|| Checking for direct PDF access through Ovid
To the Editors:“Scientific understanding proceeds by the way of constructing and analyzing “models” of the segments or aspects of reality under study. The purpose of these models is not to give a mirror image of reality, not to include all its elements in their exact sizes and proportions, but rather to single out and make available for intensive investigation those elements, which are decisive. We abstract from nonessentials, we blot out the unimportant to get an unobstructed view of the important, we magnify to improve the range and accuracy of our observation. A model is, and must be, unrealistic in the sense in which the word is most commonly used. Nevertheless, and in a sense paradoxically, if it is a good model, it provides the key to understanding reality.”1We appreciate the interest and debate by Dimitrov et al2 in responding to our article.3 Dimitrov et al call for enhanced attention to population demographics, particularly population recruitment and departures, and other parameters when modeling the effectiveness of HIV prevention interventions. If our study objective were to provide the exact estimates of HIV transmission, adding a wide range of demographic, programmatic, socioeconomic, and biologic parameters would have been helpful. However, our objective was to evaluate potential policy options to minimize new HIV infections and AIDS deaths in Vietnam. For this query, we were guided by the philosophy of Baran and Sweezy, “to single out and make available for intensive investigation those elements which are decisive,” in constructing the simplest model possible to answer our study question.1Early initiation of antiretroviral therapy (ART) is associated with a substantial reduction in HIV-related mortality, morbidity, and transmission4,5 and is a key component of national HIV responses. We considered the current epidemic dynamics in Vietnam and modeled various policy scenarios of expanding HIV testing and treatment in combination with other prevention interventions. The key messages from our model are: (1) earlier and expanded access to HIV testing and ART will likely lead to a substantial reduction in AIDS deaths and new HIV infections in Vietnam; (2) prioritizing key populations with a high HIV incidence such as people who inject drugs, sex workers, and men who have sex with men, will improve cost-effectiveness; (3) combining ART with other prevention interventions will accelerate the impact; and (4) investing in earlier and expanded access to HIV testing and ART will lead to relative cost savings in the future.Demographic parameters play a relatively important role in certain HIV epidemic phases and geographical areas. For example, migration of Vietnamese sex workers from Cambodia back into Southern Vietnam probably contributed to increased HIV transmission in the early 2000s.6 However, Vietnam's National Estimation and Projection Working Group agrees that migration among populations in Vietnam is low and its impact on the epidemic is negligible. We agree with Dimitrov et al that changes in demographics can influence the modeling outputs in the long term; however, for the concentrated epidemic in Vietnam, the size of key populations, rather than the size of the general population, will likely be more important when modeling the epidemic and intervention impact. Given the difficulties of accurately projecting the size of key populations over decades, a more meaningful approach would be to update the projections as new population and behavioral data become available.From a national programming perspective, it is important to understand how behavioral and programmatic parameters influence the impact of public health interventions. Our model assumed (1) high uptake of HIV testing and treatment and (2) high retention and adherence to ART as part of a well-functioning programme.