Location Intelligence Vital to Johnson & Johnson’s Vaccine Development Program

Johnson & Johnson combined location data with several other types of data into an advanced analytical model to derive insights into and develop a deeper understanding of the pandemic.

In February 2020, Johnson & Johnson announced that its pharmaceutical arm, Janssen Pharmaceutical Companies (Janssen), would further expedite its investigational coronavirus vaccine program through an expanded collaboration with the Biomedical Advanced Research and Development Authority (BARDA), which is part of the Office of the Assistant Secretary for Preparedness and Response (ASPR) at the U.S. