During the early 1980s scientific researchers anxious to track the course of the new disease had recruited large groups of those at risk, principally gay men, in order to collect data on the prevalence of the disease and possible routes of transmission. As the decade went on, these cohorts and others subsequently recruited were to provide hugely valuable information which continues to provide our most reliable guide to the effects of HIV.

These cohorts were able to provide information on three major concerns: the routes of transmission; how long HIV might take to cause AIDS; and whether other factors (or 'co-factors' might be involved in the disease).

Although routes by which HIV could not be transmitted were fairly clear by 1985, it was still unclear at this time how likely it was that HIV could be transmitted by routes such as oral sex, from a woman to a man during vaginal intercourse or through needlestick injuries. As new cohorts and case reports updated our knowledge on these issues, some media commentators took changes in advice as signalling that scientific opinion was, and is, deeply uncertain over the routes of HIV transmission. 'Do scientists know everything there is to know about the transmission of HIV?' was the angle of such stories, and they sought to grab attention by proposing new and improbable routes of transmission, often based on questionable laboratory evidence rather than epidemiological observation in the real world. Transmission by mosquitoes and through kissing have been two of the most persistent examples.

Another cause of confusion is the length of the incubation period between infection and the appearance of symptoms (the asymptomatic period). Confusing evidence on this subject has lent fire to the arguments of those who question the role of HIV in AIDS, and has caused much anxiety amongst HIV-positive people.

In the first years of the epidemic doctors assumed that the incubation period was no more than a year, on the erroneous assumption that individuals who reported sexual contact with those diagnosed with AIDS tended to develop AIDS within a year of sexual contact. It is more likely that these individuals had been infected for considerably longer, and that their contact with earlier AIDS cases was merely the consequence of the high prevalence of HIV within the gay population.

The predicted incubation period gradually lengthened as cohort observations were able to show that people infected in the early 1980s were taking longer than previously assumed to develop symptoms. Current cohort data suggests that this period is, on average, around 10 years (with considerable variation). However, while some will develop AIDS quickly, perhaps within one to two years of infection, others may still be healthy twenty-five years after infection.

Cohorts have also been able to answer the question of whether other factors apart from HIV might be responsible for immune deficiency in those at risk of AIDS. Cohorts have shown conclusively that HIV is the primary factor associated with immune deficiency in all groups at high risk of AIDS, that AIDS-defining illnesses occur very rarely in the absence of HIV infection, and that factors such as drug use do not cause immune deficiency. However, cohorts have also shown that a number of co-factors appear to increase the likelihood that AIDS will develop in HIV-positive people. The clearest influences on disease progression are age, income and access to care; older people and the very young develop AIDS much more quickly than those in their 20s and 30s. Those with incomes at or below the poverty line have been shown to progress to AIDS far more quickly than those with high incomes.