The Sexunzipped website was developed over an 18-month period using an iterative process. This included a review of the relevant literature to define the theoretical basis for the intervention, seeking the views of young people through extensive focus group research, conceptualizing the intervention with a Web development company, and developing the content for the site. We report the outcome of our consultation with young people in detail elsewhere (young people’s views on the content, design, and interactive features of the Sexunzipped website) [11].

A central principle of the Sexunzipped intervention was the adoption of a holistic approach to sex and relationships. This focused on a broad range of topics, including sexual pleasure and relationships, rather than just safer sex behavior. The language used on the site aimed to be gender and sexuality neutral to allow engagement with the website content regardless of gender, sexuality, or sexual preference. Our preliminary focus group work suggested that young people value honest information about sexual pleasure and sexual practices [11]. Previous research highlights the multiple influences on sexual health behavior, including individual factors (eg, motivation and affect), interpersonal factors (eg, sexual scripts and communication), systemic factors (eg, peer norms and familial factors) [12,13], and the suggestion that a focus on sexual pleasure may facilitate safer sexual behavior [14]. The website emphasized equality and sexual rights, informed by the connection between power imbalances between the genders and sexual health risk [15]. In developing the intervention, one of a number of definitions proposed by the World Health Organization [1] was adopted in which sexual health is viewed as:

Health behavior research has shown that theory-based interventions can lead to larger effects on behavior than interventions without a theoretical basis [16], with this effect also reported for Internet interventions [17]. A number of theories and models of behavior are relevant to sexual health interventions, but identifying a unitary model for application in an intervention can be difficult given the similarities between them [18]. Because of the breadth of the intervention content, an integrated behavioral model was chosen as the guiding theory for the Sexunzipped intervention [19]. The integrated behavioral model integrates concepts from the theory of reasoned action (TRA) and the theory of planned behavior (TPB), as well as concepts from other health behavior theories [19]. The integrated behavioral model suggests that behavior is influenced primarily by intention, which is in turn influenced by attitude, perceived norms, and sense of personal agency. Salient beliefs influence each of these factors, for example, beliefs about whether an action will be of benefit to the individual and beliefs about the opinion of others. The integrated behavioral model further posits that environmental constraints, behavior salience, habits, knowledge, and skills have further direct influences on behavior. Although the core aspects of the TRA and TPB have been supported by extensive research, eg, [20], some authors have suggested that these theories can be enhanced by the inclusion of anticipated affect within behavioral formulations [21,22]. Anticipated affect “refers to the prospect of feeling positive or negative emotions (eg, exhilaration, regret) after performing or not performing a behavior” [21]. The core components of the integrated behavioral model and the concept of anticipated affect were central to the development of content for the Sexunzipped website, for example, including information and activities that challenged beliefs and encouraged reflection on anticipated affect.

Integrated behavioral model principles underpinned the website content that addressed safer sex behavior (condom and contraception use and STI risk reduction). In line with the findings from our focus group work, the site covered topics such as relationship quality, sexual violence and control, and sexual pleasure [11]. These topics were included to provide information and encourage self-reflection (to support informed choices), but did not aim to change specific behaviors.

Table 1 lists the primary safer sex behaviors targeted by the website along with the determinants or beliefs salient for behavior change and ways in which these factors were addressed on Sexunzipped.

In terms of the site content and design, the initial focus group research and literature review shaped the overall scope of the content and helped to identify salient beliefs and behaviors to target; the integrated behavioral model provided the framework for identifying for relevant beliefs and factors related to safer sex behavior. The development of the content, particularly the interactive activities, drew heavily from cognitive behavioral therapy [23] and motivational interviewing [24], for example, exploring links between thoughts, feelings, and actions, and drawing on techniques such as identification and rating of thoughts and feelings, consideration of the importance or pros and cons of a particular behavior, and reviewing past experiences or future consequences.

Such behavior-change techniques were employed to differing degrees throughout the site to prompt safer sex behavior change [25,26]. The techniques most commonly utilized were “provide information on consequences of the behavior in general,” “provide information on consequences of behavior for the individual,” “prompt anticipated regret,” “barrier identification,” “provide instruction on how to perform the behavior,” “prompt practice,” and “general communication skills training” [26].

The site was organized into 3 distinct but related sections each containing a number of topics: relationships, safer sex, and sexual pleasure. The aims and rationale of each section are presented in Table 2. Content was presented in 3 formats: (1) text-based information, (2) interactive quizzes, and (3) interactive decision-making activities.

Each of the 3 website sections (relationships, safer sex, and sexual pleasure) consisted of interactive activities and detailed information (see Table 3). The information sections comprised factual information, as well as arguments or information to challenge beliefs related to safer sex behaviors. At the end of each activity or information section, users were presented with links to other relevant topics within Sexunzipped and/or links to organizations providing specific services, such as treatment of STIs or help with domestic abuse. Given the breadth of the site, the navigation was designed so that users would self-direct to the sections of interest and salience. Quotations from young people were employed throughout the site in response to focus group findings that the site should represent the views of a range of young people [11].

