Projects

How can we monitor and learn trends about Lyme disease?

Using a data coordinating center for Lyme disease surveillance

Challenge

The surveillance landscape for Lyme disease, particularly in high-incidence regions, has shifted with a recent change to laboratory-based reporting criteria no longer requiring reporting of clinician-diagnosed cases. This change has created a need for new data sources to estimate incidence and monitor trends in the epidemiology and clinical manifestations of Lyme disease in these areas.

Electronic health records (EHRs) are a promising potential source of data for monitoring trends in Lyme disease. EHR data have been used to assess Lyme disease epidemiology, and several algorithms incorporating diagnosis codes, test orders, and antibiotic dispensing have been proposed for case identification. With respect to constructing optimal case identification algorithms, there is room for improvement.

The Surveillance-Based Lyme Disease (SubLyme) Network is a collaboration between CDC’s National Center for Emerging and Zoonotic Infectious Diseases, Westat, and 5 health care systems aimed at using EHR data to supplement and enhance traditional Lyme disease surveillance.

Solution

For this surveillance study, Westat is working with CDC and 5 health care systems in areas with a high Lyme disease burden to establish a virtual epidemiology network. This network will gather EHR data from these health care systems and use these data to generate Lyme disease incidence estimates and support in-depth studies of epidemiology, clinical manifestations, and, should a product come to market, vaccine impact.

The objectives of this project are to:

  1. Develop and test EHR-based case definitions for Lyme disease.
  2. Apply these definitions across disparate health care systems to estimate local Lyme disease incidence.
  3. Use these definitions as the starting point for additional EHR-based studies of Lyme disease epidemiology.

The definitions to be tested include both traditional public health-style case definitions and machine learning-derived definitions.

Westat serves as the data coordinating center for this network and is responsible for developing protocols, harmonizing data collection efforts, analyzing data, and facilitating collaboration between network partners.

Year one of this project is focused on the following tasks:

  1. Establish a network. Westat is working with major health care systems that are directly contracted with CDC to collect EHR data on a well-defined cohort of patients and also provide a validation set (both cases and non-cases) to be used for the production and validation of a set of case definitions. Included in the data provided by sites will be dates related to Lyme disease events, such as date of encounter, date of diagnostic test, and date of medication prescribed and high-level geographic information.
  2. Establish a set of EHR-based case definitions. Using the validation data, we will produce a series of candidate case definitions drawing on EHR data. These definitions will include both traditional surveillance case definition algorithms and classification tree–based machine learning approaches. We will evaluate the performance of these definitions and decide on one or more to use for incidence estimation and defining the population to be used for later studies.
  3. Produce and routinize incidence estimates. Using the case definition or definitions established from the validation data, we will estimate Lyme disease incidence in the cohort defined by each health care system. An interim estimate will be provided during the peak Lyme disease transmission season along with a final estimate later in the year.

Following the pilot phase, we will support a series of more in-depth analyses to complement the ongoing incidence estimation. Tasks for this full network phase will include developing a framework for supplemental studies and conducting vaccine impact assessments, should products come to market.

Results

Through this effort, Westat will provide CDC with a new data stream for situational awareness of and research into Lyme disease. This will complement existing public health reporting by providing greater insight into the clinical incidence of Lyme disease without relying exclusively on laboratory-based surveillance. The goal for this work is to ultimately provide the framework for the future of Lyme disease surveillance and enhance CDC’s capacity to identify and respond to trends in Lyme disease incidence.

Insights

Deep Dive with Our Experts

view all insights

How can we help?

We welcome messages from job seekers, collaborators, and potential clients and partners.

Get in Contact

Want to work with us?

You’ll be in great company.

Explore Careers
Back to Top