Ideally, everyone should be tested for HIV and in fact, federal guidelines introduced in 2006 recommend routine HIV screening for all patients. In reality, however, only about half of U.S. adults have ever been tested for HIV and about half of the 50,000 new infections that happen every year in the U.S. are transmitted by people who are unaware of their HIV status. Such statistics recently led a group of researchers to ask if there’s a more efficient way to go about curbing the HIV epidemic.

“We strongly support the concept of universal testing and treatment to limit or control the spread of HIV,” Susan Little, faculty in the Antiviral Research Center at the University of California-San Diego, told me. “The problem with that approach is that here in the U.S., despite having the greatest health care dollar resources available, we’re not universally testing or treating. …So if we have limited prevention dollars to spend, where might we focus our resources to have a much greater impact than if we used the same resources to try to reach everyone?”

Basically, Little and her colleagues asked this question: Instead of casting such a huge prevention net and hoping that people will fall in, could they shrink the net, attach a GPS unit to it and let it lead them to hotspots of high risk and greater preventive potential? They found that they could — and moreover, all the data they needed to do it was already available. Using the unique genetic HIV sequence from individual patients as a proverbial GPS device, the researchers were able to generate maps that pinpointed high-risk transmission networks that they hope can eventually direct prevention resources where they’ll have the greatest impact. Here’s how they did it.

Little and her fellow researchers, who published their methods and findings in the June issue of PLOS ONE, analyzed HIV sequence data among nearly 500 people recently infected with HIV as well as 170 of their sexual and social contacts in San Diego between 1996 and 2011. The sequencing data was already available and centralized as a result of testing HIV patients for antiretroviral drug resistance, which has become standard protocol in HIV care. Using the sequencing data, they pinpointed viruses from different individuals with a high degree of genetic similarity, which could suggest a transmission link. And from those links, they could map a network of HIV transmission and then model the likelihood of new HIV transmissions. (Little did caution, however, that viral similarity does not alone prove that a transmission occurred, only that the individuals are both part of a closely connected transmission network.)

“Not everyone who is HIV-infected is equally likely to transmit the infection to others,” Little said in a news release about the study. “There are clusters of more active disease transmission. We can use this information to target treatment interventions to those most likely to transmit the virus to others and markedly reduce the number of new infections.”

In mapping the networks, researchers were also able to map, in many cases, the direction of infection. When putting all the components together, researchers could measure the rate at which a network and sub-networks were growing and zero in on areas in the population in which HIV is spreading at a faster rate. Within a network, researchers then calculated a transmission network score that estimated the risk of transmission from a patient who is newly diagnosed to a new partner.

According to the study findings, high scores identified a group of patients who had significantly greater predicted risks of HIV transmission within their first year of living with the virus than patients with low scores. If put into clinical practice, the concept means clinicians could more precisely direct treatment and prevention education to patients at highest risk of transmission. Little and study co-authors Sergei Kosakovsky Pond, Christy Anderson, Jason Young, Joel Wertheim, Sanjay Mehta, Susanne May and Davey Smith write:

When adequately sampled, HIV-1 sequence analysis can help characterize local HIV epidemics. This network based study in San Diego, California, corroborated previous findings that higher (viral load) was associated with transmission risk and that early (antiretroviral treatment) decreased this risk. This study went further to identify that network connections at baseline also predicted future transmission risk, and prevention efforts targeted to these individuals may be a better use of prevention resources than random implementation or targeting individuals with higher number of sexual partners or recently diagnosed with a (sexually transmitted infection).

But wouldn’t genetically linking newly infected patients to other patients violate medical privacy? No, Little told me. Clinicians would only get back a transmission network score. It works like this: When a patient’s sample is sent to the lab to test for drug resistance, the lab could automatically search for sequencing similarities within its existing database, generate a score and relay only that score to the physician, who could then more precisely act to prevent further HIV infections. The doctor wouldn’t know how many similarities were found or whom their patient might have had sexual contact with — they’d just get a score.

That score would ideally be much more telling than the kind of information that a doctor would normally collect during a patient encounter. For example, Little told me, a patient who identifies only one sexual partner may be at low risk for transmission. However, sequencing data could reveal that the one partner has dozens of potential partners, which offers unique insight into an underlying transmission network of which even the initial patient is unaware. Little admitted that while labs may be nervous about linking sequencing similarities at the moment, the scoring technique is a strategy that wouldn’t first require a revision of medical privacy laws.

“These data are already available and we could start using it tomorrow,” she said. “But we do need to figure out answers to some important questions.”

In addition to ensuring that no one’s medical privacy is breached, Little and her colleagues also want to know how much of an impact the new information would really make. Finding patients at high risk of transmission is only one part of the prevention puzzle; getting high-risk patients to adhere to a treatment regimen and modify their risk behaviors is another. Little noted that while newly infected patients do tend to engage in immediate behavior changes, risky behaviors tend to resurface over time. Still, the mapping method is a promising approach to help clinicians as well as public health practitioners maximize their prevention resources.

“Some people might read the study and think ‘why are you targeting interventions when everyone should be treated’ and I completely agree with that,” Little told me. “This isn’t meant to say that we should be targeting certain individuals because that’s a better approach. It’s that the approach of universal testing and treatment has well-documented failures. …Until we develop better methods to endorse and get people to accept testing and treatment, we’re saying that if we really want to control the HIV epidemic, we may have the data we need to do that right now.”

According to the Centers for Disease Control and Prevention, about 1.1 million people in the U.S. were living with HIV as of 2010, and about 16 percent don’t know they’re infected.

To read the full HIV network study, visit PLOS ONE.

Kim Krisberg is a freelance public health writer living in Austin, Texas, and has been writing about public health for more than a decade.