Search for more papers by this author

Address: CHESS Project, 1130 WARF Building, 610 Walnut St., Madison, WI 53705-2397.

Address: CHESS Project, 1131 WARF Building, 610 Walnut St., Madison, WI 53705-2397.

Address: 440 Henry Mall, University of Wisconsin, Madison, WI 53705-2397.

Address: CHESS Project, 1122 WARF Building, 610 Walnut St., Madison, WI 53705-2397.

Address: Nicotine Research Center, Mayo Clinic, 200 First St. SW, Rochester, MN 55905.

Address: Nicotine Research Center, Mayo Clinic, 200 First St. SW, Rochester, MN 55905.

Address: CHESS Project, 5 WARF Building, 610 Walnut St., Madison, WI 53705-2397.

Address: CHESS Project, 1119 WARF Building, 610 Walnut St., Madison, WI 53705-2397.


Computer-assisted interventions represent an innovative approach to adolescent smoking cessation that may offer advantages over traditional smoking interventions in their potential to provide adolescents with a variety of appropriate cessation-related activities, as well as interactive feedback tailored to their developmental, psychosocial, behavioral, and biological needs. As part of a larger project to evaluate the effectiveness of an Internet-based approach to smoking cessation among adolescents, this paper will describe the intervention development process. Stomp Out Smokes (S.O.S.), an Internet-based information and support system, was created to address the specific needs and experiences of adolescents who want to quit smoking. The development of S.O.S. involved an iterative process with five distinct yet often overlapping phases: Phase 1: review of the adolescent development, smoking cessation, and health literature; Phase 2: development and implementation of a needs assessment; Phase 3: construction of a site development strategy; Phase 4: development, review, and revision of content; and Phase 5: development of the website architecture and graphic design.


Adolescent smoking is a major public health concern (CDC, 1998). In addition to the estimated 3000 youth who begin using tobacco each day, many adolescent smokers express the desire to quit, but find it difficult to do so. One study reports that nearly half of high school seniors who smoke say they would like to quit; of them, nearly forty percent had tried unsuccessfully to quit (Perry, 1999).

Despite the mean age of initial tobacco use of 10.7 years in boys and 11.4 in girls, until recently, very little work has been done to provide intervention for younger smokers (Brownson et al., 1991). Targeting adolescents for cessation programs as early as age 11 may help minimize their lifetime exposure to and damage from tobacco.

The few interventions that have been studied for adolescent smokers have not addressed their unique developmental needs, nor have they been developed specifically for them. Recent studies testing pharmacological interventions among adolescent smokers found that adolescents exhibit lower abstinence rates than adults who are given the same interventions (Smith et. al., 1996; Hurt et. al., 2000). This suggests that quitting smoking is a different experience for adolescent smokers and that research on effective interventions tailored to adolescents is needed to learn how to help them stop smoking.

A comprehensive search of the literature on behavioral treatment of smoking among adolescents (Sussman, Lichtman, Ritt, & Pallonen, 1999) yielded 17 cessation studies based on a wide range of theoretical perspectives. Most of the interventions were based on adult models and were not tailored to adolescents. In addition, the majority were conducted in school settings. However, since adolescent smokers are more likely than nonsmokers to be dropouts or absent from school (Charlton & Blair, 1989; Pirie, Murray, & Luepker, 1988), school-based studies may fail to reach a substantial portion of smokers. In addition, these studies had several methodological limitations including use of single group or quasi-experimental design and small sample sizes. These research findings, and more specifically, the gaps in this body of research led us to develop an alternative approach to adolescent smoking cessation: an Internet-based intervention. A unique and important aspect of this intervention is that its development was driven by research exploring the specific needs of adolescents who were either current or former smokers.

With funding from the National Cancer Institute, smoking cessation experts at the Mayo Nicotine Research Center (Patten et al., 2001) teamed with Comprehensive Health Enhancement Support Systems (CHESS), acknowledged leaders in the field of interactive health communications (Slack, 1999), to create a CHESS-based module for adolescent smoking cessation. This module was named Stomp Out Smokes or S.O.S. CHESS has created tailored, computer-based health information and support programs based on the identified needs of the target population and theoretical models of crisis, coping, learning, and behavior change theories that are among the only systems of this sort to have shown positive outcomes in randomized controlled trials (Gustafson et al., 1999; Gustafson et al., 2001).

An Internet-based intervention has the potential to provide adolescents with a variety of appropriate cessation-related activities, as well as interactive feedback tailored to their developmental, psychosocial, behavioral, and biological needs. Even more importantly, it can give the adolescents some control over treatment decisions.

As a medium, the Internet has several characteristics that make it especially promising for an adolescent smoking cessation intervention (Borzekowski & Rickert, 2001). First, its interactive nature permits immediate, individualized feedback, which allows for tailored information and group discussions, qualities that are associated with interpersonal encounters. It also can deliver information to masses of individuals quickly, efficiently, and on demand, which gives it the capacity to provide in-depth information and the combination of audiovisual and textual content. In addition, the Internet shares with print media the ability for users to refer back to information found previously, 24-hour access, and nonlinear control over navigation through the material. All of these characteristics facilitate greater user involvement, and thus effect a more active audience than do traditional mass media (Street & Rimal, 1997). Furthermore, since an Internet-based intervention can be delivered outside of traditional health care settings and instead be used in the home or school, there is also the potential to increase access to health care services for those with geographic or financial limitations.

In order to take advantage of the many promising characteristics of the Internet as a medium while creating a behavior-change program based on a theoretical framework, we followed a carefully planned development process.

Development Approach

The development of S.O.S. involved an iterative process with five distinct yet often overlapping phases: Phase 1: review of the adolescent development, smoking cessation, and health literature; Phase 2: development and implementation of a needs assessment; Phase 3: construction of a site development strategy; Phase 4: development, review, and revision of content; and Phase 5: development of the website architecture and graphic design. Even when the site was first “complete,” we anticipated changes; in practice, the site is currently subject to continual review and revision.

The tremendous ease of publishing on the Internet (in comparison with other forms of publication) is what allows us to use this iterative approach, which is impractical with other media. This is not to suggest, however, that additions and changes are made frivolously; they are made only when there is compelling new information that is evaluated as being more beneficial to our audience than what is currently published. We consult the literature whenever questions arise that are not adequately covered in the material previously reviewed. Relevant bodies of literature are constantly monitored for new developments regarding all elements of our project, including user needs, treatments, and design, and relevant information is incorporated in our intervention as soon as it is prepared for publication.

Phase 1: Review of the Literature

We began the site development process with an in-depth review of the literature on adolescent development, health, and smoking behaviors, which yielded important information related to the unique developmental, psychosocial, biological, and behavioral factors that play a role in the adolescent smoking experience.

Developmental Issues. One aspect of smoking in adolescents that makes it particularly difficult to treat is that smoking can play a role in fulfilling key developmental needs of adolescence, including establishment of autonomy, independence, intimacy, and identity (Perry, 1999), as well as bonding with peers, maturity, and having a positive social image (CDC, 1994). Understanding and acknowledging the importance of these needs allows us to address them and attempt to provide alternatives to satisfy the underlying developmental needs of adolescent smokers.

In order to understand fully the needs met through the adolescent smoking experience, we must first ask the question: in the eyes of the adolescent, what value is produced by smoking? (Firat & Venkatesh, 1995). In a series of nine focus groups of adolescent smokers between the ages of 15 and 18 years, Balch (1997) found that the most important perceived benefits of smoking were mood control, social bonding and companionship, a more mature or “cool” appearance, and weight control. Perceived disadvantages of smoking were financial costs, conflicts with parents, social hassles, and pressure from nonsmoking friends. Balch also found that adolescent smokers prefer to be involved in their own treatment decisions, set their own goals, and progress at their own pace. Aspects of smoking cessation interventions that adolescents perceived as negative were preaching and nagging.

A study by Williams, Cox, Kouides, & Deci (1999) confirms the importance of adolescents' involvement in their own treatment in their findings related to an autonomy-supportive style of interaction. According to the autonomy-supportive style, an authority figure, such as a physician or counselor, takes into account the adolescent's perspectives when providing relevant information, offers choices and minimizes pressure, and encourages them to accept more responsibility for their own behavior. This study found that when adolescents perceived messages about not smoking as autonomy supportive, they had more autonomous motivation for not smoking, which in turn predicted a decrease in self-reported smoking.

Psychosocial Factors. Peer influence appears to be a significant predictor of the initiation and maintenance of adolescent smoking (Botvin et al., 1992; Chassin et al., 1991; Stanton & Silva, 1991; Swan, Creeser, & Murray, 1990). Similarly, adolescents who stop smoking experience effects on their social environments that they often perceive as negative, including a decline in the number of smoking friends (Chassin et al., 1994). Conversely, positive peer support can have a beneficial effect on a young person's health. According to Turner (1999), peer-led initiatives inherently acknowledge young people's skills and abilities and their constructive role in the solution to problems. In fact, several studies have shown that peer initiatives can improve knowledge, change attitudes regarding health-related behavior, and improve both self-esteem and self-efficacy, all of which are important factors in obtaining and maintaining a healthy lifestyle (Bernard, 1991). Based on normal adolescent psychosocial development, and on current research on adolescent smoking behaviors and theories, a smoking cessation intervention that involves the use of peer support should contribute to its effectiveness (Albrecht, Payne, Stone, & Reynolds, 1998).

In addition to social influences, an adolescent's psychological well-being appears to influence smoking behaviors. Several studies have shown that positive outcome expectations for the effects of smoking (e.g. concentration and improved affect) are related to the persistence of smoking (Hansen et. al., 1985; Skinner et al., 1985; Wills & Shiffman, 1985). According to a Centers for Disease Control report (1994), a substantial proportion of adolescents report smoking to reduce stress and negative affect. A strong link has also been found between smoking and beliefs about its consequences (Hansen et al., 1985; Skinner et al., 1985; Wills & Shiffman, 1985), as well as perceived ability to cope with a difficult situation (Ellickson & Hays, 1992). Adolescents low in personal resources such as self-esteem, mastery, or social support may turn to cigarette use because it is the only means available to cope with stress (Pederson et al., 1997; Wills & Shiffman, 1985). Researchers have also reported links between cigarette smoking and anxiety (Patton et. al., 1996) as well as depression in adolescents (Choi, Pattten, Gillin, Pierce, & Kaplan, 1997).

In addition, Sussman et al. (1998) concluded that the most effective approaches in reducing cigarette smoking among adolescents included cognitive-behavioral components, such as instruction in coping skills, that can help adolescents find alternative ways to cope with negative affect. Wills came to a similar conclusion (1986) upon observation that certain coping strategies (adult support, behavioral coping, and relaxation) resulted in less smoking.

Biological Factors. Biological factors have also been shown to affect the smoking status of young smokers. Like adult smokers, adolescent smokers develop nicotine dependence and exhibit withdrawal symptoms when they try to stop smoking (McNeil, West, Jarvis, Jackson, & Bryant, 1986; Smith et al., 1996; Stanton, Lowe, & Silva, 1995). Researchers administering the Fagerström Tolerance Questionnaire and the Fagerström Test for Nicotine Dependence with adolescent smokers have found scores comparable with those of adult addicted smokers (Prokhorov, Pallonen, Fava, Ding, & Niaura, 1996; Rojas et al., 1998; Smith et al., 1996). A survey by Stone and Kristeller (1992) found that 28% of daily smokers (those who smoked at least one cigarette per day for the last month) reported nicotine addiction as the most important reason they continued smoking. Recent research suggests that addiction to nicotine may occur even before an adolescent becomes a daily smoker (DiFranza et al., 2000).

Behavioral Factors. Smoking cessation in adolescents is also influenced by behavioral factors, such as readiness for change and motivation to change. Readiness for change refers to the causes, considerations, reasons, and intentions that move an individual to perform certain behaviors or sets of behaviors like quitting smoking (DiClemente, 1999). Varying degrees of motivation imply that there are several stages of behavior change, such as those identified by Prochaska and DiClemente (1992) in their transtheoretical model (also know as the Stages of Change model). The transtheoretical model, which has been applied in various health behavior contexts, including smoking cessation, conceptualizes behavior change as a five-stage process (Maibach & Cotton, 1995). These stages were adapted for use in the context of adolescent smoking (see Table 1).

According to this model, behavior change requires stage-specific interventions, based on Bandura's social cognitive theory (SCT), to motivate an individual to move from one stage to the next. SCT indicates that behavior is determined reciprocally by both internal personal factors and the external living environment; in other words, individuals and their behavior are influenced by their surroundings, yet individuals also influence their surroundings through their behavior and expectations. According to Bandura (1997), perceived self-efficacy has a great influence on a person's motivation for making behavior changes. He notes that efficacy beliefs determine the height of goals people set for themselves, how much effort they are willing to exert, their staying power in the face of difficulties, and how formidable they perceive the obstacles to be (Bandura, 1999). In fact, Bandura (1997) perceives that self-efficacy affects every phase of change in substance abuse-the initiation of changes, their achievement, recovery from relapse, and long-term maintenance of abstinence.

According to Bandura (1997), high self-efficacy can be achieved through successful self-regulation. Successful self-regulators track their behavior and the cognitive and situational conditions under which they engage in it. They set proximal goals for exercising control over their behavior in the here and now. They draw from an array of coping strategies rather than relying on a single technique. They create motivating incentives to sustain their efforts. And they apply multifaceted self-influence consistently and persistently.

Maibach and Cotton (1995, p. 44) have prescribed specific strategies for improving perceived self-efficacy and moving an individual from one stage to the next which are based on SCT. Table 1 outlines the application of Maibach and Cotton's strategies to an adolescent smoking cessation context.

Summary of Literature. Our literature review on adolescent development and smoking cessation had substantial implications for the design of our website, including fundamental decisions on content and our approach to content development. Perhaps the most important overriding theme in the literature is that smoking is not an isolated act; rather, it is a complex and variable behavior that occurs in particular and dynamic developmental, psychosocial, biological, and behavioral contexts. Therefore, throughout the site development process, we attempted to consider the multitude of factors that play a role in adolescent smoking cessation and build an intervention capable of changing dynamically to address a user's needs over time and through the various stages of behavior change.

Phase 2: Develop, Administer, and Analyze Needs Assessment

To improve our understanding of adolescent smokers and their quitting experiences, we developed and administered a comprehensive needs assessment survey. This phase began with focus groups conducted at four geographically and ethnically diverse sites: Hartford, CT; Kansas City, MO; Madison, WI; and Rochester, MN. Focus group participants were adolescents ages 11 to 17 who were current or former smokers. Of the 37 adolescents, 59% were minorities, 50% were female, and 65% were current smokers.

The focus groups were conducted to gather information on the process of quitting as defined by current and former adolescent smokers. This information was used to build a quantitative needs assessment survey. The surveys were completed by 1305 adolescents at four sites: Hartford, CT; Kansas City, MO; Madison, WI; and Rochester, MN. Of the respondents, 46 were former smokers, 234 were current smokers, and 1025 were nonsmokers. In addition to answering survey questions, participants were asked what they considered the three most important questions or concerns they had when they quit or when they think about quitting. This open-ended item allowed participants to communicate what they thought about when thinking about quitting, without being prompted by preselected survey responses, which guided our decisions about relevant topics and main ideas for the website.

The following is a general description of the needs assessment results as they apply to website development. The specific methods for survey development and administration as well as an in-depth analysis of its results can be found in Pingree et al. (2001).

Responses from current and former smokers were considered most relevant for site development purposes. Our overall finding in the extant literature—that adolescent smokers and their smoking attitudes, beliefs, and behaviors are widely variable—was confirmed by the needs assessment. It is clear that not all adolescent smokers have the same needs, and those needs might change within the same individual as s/he continues on the path of adolescent development. More detailed analysis indicated that on 29 items, the data showed distinct differences between what current smokers reported would help them quit, and what former smokers indicated helped them quit. This suggests that those who have successfully quit may have gained some insight into the circumstances of their smoking and quit attempts that could be used to help current smokers quit.

Data from the needs assessment also showed distinct differences between the responses of smokers segmented by their stage of behavior change, which suggests that the Stages of Change model is an appropriate basis for the website structure. However, DiClemente (1999) found that although the stages represent a sequential pathway through the process of change, individual substance abusers do not usually negotiate the stages in a single sequential transition; recovery and movement through the stages are marked by regression and relapse, as well as recycling. This evidence suggests that a strictly linear structure is not appropriate for this site, an approach which can be easily accommodated via standard website structuring strategies.

Phase 3: Site Development Strategy

Based on the literature review, needs assessment, our considerable clinical and research experience with adolescent smokers (Patten et al., 2001), and our experience in developing interactive information, social-support, and problem-solving systems, we constructed our site development strategy. The strategy comprises underlying theory, message content, context and meaning, information architecture considerations, graphic design plan, and voice (point-of-view). The overall site development process led to the identification of some basic objectives:

  • include content that will be meaningful to an adolescent audience;
  • use language that appeals to and is understood by an adolescent audience;
  • provide information in a context that is appropriate for an adolescent audience;
  • organize the information in a nonlinear, hierarchical structure that is easy to navigate;
  • include graphic design elements that appeal to an adolescent audience;
  • establish a voice that is accessible to an adolescent audience.

It should be noted that this does not mean that all messages, design elements, and graphics will appeal to and/or be understood by all users; rather, the intent is that every adolescent who enters the site should be able to easily find something of interest somewhere in the site. Ultimately, certain content areas and messages will be more popular than others, but to the greatest extent possible, our main objective is to address the needs and interests of all users.

The process through which these objectives are met is continuous iteration of content development, modification, user evaluation, and feedback. Online comment and feedback mechanisms are in place in S.O.S. to facilitate continuous communication between users and developers. Users can communicate what they like and don't like at any time by writing in a specific section of the discussion group entitled Website Feedback or by writing in our Comments section. Developers can then make modifications based on the comments sent through these systems. By structuring an intervention that generates true dialogue between users and developers, we attempted to downplay adult-adolescent power differentials and alienation by focusing our attention on the issues adolescents were likely to face in their smoking cessation efforts.

Construction of the site development strategy also included drafting a basic site map to guide the initial creation of content and design. This site map detailed the elements that we expected to include in the site, although final decisions for specific content were based on findings from the literature, needs assessment results, and adolescent panel feedback. The site map provided a foundation for how specific content areas would be presented and eventually connect to one another to form a comprehensive smoking cessation program. This site map defined three major categories of services we hoped to provide through the content and design of the final product: interactive feedback, mechanisms for self-expression, and informational pieces.

Interactive feedback services include Quit Plan and Quit Notes, which offer ways to interact with the computer and allow for tailoring of an individual's quitting plan as well as tools for tracking an individual's quitting progress. These services allow users to self-assess where they are in their quit attempt and how they are doing at each login. The system then uses this information to guide the user to the content most likely to be relevant to that particular user and that particular time, whether it be an interactive exercise or an informational piece. The services also allow users to input personal information regarding their smoking habits and receive feedback based on their input. Users can review alternatives to smoking and tips for getting through all of the stages of quitting, select those that they find most appealing, and save them in their Quit Notes for future reference. Most importantly, these services allow users to revise and update any information they enter into the system or any information they save in their Quit Notes at any time, creating a system that can meet user needs that may evolve dynamically.

Services that allow for self-expression include an online Discussion Group, Live Chat, a personal journaling service, an art gallery where users can post their artwork, and an Ask an Expert service that allows adolescents to communicate with smoking cessation experts through private bulletin boards. These services not only allow the user to become involved with others going through the quitting process, but they provide an automatic mechanism for keeping the site up-to-date with real-life struggles and solutions experienced in the quitting process. In addition, the social contact, built-in accountability, and ongoing communication help to keep the interest of the user and promote ongoing use of the system.

