Sir - Elliot Marseille and colleagues (May 15, p 1851),Marseille E, Hoffman PB, Kahn JG. HIV prevention before HAART in sub - Saharan Africa. Lancet 2002; 359: 1851 - 56. [Text]  used cost-effectiveness analysis to provide a simplistic and outdated proposition that prevention of HIV/AIDS should take priority over and be funded to the exclusion of treatment in Africa. Prevention is almost always cheaper than care, irrespective of a country's development status, especially if therapeutic options are dominated by patented drugs.

Prevention is also effective, thus UNAIDS and its cosponsors advocate prevention as the foundation for all HIV/AIDS programmes. Prevention and treatment have overlapping but not identical goals. A comparison based on only one metric can be instructive but seldom conclusive. No nation's health policy strictly enforces trade offs between prevention and care. By taking so narrow a view and in view of the fact that Africa has around 28 million infected people, Marseille and colleagues fundamentally mis-state the problem.

Historically, at more than US$10,000 per patient-year, treatment was clearly unaffordable. As prices plummet and resources increase, implementation capacity will rapidly replace finances as the limiting constraint. Prevention efforts undoubtedly need to be scaled up substantially. Expansion of capacity for both prevention and care needs time and incremental increases in finances. Such efforts are not simply additive, since every strategy improves the effect of the other.

Prevention and care involve different sectors and constituencies, investment in both simultaneously can achieve more than would be accomplished by separate investment. Highly active antiretroviral treatment (HAART) grants extra years of quality life, which goes beyond price and generates vast collateral benefits, including saving the health system several hundred dollars per patient-year in averted palliative and opportunistic-infection care.

At a prevalence of 5%, demand for medical care is estimated to increase faster than the public sector's capacity to cope. In many African countries, health sectors are already overwhelmed by AIDS in patients and staff. HAART could save lives, money, and the health systems themselves. In Namibia, the output per person is estimated to rise to higher than the per - person taxes needed to fund treatment programmes. Sanderson WC, Fuller B, Hellmuth ME, et al. Namibia's future: Modeling population and sustainable development challenges in the era of HIV/AIDS. Laxenburg, Austria: International Institute for Applied Systems Analysis, 2001. Treatment also has substantial positive effects on national development. Bonnel P. Economic analysis of HIV/AIDS. ADF background paper. http://www.unaids.org/publications/documents/economics/index.html##cost (accessed June 14, 2002).  AIDS destroys adults as workers, parents, and care givers in the prime of their lives.

Treatment saves children from orphanhood; keeps households, social cohesion, and businesses intact; improves returns on social investments, such as education and development; increases growth and security; and keeps to a minimum exacerbation of poverty.

As Marseille and colleagues rightly note, HIV has moved beyond public health and has become a social, economic, and security concern, without appreciating that the time scale of the larger concerns is distinct from that in which prevention operates. Prevention can help to avert such threats in the indeterminate future. However, people, societies, economies, and nations are at risk now because of the potential of millions of premature deaths of those already infected. Only treatment can change that trajectory.

Countries with the greatest infection rates are at disproportionate risk, making treatment there even more urgent. The economic justification for HAART is its leverage effect on HIV prevention and its potential to secure the future against disabling social and economic ills. Prioritisation is not an issue of lives today over lives tomorrow; the quality of the future depends crucially on the quality of life today. Marseille and colleagues and Andrew Creese and colleagues (May 25, p 1635)Creese A, Floyd K, Alban A, et al. Cost - effectiveness of HIV/AIDS interventions in Africa: a systematic review of the evidence. Lancet 2002; 359: 1635 - 42. [Text]offer static perspectives.

Prices and resources are not fixed, especially now. The commitments made by African nations and all UN Member States mark unprecedented political momentum.United Nations General Assembly Special Session on HIV/AIDS. June 2001. New York, USA.The epidemic has thrived on inabilities to anticipate dynamic effects.

Epidemiologically, how quickly and far HIV could spread was not foreseen; economically, policy decisions have frequently been based on prices that are probably obsolete before the policies can be implemented. The cost ratio of care to prevention may currently be 28 to 1. Two years ago it was more than 200 to 1. In two years time it will probably be far lower than it is today. The price today is less relevant than the expected total cost, which will decline with time. Finally, there is no arbitrary threshold at which treatment becomes of value.

The decision to treat, cannot be based only on narrow variables of cost-effective analyses but must involve humanitarian considerations; and evolve as factors, not least, prices and capacity change. Moreover, expanding treatment now may accelerate cost reductions by stimulation of greater supply and competition, and identification of cheaper administration mechanisms in resource-limited settings.