Growing in silence

selected regions and countries with expanding HIV/AIDS epidemics

Introduction

The AIDS epidemic has killed over 20 million people worldwide, including more than 3 million in 2002 alone, and an estimated 5 million people acquired HIV in 2002, bringing to 42 million the number of people globally living with the virus [¹]. The HIV/AIDS epidemic has displayed great regional variation since its inception, and it continues to do so. UNAIDS/WHO have estimated that, by the end of 2002, the HIV prevalence rate among adults was 0.1% in east Asia, the Pacific and in Australia and New Zealand, 0.3% in north Africa, the middle east and western Europe, 0.6% in south and south-east Asia, Latin America, eastern Europe, central Asia, and north America; 2.4% in the Caribbean; and 8.8% in sub-Saharan Africa [¹]. Although these regional estimates give a broad indication of how the HIV/AIDS epidemic affects different regions to a varying degree, they also hide sub-regional variations as well as differing trends over time in countries and regions.

In this overview we want to highlight expanding epidemics that are under-recognized or under- publicized. The selected regions and countries do not therefore necessarily represent the regions and countries with the fastest growing epidemics per se. Instead we have chosen to highlight countries and regions that have not received much attention in the media and the scientific literature, despite having growing epidemics. In Asia and the Pacific, 7.2 million people were estimated to be living with HIV by end of 2002. Whereas the growth of the epidemic in this region is largely attributable to the growing epidemic in China, several other countries in the region have growing epidemics. In eastern Europe and central Asia, the number of people living with HIV at the end of 2002 stood at 1.2 million. HIV/AIDS is believed to be expanding rapidly in the Baltic states, the Russian Federation and several central Asian republics. Sub-Saharan Africa is the region where the HIV epidemic started first, and it has been the worst affected by the epidemic. By the end of 2002, an estimated 29.4 million people were living with HIV/AIDS, and 3.5 million had been newly infected during 2002. Although the epidemic has stabilized in sub-Saharan Africa as a whole, several countries in southern and central Africa continue to witness increasing prevalence levels [²]. In this paper we discuss the levels and trends of HIV prevalence and key determinants in selected countries in the above three regions. We have chosen to highlight Vietnam (Asia and the Pacific region), the central Asian republics (eastern Europe and central Asia region), and Lesotho, Namibia and Swaziland (sub-Saharan Africa), often overshadowed in terms of media attention for HIV/AIDS by countries such as China and India, Ukraine and the Russian Federation, and Botswana and South Africa in the respective regions.

Vietnam

HIV testing in Vietnam started in 1988, and the first infection was reported in 1990 in Ho Chi Minh City. For many years the epidemic was primarily associated with injecting drug use in the north and injecting drug use and commercial sex in the south [³]. By the mid-1990s, prevalence had reached 70–80% among injecting drug users (IDU) in several provinces throughout the country, and was 12% among sex workers near the border with Cambodia. By 2002, the majority of sentinel provinces were reporting that more than 20% of IDU were HIV positive, and the prevalence in sex workers in at least seven provinces was above 5% (Fig. 1). There were also signs that the epidemic was spreading beyond IDU and sex workers to their clients and other sexual partners, as evidenced by data showing an increasing HIV prevalence among antenatal clinic (ANC) attendees and military recruits.

Fig. 1.: HIV prevalence levels among injecting drug users, female sex workers, military recruits and pregnant women attending antenatal clinics, by province, Vietnam, 2002. ANC, Antenatal clinics; FSW, female sex workers; IDU, injecting drug users.

