Sexually transmitted infections in the UK

Since 1998 there has been a substantial increase in reported cases of sexually transmitted infection (STI), most strikingly in the 16–24 years age group.¹ Across genitourinary medicine (GUM) clinics in the UK in 2007, young people accounted for 65% of chlamydia cases, 50% of cases of genital warts and 50% of gonorrhoea infections.¹ Chlamydia is the most common STI in under-25s. Since 1998, the rate of diagnosed chlamydia has more than doubled in the 16–24 years age group (from 447 per 100,000 in 1998 to 1102 per 100,000 in 2007). This may be because of a combination of a higher proportion of young people testing, improved diagnostic methods and increased risk behaviour.¹ Chlamydia infection can frequently go undetected, particularly in women, as it is often asymptomatic.¹ If left untreated, chlamydia can lead to pelvic inflammatory disease and infertility in women. This highlights the importance of testing this higher-risk age group to ensure prompt diagnosis and treatment.

It is estimated that 11–12% of 16- to 19-year-olds presenting at a GUM clinic with an acute STI will become reinfected within a year.² In order to minimise reinfection, preventative measures are required, including effective methods of notifying partners to ensure rapid diagnosis and treatment and reduce the likelihood of index patients being reinfected from the same source.

Current partner notification practice

Partner notification (PN) is an essential element of STI control. It supports patients by enabling diagnosis and treatment for their sexual partners and is an effective way of identifying at-risk and infected persons. PN has been defined as the spectrum of ‘public health activities in which sexual partners of individuals with STD are notified, informed of their exposure and offered treatment and support services’.³

Treatment of partners remains important for three reasons:

1. to protect the original patient from reinfection and its health consequences
2. to prevent the further spread of infection by infected partner(s)
3. to reduce transmission of STIs (at a population level) by shortening the duration of infection, which is a key determinant of onward transmission rates.^4,5

In the UK, PN has been supported mainly by specialist health advisers (HAs) based in GUM clinics. However, as a result of the growth of the National Chlamydia Screening Programme (NCSP) in England, this role has also increasingly been taken on by local chlamydia screening offices (CSOs) in the community (including general practice surgery settings).⁶

Patient referral, in which the patient contacts their sexual partner(s) to arrange testing and/or treatment, supported by the HA, is the most commonly used method of referral, as most partners prefer to contact their partners themselves. Moreover, this additional support is not available in some settings.⁷ However, provider and contract referral are also used.³ In provider referral, the HA offers to contact one or more of the index’s partners on their behalf; in contract referral, patients are asked to agree to a specialist HA informing their partner(s) if this has not been done after a verbally agreed period of time. These are important services to reach partners who might not otherwise be informed, for example casual or ex-partners. The extent to which patient referral is supplemented by health providers contacting partners on the patient’s behalf, and with their agreement (provider or contract referral), is variable.⁸ There is, to date, no three-way comparison between patient referral alone, provider referral and contract referral. One trial, situated in a service with very high PN rates, suggested no advantage of contract referral over patient referral. In this study, contract referral achieved 1.15 partners tested per index case, versus 1.27 for patient referral.⁹ However, in a setting with lower rates of successful PN, the offer of contract referral if partners did not present within 3 days achieved 0.62 partners tested per index case of gonorrhoea compared with 0.37 both for simple and for enhanced patient referral.¹⁰ Most relevant to this trial is a study by Katz et al.,¹¹ which achieved partner treatment of rates of 0.72 per index case for provider referral, and 0.22 and 0.18 for two differing forms of patient referral.

Partner notification in the primary care setting

Sexually transmitted infections are increasingly diagnosed and treated within the primary care setting,^12,13 and around a third of patients presenting to GUM clinics first seek care from their general practice surgery.¹⁴ Maximising the quality of care for patients seen in general practice with STIs has considerable potential for public health gain. However, clinical structure and process standards for PN remain poorly implemented in primary care.¹⁵

Partner notification has been shown to present particular challenges to primary care practitioners.¹⁶ There is historical evidence that only 30% of UK general practitioners would treat a partner,¹⁷ and as few as 13% of index patients have a documented attendance at a GUM clinic.¹⁸ General practitioners (GPs) may overestimate how much PN they do,¹⁹ while patients treated by a GP are more likely to require retreatment than those treated in a GUM clinic at the outset.²⁰ This may be because of a lack of established processes for PN within many general practices.

The GP or practice nurse faces several specific challenges in PN compared with sexual health services. The sexual partners of index patients are often not registered at the practice, and general practice has no mechanism for enabling STI treatment in this situation, or for following up compliance. Even if partners are registered at the practice, the duty of confidentiality to individual patients presents difficulties in PN if the index patient declines to discuss their infection with the partner. Staff in general practice may not be confident in handling common issues with PN, which can be time-consuming and require training or support from a HA.²¹ Patients diagnosed with a STI may be less willing to give frank information on number of partners, particularly casual or concurrent partners, to familiar staff than to a specialist STI service,^21,22 although this has not been shown consistently.^7,23

Recognising these difficulties, guidance from the National Institute for Health and Care Excellence recommended that all patients with a STI, regardless of the setting of diagnosis, should be offered support in PN, which may be within the primary care setting or through referral to a PN specialist.²⁴ It did not, however, specify standards for content or delivery of this support. A high-quality randomised controlled trial (RCT) has demonstrated that specifically trained practice nurses can achieve PN outcomes equivalent to those achieved by referral to attend a GUM clinic.⁷ This trial provided important evidence that PN in the form of patient referral can be undertaken within a highly motivated and specifically trained primary care setting. However, it does not provide an adequate model for a comparison between patient, provider and contract referral in a primary care setting, as it is unlikely that provider or contract referral could become routine work among all general practice surgeries in the foreseeable future, especially in those with no particular interest in sexual health.

