HIV Molecular Surveillance: Questions and Answers

What is HIV surveillance?

Surveillance is the ongoing, systematic collection, analysis, interpretation, and dissemination of data regarding a health-related event. HIV surveillance collects, analyzes, and disseminates information about new and existing cases of HIV infection (including HIV Stage-3 or AIDS) for the purposes of guiding prevention planning.

What is molecular HIV surveillance?

Molecular HIV surveillance is the collection, reporting, and analysis of HIV genetic sequences generated through HIV drug resistance testing.

Genotypic HIV drug resistance testing is routinely conducted in many developed countries as a standard part of clinical care when initiating antiretroviral treatment and when drug resistance is suspected. This testing is recommended for all persons with diagnosed HIV infection at entry to HIV care to help HIV care providers select an appropriate treatment regimen. Drug resistance testing may be repeated if treatment is not resulting in a suppressed viral load and there is a question of whether the virus is resistant to the current treatments. HIV genetic sequence data are routinely reported to the state/local HIV surveillance program and to CDC through the National HIV Surveillance System (NHSS).

Molecular surveillance is not new to public health – it has been used for years to track foodborne infections and diseases such as TB. Molecular HIV surveillance is quickly becoming a part of routine HIV surveillance and can identify transmission clusters that would otherwise go unrecognized.

Important Note: Molecular analysis examines the genetics of the virus, not the person.

How is molecular HIV surveillance improving HIV prevention in the U.S.?

A critical step toward bringing the nation closer to the goal of no new infections is the use of molecular surveillance tools to identify and respond to groups of HIV-infected persons who have a connection related to HIV transmission (i.e., HIV transmission cluster).

Early evidence shows that molecular HIV surveillance can identify transmission clusters that would otherwise go unrecognized. Detecting recent and ongoing HIV transmission clusters is critical to focusing HIV prevention efforts where they’re needed most and expanded use of molecular HIV surveillance has the potential to significantly improve HIV prevention efforts.

What is a molecular cluster?

A molecular cluster is a group of persons with diagnosed HIV infection who have genetically similar HIV strains. The genetic sequence of HIV accumulates changes over time, sometimes rapidly. Immediately following transmission of HIV between two people, the genetic sequence of the HIV strain in the recipient will be nearly identical to strains found in the transmitting person. As time passes, however, the strains infecting each person will change independently of one another and will look more and more different.

A molecular cluster does not necessarily mean there are more HIV cases than usual overall. But it might indicate that HIV is spreading rapidly and public health action is needed. This information can help us take public health action, potentially preventing future transmissions.

How are molecular clusters identified?

Molecular clusters can be identified by state health departments or by CDC.

CDC Cluster Identification

CDC conducts routine analyses to identify molecular clusters that are concerning for recent and rapid transmission of HIV.

  • These analyses are conducted using national data that is available each quarter (based on data transmitted by HIV surveillance jurisdictions to CDC).
  • Prior to analysis, all HIV sequences in the national dataset are evaluated to determine the quality of the data.
  • CDC analyzes data using a secure local installation of HIV-TRACE, a software tool developed by University of California, San Diego and Temple University, to find clusters consistent with recent and rapid HIV transmission.
    • These analyses include only cases diagnosed in the most recent 3-year period.
    • The analyses identify pairs of sequences that are very closely related (that have no more than 0.5% variation, or genetic distance, between them).
    • Each sequence is connected to all other linked sequences.
    • A group of sequences for which each sequence is linked, either directly or indirectly (through other sequences), to all other sequences, is considered a molecular cluster.
  • Molecular clusters of concern are defined as clusters with at least 5 cases diagnosed within the most recent 12-month period.
  • When a molecular cluster of concern is identified, the primary jurisdiction involved (the jurisdiction with the majority of cases in a cluster) is notified and a cluster snapshot, describing the cluster, is shared with the jurisdiction.
  • CDC’s prioritization criteria may be modified or expanded in the future, as capacity allows.
What do molecular clusters tell us?