Recent sentinel surveillance data indicated that prevalence is continuing to increase among female sex workers in both the north and the south, and although some of these women became infected through injecting drug use [^4,5], there is also evidence of heterosexual spread. In 2002, HIV prevalence among sex workers in the south was 11% in Can Tho, 15% in An Giang, and 24% in Ho Chi Minh City. Although a high proportion of HIV-infected sex workers in Ho Chi Minh City have a history of injecting drug use, there is evidence that many of them became infected heterosexually. A 2000 survey among 890 sex workers at a rehabilitation centre in Ho Chi Minh City showed that 88.4% (61/69) of women who had returned from selling sex in Cambodia were HIV positive, as opposed to 49% (187/379) of those who had a history of injecting, and 19% (19/98) of those who did not have a history of injecting [⁵]. There is also a growing epidemic among sex workers in the north, where the epidemic has been dominated by IDU until recently. HIV prevalence among sex workers reached 15% in Hanoi and 8% in Hai Phong in 2002. Regardless of whether these women became infected through injecting or through selling sex, heterosexual spread through commercial sex transactions is now likely to occur at an increasing rate. Behavioural surveillance data in 2000 indicated that street-based sex workers reported a mean of 15 clients per week in Ho Chi Minh City, and in Hanoi and Hai Phong these means were 19 and 17 respectively [⁶].

Although the transmission from IDU to their regular and commercial sex partners will certainly continue to be a driving force in the epidemic, the real potential for spread in Vietnam will be largely dependent on the amount of heterosexual spread, which is dependent upon the proportion of men who purchase sex. To understand this potential it is important to know the size of the sex worker population, their HIV prevalence, and the size of the client population. These parameters are not well documented in Vietnam; however, in other Asian countries, the proportion of men buying sex ranges from 5 to 20%. This proportion was as high as 20% in 1990 [⁷] and 15% in 2000 [⁸] in neighbouring Thailand and Cambodia, respectively. Although there is no reliable information on the proportion of men who purchase sex in Vietnam, some studies indicate that it may be significant, especially in urban areas [⁹]. Given the increasingly open society in Vietnam, as well as increased migration from rural to urban areas, this proportion may well increase, especially among young people. Since 2000, the prevalence of HIV was 1% or higher among military recruits in 10 provinces, and above 0.5% among ANC attendees in 11 provinces. The aggregate prevalence among military recruits in provinces where it was measured in both 1998 and 2002 increased nearly fivefold from 0.2% in 1998 to 0.9% in 2002, and among ANC women it tripled, increasing from 0.09% in 1998 to 0.28% in 2002, suggesting that there is already significant heterosexual spread in Vietnam.

In summary, the real impact of the epidemic in Vietnam is only just beginning to be felt. Sexual transmission from high-risk populations to commercial partners and other sexual partners will be a determining factor in how much it spreads, implying that prevention efforts need to be focused on limiting the spread from sex workers to clients, in addition to limiting spread through injecting drug use.

Central Asian republics

The central Asian republics include five countries (Kazakhstan, Kyrgyzstan, Tajikistan, Turkmenistan and Uzbekistan) with a total population of more than 58.5 million. The principal source of information on the HIV/AIDS epidemic in these countries is the HIV case reporting system. In this system, diagnosed HIV infections are centralized into national HIV case reporting data files. HIV testing was mandatory for several subgroups of the population until the mid-1990s, when the testing policy changed in some countries. In Kazakhstan, for example, testing is currently only mandatory in blood donors. Little information is available from Turkmenistan, which reported two HIV cases up to 1994. Only very recently have some countries initiated sentinel surveillance activities in high-risk population groups to allow a better monitoring of the epidemic [¹⁰]. Information based on case reports needs to be interpreted with great caution, as it is influenced by the testing strategy and access to voluntary counselling and testing. In addition, in many countries the laboratory testing equipment is outdated, and access to test kits is limited (for example, in Tajikistan 40% of donor's blood was not tested in 2002).

The number of reported HIV infections in the region has grown exponentially from 88 in 1995 to 5458 in 2002 [¹¹]. Kazakhstan was the first country of the region to report a ninefold increase in HIV cases in 1997. In 2001–2002 the other countries except Turkmenistan went through a similar experience. The rates of reported HIV infections per million of population in 2002 are similar in Kyrgyzstan (31.7), Uzbekistan (38.3) and Kazakhstan (43.3), as shown in Fig. 2. The extent to which the HIV epidemic has spread geographically differs from country to country. In Kazakhstan all regions are affected by the epidemic, whereas in Kyrgyzstan, which witnessed a rapid increase in 2001, more than half (59%) of the cases are concentrated in the Osh region, which is one of the routes of drug trafficking from neighbouring countries.