Two interactive activity templates were developed to allow for the repetition of interactive formats across the site. The first was an interactive quiz, whereby the user received feedback based on responses to questions. The second was a decision-making activity where responses were recorded and presented back to the user with a series of prompts to encourage reflection on behavior, emotions, and consequences, or to aid with decision making. A template for text-based information pages was also developed. Examples of these activities are presented later in this paper.

In the “Relationships” section, there were a total of 19 activities and 15 information pages. The content was informed by previous research identifying the beneficial effect of communication and assertiveness skills training in reducing risky sexual behavior [27,28]. Recent research on the extent of intimate partner violence among young people in the United Kingdom [29] suggested the need to address violence and coercion.

The “Safer sex” section was comprised of 15 activities and 45 information pages. Content in this section contained primarily information about STIs and contraceptive methods and activities to encourage safer sex behavior, such as better communication with partners, condom use for penetrative sex, and STI testing with new partners. The content was based on a number of empirical findings. For example, previous research has established the importance of including clear messages about risk [30] and methods to review past risk experiences and develop risk reduction plans [12]. A meta-analysis of different approaches to human immunodeficiency virus (HIV) prevention suggested that interventions for young people should provide discussion of normative behavior and attitudes, as well as condom provision and skills training [31]. Several reviews have identified the importance of including content related to increasing communication about safer sex [30,32,33], which formed a central part of the website.

The “Sexual pleasure” section contained 13 activities and 44 information pages. Content in this section focused mainly on descriptions of different sexual practices, enhancing sexual pleasure, and communication about sex and sexual pleasure. This was primarily informed by our focus group research that suggested young people wanted information on sexual practices and sexual pleasure [11]. Some commentators have suggested that a stronger focus on sexual pleasure may increase safer sex behavior and use of condoms [34,35]. Although empirical research on this is limited, there is some evidence suggesting that eroticizing safer sex messages may facilitate safer sexual behavior [14] and that condom discomfort or loss of sensation may reduce use of condoms [36,37]. The site made links between sexual pleasure and safer sex, for example, suggestions on how to deal with the reduced sensation and interruption of sex associated with condom use.

The Sexunzipped website was evaluated in an online, randomized controlled trial (RCT) that compared Sexunzipped to an information-only control website. The protocol for this study is presented subsequently.

We conducted an online RCT designed to test the hypothesis that the Sexunzipped theory-based, interactive, online intervention would be more effective in promoting sexual health in young people than an information-only website. A total of 2006 young people aged 16 to 20 years were enrolled in the trial between November 2010 and March 2011. Participants completed a baseline demographic and sexual health questionnaire online and were automatically randomized by computer to the intervention website or the control website. We measured sexual health outcomes at 3 months by repeating the online sexual health questionnaire and by asking half of the participants to return (by mail) a urine sample for genital chlamydia testing. Ethical permission for the study was granted by the University College London Ethical Committee (ref: 1023/002).

Retention in online trials can be difficult [39]: this study was a pilot (feasibility) trial with a number of substudies to test the best ways to maximize retention at follow-up. The full results of the website evaluation and trial design substudies are reported separately.

Developing an online behavior-change intervention is costly and time consuming. It requires the bringing together of different sets of skills and knowledge, primarily expertise in sexual behavior change, user involvement, and website design and technology. It is necessary to partner with technical and design experts from an early stage in order for behavior change experts to understand the technical implications and costs of different online formats and for technical and design experts to understand the processes that facilitate behavior change. A good understanding of the requirements of the target group is also required, including how they use the Internet. This may be particularly important for interventions for young people given their rapid adoption of new technology [4]. Furthermore, the process of translating behavior-change techniques and theories into an interactive format requires careful specification of the intervention aims and objectives [42]. Given the large public health potential, the cost involved in developing online interventions, and the need for attractive design, future interventions may benefit from collaborating with established sites that already have a user base, a brand, and a strong Internet presence.

The Internet is a fast-changing medium, with increasing competition for the attention of young people, and design and technology quickly becoming obsolete. This presents a number of challenges for designers of interventions. Unlike face-to-face interventions, or facilitated computer-based interventions where users may be a captive audience to some extent, Internet interventions require the development of an online presence that will attract users. It is vital to involve users in decisions about intervention content, design, and features, paying attention to aspects that will attract and retain users’ interest. Given the finding that theory-based online interventions lead to better outcomes than interventions that are not theory-based [17], a central challenge in developing effective Internet-based interventions for young people is to find effective ways to operationalize theory in ways that address the views and perspectives of young people.