Molecular clusters may represent recent and ongoing HIV transmission, which is critical to focusing HIV prevention efforts where they’re needed most.

  • Because HIV is constantly evolving, people whose viral strains are genetically similar may be closely linked – directly or indirectly – by transmission
  • Prompt detection of rapidly growing clusters makes it possible to quickly target prevention programs that can improve the health of persons with HIV and reduce new infections
As rapidly growing HIV transmission clusters are identified, communities can intervene by implementing public health strategies with proven effectiveness. Expanded use of molecular surveillance will advance high impact prevention efforts by enhancing the ability of health departments to target prevention resources where they’re needed most.
What do molecular clusters not tell us?
  • Two people may be linked because they have a similar HIV genetic sequence, but this does not mean that one person transmitted HIV to the other. The genetic information only tells us that there is a link between two people, but cannot tell who gave the virus to whom.
  • Molecular clusters generally do not identify all individuals with HIV in a jurisdiction. People can be missing from a cluster when genetic data alone is used because not every person in a cluster may have been diagnosed yet or not every person with a diagnosis has a genetic sequence available.
Do molecular sequence data reveal who infected whom?

No. It is not possible to confirm direct HIV transmission from one person to another using molecular surveillance data alone. Although two persons infected with highly similar HIV strains could be directly linked through transmission, other transmission relationships could be consistent with this sequence similarity: both could have been infected from a third source, or they could be connected indirectly through a transmission chain including one or more intermediaries.

The data the states are collecting cannot determine who infected whom. The data they receive from laboratories is a partial sequence, or series of letters, that typically represents only about 15% of a total HIV genome. This testing helps the provider know which treatment regimen will be most effective, and the sequence is not unique to a specific individual.

What is a transmission cluster?

A group of HIV-infected persons (diagnosed and undiagnosed) who have a direct or indirect epidemiological connection related to HIV transmission. Transmission clusters can represent recent and ongoing HIV transmission in a population, where public health interventions could improve care outcomes and prevent new infections.

How are transmission clusters identified?

Transmission clusters can be identified through multiple mechanisms:

  • HIV case surveillance data. Aberrant increases in HIV diagnoses in a particular geographic area or population could represent recent and ongoing HIV transmission. It is important to note, however, that an increase in the number of diagnoses may not reflect an increase in transmission. Rather, an increase in reported diagnoses may reflect an increase in HIV testing that has diagnosed people whose infection may be longstanding.
  • Molecular HIV surveillance data. Analysis of HIV sequence data reported to HIV surveillance can identify clusters of cases with closely related HIV strains. This method may be particularly useful in identifying transmission clusters that are not detected through other mechanisms. Examples include transmission clusters occurring in an area with a high incidence of HIV infection, those involving multiple jurisdictions, or those in populations for whom partner service data is limited (e.g. due to anonymous partners). Molecular data can also be used to confirm that clusters identified through other mechanisms truly represent a transmission cluster.
  • HIV partner services and contact investigations. Partner services staff [referred to as disease intervention specialists (DIS) in many jurisdictions] routinely conduct HIV contact investigations, elicit partner information, and notify partners of their possible exposure or potential risk of HIV infection. As DIS work intensively in local communities, they are positioned to notice unexpected patterns or increases in HIV diagnoses.
  • Health care providers, health department staff or community members. HIV transmission clusters are sometimes first detected through astute observations from frontline staff at the health department or clinical providers. Observations like these call for further investigation to determine if and how these persons are connected and the extent of other connections they may have in a community.
  • Changes in patterns of other diseases: Aberrant increases in diseases that share similar transmission risks such as hepatitis C or STDs could also prompt investigation to identify potential recent and ongoing HIV transmission.
What is the relationship between a molecular cluster and a transmission cluster?
  • Identification of molecular clusters provides a tool to identify transmission clusters.
  • A molecular cluster contains only those people for whom HIV genetic sequence data are available and can be analyzed. As a result, molecular clusters include persons with diagnosed HIV infection who entered HIV care, had HIV drug resistance testing, and their sequences were transmitted to the local/state HIV surveillance program for analysis. This represents a subset of the underlying transmission cluster, which can also include:
    • Persons with diagnosed HIV infection who do not have a sequence available for analysis, either because:
      • They did not enter HIV care
      • They entered care, but have not had an HIV drug resistance test
      • They entered care and have had HIV drug resistance test, but the sequence was not transmitted to the health department for analysis, or was of poor quality and could not be analyzed
    • Persons with undiagnosed infection
  • Once a molecular cluster is identified, the molecular cluster can be characterized using case surveillance information such as demographic, geographic, risk, clinical and laboratory data; however, the corresponding transmission cluster and risk networks can only be identified through investigation, which includes the assessment of other data sources.
What is a risk network?

