Sexually transmitted infection (STI) diagnoses in England have been persistently high over the past decade [1]. Although the most recent annual data show a 32% reduction in STI diagnoses, down from a 10-year high of 467,096 in 2019 to 307,901 in 2020, this likely reflects a temporary shift brought about by the COVID-19 pandemic, which placed restrictions on public behavior and reduced access to STI testing [1,2]. Young people continue to carry a high burden of infection, with 46% of all new diagnoses among 15- to 24-year-olds [1]. Furthermore, women, those of Black ethnicity, those living in the most deprived areas, and men who have sex with men, remain disproportionately affected [1]. Among young people, chlamydia continues to be the most prevalent STI [1]. Rates of chlamydia among 15- to 24-year-olds have been high but stable over recent years, whereas increases have been observed for other STIs, such as gonorrhea and syphilis [2]. The removal of COVID-19 restrictions and an increase in testing availability will likely bring with it a return to prepandemic levels of STI diagnoses and as such, sexual health continues to be a public health concern [1]. Untreated STIs can lead to serious health consequences, such as pelvic inflammatory disease, ectopic pregnancy, and infertility, which have a significant impact on health and quality of life [3]. The estimated annual cost to the National Health Service (NHS) of STI treatment is £620 (US $762.6) million [4]. The best way to avoid STIs from penetrative sex is to use a condom [5], but young people report inconsistent use [6,7].

Although condom use is the cornerstone of STI prevention, STI testing is important for treatment and to prevent onward transmission. There have been notable changes in the landscape of STI testing within England since 2017. In the period between 2017 and 2019, against a backdrop of increasing testing overall, there was a shift toward the use of internet services, and away from other nonspecialist services [8]. Between 2018 and 2019, while the number of tests carried out by specialist sexual health services (eg, genitourinary medicine services) remained relatively stable, those carried out by nonspecialist services (excluding internet testing), such as pharmacies, general practice, and community-based settings, decreased by 9%, and those by internet services increased by 69% [8]. This trend reflects recent year-on-year increases, with internet testing accounting for 4% of all testing in 2017, 13% in 2018, and 19% in 2019 [8]. The global COVID-19 pandemic has further served to accelerate the move in this direction across Europe [9,10]. In England, despite there being a reduction of 30% in expected levels of chlamydia testing in 2020, a 2-fold increase in internet testing was observed, with approximately 40% of all chlamydia testing being processed through the internet in 2020 [1]. Whether internet testing remains at these elevated levels or begins to adjust back toward those observed prepandemic, remains to be seen. Regardless, internet testing is now an established and significant route to STI diagnosis and treatment within England.

Evidence indicates that STI self-sampling websites are effective at reaching higher prevalence groups; chlamydia positivity is comparable to that of other nonspecialist services [8], and they attract equivalent numbers of males and females, and those from across the spectrum of deprivation (E Hollis, unpublished data, 2017). Repeated use of internet testing services is however common, with chlamydia positivity remaining high for those that retest (E Hollis, unpublished data, 2017), suggesting that testing online does not reliably translate into future individual prevention efforts. Furthermore, while information on STIs and contraceptive methods is typically found on self-sampling websites, this is insufficient for behavior change [11], and inconsistent with Public Health England guidance to provide those at elevated risk of STIs with preventative intervention [12]. As such, self-sampling websites provide an important but currently missed opportunity to intervene with a priority group.

In 2016-2017, with the support of Medical Research Council Public Health Intervention Development funding and in line with Medical Research Council guidance on the development of complex interventions [13], Wrapped was cocreated by young people, stakeholders, and researchers [14]. Wrapped is a fully automated, multicomponent, and interactive digital intervention developed for the users of STI self-sampling websites aged 16-24 years. It aims to reduce the incidence of future STIs by increasing correct and consistent use of condoms for penetrative sex. The users of STI self-sampling websites can be signposted to Wrapped upon order of an STI self-sampling kit (through a hyperlink). Content is tailored, with between 1 and 6 different components allocated to users in accordance with their self-identified barriers to condom use. These barriers reflect important behavioral determinants of condom use identified through existing meta-analyses [15-18], namely condom use attitudes (particularly beliefs around pleasure, enjoyment, and spontaneity), condom availability, and behavioral capability and self-efficacy for condom use and communication.