The National Chlamydia Screening Programme

Evidence from a Department of Health-funded chlamydia screening pilot study of opportunistic screening in primary and secondary health-care settings demonstrated that it was feasible and acceptable to test women for chlamydia using urine samples using these services.²⁵ In a separate Health Technology Assessment (HTA)-funded programme, the Chlamydia Screening Studies (ClaSS) project,²³ a cross-sectional study of 19,773 women and men aged 16–39 years, selected from general practice, invited participants to collect urine and vulvovaginal swab (for women) specimens at home and post to the laboratory for testing. These studies confirmed that the prevalence of chlamydia was highest amongst those aged under 25 years and was similar in both men and women. The urine and swab tests were also shown to be suitable samples for diagnostic testing with nucleic acid amplification tests (NAATs).

The NCSP in England, which was established in 2003, is an opportunistic testing programme which has been rolled out over a number of years. The key objectives of the programme are to ‘prevent and control chlamydia through early detection and treatment of infection; and reduce onward transmission to sexual partners and prevent the consequences of untreated infection’.²⁶ Screening has come to contribute an increasing proportion of primary care STI diagnoses and must be considered as part of the relevant population when conducting any trial of PN in primary care.

There has been marked geographical variation in the mix of services contributing NCSP tests, and in positivity by setting, with positivity in educational settings as low as 3%.²⁷ By 2010, primary care was increasingly identified by NCSP as a key setting. In 2007/8, the highest percentage of tests was conducted within the community contraceptive services, at 25.9%; a further 17.9% were tested in youth services, 13.4% in education and 12.6% in general practice.²⁸ Coverage of the target 16- to 24-year-old population varied between 0% and 14% in different primary care trusts (PCTs), with local NHS targets set at 15% coverage.²⁹ In 2007, over 270,729 screens were performed, with 9.5% females and 8.4% of males testing positive.¹

The Department of Health Public Health Outcomes Framework 2013–16, published in 2012, now specifies the number of chlamydia diagnoses per unit population as a public health outcome target for England, replacing coverage targets.³⁰

The NCSP targets under-25s who are sexually active to test for chlamydia and offers sexual health promotion advice. Although the programme has a national co-ordination team, the organisation of each geographical area is determined at local level. In its early years, the NCSP encouraged the development of varying models of service provision including testing in outreach settings such as colleges, prisons, youth services and even shopping centres. Possible location categories for the treatment of index patients (as specified by the NCSP) are GUM, family planning, CSO, general practice, pharmacy and other.

The organisation of PN varies markedly, with GUM clinics providing this service in some areas, and community-based local chlamydia co-ordinators in others, while some high-volume areas have not made specified provision to support PN in primary care.

The commissioned research and its implementation

Given the increase in reported STIs over the last 10 years¹ and variable management of PN in the UK,³¹ there was an evident need for further robust, evidence-based research on the effectiveness of different methods of PN. The National Institute for Health Research (NIHR) HTA programme therefore sought to commission a RCT to evaluate the clinical effectiveness and cost-effectiveness of different existing approaches to PN, specifically in primary care. The HTA tendered for a RCT comparing patient referral, provider referral and contract referral for patients diagnosed with a bacterial STI in the primary care setting. This study was commissioned as a full-scale RCT, and included health economics, mathematical modelling and patient factor components. It also included the building of a web-based referral tool that could both facilitate referral from primary care to specialist GUM services where HAs were based and collect summary outcome data on PN. It was envisaged that this web tool could subsequently have a role in NHS clinical care.

At the outset of the research, we undertook initial pilot work (phase 1) on the assumption that the RCT would scale up in the form in which it had been commissioned. However, it became clear that this was not likely to be feasible because of recruitment problems. We then undertook additional work aimed at improving recruitment. This included exploring potential for improvement in our existing recruitment strategy through literature review and consultation (phase 2), implementing the new recruitment strategy (phase 3) and, finally, implementing an additional novel recruitment strategy being used in a related study elsewhere (phase 4). Unfortunately, these efforts were ultimately unsuccessful, but they provide some interesting lessons relevant to the future planning of certain types of trial in primary care.

Phase 1 also identified some interesting conceptual and operational challenges to the planned comparison between the three proposed arms of this trial. As we sought to develop standardised approaches to provider referral and contract referral in consultation with practitioners, it emerged that practitioners found clear-cut distinctions between these approaches problematic. We explored and addressed this using qualitative methods.

Because of recruitment failure, we were unable to compare the costs of achieving our proposed outcomes. We are, however, able to present the costs of the intervention arms used in this study, and also of the intensive recruitment strategy used in phase 4, which may inform recruitment plans for other studies.

Unfortunately, because of insufficient recruitment, the patient factors and modelling work were not possible and they are not reported.

Outline of report

This report presents the various phases of recruitment, focusing on findings of relevance to other future studies in the fields of PN, sexual health in primary care or recruitment challenges. The four phases of recruitment mentioned above are presented, alongside methodological chapters exploring the practice of PN and the economic findings, and a chapter setting out the health landscape of chlamydia screening and its implications for research in sexual health.