How you offer patients an HIV test has a significant impact on the likelihood of them accepting tests, finds a study published by The BMJ.

In the first randomized controlled trial to evaluate consent for HIV testing, the results show that opt-out HIV testing can substantially increase the number of patients accepting tests, while opt-in schemes may reduce testing.

However, the researchers suggest that active choice testing (directly asking patients if they would like an HIV test) may best reflect patients' true preferences, and they call for further work to assess the effects of different approaches on patient behaviour.

Worldwide, approximately 37 million people are infected with HIV and approximately 46% of infections remain undiagnosed.

Testing for HIV infection remains a critical first step in controlling the epidemic. US guidelines endorse opt-out testing, and Europe has seen a trend toward this testing scheme.

However, over the past decade, relatively little research has been conducted to help us understand the effect of various consent methods on acceptance of testing.

To examine this important matter, researchers at the University of California decided to evaluate acceptance of three distinct approaches of offering HIV testing: "opt-in" (notifying patients that HIV testing was available but requiring them to ask specifically for testing), "active choice" (directly asking patients if they would like an HIV test), and "opt-out" (patients are told that they will be tested unless they decline).

They randomized 4,800 emergency department patients, who were aged 13-64 years, had a variety of medical complaints, and were not already known to be infected with HIV, to one of the three groups, with acceptance of a test as the outcome.

A total of 38% of patients in the opt-in group accepted a test, compared with 51% in the active choice group, and 66% in the opt-out group.

Furthermore, they found that patients identified as being at intermediate and high risk were more likely to accept testing than were those at low risk in all groups.

The opt-out effect was significantly smaller among those reporting high risk behaviors, but the active choice effect did not significantly vary by level of reported risk behavior.

"Our study provides evidence that small changes in wording can significantly affect patients' behavior and thus our understanding of their preferences and is crucial to providing patient centered care," conclude the authors.

In an accompanying editorial, researchers based in Denver, Colorado argue that to maximize test acceptance and subsequent new HIV diagnoses, "we must use evidence to drive decisions about the best way to conduct testing procedures."

They commend the study authors for helping to improve our understanding of how best to offer an HIV test, and say their results support the notion that "the ask" is a critical piece of the equation and is probably as important as "the test."