In 2019, funding was awarded by the National Institute for Health and Care Research Public Health Research program to determine whether it is feasible to run a randomized controlled trial (RCT) to test the effectiveness and cost-effectiveness of Wrapped. This paper presents the protocol for this feasibility randomized controlled trial (fRCT).

The aim of this study is to assess whether and how it is possible to carry out a future definitive RCT to evaluate the effectiveness and cost-effectiveness of Wrapped plus usual care, in comparison to usual care only.

The primary objectives are to estimate the following parameters for planning a definitive RCT:

The secondary objectives are as follows:

A 2-arm parallel group fRCT of Wrapped plus usual care compared to usual care alone with a nested qualitative study.

Recruitment, intervention delivery, and data collection are conducted entirely on the internet. Participants will be recruited from an existing chlamydia self-sampling website (freetest.me [20]). Freetest.me provides a web-based chlamydia testing service to local authority areas across England (typically subcontracted to and administered through NHS trusts) and is free at the point of use. For the purposes of this study, users of freetest.me residing in 1 of 4 local authority areas will be invited to participate.

The inclusion and exclusion criteria are outlined in Textbox 1.

Table 1 gives an overview of participant activities and timing.

To recompense for time spent in completing activities, participants will receive payment in the form of vouchers (see Multimedia Appendix 2 for the schedule of payments).

A study invitation will be placed on the “thank you” page of the freetest.me website displayed to users after they have placed an order for a chlamydia self-sampling kit. Every 2 weeks, the advert will change to emphasize a different value proposition (see Multimedia Appendix 3 for the list of adverts). A hyperlink within the advert will take users to Research Electronic Data Capture (REDCap), a secure data capture and management platform, for participant information and consent (see Multimedia Appendix 4 for a copy of the participant information and consent form). Each individual’s freetest.me order number will be appended to the URL. This order number will be captured by REDCap enabling the research team to ensure that each participant is a unique and genuine user of freetest.me.

REDCap is being used to monitor and manage all survey data collected as part of this study. Surveys will be conducted at baseline (month 0 [M0]), month 3 (M3), month 6 (M6), and month 12 (M12) and used to collect data on condom use attitude, behavioral capability, self-efficacy, and intention, along with details of any partnered sexual activity and condom use that has occurred in the previous 3-month period. The surveys will also be used to measure the experience of partner resistance to condom use and to collect health economic data. There are between 63 and 76 core items per survey (depending on time point), with data largely being collected using fixed response items and Likert scales (see Multimedia Appendix 5 for survey items). The baseline survey will additionally collect participant’s contact details to enable communication with participants regarding the completion of follow-up surveys, and basic demographic data. The M3 survey will also be used to collect self-reported data on participant engagement with the intervention.

Completion of the baseline survey is scheduled immediately after consent is provided. All other survey completion will be triggered by email. Participants who do not complete the survey will be sent up to 3 reminders (delivered by SMS text message). Subsequently, if no response is received, 1 email and 1 telephone call will be made to prompt completion. During the telephone call, participants will be given the option to verbally respond to a reduced set of survey questions. Finally, if there is still no response, a paper-based version of the survey will be posted out with a stamped addressed return envelope. Those who have not provided any data 20 days after the initial survey is sent out will be classed as nonresponder. Surveys will continue to be sent to nonresponders (as well as chlamydia tests––see below) at subsequent time points unless a participant asks to be withdrawn from the study.

Chlamydia tests will be processed by freetest.me at M0, M3, and M12. The chlamydia test at M0 is triggered by participants when they order a self-sampling kit through the freetest.me website immediately before receiving the study invite (see “recruitment” above). Participants will be asked to self-report their result, and any treatment received if applicable, through SMS text message. A single SMS text message reminder will be sent to participants who do not respond. Those who have not provided this data 4 weeks after the SMS text message contact is made will be classed as nonresponders.

