Pregnant women from Northern Uganda who undergo voluntarily counselling and testing (VCT) in order to access services to prevent mother-to-child transmission (PMTCT) are testing positive for HIV roughly at the same rate (~11%) as women who are tested anonymously at the same site for the national surveillance program, according to a study in the June 10th edition of AIDS. The findings suggest that data from VCT for PMTCT programmes could be used for surveillance purposes.
However, less than half of the women who could have accessed VCT actually got tested, which means PMTCT was not accessed by at least half of the women who could have used it to protect their infants from becoming HIV-infected.
HIV surveillance and VCT
In sub-Saharan Africa, the epidemic is commonly being monitored with the help of anonymous surveillance testing amongst pregnant women attending antenatal clinics. Numerous studies suggest that prevalence in the antenatal population approximates that of the general (sexually active) population in settings where HIV is primarily spread via unprotected heterosexual intercourse.
Antenatal surveillance testing is necessary in order to monitor trends in the epidemic and evaluate prevention efforts,as well as for health service planning. However, it does consume resources, and in settings, the same population is already being tested voluntarily for PMTCT programmes.
If VCT uptake is fairly consistent across a population (i.e., there are no factors that might bias findings) and VCT testing results are generally consistent with surveillance testing, then ongoing anonymous testing may not be necessary.
The study
St. Mary's Hospital Lacor in Uganda conducts antenatal testing for the Gulu Health District (in the north of the country, bordering Sudan), and since the year 2000 has also offered the same population VCT and PMTCT services. The hospital gathered information on socio-demographic characteristics and reproductive history of pregnant women attending its antenatal clinic between 2001 and 2003, and compared the data from the women screened for their PMTCT programme with those tested anonymously.
A total of 14,040 women attended the antenatal clinic for at least once. HIV testing was conducted on a random sample (stratified by age) of 3580 for the surveillance programme, while 6785 agreed to be tested for HIV-1 infection as the first step in enrolling in the PMTCT programme.
Uptake of VCT was slightly more common among women residing in urban areas; those who had been residing at their current address for two years or less; those with less than seven years of education; those who were cohabitating but not married; and those whose partner had a 'modern' occupation. But there were only weak associations between VCT uptake and testing positive (for time at their current address, marital status, and partner's occupation). In other words, these factors would not strongly bias the findings if VCT data was used to estimate HIV surveillance. What may also be equally important is that HIV did not appear to be significantly more common among any of the subgroups that avoided testing. HIV prevalence findings were fairly consistent for each age group over the three years the study was conducted. Overall, 11.1% of those tested anonymously and 10.9% of those who tested for PMTCT were found to be HIV-positive.
Discussion
While the study might be considered a success - since it found that at this site at least, VCT data can be substituted for anonymous antenatal testing for surveillance purposes - it also shows how VCT is not reaching everyone it should. This is not because the service was not being offered - VCT was offered to every women attending the antenatal clinic more than once (12199 out of 14,040). In fact, a total of 5414 women refused to be tested.
Out of the total study population, 7255 did not access VCT. At a seroprevalence rate ~11%, 800 of these women were likely to be HIV-positive. Without PMTCT, about one third of these women were likely to have passed HIV onto their infant during childbirth, and as a result more than 200 children whose mothers received antenatal care through this hospital over three years were born with HIV. Most of those infections could have been preventable.
One would hope that the opportunity to prevent the virus from being passed on to their infants would be a good incentive for women to get tested, but it clearly is not enough. The reasons for the women not to get tested are complex, including fear of stigma and fear of their husbands reactions, should they test positive. Women in many parts of Africa are generally blamed for being infected whether their husband was infected first or not. Many may face violence or eviction should their status become known.
Nevertheless, Uganda's prevention efforts are frequently cited as a model for the rest of Africa. Whether the relatively low prevalence rate (compared to southern Africa) is due to Uganda's ABC (abstinence, be faithful, condoms) prevention programme; due to differences in the virus subtype infectivity or being at a different spot on the bell curve of infection, AIDS is not new here. The fact that so many are afraid to be tested and to access health services for more than twenty years into the epidemic is a testament that something is still not working.
Reference:
Fabiani M et al. Using prevalence data from the programme for the
prevention of mother-to-child-transmission for HIV-1 surveillance in
North Uganda. AIDS 19:823-827, 2005.