Bangladesh is a low prevalence nation for HIV and therefore, according to the guidelines of the Second Generation Surveillance system for HIV, surveillance should concentrate amongst selected groups of individuals who are known to be most at risk to HIV and some of the population groups that may bridge the epidemic into the general population. Therefore during all rounds of surveillance conducted so far in Bangladesh, including the 5th round, the population groups selected were confined to those considered to be most vulnerable and some bridging populations.

Most at risk populations
Injecting drug users: Injecting drug users (IDU) had the highest rate of HIV infection with 4 per cent prevalence in Central Bangladesh and in one neighbourhood of Central-A, 8.9 per cent of IDU were HIV positive. While HIV prevalence in IDU increased significantly in one specific area (Central Bangladesh), there were no changes between the 4th and 5th rounds in the rest of the country. Bangladesh, therefore, still remains a low prevalent nation for HIV. Furthermore active syphilis rates declined significantly in IDU over the rounds in Central Bangladesh. However, Hepatitis C (HCV) prevalence in IDU remained high.

HCV prevalence was high in IDU from most sites. The highest rates were recorded in IDU from Northwest-B2. The HCV rates were surprisingly low in two sites Northwest-F and Central-H. In Central-A where the HIV prevalence was 4 per cent, HCV prevalence was 59.2 per cent. Overall, out of 1619 IDU sampled, 54.2 per cent tested positive for HCV .

The 5th round BSS showed that needle/syringe sharing continued to be routine among IDU especially among those in Central-A. However, sharing was comparatively lower in the Northwest region than in Central-A. Most IDU used other modes of taking drugs before they started injecting. A large proportion of IDU had commercial and non-commercial female sex partners and condom use was infrequent. A proportion of IDU (4.3-6.7 per cent) sold blood in the last year. IDU were highly mobile. IDU from other cities traveled to Central-A where they injected drugs. Injecting drugs while abroad was more commonly reported by IDU from Northwest-B and B1 (10-12 per cent).

A considerable proportion of heroin smokers injected in the last six months and most shared needle/syringe during their last injection. More than half of the heroin smokers had commercial and non-commercial female sex partners in the last year and those who did had multiple sex partners. Condom use, both in the last sex act and consistently in the last month, was very low with both commercial and non-commercial partners.

Sex workers: HIV prevalence remained low in female sex workers. Casual female sex workers in one of the northwest border areas had the highest rate of HIV prevalence, which is 2 per cent but the total numbers of sex workers sampled were less than 400 (n=101). Active syphilis rates remained high among the female sex workers. However, active syphilis rates declined significantly over the rounds in most of the brothels and in the street based sex workers from Central-A. Female sex workers in the border areas were considerably mobile and sold sex across the border to India.

Hotel based sex workers were comparatively younger and had highest number of clients among all female sex worker groups. Consistent condom use in female sex workers remained low in all groups.

Males who have sex with males: HIV prevalence was low in males who have sex with males (MSM), male sex workers (MSW), Hijra and partners of Hijra. In Central-A, changes in the active syphilis rates in MSM and MSW over the rounds were not significant.

Almost all of the MSW and Hijras reported that they had new clients in the last week. Some MSW sold sex to females in the last month. Hijras reported more clients in the last week than MSW while condom use was low in this group.

All groups of sex workers reported violence in the last year. Both being raped and beaten was most commonly reported by Hijras and female street based sex workers from Central-A. Violence was reported to have been committed by both police and mastans.

Summary
The relatively low level of HIV in Bangladesh today does not guarantee low prevalence tomorrow. Experience teaches us that early epidemics do not show their magnitude at first and place few demands on the health sector. All the risk factors which give birth to explosive HIV epidemics are present in Bangladesh today. In the absence of good quality and high coverage intervention programmes, HIV prevalence may jump to very high levels within months. Once HIV prevalence crosses the 10 per cent level, epidemics become very difficult to control. Policy makers and programmers within the Government of Bangladesh, bi-lateral agencies, national and international NGOs have a key role to play in recognising the urgency of the situation and taking immediate action.

The data from the 5th round of the serological surveillance confirm the fears from the pervious 4th round that there is an impending epidemic among the injecting drug users in Central Bangladesh and one neighborhood in that city is already experiencing an epidemic.

Source: ICDDR,B