Integrating Genomics into Public Health Surveillance: Ushering in a New Era of Precision Public Health

Public health surveillance has been defined as “the ongoing systematic collection, analysis, and interpretation of data, closely integrated with the dissemination of these data to the public health practitioners, clinicians, and policy makers responsible for preventing and controlling disease and injury.” Surveillance provides an essential scientific foundation for both clinical and public health practice.

In a recent CDC paper, Richards and coauthors show how advances in information technology, data science, analytic methods, and information sharing are providing more opportunities to substantially enhance public health surveillance. They described recent enhancements in data analysis, visualization, and dissemination at CDC and identified several challenges ahead.

How can the tools of genomics be used in public health surveillance? Through CDC’s Advanced Molecular Detection initiative, whole genome sequencing of pathogens is already changing infectious disease surveillance and outbreak investigations. As cited in Richards’ paper, a recent outbreak investigation of human immunodeficiency virus (HIV) in Indiana showed how genome sequencing can help us understand origin and transmission patterns. The use of whole-genome sequencing is also increasing the speed of detection of even small Listeria foodborne outbreaks. Genome sequencing is now poised to transform PulseNet, the major foodborne outbreak detection system, which until recently relied on pulsed-field gel electrophoresis. Genome sequencing is also being integrated in international PulseNet activities.

Advances in human genomics will continue to have an increasing influence on public health surveillance, as we have written in previous blogs (read here and here). More precision in surveillance is now needed as new genetic tests become available in clinical research and practice. As of June 19, 2017, the NIH maintained Genetic Testing Registry listed 49,817 tests, 10,745 conditions, 16,232 genes, and 494 labs. In addition, next generation sequencing is increasingly used in clinical practice such as in diagnosis of rare diseases, and in large scale research settings, such as the U.S. Precision Medicine’s one-million person cohort initiative (“AllofUs”).

For many genetic tests in clinical practice, surveillance is needed to monitor the public health burden of diseases diagnosed by these tests and the impact of evidence-based interventions. For example, Lynch syndrome is a genetic condition affecting more than a million people in the U.S., which increases the risk of colorectal and other cancers. Lynch syndrome affects 1 to 3% of colorectal cancer patients, but most patients are not diagnosed. A diagnosis of Lynch syndrome will not only enhance medical management of these patients, but will also inform their relatives, through cascade screening, of their increased risk of preventable cancer. Currently, state cancer surveillance systems do not specifically identify Lynch syndrome as a diagnosis on a reported colorectal cancer. Collecting Lynch syndrome diagnoses in public health surveillance will allow a more precise assessment in the population of its incidence, morbidity and mortality, and geographic or racial/ethnic disparities in disease burden and access to interventions.

Similarly, familial hypercholesterolemia (FH), a relatively common genetic disease of cholesterol metabolism, affects about 1 million people in the United States, and most don’t know they have the condition. People with FH are at markedly increased risk of premature heart attacks. Diagnosis of FH could lead to earlier and a more aggressive lowering of cholesterol levels as well as screening in first degree relatives, who have a 50% chance of being affected. Until recently, surveillance for FH could not be tracked within medical records or death certificates, as it accounts for only a small fraction of nonspecific hypercholesterolemia. However, in 2016, new ICD 10 codes became effective for coding. Over time, this will allow a more precise tracking of population incidence, morbidity and mortality, variations in subpopulations, and gaps in implementation of evidence-based interventions.

In addition, integrating genetic information, including simple family history, in national surveys and surveillance systems will allow tracking of behavioral risk factors and health interventions stratified by population genetic subgroups. Currently, demographic variables such as age, gender and race/ethnicity are used to stratify prevalence of various risk factors (such as smoking, or drug use) as well as adherence to recommended interventions (such as colorectal cancer screening and mammography). People with certain genetic risk factors and/or family health history are at increased risk of many diseases. Population surveys will be able to assess how genetic and family health history information influences prevalence and trends of disease risk factors and health interventions. For example, in the mid 2000s, we measured dozens of genetic risk factors in the National Health and Nutrition Examination Surveys (NHANES). This has allowed rich survey dataset that includes various population health characteristics to be stratified by relevant genes. See a list of relevant genetic papers published using NHANES here. More generally, the use of family history in NHANES has been valuable in assessing the prevalence and association of family history with risk factors and diseases of public health significance. Examples of analyses of the role of family history in the prevalence of diabetes and adoption of risk-reducing behaviors are shown here and here.

Ultimately, as genetic and environmental data become more accessible and granular, combining them in public health surveillance will provide a more complete picture of the determinants of health and disease and the impact of health interventions, ushering in a new era of precision public health.

Submit your comments here.