Acute presentations of HIV are still missed in low prevalence areas L Ratcliffe, 1 S Thomas, 1 N J Beeching, 1,2 P A Phillips-Howard, 3 M Taegtmeyer 2 ABSTRACT Objectives To evaluate missed opportunities and delays in the diagnosis of HIV in a low prevalence setting over a 24 year period. Methods Patients with acute presentations of HIV were included in a retrospective note based review. Data were compared from acute presentations in 1985e2001 (88/ 241 new patients) with 2005e2007 (99/136 new patients). The number of recorded clinical and laboratory clues to infection and subsequent time delays to diagnosis of HIV were evaluated. Results The findings reflect the shifting demographics of HIV in the UK over the past two decades, exemplified by an eightfold increase in tuberculosis at presentation. Despite recording clinical stigmata of HIV (clues) in the notes, the number of missed clues increased, and many clinicians failed to request HIV testing. The median delay between presentation and diagnosis reduced from 5 to 1 day (p<0.001), and mortality dropped from 14% to 4% among patients presenting with acute symptoms. However, there was still a delay of more than 30 days before diagnosis for almost one in five patients. Conclusions Despite some improvement and better awareness, there are still significant delays before hospital doctors consider the diagnosis of HIV for patients in low prevalence areas, even among some patient groups with high risk. Hospitals should consider moving to opt-out routine HIV testing of all medical admissions. INTRODUCTION An estimated 83 000 people infected with HIV were reported in the UK in 2008, of whom the highest proportion live in London. HIV prevalence in North West England is low, with an average of 0.79 per 1000 persons reported infected. 1 Among adults aged over 15 years who had a new diagnosis of HIV made in 2008, 32% had a CD4 cell count <200310 6 /l within 3 months of diagnosis. 1 Patients who present late in the course of HIV disease and with low CD4 cell counts (usually <200310 6 /l) have higher mortality, 2e4 prolonged hospital stays, and increased hospital related costs. 5 These acute presentations to healthcare profes- sionals may not be recognised by the non-specialist, prompting the Department of Health in the UK to encourage HIV testing in all settings, 6 supported by new guidance on HIV testing issued by national specialist societies and patient groups in 2008. 7 Current guidance recommends that HIV tests be offered to all persons living in areas of the UK where prevalence exceeds 2 per 1000 and that targeted testing, based on symptoms, should be conducted elsewhere. 6 General physicians may be confronted with patients who present with acute disease either from seroconversion illnesses or from advanced HIV disease. While seroconversion illnesses may be diagnostically challenging, advanced HIV generally results in a number of obvious clues or ‘stigmata’ of immunosuppression and tables of suggestive conditions are available in HIV guidance for non-specialists. 8 Missed opportunities for the diagnosis of HIV have been described elsewhere, particularly in low prevalence areas 9 and when patients present with HIV seroconversion illnesses. 10 11 In this paper we report on a study conducted in the Liverpool Tropical and Infectious Disease Unit (TIDU) (box 1), a tertiary referral unit in North West England, to identify the impact of enhanced training on the timely diagnosis of HIV among non- specialist, general medical healthcare professionals. Using patient data from a 24 year period, we set out to assess the number of diagnostic clues that were recorded and missed at the first HIV related illness presentation to general physicians, to docu- ment delays to diagnosis of HIV and to identify factors associated with delay. We compared two time periods to see if the number of missed clues and delays in diagnosis had improved in recent years. METHODS Data collection A retrospective case notes based review was performed of all HIV related presentations in Merseyside that were referred to the TIDU in Liverpool. Two cohorts of patients were evaluated and the results from the cohorts were compared. Period 1 included patients found to be HIV positive and who were referred to the TIDU, as an inpatient or outpatient, between 1 January 1983 and 31 December 2001 (figure 1). The data for this period were collected in 2002. In period 2 we reviewed case notes from a similar sized cohort of HIV patients presenting to the TIDU between January 2005 and December 2007. The cohort again consisted of newly diagnosed HIV patients and the data for this group were collected retrospectively in 2009. Patients referred solely to the clinics of the Liver- pool Sexual Health Centre were excluded from either cohort. A pretested proforma was used to gather baseline data about patients on their sex, sexuality, age at HIV diagnosis, country of origin, previous HIV testing, experience of foreign travel, and intravenous drug use, together with clinical and laboratory features of infection and details of the time between presentation and diagnosis of HIV. The same proforma was used for both cohorts. 1 Tropical and Infectious Disease Unit, Royal Liverpool University Hospital, Liverpool, UK 2 Liverpool School of Tropical Medicine, Pembroke Place, Liverpool, UK 3 Centre for Public Health, Liverpool John Moores University, Liverpool, UK Correspondence to Dr Miriam Taegtmeyer, Liverpool School of Tropical Medicine, Pembroke Place, Liverpool L3 5QA, UK; m.taegtmeyer@liverpool.ac.uk Received 5 September 2010 Accepted 5 December 2010 Published Online First 21 January 2011 170 Postgrad Med J 2011;87:170e174. doi:10.1136/pgmj.2010.109801 Original article group.bmj.com on March 15, 2011 - Published by pmj.bmj.com Downloaded from