• HIV
• HIV testing

The UK Health Protection Agency (HPA) estimates that 26% of the UK population infected with HIV are undiagnosed.1 This is estimated using data sources including HIV testing of unlinked anonymised blood samples taken for syphilis testing from attenders at 16 UK genitourinary medicine clinics, the GUM Anon survey.2 This estimated proportion of undiagnosed HIV infections may be an overestimate caused by individuals who are aware of their HIV infection not reporting this when testing for sexually transmitted infections (STIs).3 We conducted a survey to establish whether HIV-infected individuals who attend our centre for HIV care go to a different centre for STI screening and if so, whether they report their HIV status.

All HIV-positive individuals attending the HIV clinic at the Mortimer Market Centre (MMC), a central London sexual health and HIV service, for routine outpatient clinic appointments during a 10-day period were invited to complete an anonymous questionnaire. The questionnaire asked about age, sex, whether participants had ever attended a different UK sexual health clinic for an STI screen since starting to receive their HIV care at MMC and whether the clinic knew they were HIV positive.

We obtained 102 completed questionnaires from an eligible population of 262 individuals attending the clinic during the study period, a response rate of 39%. The median age of the responders was 41 years (interquartile range: 35–49 years). Eleven responders (10.8%, CI 5.5% to 18.5%) indicated that they had ever had tests for STIs at a different UK sexual health clinic since receiving their HIV care at MMC. Two responders (1.9%, CI 0.2% to 6.9%) stated that the sexual health clinic at which they were screened for STIs was not aware of their HIV status.

Although the response rate was low, we have confirmed that individuals who attend a central London combined sexual health and HIV service do sometimes attend other centres for STI screening and some do not reveal their HIV status. The survey recruited patients from one central London clinic and the behaviour of this population may not be representative of the whole UK HIV-infected population. While this study was too small to provide a precise estimate of the frequency of this behaviour nationwide, we have established that it does occur. This has consequences for our estimate of the true undiagnosed fraction of HIV. To improve the validity of the estimates of the undiagnosed fraction would require extending the study to a larger sample and including other HIV treatment centres.