Identifying populations at an increased risk for HIV and preventing outbreaks remain a challenge, but researchers are using tracking technology to find these at-risk individuals and provide treatment, as shown by a recent study and a new research program.
More effective strategies are urgently needed to maximize the dissemination and impact of HIV biomedical interventions among young Black sexually minoritized men (SMM) and transgender women (TW), who bear a significant burden of new HIV diagnoses in the United States, said Yen-Tyng Chen, PhD, of Rutgers University, New Brunswick, New Jersey, in an interview.
“SMM and TW aged 25-34 years contribute the highest number of new HIV infections, and Black SMM and TW are more vulnerable to HIV than other racialized groups of SMM and TW,” said Chen. “These disparities are largely due to structural barriers, such as access to health and stigma related to PrEP [pre-exposure prophylaxis] use or toward racial and sexual/gender minorities,” she said.
In a recent study of more than 200 individuals, Chen and colleagues used GPS data to identify venue-based networks to provide HIV prevention interventions and identify more potentially affected partners for Black SMM and TW.
Venues where individuals socialize not only can play a role in the transmission of HIV but also have potential for providing HIV and sexually transmitted infection prevention strategies, the researchers noted.
In the new study published in Annals of Epidemiology, the researchers reviewed data from the Neighborhoods and Networks Study, an ongoing cohort study of the relationships between neighborhood- and network-level characteristics and HIV care engagement outcomes among Black SMM and TW in Chicago, and in the southern United States. The current study focused on individuals in Chicago.
The researchers mapped GPS data to nearest SMM- and TW-friendly venues that would yield the highest reach to local individuals who were candidates for PrEP intervention. In the analytical sample of 272 individuals, 54% were not engaged in HIV care (PrEP or antiretroviral therapy); 86% identified as SMM and 11% as TW. More than half of participants (58%) self-identified as gay, homosexual, or lesbian. Most participants were single (64%) and 68% reported annual incomes below $20,000.
Participants carried a GPS device for a median of 11 days. The study included a total of 222 pre-identified SMM- and TW-friendly venues. Participants were within 50 miles of 75.5% of all venues during the study period.
Real-Word Implications and Limitations
“We use big data, including GPS, survey, network data, to identify optimal sets of venues for HIV prevention,” Chen told Medscape Medical News. Venue-based affiliation networks could help identify areas to reach the priority population or connect many other venues frequented by the priority population, she said. “By identifying important, feasible, and contextually acceptable venues for HIV interventions, we can maximize the impact of HIV PrEP intervention,” she added.
The use of GPS is not without challenges, Chen noted. The current approach does not account for the length of time spent at each location, she said. “Also, we matched participants’ GPS data with a pre-identified list of SMM- and TW-friendly venues, which may not have been complete; therefore, venue affiliation may be underestimated,” she noted. However, “we may overcome these barriers by doing a real-time ecological momentary assessment, whereas we can obtain data on where the participants spend time and why they are there,” Chen added.
“Future studies are warranted to develop and evaluate implementation strategies based on this approach, especially among highly marginalized populations that are not typically reached through convention outreach interventions,” Chen said.
Additional Algorithms to Find and Assist
A new initiative in development by researchers at Yale University will use the latest technology to identify early signs of HIV outbreaks based on temporal patterns, according to a press release from the university.
“To maximize the impact of our work, we met with health officials, local clinicians, and community-based organizations to understand the needs of those on the ground as they seek to prevent and manage HIV outbreaks and epidemics in their locales,” said Gregg Gonsalves, PhD, associate professor of epidemiology at the Yale School of Public Health, New Haven, Connecticut, in the press release.
More than a dozen outbreaks of HIV among individuals who use drugs have occurred in the United States since 2014, Gonsalves said in an interview.
“These outbreaks have clinical, epidemiological, and economic implications,” he said. Infections associated with drug use (not only HIV infections but also skin and soft tissue infections) and overdose require treatment, and individuals with late-stage hepatitis C virus (HCV) and endocarditis may need valve replacements and liver transplants, he said.
“Epidemiologically, if we are going to end the AIDS epidemic in the US, these outbreaks need to be prevented, detected earlier, and managed well and efficiently,” said Gonsalves. Identifying and targeting the often hard-to-reach populations at an increased risk for HIV remain a challenge, but the Yale team is designing algorithms to assist with the prediction of high-risk places for HIV outbreak to early, quick detection of nascent outbreaks, and identification of undiagnosed cases in the community, Gonsalves explained.
High-impact HIV prevention requires containment strategies across the lifecycle of an outbreak, said Gonsalves. The research team’s comprehensive plan involves pre-deploying resources to places correctly identified as high risk, even before an outbreak occurs; shortening the time to detect an incipient outbreak once it has happened and to distinguish it from a false alarm; and allocating case identification and containment resources efficiently to engage an outbreak already in progress, he said.
The research project will focus on three areas: Temporal pattern recognition and outbreak vulnerability, refined outbreak detection, and adaptive case-finding strategies.
To improve traditional HIV surveillance, the research team will explore temporal patterns to identify early indicators of HIV outbreaks. The team plans to use distributed lag interaction models to link current HIV case counts and historical data on variables such as other infectious diseases and overdose rates.
The team also plans to take a page from engineering’s playbook with an adapted technique to identify abrupt changes in system processes; the goal is more accurate identification of emerging outbreaks and fewer false alarms, according to the press release.
With regard to case-finding, the researchers plan to create algorithms to help deploy mobile units and other resources for HIV intervention in the event of an outbreak.
Potential for Public Health Impact
Physicians and other healthcare providers are being confronted by a converging public health crisis of substance use and infectious diseases, Gonsalves said. Primary care physicians and others outside of infectious disease and addiction medicine practices will see more patients in need of diagnosis and treatment of an opioid or other substance use disorder, he noted.
Even in places with access to specialized care, clinicians will see more skin and soft tissue infections, including endocarditis, HIV, and HCV, which in some places will stretch clinical capacity, said Gonsalves. “With better prediction and detection, we hope to minimize cases of these clinical sequelae of drug use,” he said. “By finding undiagnosed cases of disease, notably HIV, we can get people into care earlier, before they develop symptoms of disease and minimize the chance of them passing on the disease to others,” he said.
The current study by Chen and colleagues was supported by the National Institute on Drug Abuse. The Yale research was supported by the National Institutes of Health’s Office of AIDS Research and the National Institute on Drug Abuse. The researchers had no financial conflicts to disclose.
