Non-targeted HIV testing in health settings – worthwhile?

A large, recently published French study, based in metropolitan Paris, places a question mark over the value of non-targeted HIV screening as a strategy to lower the number of undiagnosed infections and improve early detection.

Late diagnosis of HIV remains a common problem both in France – where, despite accessible testing, one-third of diagnoses are in conjunction with CD4 counts less than 200/μL – and in other countries.  National health authorities have therefore promoted untargeted testing in the US, the UK and France.    However, the strategy remains controversial.  The authors of this study claim that it is the first large-scale study to have assessed its effectiveness.

The testing was conducted in 29 Paris Emergency Departments (ED) – an ideal setting for assessing the impact of untargeted testing, given that 25% of French population visit an ED at least once a year, including low-income, uninsured and other subgroups that might not be reached in other health care settings.  Of the 20,962 eligible patients who visited the EDs during the periods when screening was operative (i.e. during the 6-week period randomly assigned to each participant ED), a total of 12,754, or 63% of those eligible, consented to the test.

The number of HIV cases newly diagnosed (18 or 0.14%) as a result of the study, though small, fell within the authors’ expectations.  The significant finding relates to the characteristics of those newly diagnosed.  With one exception, they all belonged to high risk groups – sub-Saharan African or men having sex with men (MSM).  8 were being seen for HIV related symptoms, 7 of them with advanced-stage disease.  Of the 18 newly diagnosed, 6 did not return for a follow-up visit despite repeated calls, 4 were hospitalized immediately, 8 returned for the follow-up.  Among the 12 patients successfully linked to care, only 4 had CD4 counts greater than 350μ/L.

Certainly, this intervention does not appear to have reached infected persons not belonging to high-risk groups.  The authors point out that an ED-based screening strategy limited to men aged 18-45 and African-born persons would have identified all new HIV infections for 50% fewer HIV tests!   Questionnaires administered to consenting patients demonstrate characteristics that seem broadly representative of the Paris population, and findings of a complementary study of covariates of HIV refusal in 7 EDs suggest that refusal-associated factors did not unduly bias the results of the study.  The authors conclude that their observations do not support the implementation of non-targeted screening in the ED setting.  Given this study group appears highly representative of the general population, the authors are also very sceptical of the likely value of untargeted screening in other healthcare settings – at least in France.

The applicability of these findings to other countries cannot be assumed.  But the study certainly highlights the need for additional country-specific studies of the effectiveness of the strategy of untargeted HIV testing.

Kayigan Wilson d’Almeida, Anne-Claude Crémieux et al., “Modest Public Health Impact of Nontargeted Human Immunodeficiency Virus Screening in 29 Emergency Departments”, Archives of Internal Medicine, published online 24th October 2011

http://archinte.ama-assn.org/cgi/content/full/archinternmed.2011.535

For national strategies supportive of untargeted HIV testing, see the following:

In the US:  http://www.cdc.gov/mmwr/preview/mmwrhtml/rr5514a1.htm;  in the UK http://www.hpa.org.uk/Publications/InfectiousDiseases/HIVAndSTIs/1011TimetotestHIVtesting/; in France http://www.sante.gouv.fr/plan-national-de-lutte-contre-le-vih-sida-et-les-ist-2010-2014.html.