Using the premarital examination for population-based surveillance for HIV in China

All couples attending the premarital examination at four selected sites in Zhejiang and Yunnan provinces in China were included in this study. Blood from 9952 individuals was tested for HIV-1 using a gelatin particle agglutination technique. There were no HIV-positive individuals in Zhejiang. In Yunnan, 28 of 3742 individuals were positive, a prevalence of 0.75%. The premarital examination should be used for voluntary counselling and testing and anonymous surveillance for HIV in high prevalence areas. Providing greater opportunities for testing is one of the essential prerequisites for the reversal of the AIDS pandemic [1]. Testing is especially important in China: over 95% of the estimated one million HIV-positive individuals there are unaware of their status [2,3]. With sentinel surveillance targeting high-risk groups [2], little is known about the prevalence in the general population. This is of concern, because it is widely believed that conditions are now present for spread in the general population; for example, there has been a 16-fold increase in the prevalence of sexually transmitted infections nationwide (to 860 000) over the past 15 years [4,5]. It is now predicted that there will be 10–15 million HIV-positive individuals by the year 2010 [6,7], increasing the urgency to widen the surveillance system to include low-risk populations. China already has a vehicle for population-based surveillance: the premarital examination (PME). Over 97% of the population get married, and the health examination is a legal requirement for marriage in most of China. It is used to screen for a range of health and reproductive disorders, generally with a view to treatment before marriage takes place [8]. With the advent of dried blood spot (DBS) testing, based on particle agglutination, there is now a cheap, easy and reliable test for antibodies to HIV, which is ideal for population studies [9–11]. This study was carried out to assess the utility of DBS testing for surveillance purposes at the PME, and to describe the epidemiology of HIV at the PME in an urban and a rural area in Yunnan and Zhejiang Provinces, representing high and low-risk areas for HIV infection, respectively. Yunnan is located in the south-west of China, north of the Golden Triangle and has a long border with Myanmar. Sentinel surveillance shows that Yunnan accounted for 40% of all HIV cases nationwide in 2001 [2]. Zhejiang is a wealthy province, located on the east coast, just south of Shanghai. It is a low HIV prevalence province, but has the highest rate of sexually transmitted infections in the country [12]. The four sites were selected to be ‘typical’ of urban and rural areas in the two provinces: the urban areas were districts in the provincial capital and the rural, middle income counties. Before the commencement of data collection, two laboratory technicians (one from each province) were trained in DBS techniques (a gelatin particle-agglutination technique using Serodia reagents) [9] in the Virology Department of Great Ormond Street Hospital, London, UK. Laboratory facilities were then established in the capital cities of the two provinces, in university-affiliated laboratories. All couples attending the PME at the four sites during the study period of January 2001 to March 2002 were recruited into the study. Routinely taken blood was collected onto filter paper, dried, stored in moisture-free bags, and transported to the respective laboratory. All positives were confirmed by Western blot, yielding a 100% specificity rate. Quality control was ensured through participation in the Dried Blood Spot HIV Quality Control programme conducted by the Centers for Disease Control and Prevention, Atlanta, GA, USA. Tests were carried out anonymously, as required by local regulations at the time of the study. Results were linked to demographic and risk factor variables. The results are shown in Table 1. A total of 9952 individuals (4976 couples) were tested. In the two sites in Zhejiang Province there were no HIV-positive individuals among the 6210 tests carried out. In Yunnan overall 28 out of 3742 [0.75%, 95% confidence interval (CI) 0.48–1.1%] tested positive, 16 out of 2330 or 0.68% (CI 0.35, 1.05%) at the urban site) and 12 out of 1412 or 0.84% (CI 0.39, 1.3%) at the rural site. Of the 28 who tested positive, the age range was 22–32 years (mean 25.5), 22 were men (male : female ratio 3.7 : 1), 14 were known current or former drug users, and two were sex workers (both female). There was only one concordant couple. The overall cost per test was estimated to be US$0.6, excluding labour costs.Table 1: Sample by province and area.This is a pilot study, intended mainly to assess the utility of the screening technique, and the epidemiological findings are limited by the small sample size and limited number of sites to date. Therefore, its wider extrapolation to provincial populations of over 40 million needs to be cautious. However, the contrast between the two provinces is indicative of emerging epidemiological patterns in different areas of China. With the zero-positivity rate in Zhejiang the pilot is continuing, in order to determine the utility of introducing routine testing at the PME there. In contrast, even allowing for some underreporting of risk behaviours, our data suggest that HIV may already be affecting those outside high-risk groups in Yunnan. Of 28 positive individuals, 14 were known to have used intravenous drugs and two were known sex workers, so 12 (or 43%) were classified as ‘low risk'. This study has clear implications for the policy on testing. First, it has demonstrated that DBS testing at the PME is a feasible means of carrying out population-based surveillance at low cost. Combined with the sentinel surveillance of high-risk groups, the epidemic can then be monitored in a systematic and comprehensive way. The access it provides to men is particularly desirable, given the reported male : female ratio of approximately 4 : 1 in China [2]. Second, the large number of cases of undiagnosed HIV in Yunnan, and the high proportion of non-concordant couples, shows that HIV testing at the PME must be introduced in the context of a twin-track system of anonymous surveillance and voluntary counselling and testing in order to reduce transmission. Such testing is now even more relevant given that cheap, locally produced antiretroviral drugs have just become available, and that the government has recently committed to financing some treatment. Voluntary counselling and testing programmes at the PME would be particularly desirable in the higher prevalence provinces, and such programmes are under discussion. They must be carried out against a background of awareness-raising, destigmatization of the disease, and the expectation of benefit from testing [1]. Guidelines are currently being developed for the management of test-positive individuals, the management of non-concordant couples, and the safeguarding of confidentiality. Contributors T Hesketh and A Tomkins designed the study. T. Hesketh was responsible for the overall execution and wrote the paper. X.M. Huang, Z.B. Wang and W.X. Zhu organized the data collection. D. Cubitt was responsible for the laboratory aspects of the study.