The chlamydia tests at M3 and M12 will be processed differently to that at M0. For this purpose, chlamydia self-sampling kits will be sent directly from the study team to participants. One week before posting the kits, participants will be sent an SMS text message asking them to confirm or amend their postal address. The kits that participants receive at M3 and M12 will contain the usual user instructions to collect the sample and return by Freepost to freetest.me for processing. Participants will be sent 3 SMS text message reminders to complete the test kit. If the test kit still has not been returned after this point, 1 phone call will be made to prompt return. Those who have not returned their kit 30 days after it is posted out will be classed as a nonresponder. Participants who have a positive test result at M0, M3, or M12 will receive 1 item within SMS text message survey 2 weeks post results to record whether their infection has been treated or not. If treatment is incomplete, or the participant does not respond, they will be sent 1 reminder SMS text message to obtain this information. Each local NHS trust will be informed of any positive cases (for participants residing in their local authority area) so that they can provide appropriate treatment and follow-up.

Randomization of participants to trial groups in a ratio of 1:1, is fully automated, and fully concealed within REDCap software, and will take place following completion of the M0 survey. Before randomization of each participant, the research team will manually check that each individual is unique, by cross-checking their identifying information (name, email address, postal address, mobile phone number, and freetest.me order number) with participants already within the sample. They will also check that the participant has answered the baseline survey with reasonable care (ie, no evidence of random responding or completion under the expected minimum time). Those not meeting these criteria will not be randomized. Participants will be blind to group allocation and the research team will be unable to influence any aspect of this procedure. The randomization list will be generated by the study statistician who will monitor randomization on a weekly basis during active recruitment to ensure that the algorithm is being correctly applied. Randomization will be at the individual level as control group contamination is not anticipated (although this will be monitored to determine the suitability of this approach for the definitive trial). Stratification across groups, using randomly permuted blocks (ethnicity, sexual identity, deprivation) will be performed to balance participants across the trial arms.

Participants will be individually randomized within the stratifications into 1 of 2 groups: usual care or Wrapped plus usual care (see “Wrapped website” and “comparator website” below for further details on the respective interventions). Participants will be directed to their allocated intervention following randomization through a hyperlink. All participants will be free to interact as much or as little with the respective materials; participants will be emailed the hyperlink to enable repeat visits. An email reminder to visit the intervention will be sent to all participants who have not done so 2 weeks after first access is provided.

Primary outcome measures are as follows:

Secondary outcome measures are as follows:

At least 60 participants per arm are recommended to allow proportions to be estimated with good precision [24]. This is inflated to allow for an estimated 25% nonreturn of initial chlamydia self-test sample (based on freetest.me statistics; 60/0.75=80) and a further estimated 30% dropout over the course of the feasibility trial (a conservative estimate based on 19% dropout achieved in Free et al [25] sexual health feasibility RCT (80/0.7=114 per arm). Based on this, the total sample size required for the present fRCT has been calculated as 230 participants.

The following criteria will be used to inform progression to a full trial:

The progression decision will be determined by the independent Study Steering Committee (SSC), in conjunction with the Data Monitoring and Ethics Committee, based on achievement of all the above criteria, or convincing evidence that any 1 criterion is amenable for necessary improvement.

Ethical approval was obtained from the NHS East Midlands—Leicester Central Research Ethics Committee (reference 20/EM/0275) and the trial was registered with the ISRCTN (ISRCTN17478654). To participate, all individuals will be required to provide informed consent. All participants will receive a £65 (US $79.95) payment in the form of vouchers to recompense for time spent completing the research activities. All research data will be deidentified before analysis and reporting.

A nested qualitative study will be conducted during the fRCT. The aims of this study will be to explore participant’s views and experiences of the Wrapped intervention, experience of trial procedures and materials, and perceptions of whether and how the intervention has changed their condom use beliefs or behavior.

Approximately 30 individuals (number to be determined by data saturation) participating in the fRCT will be invited to take part. All participants will receive a £20 (US $24.6) voucher in recognition of their time. Participants will be asked at baseline (single question in survey) whether they are willing to take part in a semistructured telephone or video call interview and if so, to provide suitable contact details.

Participants will be purposively selected to create a varied sample representing different demographic criteria, completers and noncompleters from both trial arms, those with varying levels of intervention engagement, and those thought to have experienced specific events that could influence dropout. Selected individuals will be contacted and directed to a new project page on REDCap where they will be asked to provide informed consent to take part in this additional aspect of the study.