Informational pieces provide accurate, up-to-date information on smoking and quitting and other topics relevant to an adolescent smoker. It should be noted that specific content areas often combined two or three of the major categories of services. Table 2 presents the site map which lists the content areas and shows which categories of services each content area provides.

Phase 4: Content Development, Review, and Revision

For all practical purposes, Phases 4 and 5 (Information Architecture Development and Graphic Design) occurred simultaneously. While content was being developed piece by piece, the planning and programming of the site architecture was also being developed using the draft site map to determine an inherently logical order and categorization of the elements that were to be included. During this highly creative process, sometimes the categories suggested a new piece of content, and sometimes the content suggested a new categorization scheme. Content and architecture development are excellent complements to the other, and developing them more or less simultaneously is a logical, dynamic, synergistic, and beneficial approach.

Once drafted, content and design ideas were presented, in written form, to adolescent panels from Madison, Rochester, and Hartford, who provided frank feedback on writing style, word selection, and concept presentation. A core group of approximately 17 adolescents was involved in the panels. It should be noted that the purpose of the panels was not to evaluate the effectiveness of the system as a whole, but rather to evaluate content and design choices. Evaluating the system for its effectiveness in smoking cessation is currently underway and will be discussed in future papers.

In the writing process, we made a concerted effort to follow additional guidelines that emerged from our team knowledge and interactions with our target audience:

  • Use colloquial language without trying to impersonate adolescents.
  • Use direct, clear language.
  • Use non-judgmental language.
  • Use pictures rather than text whenever possible.
  • Use specifics rather than abstractions.
  • Use an active voice both in sentence structure and ideas.
  • Simplify and contextualize the language.

Content development included extracting the primary implications of our literature review and needs assessment and finding ways to present the information with appealing graphics, language and clarity. The following gives examples of how we applied our findings to create specific content areas.

1. Implications of developmental factors.

  • Acknowledge that adolescents perceive some aspects of smoking as beneficial and identify alternative sources for the perceived benefits of smoking.
  • Emphasize negative aspects of smoking as perceived by adolescents.
  • Avoid tactics that adolescents perceive as negative such as preaching and nagging.
  • Allow adolescents autonomy in cessation process and decisions.

Allowing adolescent users autonomy in their quit attempt is achieved through the S.O.S. Quit Plan and personalized Quit Notes (e.g. “Steve's Quit Notes”). The Quit Plan is an interactive component of S.O.S. that allows the user to choose a quit date and then offers, on a daily basis, different tailored information and links to specific S.O.S. content relevant to planning their quit attempt. The system also allows users to track their progress and receive tailored tips and problem-solving strategies for up to six weeks following their quit date. Unsuccessful quit attempts (based on user self-report) trigger supportive tailored feedback and an opportunity to review possible reasons for the slip or relapse and set a new quit date. This component of the system provides some guidance while at the same time allowing users to progress at their own pace and design their own quit plan. Figure 1 is an example of a Quit Plan screen.

Quit Notes let adolescent users tailor their smoking cessation intervention to their unique needs. These tailored Quit Notes allow users to select and store (and retrieve with a single mouse click) only those quitting tips, coping strategies, names of supporters, and planning strategies that are meaningful to them. Users can find information to personalize and place in their Quit Notes throughout the various content areas of S.O.S. All interactive content can be revised and/or updated at any time. See Figure 2 for an example of a Quit Notes screen.

Content is provided to address the aspects of smoking that adolescents may perceive as beneficial. For example, if an adolescent smokes to portray a certain image, information in the section “Me as a Nonsmoker” can help them find ways to portray that image without smoking. For adolescents who have not identified their motivation for smoking, there is an interactive component that allows them to answer questions about their smoking, identify their smoking behaviors, and consider their motivations for smoking. It also helps them to identify the negative consequences of their smoking and offers positive alternatives to smoking based on their input.

Research with adolescent smokers has shown that a significant predictor of planning to stop smoking or having actually stopped is believing that secondhand smoke harms nonsmokers (Glantz & Jamieson, 2000). This negative aspect of smoking is emphasized in content on the health effects of smoking and secondhand smoke. Figure 3 is a screen from the section on the health effects of smoking.

The financial cost of smoking, also cited by adolescents as a negative aspect of smoking (Balch, 1997), is discussed in a section that automatically calculates the cost of their smoking habit, and provides a list that includes popular items that cost approximately the same amount as the adolescent is spending on cigarettes.

2. Implications of psychosocial factors.

  • Reinforce positive peer influences; suggest ways to resist negative peer influences.
  • Include content on how to cope with negative affect and stress.
  • Provide content about smoking beliefs and consequences.

Several sections of S.O.S. encourage adolescents to support each other in their quit attempts. An online support group, as well as Live Chat, are available for adolescents to support each other through their quitting process. These services offer a venue for anonymous, non-threatening communication among adolescents facing similar issues and concerns. Adolescents can also support each other by sharing tips and suggestions for staying smoke-free that have worked for them. In addition, we have developed content to address dealing with others who smoke and dealing with supportive and non-supportive friends and family.