A risk network includes the group of persons among whom HIV transmission has occurred and could be ongoing. This network includes persons who are not HIV-infected but may be at risk for infection, as well as the HIV-infected persons in the transmission cluster.

Networks of concern include:

  • Networks in which HIV transmission occurred rapidly (with multiple new infections occurring within months of one another) and within a recent time window (within ~1-2 years). Recent, rapid transmission could indicate ongoing transmission or a potential outbreak for which public health intervention could interrupt transmission and prevent future infections.
  • Networks characterized by poor outcomes, such as late diagnosis or unsuppressed viral loads; this could suggest poor or limited access to care, and could indicate a network in which persons with HIV infection that has not yet been diagnosed or persons with high viral loads could be contributing to ongoing transmission.
  • Networks representing vulnerable or underserved populations, such as pregnant women, adolescents, rural populations, people who inject drugs, foreign-born persons, or other groups defined by local epidemiology and context.
  • Networks in which drug-resistant strains of HIV are being transmitted; particularly networks with resistance to first-line antiretroviral medications or pre-exposure prophylaxis (PrEP) regimens.
  • Networks including persons in stage 0 of HIV infection, which includes acute infection or recent seroconversions (negative HIV test within 180 days of HIV diagnosis), indicating very recent transmission.
  • Networks not reached by testing efforts, as evidenced by large proportions of cases that were diagnosed through incidental testing, such as screening in plasma centers, emergency departments, or correctional institutions; this could indicate other cases in the network that have not yet been diagnosed and could be contributing to ongoing transmission.
How does identifying transmission clusters impact HIV prevention among risk networks?
  • Investigation of transmission clusters can identify risk networks that are concerning for ongoing transmission, early infection, poor health outcomes, or other reasons, such as transmission in a particularly vulnerable or underserved population, or transmission of drug resistance.
  • Investigation of transmission clusters can identify key characteristics of the underlying risk network to guide intervention efforts to improve health outcomes and prevent additional infections.
    • Investigation includes the examination of existing data, including partner services data, or collection of new data, to identify factors associated with transmission.

    Intervening in risk networks can improve health outcomes and interrupt transmission through activities that could include:

  • Identifying persons with diagnosed HIV infection in the transmission cluster who are out of care, and ensuring that these cases are linked to or re-engaged in care.
  • Identifying persons with undiagnosed infection who are part of the transmission cluster, and linking these persons to care.
  • Identifying HIV-negative persons in the risk network who are at risk for acquiring HIV and offering effective prevention interventions, such as PrEP and other prevention services.
  • Identifying potential venues, communities, or geographical areas in need of services or broader community interventions.
  • Other interventions tailored to the factors facilitating transmission or slowing diagnosis in the particular cluster.
  • By expanding our knowledge of transmission dynamics, transmission cluster data can be a powerful tool to target the interventions we know are effective (engagement in care, HIV testing, PrEP).
What are the steps in assessing and responding to transmission clusters and risk networks?

