Naeem Hassan Saleem  ( Canada-Pakistan, HIV/AIDS Surveillance Project (HASP), Pakistan. )

In response to article "Institutional and Surveillance Database Use in Epidemiologic Research in Developing Countries: Revisiting Some Limitations" published in Vol. 58, No.3, March 2008, we agree that surveillance data must be cautiously used, however, we differ in opinion that surveillance system in Pakistan is not well developed especially for HIV.  As we know that surveillance data can contribute to planning, implementation and evaluation of public health response, Pakistan has developed two arms of surveillance for HIV.  First arm quite commonly known as sentinel surveillance constitutes surveillance centers, blood banks, STI clinics and Voluntary Counseling and Testing Centers (VCTs).  Surveillance centers spread throughout the country more often serve as diagnostic centers and have diminished clarity about primary clients of the centre and have poor overall reporting 'culture'.  Blood banks lack enforcement of existing blood safety laws leading to much unscreened blood for HIV especially in private sector.  Similarly, government clinics capture only a small and most likely non-representative sample of men and women with STIs as majority opt for private clinics or take self-medication for STIs.  VCT centers are managed by NGOs and usually provide counseling and private or public sector facilities and HIV testing but contributing very little to surveillance data. Majority of these data sources do not indicate persons at risk of acquiring the HIV, instead indicate the person who has already acquired the infection.
As Pakistan has concentrated HIV epidemic, the infection remains concentrated in vulnerable populations while the first arm of surveillance system is focused on the general population.  Trends in infection in vulnerable population are not captured by first arm of surveillance which is mainly designed to track infection in the general population. Needless to say that it is very cost-effective method of estimating prevalence and provides both spatial and temporal trends of HIV infection in the general population. Keeping in view the above facts, the Second arm of Surveillance has recently been fully developed in the light of international guidelines. Second Generation Surveillance (SGS) recruits representative sample of vulnerable population comprising of commercial sex workers and injecting drug users (IDUs).  It takes into account of existing risky behaviours and HIV sero-prevalence.  Due to the varied nature of both arms of surveillance, if intelligently used, it can be used to monitor HIV/AIDS epidemic and guide policy decision.

1.  Schwartzbaum J, Ahalbom A, Feychting M. Berkson's bias reviewed. Eur J Epidemiol 2003; 18:1109-12.
2.  Emmanuel F, Bokhari A.  Developing Surveillance System For HIV/AIDS In Pakistan, JCPSP 2006; 16: 676-9.
3.  Shukla RS, Bhuyan KK. Can data on HIV sero-reactivity among blood donors provide an insight into HIV prevalence in the general population? Indian J Public Health 2007; 51: 14-21.
4.  Second Generation Surveillance for HIV: the next decade. Geneva: World Health Organization and Joint United Nations Programme on HIV/AIDS 2000.
5.  Sun X et al. The development of HIV/AIDS surveillance in China.  AIDS 2007; Suppl 8: S 33-8.