In focus groups conducted by Balch (1997), adolescents stated that they preferred advice about smoking from a successful quitter. S.O.S. contains Personal Stories, which are real-life accounts of people who have quit or are trying to quit smoking. These first-person accounts are of past or current smokers with different racial, socioeconomic, and educational backgrounds, and who have had both positive and negative experiences with different treatment options. These stories address barriers to quitting and how the barriers were managed, successfully or unsuccessfully. See Figure 4 for an example.

Content areas on reasons for smoking and reasons for quitting were developed to address the perceived benefits of smoking. For example, several sections of S.O.S. directly address how to deal with negative affect, including stress, depression, anger, and boredom, without smoking. These content areas give specific examples of typical situations adolescents often face and concrete suggestions for dealing with them. Interactive components on the site allow users to enter and save personal plans for dealing with negative affect. Also, the common belief among adolescents that cigarettes offer some physical benefits, such as increased concentration, increased energy and decreased agitation, is examined in a section on the health effects of smoking that explains temporary and lasting physiological effects of smoking. Content also addresses smokers' desire to control their weight and to create a desired image. Myth Quizzes provide a fun way for users to learn the true consequences of smoking on their health and relationships.

3. Implications of biological factors.

  • Address cravings and withdrawal symptoms that tend to occur during smoking cessation when a smoker is addicted to nicotine.

S.O.S. addresses the most common withdrawal symptoms reported by ex-smokers and offers ways to deal with them without smoking. These tips are offered throughout the site as well as automatically at login, at appropriate times, dictated by the user's quit date. Information on medications that can be used to help smokers overcome cravings and withdrawal symptoms, as well as how to use the medications properly, is provided. There is also a section dealing with cravings that discusses what to expect, as well as ways to get them under control. There is a comprehensive list of smoking substitutions, including things to do with one's hands and mouth when the urges strike. Adolescents can also enter their unique withdrawal symptoms, cravings, substitutions, and strategies for dealing with them, and save them in their personal Quit Notes to refer to when needed.

4. Implications of behavioral factors

  • Acknowledge that adolescents will be at different stages of change and provide content relevant to all stages.
  • Help adolescents become successful self-regulators to improve their self-efficacy and thus increase the motivation for moving through the Stages of Change with the goal of reaching the maintenance stage.
  • Provide concrete suggestions for making behavioral changes that will benefit smokers at all stages of change.

S.O.S. contains content for an adolescent at any stage in the quitting process from thinking about quitting to staying smoke-free. This is evident when you look at the side menu of any page of the site (see Figure 5). Relevant information and interactive components that offer tailored feedback are provided for every stage. Self-regulation is encouraged by daily tailored feedback provided by the S.O.S. system. It helps adolescents track their quitting process by giving them daily feedback specific to their situation, including the number of days they have been smoke-free, praise for success, support for setbacks, withdrawal symptoms they may be feeling and how to handle them, the positive effects quitting has had on their body so far, reminders of how they said they would handle negative affect, and rewards they chose for themselves to reinforce their successes. In addition, users who relapse have the opportunity to write about their relapse experiences and save them to refer to when needed. Users can also enter information into their Quit Notes such as smoking triggers to remind them of what to avoid (see Figure 6 for an example)

5. Implications from our needs assessment

  • Include content that touches on a wide variety of topics, issues, cessation tactics and strategies.
  • Present the point of view of former smokers.
  • Acknowledge and address the wide range of attitudes, beliefs, and behaviors of current smokers.
  • Include comprehensive information on strategies and methods for quitting.

Content covers a wide variety of topics related directly and indirectly to smoking. Some content areas are more comprehensive than others based on the findings of the needs assessment. For example, the information on various strategies used for smoking cessation and relapse prevention is very detailed because so much interest was expressed on this topic in the needs assessment. Other issues that were mentioned in the needs assessment, but with less frequency, are addressed with less detail.

Because the needs assessment showed a difference between the attitudes and beliefs of current and former smokers, S.O.S. provides the point of view of each. Adolescents can read about and learn from the experiences of former smokers in Personal Stories. They can also express their own attitudes and beliefs by displaying artwork, writing in a journal, and conversing with others through an online Discussion Group. Common issues and concerns of current smokers are also addressed by an extensive list of expert-answered Frequently Asked Questions (for an example of a FAQ see Figure 7).

Although the needs assessment identified differences in needs by sex and race, this website does not segment the audience on a group level; rather, tailoring occurs on an individual level to better address the needs of all users.

Phase 5: Information Architecture Development and Graphic Design

Designing the S.O.S. website took place in three stages: 1) information architecture development, 2) interface design, and 3) content layout and structure programming.

