Opt-out HIV testing not efficient in US, suggests researcher

A researcher at Johns Hopkins University has calculated that well-funded, carefully targeted HIV testing and counselling programmes will detect significantly more undiagnosed HIV infections, prevent more incident HIV infections, and be much more cost effective than the opt-out testing policy being proposed by the US Centers for Disease Control and Prevention (CDC).

In an article published in the June edition of PLoS Medicine, Dr David Holtgrave estimates that an HIV testing and counselling programme costing the same amount as opt-out testing, but highly targeted at those with the greatest risk of HIV, would detect almost 75% of individuals with undiagnosed HIV in the US; help prevent over a third of all new HIV infections each year; and cost only $59,000 per infection averted compared to a cost of almost a quarter of a million dollars per infection averted for opt-out testing.

However, a separate study, published in the June 22^nd edition of Morbidity and Mortality Weekly Report did find some value in opt-out testing. Of the individuals who tested HIV-positive after being offered an opt-out test in a hospital emergency department, almost 50% belonged to a group with a recognised HIV risk but remained undiagnosed.

Opt-out or targeted testing?

In September 2006, the CDC recommended opt-out HIV testing as part of routine healthcare for all individuals aged between 13 and 64. To help remove potential barriers to HIV testing, the CDC also proposed that an individual risk assessment, pre-test counselling and specific informed consent for an HIV test would no longer be required. Post-test counselling to individuals with a high-risk of HIV will only be offered “so long as the counselling does not become a barrier to routine testing.”

The CDC’s recommendations received a mixed response. They were welcomed by the American Medical Association, which expressed the hope that opt-out testing would lead to the prompt diagnosis of HIV and help detect some of the 250,000 individuals with undiagnosed HIV in the US. Others, however, were less enthusiastic. In particular, the National Association of People with AIDS expressed its concern about issues of informed consent and counselling. Indeed, studies sponsored by the CDC have shown that client-centred counselling accompanying an HIV test can reduce the incidence of sexually transmitted infections by a fifth.

Dr Holtgrave noted that neither the CDC, nor any of the organisations endorsing routine, opt-out testing, had conducted any analysis to determine the costs and consequences of the policy. He therefore developed a model to estimate the potential impact of four different testing scenarios on:

• The number of undiagnosed infections detected.
• The number of infections averted each year.
• Cost per-infection averted.

The four scenarios examined were:

• The CDC’s recommended opt-out testing.
• Opt-out testing leading to higher levels of risk behaviour because of the lack of counselling.
• Opt-out testing accompanied by client-centred counselling.
• Using the resources devoted to opt-out testing for a programme of HIV testing and counselling targeted at settings where there is a high HIV prevalence.

Dr Holtgrave estimated that routine, opt-out testing would lead to 65.5 million people being screened for HIV in the first year. Of the 250,000 individuals in the US who are HIV-positive, but unaware of this infection, some 56,000 would have their infection diagnosed. Approximately 3,600 of the 40,000 incident HIV infections per year in the US would be averted. The programme would cost over $864 million at a cost of $237,000 per infection averted.

Not offering counselling as part of opt-out testing, would, Dr Holtgrave calculated, reduce the number of infections averted by over 500 and mean that the cost per infection prevented would increase to over $280,000.

Although offering client-centred counselling as part of opt-out testing would increase the annual costs of the programme to over $1,419 million, it would prevent an additional 1,700 incident infections at a cost of $269,000 per infection averted.

However, the most successful programme by far would be one that used additional funds for HIV testing to target those most at risk of HIV. Dr Holtgrave calculated that such a programme could lead to the diagnosis of over 188,000 of the 250,000 people with undiagnosed HIV in the US, prevent 12,000 of the 40,000 new HIV infections per year in the US, and cost only $59,000 per infection averted. Even if HIV prevalence were as low as 0.3% and there was no benefit from counselling, targeted testing programmes would still be preferable to opt-out testing without counselling.

“If $864.2 million could be made available for a testing or a counselling and testing programme, it would appear that the better investment would be a highly targeted program. This testing strategy could identify roughly three-fourths of persons in the US now unaware that they are living with HIV infection, and prevent about 36.4% of the incident HIV infections”, concludes Dr Holtgrave.

Opt-out testing in emergency departments

A separate study published in the June 22^nd edition of Morbidity and Mortality Weekly Report (MMWR) found, however, that 1% of patients offered a routine, opt-out HIV test when attending a hospital Accident and Emergency department, tested HIV-positive. The study included three hospitals, one in Los Angeles, one in New York City and one in Oakland, California. Between early 2005 and spring 2006, 19,500 patients were offered an HIV test. All reported being HIV-negative or unsure of their HIV status. A total of 9,365 patients accepted a rapid HIV test and of these 97 tested HIV-positive. Nearly all (85, 88%) then entered HIV care.

Almost half of these individuals diagnosed with HIV after routine testing in accident and emergency were not in a group with a recognised high risk of HIV.

The same edition of MMWR contains a study looking at HIV testing targeted at racial and ethnic minority men at gay pride events in the US. A total of 133 men had rapid HIV tests at eleven different events and eight men (6%) tested HIV-positive.