Fig. 2.: Annual new HIV infection case report rates in the central Asian republics, 1994–2002. —♦— Kazakhstan; —□— Kyrgyzstan; —▵— Tajikistan; —×— Uzbekistan.

Over 80% of reported HIV infections were related to injecting drug use in 2001, reaching 97% in Kyrgyzstan. Sentinel surveillance data show HIV prevalence among IDU in April–June 2002 from 1 to 9% [^12,13]. All five countries serve as drug trafficking routes from Afghanistan to Russia and western Europe. Drug users are switching from alcohol to heroin, which is cheaper, and heroin users are starting to switch from smoking or snorting to injecting [¹⁴]. The retail price of a single dose of heroin in Kyrgyzstan is as low as US$0.50–1.00. The estimated number of IDU in Tajikistan is 135 000 and perhaps 200 000 in Kazakhstan [¹⁴].

The most affected part of the population is young people. In 2002, young people between 15 and 29 years made up over 60% of all newly reported cases in Kazakhstan, and approximately 74.5% of these young people living with HIV/AIDS were unemployed. The proportion of women among the reported HIV infections in 2002 does not exceed 23%. Although the number of babies born to HIV-infected mothers is less than 50, high birth rates and limited access to preventative mother-to-child transmission suggest that the significance of mother-to-child-transmission is set to increase in the near future.

Sexually transmitted infections, such as syphilis and gonorrhoea, have increased 100-fold since 1991, and although they have decreased in recent years from their peak level, they remain very high. For example the rate of syphilis in Kazakhstan was 140 per thousand in 2001. Approximately 40% of female IDU are engaged in sex work, thus providing a bridge between the high-risk and the general population [¹⁵]. Sentinel surveillance results show that HIV prevalence among sex workers in Kazakhstan in 2002 varied from 0.7 to 35.3% [¹³]. In 2002 in all the countries the proportion of heterosexual transmission has grown in comparison with 2001, e.g. in Kazakhstan from 5.5 to 22.2%, in Uzbekistan from 5.5 to 11.2% [¹¹]. It is as yet unclear what proportion of these cases are sexual partners of IDU and what proportion are unrelated to IDU.

In conclusion, the recent rapid increase in new HIV infections in central Asia is alarming. As the epidemic is very recent, with most new infections occurring among vulnerable groups including IDU and sex workers, prevention efforts should target these groups and aim to achieve high coverage. Such a strategy supported by prevention efforts for young people have the potential to avoid a much larger epidemic in this region.

Lesotho, Namibia, Swaziland

In Lesotho, the most recent sentinel surveillance among pregnant women was run in 2000 in six ANC. Prevalence levels among pregnant women were highest in the capital city Maseru (43%), followed by Leribe (26%), Quthing (22.8%), Mafeteng (19%), Maluti (19%), and Mokhotlong (12.3%) [¹⁶]. Fig. 3 shows the increase in prevalence over time in Maseru. HIV prevalence levels among sexually transmitted infection patients at the six sentinel sites used for the antenatal survey were much higher in 2000, ranging from 39 to 65% [¹⁶]. In the 2001 behavioural survey that included in-school and out-of-school youth, it appears that young Basotho are at high risk of HIV infection [¹⁷]. Comprehensive knowledge about HIV/AIDS was somewhat higher among in-school youth at 24 and 26% than among out-of-school youth at 10.3 and 18.8% for 15–19-year-old boys and girls, respectively. In-school boys start sex earlier than girls, with 52.5–60% of 15–19-year-old boys having had sex compared with 18.7–24.5% of girls. For out-of school youth, these proportions are similar for boys at 58.1%, but higher for girls at 40.6%. Reported condom use with a non-regular partner among in-school youth ranged from 57 to 69% for boys and 71 to 73% for girls, whereas it was lower among out-of-school boys at 37%, and girls at 46%, respectively. The 2000 Multiple Indicator Cluster Survey showed that 33% of women knew three major ways to prevent HIV infection, 58% knew where to get an HIV test, and 12% had actually had an HIV test [¹⁸].

