Behavioural intervention to reduce sexually transmitted infections in people aged 16–24 years in the UK: the safetxt RCT

Abstract

Background The prevalence of genital chlamydia and gonorrhoea is higher in the 16–24 years age group than those in other age group. With users, we developed the theory-based safetxt intervention to reduce sexually transmitted infections. Objectives To establish the effect of the safetxt intervention on the incidence of chlamydia/gonorrhoea infection at 1 year. Design A parallel-group, individual-level, randomised superiority trial in which care providers and outcome assessors were blinded to allocation. Setting Recruitment was from 92 UK sexual health clinics. Participants Inclusion criteria were a positive chlamydia or gonorrhoea test result, diagnosis of non-specific urethritis or treatment started for chlamydia/gonorrhoea/non-specific urethritis in the last 2 weeks; owning a personal mobile phone; and being aged 16–24 years. Allocation Remote computer-based randomisation with an automated link to the messaging system delivering intervention or control group messages. Intervention The safetxt intervention was designed to reduce sexually transmitted infection by increasing partner notification, condom use and sexually transmitted infection testing before sex with new partners. It employed educational, enabling and incentivising content delivered by 42–79 text messages over 1 year, tailored according to type of infection, gender and sexuality. Comparator A monthly message regarding trial participation. Main outcomes The primary outcome was the incidence of chlamydia and gonorrhoea infection at 12 months, assessed using nucleic acid amplification tests. Secondary outcomes at 1 and 12 months included self-reported partner notification, condom use and sexually transmitted infection testing prior to sex with new partner(s). Results Between 1 April 2016 and 23 November 2018, we assessed 20,476 people for eligibility and consented and randomised 6248 participants, allocating 3123 to the safetxt intervention and 3125 to the control. Primary outcome data were available for 4675 (74.8%) participants. The incidence of chlamydia/gonorrhoea infection was 22.2% (693/3123) in the intervention group and 20.3% (633/3125) in the control group (odds ratio 1.13, 95% confidence interval 0.98 to 1.31). There was no evidence of heterogeneity in any of the prespecified subgroups. Partner notification was 85.6% in the intervention group and 84.0% in the control group (odds ratio 1.14, 95% confidence interval 0.99 to 1.33). At 12 months, condom use at last sex was 33.8% in the intervention group and 31.2% in the control group (odds ratio 1.14, 95% confidence interval 1.01 to 1.28) and condom use at first sex with most recent new partner was 54.4% in the intervention group and 48.7% in the control group (odds ratio 1.27, 95% confidence interval 1.11 to 1.45). Testing before sex with a new partner was 39.5% in the intervention group and 40.9% in the control group (odds ratio 0.95, 95% confidence interval 0.82 to 1.10). Having two or more partners since joining the trial was 56.9% in the intervention group and 54.8% in the control group (odds ratio 1.11, 95% confidence interval 1.00 to 1.24) and having sex with someone new since joining the trial was 69.7% in the intervention group and 67.4% in the control group (odds ratio 1.13, 95% confidence interval 1.00 to 1.28). There were no differences in safety outcomes. Additional sensitivity and per-protocol analyses showed similar results. Limitations Our understanding of the mechanism of action for the unanticipated effects is limited. Conclusions The safetxt intervention did not reduce chlamydia and gonorrhoea infections, with slightly more infections in the intervention group. The intervention increased condom use but also increased the number of partners and new partners. Randomised controlled trials are essential for evaluating health communication interventions, which can have unanticipated effects. Future work Randomised controlled trials evaluating novel interventions in this complex area are needed. Trial registration This trial is registered as ISRCTN64390461. Funding This project was funded by the National Institute for Health and Care Research (NIHR) Public Health Research programme and will be published in full in Public Health Research; Vol. 11, No. 1. See the NIHR Journals Library website for further project information.

A qualitative study exploring experiences of the safetxt digital health intervention to reduce sexually transmitted infections in young people in the UK

“…At the later stages of thematic analysis, Microsoft Word 2019 was used to integrate and triangulate themes developed by both researchers from both data sources, based on comparisons and team discussions. During analysis meetings with the research team, results from open feedback comments (source 1) and interviews (source 2) were triangulated with quantitative trial data (including primary, secondary and intermediate outcomes)19 20 and data from telephone interviews conducted as part of the 2013 feasibility trial 2–3 weeks after starting messages17 looking for consistencies and inconsistencies across the different data sources and searching for deviant cases.…”

“…Control group participants received a monthly untailored text message asking for information about changes in postal or email addresses. The safetxt intervention did not reduce STIs, there were slightly more infections in the intervention group with 22.2% (693/3123) versus 20.3% (633/3125) in the control group (OR 1.13, 95% CI 0.98 to 1.31) 19 20. There were some increases in self-reported precautionary behaviours such as condom use at last sex, OR 1.14, 95% CI 1.01 to 1.28 19 20.…”

“…The safetxt intervention did not reduce STIs, there were slightly more infections in the intervention group with 22.2% (693/3123) versus 20.3% (633/3125) in the control group (OR 1.13, 95% CI 0.98 to 1.31) 19 20. There were some increases in self-reported precautionary behaviours such as condom use at last sex, OR 1.14, 95% CI 1.01 to 1.28 19 20. Although our intervention did not target sexual partnerships, we assessed at 1-year follow-up the proportion of people who had two or more partners since joining the trial and found that it was slightly higher in the intervention vs control group (56.9% vs 54.8%, OR 1.11, 95% CI 1.00 to 1.24).…”

“…This result, however, was not statistically significant (p=0.06) but could have contributed to the unexpected trial outcome. Other quantitative results, including on intermediate outcomes, did not clarify either, why a statistically significant effect was shown for the condom use outcome, but not for the biologic trial outcome 19 20…”

“…Intervention text messages were sent with decreasing frequency over the period of 12 months (online supplemental file 1). 18 Our randomised controlled trial to establish the effects of the intervention on STI, condom use, partner notification and STI testing before sex with new partners was conducted among 6250 people aged 16–24 diagnosed at UK sexual health clinics with chlamydia, gonorrhoea or non-specific urethritis 19 20. Control group participants received a monthly untailored text message asking for information about changes in postal or email addresses.…”