Is justified in areas of low HIV prevalence

EDITOR—The recent papers by Postma et al and Simpson et al highlight the difficult issues in establishing a policy to test for HIV infection in antenatal clinics.1 2

Postma et al's paper examines the cost effectiveness of universal, voluntary testing of pregnant women in England in terms of healthcare costs to the NHS. Although no cut off point at which the cost for each life year gained becomes acceptable has been defined for England, a cut off point of around $50 000 is suggested in the United States They conclude that in areas of high prevalence, such as London, universal, voluntary antenatal screening of pregnant women is cost effective; how- ever, in areas of low prevalence, screening may not be justified in terms of cost effectiveness.

Screening for HIV infection in antenatal clinics fulfils most of Wilson and Junger's criteria as a good test.3 HIV infection can be asymptomatic; the tests are simple, relatively pain free, sensitive and specific; and there is effective treatment that can substantially reduce the risk of infection in the fetus Yet universal testing is not performed in most antenatal clinics. This contrasts with the ad hoc way in which universal screening for Down's syndrome has been introduced in most antenatal clinics—yet none of the individual tests currently available fulfil many of Wilson and Junger's criteria. Tests which are of low risk to the fetus are not very sensitive or specific. In spite of their unproved record, tests are available, sometimes with the patient bearing the cost of testing.

There is stigma surrounding HIV testing; an opt-in policy serves only to reinforce this by testing only those in obvious high risk groups. In low prevalence areas, where there are not large numbers of “high risk” women, those with HIV infection may be even harder to detect from a screening questionnaire as they mingle with the rest of the population. In Portsmouth a third of our HIV positive patients do not fall into any recognised high risk group. Simpson et al's paper shows that for an effective screening programme to be instigated, an opt-out policy of testing is the only model which is effective in antenatal screening; the uptake of the test increased to 88% compared with a 35% uptake with an opt-in policy. This policy should be adopted not only for areas of high prevalence of HIV testing but also for areas where the prevalence of HIV infection is low, so that the opportunity to reduce infection in the neonate and treat the asymptomatic mother is not missed.

Is acceptable to women

EDITOR—Simpson et al describe a large increase in the uptake of antenatal HIV testing following a change from an opt-in approach to routine testing unless the woman declined.1 Their study took place in Edinburgh, and they questioned whether the outcome would be similar in London, with its greater cultural and linguistic complexities.

We would like to report our experience of changing the process for antenatal HIV testing in a deprived, multicultural London population with a high prevalence of HIV infection among pregnant women (0.86% in 1998;Public Health Laboratory Service, unpublished data). An antenatal HIV testing programme was originally introduced in June 1995; women were offered a test, with signed consent, after discussion with midwives Despite staff training and support, uptake of testing increased only from 25% to 33% by 1998 and uptake varied widely between individual midwives.2

In 1998 HIV testing was integrated into the recommendation of all routine antenatal screening for infection. The new policy was launched on World AIDS Day, 1 December, resulting in supportive local publicity. Currently an updated leaflet giving information about hepatitis B, syphilis, rubella, and HIV infection is sent to women before booking. Discussion and women's decisions about the tests are summarised without a requirement for signed consent. Copies of all negative results are sent to women for inclusion in their handheld notes. Positive results of tests for HIV antibody are given by a consultant obstetrician in conjunction with a health adviser.

The new policy resulted in a rise in uptake in December 1998 to 197/241(82%) of women booked through the antenatal clinic. This has been sustained, with an overall uptake of 1440/1605 (90%) in the first six months for antenatal clinic attendees; community figures are currently being analysed. Antenatal HIV testing identified six infected women between December 1998 and April 1999; all are accepting interventions to minimise vertical transmission.

Women's views were assessed by a questionnaire sent to all women who booked in December 1998. The response rate was 39%, with 88% of responders in favour of the HIV test being recommended as a routine blood test in pregnancy. The only negative comments related to the length of time for results to be received by post, and this has since been dealt with.

HIV testing should be a routine part of antenatal care. In our experience this is acceptable to women and results in a dramatic increase in uptake.