Interviews will last up to 1 hour and will be fully transcribed before analysis. Framework analysis [27] will be used to analyze the data.

At the end of each interview, a second aspect of the qualitative study will be discussed with the participants. This is interviewing sexual partners to better understand any changes in condom use decision-making in light of the intervention. After the interview, an email invitation will be sent to the participant which they can choose to forward to their partner if they wish. It will then be up to the sexual partner to contact the research team if they want to participate. They will be given the option to email or telephone the research team if they have any questions, or simply to follow a hyperlink straight to the relevant project page on REDCap. Here, they will be required to provide informed consent to take part in the study. All data will be collected, processed, and analyzed as described above for primary interviewees. Each individual who participates will be given a £20 (US $24.6) voucher in recognition of their time.

A patient and public involvement (PPI) group will be formed for this study. Members will be existing users of web-based STI self-sampling services. The research team will work with the PPI group to determine the different value propositions included in the study advertisements; determine strategies for effective retention of participants (eg, the level and timing of voucher payments provided, communication methods, and tone of messaging); draft the participant information and consent statements for the fRCT and nested qualitative study to ensure that they are clear and easy to understand; iteratively test the surveys before their use to maximize content validity, readability, clarity, and comprehensiveness; review fRCT findings and reflect on changes that should be made to recruitment and retention strategies before a definitive trial; and two members of the PPI group will also support the monitoring of the fRCT by serving as members of the SSC.

This fRCT is designed to assess whether and how it is possible to carry out a future definitive RCT to evaluate the effectiveness and cost-effectiveness of an intervention to increase condom use among young people. The rapid increase in use of web-based STI self-sampling services, accelerated by the COVID-19 pandemic, means that young people are increasingly required to self-manage their own sexual health screening. This, along with persistently high rates of STIs among this age group, makes the provision of behavior change support in this web-based context imperative.

This fRCT employs the proposed design and methods that would be used in a full RCT. This will enable the research team to develop, assess, and optimize all the systems and procedures required to run the full trial before time and resource commitments are allocated. Furthermore, the study has been setup to inform important aspects of this future trial, such as which advertisements results in the highest recruitment rate? Are there any friction points within the intervention content that may contribute to dropout? Could the tone, content, or frequency of communication with participants be improved? Should the progression criteria for this feasibility fRCT be satisfactorily met, the research team will refine intervention content and study procedures in line with this to further enhance participant recruitment and retention, and study efficiency.

As this is an fRCT, the study is being conducted on a much smaller scale than a future trial. Some technical aspects of the study that are manageable at this stage will likely need to be revised. In the present study, for example, the research team are reporting positive chlamydia test results (arising from M3 and M12 testing) to each relevant trust which is time-consuming and represents an avoidable data protection risk. An alternative would be for this information to be passed directly to the trusts by the self-sampling service provider. In the future-full trial, the team would seek to overcome such technical issues.

The unusual context in which this fRCT is being run should be acknowledged. The COVID-19 pandemic has affected sexual health services and sexual behavior in ways that could not have been imagined at the time this study was conceived. When recruitment to the fRCT was initiated, restrictions on public behavior imposed to control the spread of infection were still in place but were lifted approximately 3 months later. The impact of the environment at that time on STI testing and diagnoses will not be known until the release of annual STI data in Autumn 2022. Although an increase in the use of internet testing will serve to increase the size of the sampling pool drawn upon for this study, thus potentially accelerating recruitment, it should not impact upon the recruitment rate, as this is merely a reflection of the rate at which those who see the advert are converted into participants. It should, however, be acknowledged that the pandemic is likely to have both positively and negatively impacted on people’s time and/or willingness to engage in research, factors which for some will also have changed with changing restrictions and may serve to influence study recruitment and retention. It is also important to consider whether changes in sexual risk behavior during this time could lead to fewer cases of chlamydia being identified in the present study than anticipated. This concern is, however, not supported by community surveys, which indicate that a substantial proportion of people continued to engage in at-risk behavior during 2020 [28-30], and the most recent STI data which shows that among those seeking internet testing, chlamydia positivity was comparable between 2019 (23,426/284,050, 8.2%) and 2020 (33,294/381,744, 8.7%) [1].