Cost-effectiveness in the developed world

The interventions available to pregnant women can only be implemented in women who are aware of their HIV-positive status. Raising the uptake of antenatal HIV testing amongst women in the United Kingdom has therefore been a key issue amongst policymakers and healthcare professionals in recent years.

In 1999, a research group from London estimated the cost effectiveness of a universal, voluntary HIV screening programme for all pregnant women in the United Kingdom (Simpson 1999). Effectiveness was measured in terms of life years gained among children whose HIV infection would be avoided, and in life years that would be gained by mothers through the use of anti-HIV treatments. Costs analysed included those for HIV antibody testing, pre- and post-test counselling, anti-HIV treatment, elective Caesarean section, and formula feed.

The lifetime costs of care for a child infected with HIV were estimated at £178,300. Hence in high prevalence areas such as London, a programme of universal voluntary HIV screening was estimated to be cost-effective, with a net cost of £4000 for each life year gained. In lower prevalence areas, this figure might be less than £20,000, depending on the number of women unaware of their status, and the local costs of screening.

The authors based their analysis on a series of assumptions, some of which may no longer be appropriate. For example, a transmission rate of 6% was assumed for women treated with a standard antenatal course of AZT (zidovudine, Retrovir) who also delivered by elective Caesarean section. Recent research suggests this rate may be closer to 1%. Similarly, modern anti-HIV regimens tend to involve combinations of drugs which, though more costly than the use of AZT alone, are more effective in preventing transmission.

In a second study, also based on United Kingdom health service costs and surveillance data, costs and benefits to the mother of an earlier HIV diagnosis through antenatal screening were estimated. Costs per life year gained by antenatal diagnosis of women were £51,258 assuming untested women were diagnosed a median of 20.4 months later than tested women. The model assumed that therapy involved three antiretroviral drugs, was begun at a CD4 cell count of 350 cells/mm³, and that treatment slowed the rate of decline in a womans CD4 count by 40% for life (Gibb 1999).

A research team from Scotland reported that the implementation of a routine, voluntary antenatal HIV screening programme more than doubled the uptake of the HIV test by pregnant women, compared with an earlier opt-in scheme (Postma 1999). The authors concluded that a similar programme may be appropriate in other high prevalence areas of the United Kingdom, but excluded London, where there are more complex issues of language and cultural heterogeneity.

United Kingdom policy shift in 1999

In August 1999 the Department of Health announced that all pregnant women in England would be offered, and advised to take up, an HIV test during pregnancy, in low- as well as high prevalence areas.

The intention of this policy was to increase the uptake of antenatal HIV testing to 90% and to increase the proportion of HIV infections diagnosed prior to delivery to 80%. The hope was that, by December 2002, this would result in an 80% reduction in the proportion of infants acquiring HIV infection at birth. Recently published reports from the Health Protection Agency state that these targets were met by 2004 (Goldberg 2005).

Pregnancy following HIV diagnosis

A large study involving European women has reported information on changes in reproductive planning before and after a diagnosis of HIV infection. In 449 women, the pregnancy rate fell from 8.6 per 100 person-years before HIV diagnosis, to 8.2 in the first four years after diagnosis and 6.0 thereafter. The rate of induced abortions increased significantly after diagnosis from 42 to 53%, though the risk of spontaneous abortion did not change. Single women were less likely to become pregnant following diagnosis than women with a steady partner, and women from northern and central Europe were more likely to compared with southern European women (van Bentham 2000).

It seems that the introduction of triple combination antiretroviral therapy has had little impact on the proportion of HIV-infected women having babies. According to American data on over 7000 women, pregnancy rates among HIV-infected women did not change significantly between 1991 to 1994 and 1995 to 1998 (Blair 2001). In addition, women taking antiretroviral therapy were no more likely to become pregnant than women who were not taking HIV treatment.

Testing during labour

Several studies have looked at rapid HIV testing of identified high-risk women during labour. The rationale is to allow administration of treatment during labour to reduce the risk of HIV transmission.

The focus of this research seems to have been the accuracy of the tests, with effectiveness ranging from 50 to 100% (Nogueria 2001; Santos 2001; Smith 2001). The efficacy of treatment during labour following detection of HIV infection using rapid testing remains unproven. Despite potential benefits in terms of reduced transmission, rapid testing during labour presents ethical dilemmas about the nature of pre-test counselling and the ability of women to give informed consent to testing at that time.

References

Blair JM et al. Have pregnancy rates in HIV-infected women changed in the era of effective ART for perinatal transmission? Eighth Conference on Retroviruses and Opportunistic Infections, Chicago, abstract 700, 2001.

Duong T et al. Vertical transmission rates for HIV in the British Isles: estimates based on surveillance data. Br Med J 319: 1227-1229, 1999.

Gibb DM et al. Costs and benefits to the mother of antenatal HIV testing: estimates from simulation modelling. AIDS 13: 1569-1576, 1999.

Goldberg D ET AL. Unlinked anonymous testing indicates antenatal HIV testing in England and Scotland is being successfully implemented. Eurosurveillance Weekly 10, 2005.

Nogueria SA et al. Assessing rapid HIV testing in labor and delivery: a pilot study from Rio de Janeiro. Eighth Conference on Retroviruses and Opportunistic Infections, Chicago, abstract 696, 2001.

Postma MJ et al. Universal HIV screening of pregnant women in England: cost effectiveness analysis. Br Med J 318: 1656-1660, 1999.

Santos VV et al. A prospective study of the feasibility of rapid HIV testing in pregnant women during the peripartum period in Rio de Janeiro, Brazil. Eighth Conference on Retroviruses and Opportunistic Infections, Chicago, abstract 695, 2001.

Simpson WM. Antenatal HIV testing: assessment of a routine voluntary approach. Br Med J 318: 1660-1661, 1999.

Smith L et al. Expedited HIV testing of pregnant women at delivery in New York State. Eighth Conference on Retroviruses and Opportunistic Infections, Chicago, abstract 694, 2001.

Stratton P et al. Obstetric and newborn outcomes in a cohort of HIV-infected pregnant women: a report of the women and infants transmission study. J Acquir Immune Defic Syndr Hum Retrovirol 20: 179-186, 1999.

Van Bentham BHB et al. Pregnancies before and after HIV diagnosis in a European cohort of HIV-infected women. AIDS 14: 2171-2178, 2000.