Background
Sexually transmitted infections
In 2012, there were 448,422 new diagnoses of sexually transmitted infections (STIs) in England.1 This declined to 417,584 diagnoses in 2016, mostly cases of gonorrhoea and genital warts. However, a concurrent 12% increase in syphilis diagnoses was seen, and chlamydia incidence has remained stable.2 There was also a considerable decline in the number of new human immunodeficiency virus (HIV) diagnoses between 2015 and 2016, with a 23% reduction among men who have sex with men (MSM) largely attributed to increased HIV testing and pre-exposure prophylaxis (PrEP).3,4
Despite improved diagnostics, widespread service provision for the treatment of curable STIs in England and greater emphasis on partner notification, infection rates remain high. STI rates are particularly high in subgroups of the population, with young adults (aged 16–25 years) and MSM having the highest rates of infection.1,2 Many STIs in young people will go undiagnosed, and, for young women in particular, this may have consequences for their future fertility and consequent costs for the health service. Individuals may be at risk for a variety of reasons, such as lack of knowledge about STIs, low self-efficacy (lacking belief that one can successfully meet a goal or perform a particular task such as negotiating the use of condoms) and poor condom use and/or sexual negotiation skills. Risk-taking may also be influenced by peer group norms. Some groups of young people, often characterised by factors associated with the broader determinants of social and health inequalities (e.g. education and literacy), are disproportionately affected by STIs.5
This has led the Department of Health and Social Care (DHSC) to develop a Sexual Health Framework,6 which recommends the prioritisation of prevention and support for behaviour change, alongside access to sexual and reproductive health services, particularly for those most vulnerable to poor sexual health (SH). This is especially crucial in the context of increasing antibiotic resistance, for example in gonorrhoea,2 the prevention of which is as important as effective treatment if we are to bring transmission rates down.
Sexual health services
Sexual health services are provided through a range of clinics in England, including general practitioner (GP) practices, genitourinary medicine (GUM) and contraception clinics, young people’s services (including third-sector providers such as the health-care charity Brook) and pharmacies. The types of care that these different providers can offer varies. Level 3 services (e.g. GUM clinics) provide the full range of STI testing, treatment and management for all patient groups, usually with contraceptive services, in which case they are referred to as integrated SH services. Level 2 services (e.g. some community SH services, Brook or enhanced GP services) provide more limited STI testing and the management of uncomplicated or asymptomatic infections. Level 1 services offer limited STI screening for asymptomatic patients only.7
Since 2013, the commissioning of NHS services for SH has changed, becoming more complex and fragmented. Most services are now commissioned by local authorities as part of their remit to provide public health services. NHS England remains responsible for HIV treatment and care, but also commissions some components of SH, such as national screening programmes. Clinical Commissioning Groups also have a role through their responsibility for commissioning primary care services.8–10 Clinics that provide sexual and reproductive health services offer an opportunity to engage those at risk in sexual risk reduction interventions.11,12 However, introducing complex triage and sexual risk reduction interventions into busy clinical settings on a large scale, essential if there is to be a population-level impact, is challenging and will have resource implications. It is imperative to show that a complex intervention is clinically effective and deliverable, as well as cost-effective, in a sufficient proportion of the diverse range of services provided in England.
Risk groups and behaviours
Young people and MSM are at higher risk of STIs. MSM specifically are at risk of HIV infection, with more than one-third of newly diagnosed HIV cases in Western Europe being among MSM.13 In addition to age and sexual orientation, several demographic, geographical and behavioural characteristics that are associated with an increased risk of STIs have been identified. This potentially allows for targeted provision of prevention services to subgroups at highest risk of STI acquisition.
Among young people, factors associated with an increased risk of a STI diagnosis include multiple partners, previous STI diagnosis and reported lack of condom use. In England specifically, the relative level of deprivation and the geographical region where someone lives are also associated with STI risk.2,14,15 Similar patterns are seen among MSM, with associations with multiple partners and geographical variations observed.16,17 There are also differences in STI diagnoses according to ethnicity, with gonorrhoea rates being considerably higher among black ethnic minorities.18 In addition, the use of drugs during sex (‘chemsex’) is a pronounced risk factor among MSM groups, although it is also seen as a risk in heterosexual groups.16,19,20 Information about these factors is often routinely collected within SH services as part of taking a clinical history and could, therefore, be used as part of triage processes.
