Earlier diagnosis of HIV infection through visual HIV testing prompts.

Abstract

Listen

Translate article

Late diagnosis of HIV infection remains a major issue, with implications for both patient outcomes and health care costs, with around 60% of patients in the north east of England diagnosed late with CD4 counts < 350 cells/μL.1 There are national guidelines available, which advise on medical conditions that should prompt consideration of HIV testing in patients.2 However, research by Lazarus et al.3 shows that in up to 82% of patients presenting with an HIV-related symptom, testing is not carried out at the earliest opportunity. Vijeratnam et al.4 proposed that HIV testing prompts on reports containing indicator conditions could help reduce the incidence of missed HIV screening opportunities. Research on computerized reminders in general practice has shown them to have a positive effect on clinician behaviour.5 The recently published National Institute for Health and Care Excellence (NICE) guideline HIV Testing: Encouraging Uptake (quality standard 157)6 advises that ‘Young people and adults newly diagnosed with an HIV indicator condition are offered an HIV test’ as this helps to reduce late diagnosis of HIV infection, improves HIV treatment outcomes and may improve the response to treatment of the indicator condition. In 2016, we investigated the likely impact of introducing visual testing prompts on earlier diagnosis of HIV infection. We found five patients (11%) in a 17-month period who had previous laboratory requests with clinical details suggestive of an HIV indicator condition. Visual testing prompts are likely to have led to an earlier HIV diagnosis in three of these patients.4 Subsequently, over a 4-day period, in autumn 2017, we reviewed the clinical request details of all specimens received in our laboratory. Of 241 sputum samples that were processed, nine (3.7%) included clinical details suggestive of HIV indicator conditions. Only one of these patients (11%) had previously been tested for HIV. Similarly, of 227 samples sent from general practice, 12 (5.3%) included clinical details suggestive of HIV indicator conditions. Of these patients, only three had been tested for HIV (25%). Our findings suggest that implementing an automated text-based visual HIV testing prompt on reports from microbiology samples may be an effective means of reducing the incidence of late HIV diagnoses in patients with HIV indicator conditions.