Deadly Sex: You Could Be Just A Few Degrees Away From Contracting HIV

How many sexual partners have you had in your lifetime?

That’s the first question a 1994 sex survey asked to 2810 random Swedish subjects.

College of Arts and Sciences research associate Luis Amaral and CAS Physics professor H. Eugene Stanley have been studying the results of this survey and others and have developed a model that can be used to track the spread of HIV, a deadly virus that attacks the immune system.

The 1994 Swedish survey, conducted by Christopher Edling, Fredrik Liljeros and Yvonne Aberg, unveiled that among the participants, who were pooled from different age, ethnic and income groups, the average number of sexual partners for men was 15 and for women was 7.5. The person with the greatest number of sexual partners reported having sex with 800 people in his lifetime.

Stanley, who graduated from Harvard and has worked at BU since 1976, and Amaral, who earned his Ph.D. from BU in 1995, examined the data from the Swedish survey and used the info they extrapolated to create a complex social, sexual network that traces the path of HIV.

“The studies ask all sorts of things; they ask questions like: number of sexual partners, age difference between partners, sexual dysfunctions, etc,” Amaral said. “Those studies don’t give you the actual network. We don’t know who they had sex with, we only know how many partners they have had.”

By analyzing how the data from the study was distributed, Stanley and Amaral found that people, even those who have had relatively few sexual partners, are sexually connected to a broad range of people, most of whom they have never met.

According to Amaral, it was believed HIV would spread like other viruses, in a bell curve distribution, pitting most results in the middle, while very few people fall outside the curve.

“One of the classical distributions is the bell shaped; one example of something that follows a bell shape distribution is peoples’ height,” Amaral said. “Essentially everyone that you know has somewhat the same height: [ranging from] four feet to seven feet; it is a small range. If I were to tell you I know someone who is 50 feet tall you wouldn’t believe me.”

A bell curve indicates the probability of contacting a sexually transmitted disease is small and limited, almost as unlikely as winning the lottery, Amaral said.

However, Amaral said the results did not form a bell curve but a curved shape called a power law distribution. A power law distribution indicates that the study had no numerical limit, meaning a person can have no sexual partners or an astronomical number of partners.

“The power law distribution doesn’t have a scale,” Amaral said. “That’s why in the study we found people who have had 600, 800 partners, which is a very large factor. When you have such a broad range of possibilities you can no longer talk about a scale.”

No scale suggests that HIV can be spread to virtually anyone. Most people in the study fell in the lower range of the curve, indicating they have had a comparatively low number of sexual partners. However, although many participants had a minuscule number of sexual partners, Stanley and Amaral concluded that they can connect each person to large numbers of other people through an extensive chain of sexual relations. Stanley and Amaral’s findings were published in Nature, an elite science magazine, in 2001.

“It was not expected that the distribution was so broad that there can be a large number of partners,” Amaral said. “It tells you that some people have so many connections.”

Since accessing the data gathered by the Swedish researchers, Stanley and Amaral have further analyzed other studies, including a sex study conducted in 1994 in the U.S., and have constructed an intricate model that explains the virus’s network — the path of HIV’s spread from one person to others across the country.

“We can actually calculate the length of the [sexual] chains,” Amaral said. “We know that the length of these chains is very short.”

Amaral compared the chain of sexual relations to the Six Degrees of Kevin Bacon game, which tries to connect any actor to Kevin Bacon by naming movies that actor has starred in and finding co-stars who have starred in a Kevin Bacon movie.

“You can connect almost any actor to Kevin Bacon in very few steps and usually it takes only 3,4,5, or 6 steps. Here is the same principle. Instead of actors think about people and instead of movies think about sex. Some people have so many connections so you can actually connect anyone to a small chain. This is true even for someone that doesn’t have many partners, but [one of their] partners has sex with many partners.”

The network results suggest that a person who has sex with just one or two people is at just as much risk of contracting the virus as somebody who has had a larger number of partners. Any one person with HIV could be in contact with somebody through the network even if the two have never met or engaged in sexual activities.

“The only possibility that would make it not work is if people were having partners within close clusters,” Amaral said. “When you look at the data you see that there is always some links between clusters, and then the distances [between people] becomes very short.”

Stanley and Amaral’s results predict the rampant spread of HIV over the coming years and debunk common myths that the virus is only contracted by promiscuous people.

“Because of the structure of this network, you are not very far away from someone with the disease,” Amaral said. “There is this image that people who get AIDS are drug addicts or prostitutes. It will be spreading to the general population, and that is indeed what is happening.”

Since the network can be used to make simulations of the propagation of the disease, Amaral said it will help researchers who are trying to develop strategies to contain the disease of prevent its spread.

“Imagine if you develop a vaccine, it’s very expensive and you can only give it to only a few people,” Amaral said. “You need the model of how the disease propagates to know what is the most efficient immunization program.”

Amaral said the probability of being in a network does not only apply to the path of HIV, but is also prevalent among different types of natural disasters. Many models of networks seem to have similar properties including the power law distribution property.

“There seem to be a number of phenomena in social, biological and physical sciences in which you find this unusual power law,” he said.

According to Stanley, studying the path of the HIV network could help researchers understand the paths of terrorist networks such as Al Qaida. Identifying and intercepting the routes of terrorist networks would ensure “the survival of the species” in spite of attacks.

Developing a model to explain the network may also allow researchers to identify patterns among brain neurons and could contribute to the mapping of the human genome.

“This is completely new work,” Stanley said.

Amaral said the Swedish study was not foolproof and there were some discrepancies in the results. While female prostitutes refused to participate in the study, men who had sexual intercourse with prostitutes were studied. Another complicating factor is that men have a tendency to exaggerate the number of sexual partners they have had.

However, Amaral said that if men did exaggerate their number of sexual partners it would most likely not be much higher than their true number. He added that even if some men doubled their number of sexual partners it would not impact the study or alter its results.

Anaral said he and Stanley are currently analyzing additional data from other national sex studies and will continue to thoroughly test their model.

“We have submitted a paper for publication,” Amaral said. “We are doing data analysis. We found data from the United States and have been analyzing that data since.” Despite the frightening results of Stanley and Amaral’s diagram, Amaral said to remember that only a small population was studied and it is not representative of the whole population.