Ever sat with a hanky in hand, wiping away tears as Edward climbs the fire escape with a rose in his teeth to profess his love to Vivienne, in the final scene of Pretty Woman? Or ever had a laugh at Belle and her hilarious antics in Secret Diary of a Call Girl?

Unfortunately, in reality, life for female sex workers is far from glamorous, romantic or funny. I originate from Bangalore, a thriving city in the state of Karnataka in southern India, popularly known as the ‘Garden City of India’, due to its pretty old colonial parks, or the ‘Silicon Valley of India’, due to its booming IT industry. Growing up there, in my privileged home in central Bangalore, sex workers certainly never crossed my mind. So eleven years later, when asked to do my Master’s project on HIV and sexually transmitted infections in female sex workers and their clients in various regions in southern India, including Bangalore, I had to look at my hometown from a slightly different angle.

The Avahan project is funded by well-known philanthropists Bill and Melinda Gates, and aims to prevent the spread of HIV around India. It turns out that HIV is highest in six states in India – Andhra Pradesh, Maharashtra (home of course to Bombay and Bollywood), Tamil Nadu, Karnataka, Nagaland and Manipur. Of these, the last two can attribute the spread of the virus mainly to injecting drug use, whilst the other four states to sexual transmission.

Anyone who has tried to conduct a survey into the lives of sex workers will know that pitfalls abound in this kind of research. In my project, I was making a model looking at HIV and genital herpes. I was essentially trying to figure out whether having one virus made you biologically more susceptible to the other, or whether if you had one, you were just more likely to get the other, because you catch both in the same way: through having unprotected sex. Collecting data on sex workers and in particular HIV is never easy, due to the huge stigmas associated with them. Sex workers themselves are relatively easy to find, if you know where to look, where to find the brothels. Their clients – normal working men, often married ones – are perhaps less forthcoming.

Condom use is a really important parameter in most HIV and STI models, as of course the rate of condom use can really tell you how well condoms are preventing the spread of the virus, and how good the prevention services are at getting them to the relevant people. One interesting fact we found in our data was that whilst clients reported condom usage of between 54–83%, the sex workers reported usage of 76–99%. But hang on – aren’t the sex workers and the clients having sex with each other? So, if they are telling the truth, shouldn’t their reported figures be exactly the same?

The massive difference between the reported condom usage from sex workers and from their clients is likely to be a form of response bias. This is where someone answers a question in the way they think the questioner wants them to answer. So, imagine for a second that you’re a sex worker in Bangalore. The Avahan project is doing everything it can to give you condoms and to encourage you to use them as much as possible. Then a researcher from the Avahan project turns up with a clipboard and asks you, expectant look on his face, how many times you have had unprotected sex with a client in the last week. You’re pretty unlikely to say “several times”. In some districts sex workers had reported condom usage of 99%, which is highly unlikely to be true, as if this was the case there would be no HIV problem in the region at all.

So how does one create a model when you can’t trust the data? Well, you just have to use your best guesstimate, which is often just the average of what the sex workers have reported and what the clients have reported. Other behavioural data, which may or may not be more reliable, include the length of time the sex worker has been active in that role and the number of sexual partnerships she has had in the last week.

The results from the models seem to show that HIV and genital herpes do have some biological interaction, and the fact that one increases with the other isn’t just down to the sex workers’ behaviour patterns. The whole study of HIV shows just how complex the spread of this virus is – down to behaviour, biological interactions with other STIs, and the problems that arise when trying to get information on these particular groups of people. The Avahan project has dramatically increased outreach to high-risk people in these Indian states; its work on condom distribution, peer-to-peer outreach, and STI testing and treatment is a great step towards combating the spread of HIV in India.