A large randomised controlled trial of rapid HIV testing,
with or without risk-reduction counselling, has shown that the counselling is
unnecessary and that testing can be provided to more people without it. The
study, published in the October 23 issue of the Journal of the American Medical Association, lends support to
existing recommendations that only a brief discussion is provided before HIV
Risk-reduction counselling is an interactive process of
assessing risk, recognising specific behaviours that increase the risk for
acquiring or transmitting HIV, and developing a plan to take steps to reduce
risks. However, there is controversy over how effective brief risk-reduction
counselling is when it is provided in the context of HIV testing. Effectiveness
may vary according to the style of counselling provided, how long it lasts, where
it is provided and who receives it.
And there are concerns that providing it may act as a
barrier to significantly increasing the number of people who test for HIV and
do so regularly. Although the Centers for Disease Prevention and Control (CDC)
has long stated that prevention counselling is not always required before HIV
testing, in remains standard practice in some settings. While some healthcare
providers see it as difficult to provide, it is not always done well.
Lisa Metsch of Columbia University and colleagues therefore
designed a randomised control trial, comparing the outcomes of providing rapid,
point-of-care HIV testing with either risk-reduction counselling, or with
In the intervention group, participants received a 30-minute
counselling session, based on the previously validated RESPECT-2 model. The
counselling included a discussion of the individual’s risk behaviours and
negotiation of a realistic risk-reduction plan that the individual committed to. There was also a brief explanation of the HIV test and how its results should
In the control group, participants just received the
explanation of HIV testing – this took between two and four minutes.
To measure the impact on sexual behaviour, participants
returned six months later and were tested for syphilis, herpes, HIV,
gonorrhoea, chlamydia and (in the case of women) trichomonas vaginalis. They
also answered questions about their sexual behaviour.
It’s worth noting that the counselling intervention was not
intended to be a discussion of the pros and cons of testing, leading to the
individual giving informed consent. The study did not assess any outcomes in
relation to this.
The trial was conducted at sexually transmitted infection (STI)
clinics in nine American cities in 2010 and 2011. A total of 5012 individuals
were randomised – two-thirds were under the age of 25, two-thirds were male, and
one quarter were men who have sex with men.
At baseline, 42.7% had a sexually transmitted infection,
including 1.0% with previously undiagnosed HIV.
Six months later, 12.3% of those who had received counselling
and 11.1% of those in the control group had acquired a new infection. The
difference between these figures is not statistically significant – in other
words there was no real difference between the two groups.
There were also no significant differences when looking at
any specific STI, or when sub-groups were analysed in terms of age, ethnicity,
gender or substance use. However, men who have sex with men who received the
counselling intervention had more
infections (18.7%) than men who have sex with men in the control group (12.5%).
Furthermore, there were no differences between the
intervention and control groups in terms of the number of partners and amount
of unprotected sex that they reported.
A financial analysis showed that it cost $23 to test each
person in the control group, rising to $56 in those receiving counselling. This
was due to higher training costs, extra staff time, quality assurance
activities and other overheads.
As the counselling intervention was not effective and cost
considerably more, providing it is not an efficient use of resources, according
to the investigators.
“A more focused approach to providing information at the
time of testing may allow clinics to use resources more efficiently to conduct universal
testing, potentially detecting more HIV cases earlier and linking and engaging
HIV-infected people in care,” they say.
In an editorial accompanying the study, Jason Haukoos and Mark
Thrun of the University of Colorado argue that a shift away from risk-reduction
counselling is integral to reducing the number of people with undiagnosed
HIV. “At a population level, to reach a
tipping point at which most clinicians screen all their adult patients for HIV
infection, the perception that counseling is necessary must not trump the reality
that in practice, counseling is typically ineffective,” they say.