Information architecture development. Creating the overall information architecture was the first step in making the site logically and consistently organized. The architecture is the backbone of the site design and is the map used to develop the interface of the website. Based on our findings that there are distinct differences between adolescent smokers from stage to stage, we chose Prochaska and DiClemente's Stages of Change model (1992) to provide a basic structure for site architecture. We organized the site using a hierarchical structure, while also considering how adolescents may look for information as they go through the quitting process. By looking at the navigation bar on the left of every screen or at the content area menu in the center of the screen, adolescents can get an overview of the content of the site and choose what section they wish to navigate (see Figure 5)

Our team of researchers, graphic designers, computer programmers, and adolescents then transformed written documents with important information into interesting, eye-catching, interactive web pages that would appeal to an adolescent smoker. Our adolescent panel and other adolescent reviewers were critical to this process, providing invaluable feedback.

Interface design. To design the overall look of the interface, graphic designers first reviewed existing websites with popular culture themes that were aimed at an adolescent audience. A primary consideration was getting and keeping the user's attention. We created prototype designs, including several color palettes, based on the popular culture influences we identified, which were then reviewed and critiqued by panels of adolescents. Through our research and feedback from the adolescent panels, we concluded that our audience prefers graphical elements such as mouse rollovers, videos, photos of other teens, with bits of animation and sound distributed throughout. We were also able to determine that they preferred icons and other visual elements to words, which led to our translation of much of the content into visual representations, such as use of a bomb to identify sulfur as one of the components of tobacco smoke (see Figure 8). Once final design decisions were made, we translated the documents and interface into html and dhtml code, active server pages, Flash, javascript, and style sheets. We then programmed the content of the website.

An important consideration in the design and programming process was download time; if the site lacks a cutting-edge multimedia design, adolescents may lose interest in the site, but if it takes too long to download, they will move on to another site (or non-S.O.S. activity) before it ever comes up.

Content layout and structure. Content was reviewed by designers and laid out in ways that truncated and summarized large textual documents. In some cases, content was transformed into multimedia presentations using Flash software to help keep the audience engaged. Programmers constructed interactive applications such as Myth Quizzes, Quit Notes, Quit Plan, Discussion Groups, Ask an Expert, Live Chat, Art Gallery, and a journaling system to encourage active participation with the website. Internal editing and testing of the site was done by staff writers, designers, and programmers.

As the web pages were developed, the adolescent panel met frequently to thoroughly examine and experiment with the system. Each individual section was rated by adolescents on the overall look, ease of use (navigation), and content. Adolescents were then given the opportunity to suggest changes and express any other comments. As previously mentioned, this was not intended to provide a thorough evaluation of the effectiveness of the system to help teens quit smoking. Our intent is not to test any one message, but rather the overall value of the intervention in behavior change. Also, individual use patterns will determine which content is presented to a user and that content will be tailored to fit with an individual user's situation. Therefore, the effectiveness of the website as a smoking cessation tool will be determined in a clinical trail which is currently underway.  Results will be presented in a future paper.

In addition to the sections described, S.O.S. also offers the opportunity to incorporate new developments in the study of smoking cessation on an ongoing basis. The ability to respond to new developments in the field enriches the multidimensionality of the intervention and enhances the likelihood that S.O.S. users will find content in the site that resonates with their worldview and is compatible with their values and preferences.

Challenges, Limitations and Practical Considerations

Developing S.O.S. proved to be an especially challenging process for many reasons. First, it differs greatly from other CHESS modules. The greatest differences were a result of the audience we targeted.  S.O.S. is for people who want to change an addictive behavior, rather than those in a health crisis, who have been the target audiences of existing CHESS modules. Without an imminent health crisis, such as a diagnosis of breast cancer, to motivate use, emphasis was placed on the integration of motivational tactics to prompt use of the site.

The age of our targeted population (11 to 17 years) also presented a new challenge for CHESS developers. There was very limited research on smoking cessation interventions designed specifically for this age range to guide us. In fact, there were no previous studies specific to web-based interventions and those websites that were generally available on the Internet were not research-driven, not created for a young audience, not tailored for the individual, and/or not designed for sustained use through the quitting process. However, the characteristics of the target population did provide great latitude for design and structure of the site because, unlike other CHESS audiences, adolescents tend to be quite familiar with the Internet and computer technology. It also forced us to push the boundaries and create a more fresh, engaging, graphics-intensive and interactive program that could function within the constraints of the hardware available.

Another challenge was balancing the need to give users independence and control of their quitting process with our responsibility to provide protective measures for adolescents using the system. S.O.S. offers several avenues for adolescent discourse, dialogue, and expression, not only among themselves, but with smoking cessation experts and the site developers. Both unmediated and mediated opportunities for verbal interaction with others using the site are offered. Because adolescents have a higher degree of vulnerability due to their age and developmental status, it is necessary for trained facilitators to observe all communications that go through the system. Facilitators monitor exchanges for expression deemed inappropriate for the purpose of the site (e.g. planning or endorsing criminal activity, attempting to use it to sell goods or services, expressing statements that are demeaning to an individual or group, expression of suicidal or homocidal ideation or intent, etc.). In the S.O.S. module, facilitators included adolescents who have quit smoking and can more closely relate to the adolescents using S.O.S. The official policy of S.O.S. is to keep disruption of such interactions to a bare minimum and to always communicate through the adolescent facilitator.  Participants of the site are informed that all communication is being monitored. We recognize that although necessary, this could potentially inhibit their use of the intervention.

