Designing an intervention to improve sexual health service use among university undergraduate students: a mixed methods study guided by the behaviour change wheel

Both focus groups and interview participants described student’s limited knowledge and awareness of sexual health services as an important barrier to service use. Further, student and health care provider participants identified a lack of understanding on LGBTQ students’ sexual health. Improved visibility of the services was identified as a facilitator to sexual health service use [6].

Student participants described physical opportunity, including service accessibility and the campus culture, as both a barrier and enabler to sexual health service use. Due to a campus environment that promotes risky behaviours, student participants described the importance of countering this culture with safe and accessible sexual health services, including flexible hours of operation, convenient location, and mobile clinics (known as STI testing clinics that are offered around campus) [6].

Survey, focus group, and interview data illustrated the importance of peer influence on student behaviour. Participants described the stigma associated with sexual health service use and the influence of peer support on health promotion behaviours. These positive and negative peer influences were found to be important barriers and enablers for accessing sexual health services [5, 6].

We found that the social influences described above directly affected students’ motivations for accessing sexual health services. Participants stated that accessing the services could jeopardize their privacy and confidentiality and lead to negative emotions (e.g., discomfort, shame, awkwardness). Further, university students are in a developmental period of exploration and experimentation and as a result, felt motivated to access sexual health services while experimenting with high-risk behaviours [6].

While the barriers and enablers to sexual health services were applicable to both universities, we found a number of important contextual differences including: size of student population; clinician knowledge on LGBTQ health; structure of health service delivery; financial resources; and location of services. These contextual elements were important factors to take into consideration when designing interventions for the two universities.

To address the social and physical opportunity barriers and enablers, we found enablement, modelling, and environmental restructuring intervention functions to be most relevant. The following nine BCTs were identified: prompts/cues, goal setting (behaviour), problem solving, action planning, restructuring the social environment, restructuring the physical environment, demonstration of the behaviour, adding objects to the environment, and social support (unspecified). Potential modes of delivery include: mobile STI testing clinics; peer outreach; flexible hours of operation; and creating a friendly and welcoming space.

Intervention functions to address the barriers and enablers under automatic and reflective motivation include: education, persuasion, modelling, enablement, and incentivisation. The following 12 BCTs were identified: information about health consequences, information about social and environmental consequences, feedback on behaviour, feedback on outcome(s) of behaviour, prompts/cues, self-monitoring of behaviour, credible source, demonstration of the behaviour, social support (unspecified), goal setting (behaviour), problem solving, and action planning. Potential modes of delivery include: peer support groups and student outreach; health care providers and students present during orientation; email or text message reminder of sexual health services and upcoming mobile clinics.

Following these three stages, we created a toolbox for our stakeholders to use in future sexual heath intervention design and program planning (Additional file 1). The behaviour change toolbox includes: the barriers and enablers to sexual health service use among university students under the COM-B components; six intervention functions most likely to bring about change; 15 BCTs to include as active ingredients in interventions; and a list of potential modes of intervention delivery. An electronic copy of the toolbox was sent to the participants of each stakeholder consultation meeting.

The theory- and evidence-based toolbox will likely be an improvement from the traditional atheoretical approach to intervention design in this context; however, the effectiveness of the six intervention functions and 15 BCTs to improve sexual health service use among university students is not yet known. Several studies have examined some of these BCTs in the context of sexual health services and found significant effects. Wolfers, de Zwart, and Kok [24] and Newby et al. [25] used intervention mapping [26] to design an intervention aimed at improving STI testing rates and increasing sexual health service uptake, respectively. These interventions include eight of the 15 BCTs identified in this study (information about health consequences, information about emotional consequences, adding objects to the environment, feedback on outcomes of behaviour, social support (unspecified), information about health consequences, demonstration of behaviour, credible source). Both interventions yielded significant positive results, including higher STI testing rates [24] and significant improvement in beliefs related to service access (i.e., service access being important and normal) among females, and a significant increase in the behaviour of visiting sexual health services among males [27]. The effectiveness of the BCTs used in these interventions shows promise for the similar BCTs identified in this current study. However, other than these few studies, the body of intervention literature on improving sexual health service use is scarce. Additional research is needed to test the effectiveness of the BCTs and intervention functions outlined in the toolbox.

University health care providers and administrators can use the intervention functions described in the toolbox to translate the 15 BCTs into intervention content. Our results show that the Education intervention function maps onto five barriers and enablers to sexual health service use among university students. Studies have demonstrated that education interventions have moderate impact on sexual knowledge and attitudes [28, 29]. However, an increase in knowledge alone does not always lead to behaviour change [28]. It is important to use targeted, multi-component interventions to combine education with other key elements to maximize the potential for behaviour change [30, 31]. As such, it may be beneficial to target students with educational interventions that include multiple BCTs, such as information about health consequences, information about social and environmental consequences, and demonstration of the behaviour. University students may benefit from a sexual health education intervention that also includes the prompts/cues BCT in the form of electronic reminders. Our student participants recommended email and text message reminders to increase their awareness of sexual health services and the reasons to access them. Studies have shown that interventions delivered by mobile technologies increase the uptake of sexual health services and STI testing, particularly for tech-savvy young adults [32‐34]. This is a widely available and accessible approach for university health centres to offer a confidential means of communicating sensitive or personal information with students [34]. Further, studies have found that using social media for sexual health education can help promote STI testing behaviours [35]. As such, there is an opportunity to leverage social media to support educational interventions that include BCTs aimed at increasing students’ capability and motivations for accessing sexual health services, such as information about health consequences, information about social and environmental consequences, feedback on behaviour, prompts/cues, and self-monitoring of behaviour [36].

