PLoS By Category | Recent PLoS Articles

Critical Care and Emergency Medicine - Infectious Diseases - Public Health and Epidemiology - Urology

Twelve Months of Routine HIV Screening in 6 Emergency Departments in the Paris Area: Results from the ANRS URDEP Study
Published: Tuesday, October 02, 2012
Author: Enrique Casalino et al.

by Enrique Casalino, Bruno Bernot, Olivier Bouchaud, Chakib Alloui, Christophe Choquet, Elisabeth Bouvet, Florence Damond, Sandra Firmin, Aurore Delobelle, Beatrice Ename Nkoumazok, Guillaume Der Sahakian, Jean-Paul Viard, Olivier Zak Dit Zbar, Elisabeth Aslangul, Anne Krivine, Julie Zundel, Jade Ghosn, Patrice Nordmann, Yann-Erick Claessens, Tassadit Tahi, Bruno Riou, Agnès Gautheret-Dejean, Christine Katlama, Pierre Hausfater, Françoise Brun-Vézinet, Dominique Costagliola


In October 2009 the French National Authority for Health recommended that HIV testing be proposed at least once to all persons aged 15 to 70 years in all healthcare settings. We examined whether routine HIV screening with a rapid test in emergency departments (EDs) was feasible without dedicated staff, and whether newly diagnosed persons could be linked to care.


This one-year study started in December 2009 in 6 EDs in the Paris area, using the INSTI™ test. Eligible individuals were persons 18 to 70 years old who did not present for a vital emergency, for blood or sexual HIV exposure, or for HIV screening. Written informed consent was required.


Among 183 957 eligible persons, 11 401 were offered HIV testing (6.2%), of whom 7936 accepted (69.6%) and 7215 (90.9%) were tested (overall screening rate 3.9%); 1857 non eligible persons were also tested. Fifty-five new diagnoses of HIV infection were confirmed by Western blot (0.61% (95% CI 0.46–0.79). There was one false-positive rapid test result. Among the newly diagnosed persons, 48 (87%) were linked to care, of whom 36 were not lost to follow-up at month 6 (75%); median CD4 cell count was 241/mm3 (IQR: 52–423/mm3).


Screening rates were similar to those reported in opt-in studies with no dedicated staff. The rate of new diagnoses was similar to that observed in free anonymous test centres in the Paris area, and well above the prevalence (0.1%) at which testing has been shown to be cost-effective.