Ensuring participants' relative anonymity is an important consideration for the success of this type of intervention. Since this intervention is still in the testing phase, the website is only available to study participants. To ensure anonymity, adolescents choose a user name and password when they register with the study coordinator. Any information that can link a user to his or her identification is kept in a secure file. Participants are identified in the system only by their user name and there is no way for one user to link another user's name to the adolescent's true identity. Users are thus allowed to interact with others trying to quit smoking as well as experts in smoking cessation without being identified. Adolescents are also asked not to reveal their real names or contact information when using S.O.S. The ability to remain anonymous proves to be another advantage to an Internet-based intervention. It may, however be a challenge if S.O.S. becomes available to anyone on the Internet.

A practical consideration we took into account as we developed S.O.S. was that in order for an adolescent smoking cessation intervention to be truly effective it must be accessible and desirable in a natural setting. It is important that adolescents make the decision to seek help for smoking cessation and make the commitment to follow a quitting plan on their own. This had implications for our study design, but also challenged us to create an intervention that adolescents could and would seek out on their own, at any time, and feel at ease using without adult guidance. This limits our abilities to improve compliance to our intervention, but challenged us to find ways to capture our audience creatively.


Although studies of research-based Internet interventions for adolescent smoking cessation are scarce, our review of the literature and experience in adolescent smoking cessation and interactive health communications led us to believe an intervention of this sort has great potential. Previous work indicates that an effective adolescent smoking cessation intervention should acknowledge that smoking is not an isolated act and address the multitude of developmental, psychosocial, biological, and behavioral factors that play a role in adolescent smoking. The intervention should be based on the needs of adolescent smokers, have the ability to evolve dynamically as the individual's needs change, and it should be available when and where it is wanted.

The development of S.O.S. was based on the needs expressed by adolescent smokers and nonsmokers. It has the ability to tailor information for individuals, evolve dynamically as the user moves thorough the stages of quitting, is accessible at any time, and can be easily and quickly updated.

Internet-based interventions must be maintained continuously. Because S.O.S. targets an audience known for setting trends and demanding cutting edge innovations, maintenance of the website will, in many ways, resemble the development process. Implementing fresh design and content as well as keeping up with technological advances in computer software and hardware will be necessary in order to present a site that remains interesting to an adolescent user. Even more importantly, we must continue to offer the latest on smoking cessation information and methods. However, before we reach the maintenance phase, testing the effectiveness of the current site will provide the initial guidance for revisions and improvements.

Implementation of the Internet-based smoking cessation intervention has begun with a randomized clinical trial to study the effectiveness of the system in helping adolescents (ages 11 to 17) to stop smoking. Participants are randomized to one of two treatments: a brief office intervention or the Internet home-based intervention. All subjects randomized to S.O.S. are given a computer for in-home use, Internet access for 6 months, and are trained to use S.O.S. The primary outcome is the point-prevalence smoking abstinence rates at week 24 verified with expired air carbon monoxide. The effectiveness of the website will be determined indirectly through the study's primary outcomes which include smoking cessation and/or reduction in smoking. In addition, we can link the study outcomes to use data collected on each user. This will allow us to look at patterns of use that correlate with smoking cessation. Issues related to the clinical trial and implementation of this intervention will be discussed in future papers.

While it is still too early to report any results, we are hopeful that S.O.S. will have an impact on the growing number of adolescents seeking help to stop smoking. We also hope that our experiences with the development process of an Internet-based intervention will provide insight for further development of interactive health communication systems. We will conclude with a few initial comments we have received from adolescent users.

“There was a lot of good advice and a lot of good points…”
“I feel that S.O.S. is very well put together and thought through. I have no objections to anything on this site and think it will help many teens.”
“Cool pictures and image effects”
“I think this program is pretty cool. It has sound effects”
“I thought it was a pretty cool program. You guys really stuck to it and really wanted to help us quit. That was the coolest part.”

And as many of our Discussion Group users would say in departure, “Peace out.”


This study was supported by grant # RO1 CA80323 from the National Cancer Institute. The authors would like to acknowledge and thank the following people for their work on this project: Ivana Croghan, Ph.D., Mayo Clinic; Richard Hurt, M.D., Mayo Clinic; Josiah Offord, Mayo Clinic; Nakia Mainor, Mayo Clinic; Amanda Drews, Mayo Clinic; Jill Daniels, Ph.D., Mayo Clinic; Rhonda Baumberger, Mayo Clinic; Ellen Dornelas, Ph.D., Hartford Hospital; Jeremy Barbagallo, M.A. Hartford Hospital; Jennifer Waldburger, M.S.W., University of Wisconsin-Madison; Ann Schensky, University of Wisconsin-Madison; Kari Harris, University of Kansas Medical Center; Jasjit Ahluwalia, M.D., M.P.H., M.S., University of Kansas Medical Center.