The Enablement intervention function aligned with six barriers and enablers to students’ use of sexual health services. Enablement is described as “increasing means/reducing barriers to increase capability (beyond education and training) or opportunity (beyond environmental restructuring)” [12]. Several BCTs can be included in enablement interventions, such as social support (unspecified), goal setting (behaviour), problem solving, action planning, adding objects to the environment, self-monitoring of behaviour, restructuring the physical environment. Our stakeholders stated that enablement interventions related to a main priority at both universities: building capacity and resiliency among their student population. Strengthening students’ sexual resilience provides them with the tools needed to prevent negative outcomes from their sexual behaviour and take control of their physical, sexual, and mental health and well-being [37]. However, compared to education, enablement interventions have not been as extensively examined in the literature. Enablement interventions with the social support (unspecified) BCT are especially relevant in this context, as our behavioural analysis illustrated how peer influence can act as a barrier and enabler to sexual health service use. Studies have shown that perceived social norms affect sexual behaviours [38‐40]. Young and Jordan [40] examined the influence of social networking photos on social norms and sexual health behaviours with a sample of college students in the United States. They found that students who viewed Facebook images with a low prevalence of sexually suggestive content estimated a larger percentage of peers used condoms and reported a greater intention to use condoms themselves in the future. In the context of university sexual health services, stakeholders could employ a similar approach with existing social media networks and curate positive images of peers accessing sexual health services to tap into students’ intentions for sexual health promotion behaviour.

The BCW includes seven broad policy categories to leverage behaviour change on a wider scale (e.g., changing legislation to encourage behaviour change at a population level) [12]. Similar to other intervention design researchers, the policy categories were found to be less practical than other BCW steps in this context [18, 41]. The selection of BCTs flowed logically from the COM-B model analysis and intervention functions. As such, we did not identify policy categories at this stage in the intervention development. Similar to Mc Sharry et al.’s recommendations [18], we believe that the policy categories will likely be more useful for broad, process-level guidance when designing implementation strategies for future sexual health service interventions.

The influence of context on intervention effectiveness is often overlooked in the intervention design process, particularly when focusing on individual-level behaviours [42]. The BCW recommends gathering input from a diverse group of stakeholders to examine the influence of context at multiple conceptual levels. Moore and Evans [42] recommend using this co-production approach with stakeholders with intimate knowledge of the context. In this study, we included stakeholders at the barriers and enablers assessment stage, as well as the intervention design stage. This helped us to move from a theoretical exercise of listing intervention functions and BCTs to a hands-on approach with our stakeholders to address the question of “What is likely to work in this situation for these people in this organization with these constraints?“ [43]

We identified several barriers and enablers directly related to the social and physical context of sexual health behaviours on campus. From this, we identified several system-level BCTs, including restructuring the social environment, restructuring the physical environment, and adding objects to the environment. A limitation of the BCW is its lack of guidance on how contextual mechanisms function across different settings and its limited detail on the characteristics of system-level BCTs. Other researchers have had similar experiences in using the TDF to examine multi-level behavioural problems [15, 43‐47]. To address this issue, some researchers have paired the TDF with organizational context frameworks, such as the Consolidated Framework for Implementation Research (CFIR), which elaborates on organizational-level determinants [48]. Future sexual health service intervention research would benefit from a similar approach to provide a more in-depth examination of the organizational context and how it influences service delivery. Similarly, we echo recent calls for future methodological research to elaborate upon system-level BCTs and characterise their meanings in more detail [49].

Traditionally, behaviour change interventions are inadequately reported which hinders the reader’s ability to accurately understand, evaluate, or replicate interventions [50, 51]. When theory is used to describe the plausible mechanisms of action, findings can be synthesized with existing literature to inform future replication and evaluation studies [52]. Recent efforts to improve the implementation and replication of effective interventions have led to the development of reporting guidelines, such as the Template for Intervention Description and Replication (TIDieR) – a 12-item checklist that aims to standardize intervention descriptions [53]. Further, the BCTTv1 was developed to offer a shared language for clearly labelling and defining BCTs to ensure that behaviour change interventions are interpreted in the same way by different readers [14, 51]. The clear reporting of BCTs in this study will inform the science on sexual health behaviour change interventions. Researchers, administrators, and sexual health program planners can use the toolbox to identify intervention functions and BCTs that apply to their context and test them in implementation and evaluation studies. This will further aid in building a repository of effective sexual health service interventions and intervention components.