Fig. 3.: HIV prevalence among pregnant women in capital cities of Lesotho, Namibia and Swaziland, 1991–2002. ▵ Hhohho region, Swaziland; ♦ Maseru, Lesotho; □ Windhoek, Namibia. Hhohho region includes the capital city Mbabane.

In Namibia, sentinel surveillance among pregnant women attending ANC has been conducted every 2 years since 1992. Fig. 3 shows the increasing trend over time in the capital city, Windhoek. Country-wide, HIV prevalence continues to increase: comparing 2000 with 2002, an increase was observed in 12 sites, and a decrease was observed in six sites [¹⁹]. In 2002, HIV prevalence ranged from 9 to 43% across the 21 sites. Prevalence among 15–24-year-old pregnant women is not yet declining, with an apparent recent non-significant decline among 13–19-year-olds (6, 11, 12, 12 and 11% in bi-annual surveys between 1994 and 2002), off-set by a sustained increase in the 20–24-year-olds (11, 18, 20, 20 and 22%, respectively). HIV prevalence among sexually transmitted infection clinic patients ranged from 10% in Opuwo to 65% in Oshakati. The prevalence of syphilis among pregnant women ranged from 4.7% (Kavango region) to 27.8% (Hardap region) in 2001 [²⁰]. The 2000 nationally representative Namibia Demographic and Health Survey showed that 24% of all women know three major ways to prevent HIV infection, 79% know where to get an HIV test, and 24% had actually had an HIV test [²¹].

In Swaziland, the 2002 round of surveillance was conducted at 17 ANC distributed over the four regions. Fig. 3 shows the prevalence over time among pregnant women in the Hhohho region (which includes the capital city Mbabane), increasing from 15.5% in 1994 to 36.6% in 2002 [²²]. Similar trends occurred in the other three regions, and the overall prevalence in Swaziland was 38.6% in 2002, up from 34.2% in 2000, and 3.9% back in 1992. Of note is the fact that HIV prevalence in urban sites (40.6%) was very similar to that in rural sites (35.9%). Syphilis prevalence among pregnant women (using the rapid plasma reagin test) was evenly distributed across the four regions in 2002, ranging from 3.2 to 4.7%. Over the years a declining trend in syphilis prevalence has been observed, from 11.4% in 1994, to 5.5% in 1996, 5.2% in 1998, 6.1% in 2000, and 4.2% in 2002. As treatment can reverse rapid plasma reagin positivity, these trends indicate the success of recently improved control efforts based on syndromic management [²²]. The Multiple Indicator Cluster Survey in 2000 showed that 33% of women know three major ways to prevent HIV infection, 60% of women know where to obtain a test, and 17% have actually been tested [²³].

Southern Africa, as a region, has the highest prevalence levels in the world. In this region, Lesotho, Namibia and Swaziland have reached extremely high levels of HIV prevalence, without there being signs of stabilization. Although some indicators of knowledge and risk suggest that prevention efforts have had an effect, they have yet to show an impact on prevalence levels. Compared with Botswana and South Africa, Lesotho and Swaziland have far fewer domestic resources to fight HIV/AIDS, and these countries may need to mobilize a large part of the needed resources from external sources. Other countries in this region, notably Angola and Mozambique, also have expanding epidemics, although still at much lower levels of prevalence [²].

In conclusion, the countries and regions highlighted in this review have all witnessed increasing levels of HIV prevalence in recent years, although they may not attract as much attention of the scientific and popular media as some other countries in the same region. Whereas the countries and regions featured in this review have reached very different levels of HIV prevalence, each of them requires strong leadership and the allocation of appropriate resources to respond to their specific HIV/AIDS crisis.

Acknowledgements

The authors would like to acknowledge the following people for their contribution to this review: A. Belonog and A. Kossukhin, Republic of Kazakhstan; M. Beknazarov, Republic of Tajikistan; B. Shapiro, Republic of Kyrgyzstan; and A Khudoberdiev, Republic of Uzbekistan.