Risk reduction interventions
Behaviour change interventions seek to promote changes in behaviour patterns associated with STI acquisition. The white paper Healthy Lives, Healthy People21 emphasises a commitment to behaviour change approaches as a solution to reducing preventable illness and death. Research has shown that behaviour change interventions can help people adopt health-promoting behaviour patterns, including safer sex practices.22,23 However, intervention effectiveness varies in relation to intervention type and target audience.24,25 Attendance at a SH clinic provides an opportunity for intervention delivery at a potentially ‘teachable moment’, when people are primed to think about their sexual behaviour and the consequences for their health.26
Sexual risk reduction interventions are complex interventions, which need to be integrated into routine service provision alongside STI testing and treatment, repeat testing and partner notification. These interventions can take a number of forms and have different objectives, such as increasing knowledge of STIs, changing cognitive antecedents such as attitudes and/or beliefs (including normative beliefs), or increasing self-efficacy.27 The mode of delivery can vary widely. The US Centers for Disease Control and Prevention community guide27 concluded that community-based individual-, group- and community-level interventions can be effective in reducing the risk of STIs in MSM. Digital approaches, such as applications (apps) or online interactive interventions, have also be shown to be effective and offer novel delivery that can be done outside the clinic environment.28 Other intervention formats, such as waiting room videos, can be provided to all clinic attendees, although they may need to be adapted to different clinic settings; the videos have been shown to reduce incident STIs.29 However, there is also evidence that tailoring interventions to individuals’ preferences and needs leads to greater uptake and makes the interventions more likely to be effective.30
As clinic resources are limited, the use of triage or risk assessment can ensure that more resource-intensive interventions are targeted at those who need them most, or those for whom the intervention has been designed and who are more likely to find it effective. As electronic patient record (EPR) systems become more widely used within clinical settings, using the coded data collected as part of routine care to inform a triage algorithm could provide a mechanism to target different risk groups with appropriately tailored risk reduction support.
Evidence gap
Currently, multiple interventions have been trialled, but there is a lack of clarity about which intervention would best be implemented in which context, by whom and for whom. Therefore, a clear understanding of the factors that influence implementation in particular contexts is needed. In addition, such interventions, although tested individually and in most cases showing a modest but consistent positive effect, have not been implemented systematically in a way that could have a population-level impact in the UK. An additional challenge is that any implementation can be achieved only if it can be delivered at minimal overall cost. As funding for health-care services, particularly those commissioned by local authorities, is under pressure, any demand for additional resources across a large number of services is unrealistic. In this context, research is required to identify brief, pragmatic, non-labour-intensive interventions that can be tailored to the level of risk of the individual attending any of a range of different SH services. The characteristics of those in the higher-risk groups will differ by clinic setting, gender, sexual orientation and other factors that will need to be incorporated into the intervention model. The Health Technology Assessment (HTA) call that led to the work described in this report was intended to address a first, important, step by determining whether or not it would be feasible to conduct a definitive trial of effectiveness of brief behavioural interventions in SH services, incorporating the development of triage processes to ensure that the interventions were most efficiently delivered to those most likely to benefit, and whether or not such a strategy would meet cost-effectiveness criteria.
Project overview
The project was a multistage, mixed-methods study design that incorporated a systematic review; a secondary analysis of national surveillance data; interviews and surveys with clinic staff; semistructured interviews, a discrete choice experiment (DCE) and focus groups with clinic attendees; monitoring of intervention offering, uptake and completion, and follow-up questionnaires; and capturing the clinical resources used. In order to achieve this, the study was organised into six overlapping work packages (WPs), summarised in :
a systematic review of sexual risk reduction behavioural interventions focusing on UK-relevant evidence (objectives 1 and 2)
development of a sexual risk assessment/triage tool to identify individuals at increased risk in SH settings (objective 3)
a mixed-methods study to describe sexual risk reduction practices and preferences in SH clinics in the UK, and to identify opportunities for intervention (objectives 4 and 5)
the selection and adaptation of a suite of evidence-based interventions suitable for delivery in SH settings and acceptable to patients and staff (objective 6)
a pilot study of the feasibility of implementing interventions to assess their acceptability, practicality and cost of implementation, and to comment on the feasibility of a future RCT (objectives 7–9)
the refinement of the triage tool and manuals of the interventions to ensure that the triage tool can be incorporated into routine care (or derived from routinely collected data) and to ensure the fidelity of the interventions, and an outline of the trial design for a full evaluation (objective 10).