The steps to investigating and intervening in a transmission cluster include:

  1. Identifying known transmission cluster and risk network
    • Systematically reviewing partner services data for all persons in the molecular cluster to identify known members of the transmission cluster and risk network, and assess completeness and outcomes of partner services investigations.
    • For HIV-positive persons in the transmission cluster, using surveillance and other data sources to determine viral suppression information and early infection status.
    • For HIV-negative persons in the risk network, determining HIV testing history and whether person was referred for PrEP.
  2. Assessing priority level and need for enhanced investigation and response
    • Reviewing data collected in step 1 for the full transmission cluster and risk network determination to assess:
      • How effectively the transmission cluster and risk network have been identified.
      • Potential for ongoing transmission, poor outcomes, and other concerning factors.
    • Determining whether enhanced investigation and/or intervention activities are needed.
  3. Initiating critical interventions for all priority clusters
    • Initiating strategies to promote viral suppression for all HIV-positive persons who do not have evidence of viral suppression, particularly those with early infection.
    • Initiating testing and re-testing activities, including PrEP referral for uninfected persons in the risk network.
  4. Conducting enhanced investigation and intervention activities, as appropriate
      • Considering strategies to identify previously unrecognized partners for testing and provision of prevention interventions, such as patient interview, or social network interviewing.
      • Considering strategies to better understand factors associated with delays in diagnosis or interruptions in care, or other factors contributing to transmission in the cluster; strategies can include medical record review or qualitative interviews.
      • Considering additional interventions depending on the characteristics of the cluster, which could include but are not limited to targeted outreach at venues, communication campaigns through media or apps, health alerts, and scale-up of PrEP or testing services.
  5. Monitoring progress, and determining when a cluster response should be closed
    • Continuing to monitor cluster growth over time in order to determine whether transmission has been interrupted, or if it is ongoing.
    • Assessing the impact of the investigation, and subsequent interventions.
    • Identifying lessons learned for future cluster response activities.
    • Determining when a cluster response should be closed.
Does the identification of a molecular or transmission cluster indicate that an outbreak of HIV is underway?

Not necessarily. The term ‘outbreak’ can be defined in different ways. While the textbook definition of an outbreak is ‘an increase, often sudden, of disease above what is normally expected in that population or area’, the term is often used to describe situations when an urgent or emergency-level public health response is needed. Deciding if a transmission cluster calls for an urgent response is a continuous process, and multiple factors should be considered, such as the size of the cluster, potential for ongoing transmission, and potential for poor outcomes. These same factors are key considerations in determining whether to describe a situation as an outbreak.

How should cluster data be handled?

Cluster data is sensitive, and should be handled in accordance to state/local public health law and procedures outlined in the CDC National Center for HIV, Viral Hepatitis, STD, and TB Prevention (NCHHSTP) Data Security and Confidentiality Guidelines, including, but not limited to:

  • All data should be stored securely whether in electronic or paper form.
  • Access to identifiable information should be limited to authorized persons.
  • Any electronic output that could breach confidentiality (e.g., line listings, STATENOs) should be stored on a secure server. Hard copies should only be produced when necessary; when produced, they should be locked up and not taken out of the office. Paper copies should be shredded when no longer in use.
  • All confidential data (including line lists and any documents including STATENOs) should be marked as confidential and encrypted for transfer or when not in use.
  • Any information taken into the field as part of field investigation or service provision should include only the minimum amount of information necessary and be maintained securely at all times. Areas should develop specific procedures for securing information during field investigations.
  • Data should only be shared with staff who have a need to know the information.
  • Any breach of data security protocol, regardless of whether personal information was released, should be reported to the overall responsible party and investigated immediately.
What should be done to minimize the risk that molecular surveillance data will be misused or misinterpreted?

To minimize risks associated with molecular surveillance data, state and local jurisdictions should:

  • Follow detailed CDC guidelines related to privacy, confidentiality, and data security, as well as relevant state and local laws
  • Ensure that information about transmission clusters is used for public health purposes only
  • Develop plans for communication with local communities, and crafting messages that avoid creating or perpetuating stigma
  • Ensure cluster data are adequately protected
  • Consider evaluating any laws that criminalize HIV transmission
CDC is working with national stakeholders to identify and address ethical concerns and will continue to address any issues that emerge as more health departments implement this work.

Note: Health departments never report names or other identifying information to CDC.