The project focused on understanding factors that influence the implementation of interventions in complex SH clinic settings, from both patient and provider perspectives. We used co-creation approaches to intervention identification and adaptation with health advisors (HAs), clinicians and service users as an essential part of ensuring that the intervention would be acceptable and deliverable. We followed the Medical Research Council revised guidance31,32 on developing and evaluating complex interventions, taking the results of the systematic review and selecting and developing the most promising package of interventions to reduce sexual risk. The National Institute for Health and Care Excellence (NICE) and the DHSC recommend user input when designing services,33 as it leads to services that are more responsive to the needs of users as services are less likely to be designed inappropriately and more likely to be used.
The methodological steps to delivering the objectives set out earlier were informed by the intervention mapping (IM) approach to intervention design.34 The IM approach is iterative and can be described as six steps (Box 1). As described, the views of stakeholders, including service users and staff who deliver interventions, are captured and are vital to the successful design and implementation of interventions that are acceptable, practically feasible and sustainable over time.35 Co-creation of interventions with stakeholders is integral to the IM approach.34 Although in part already defined by the project brief, the project included a consideration of the needs of the population and identified specified behaviour change outcomes corresponding to those needs. Regulatory processes underpinning specified behaviour changes were identified from relevant research and matching change techniques were selected.
Intervention mapping: iterative steps applied to the Santé Project
In the case of the interventions, it was anticipated that, as these were to be selected from those for which efficacy evidence already existed, we would be able to select practical intervention components designed to change defined behaviour regulation processes that work in situ.
This IM process facilitates the identification of primary and secondary outcome measures, specified as needs and target behaviour changes, thus anticipating the evaluation design. Prototype interventions should be tested and adapted to ensure the fidelity of delivery in context prior to finalising an intervention manual.
The IM process combines an ecological approach with participation from all stakeholders, a focus on specification of the underlying mechanism (in a clear logic model) and a research-based approach to ensuring the fidelity of implementation. A key part of this process is to identify change techniques (e.g. Abraham and Michie,36 Abraham,37 Michie and Johnston38), modes of delivery and delivery competencies that maximise intervention effectiveness in real-world contexts.39 All of this underpinned the approach taken to deliver the project.
Project management
Structured oversight of the project was conducted by a Project Steering Committee (PSC), with an independent chairperson, and was convened in accordance with guidelines for the National Institute for Health Research (NIHR) HTA programme. This committee also served the function of the Data Monitoring Committee.
The Project Management Group (PMG), chaired by the chief investigator, oversaw the work of all WPs. It agreed the details of project set-up and the design, initiation and supervision of the study. Each WP had a working group responsible for the day-to-day management of the work, and these groups reported to the PMG monthly.
Patient and public involvement
We embedded patient and public involvement (PPI) into the research programme at key points, including at the proposal development stage. It was essential that the developed intervention packages were endorsed by service users; therefore, PPI and service user input throughout the proposal was sought for the translation of the research outcomes into improvements in current service provision. We set up a PPI group specifically for this project and liaised with the group throughout the project. The PPI group (including our target groups of young people and MSM) helped in writing and approved the patient information sheets for this study.
Ethics
Ethics approval for the study was granted by Westminster National Research Ethics Committee (reference number 15/LO/0690) for work conducted in WPs 3 and 4, and the Chelsea National Research Ethics Committee (reference number 16/LO/0673) for work conducted in WP5. The use of data in WP2 was approved by Public Health England’s Associate Caldecott Guardian. The anonymised data were collected as part of a pilot of enhanced routine STI surveillance, and their use was not considered to require an ethics review. All service user participants provided written informed consent for interviews, focus group discussions and surveys. Health-care providers provided written informed consent for focus group discussions, verbal consent for interviews and implied consent for web surveys. Process data from clinics were anonymised, and posters informing service users that the clinic was currently part of a study were displayed informing participants that they could opt out of their process data being used in analysis.
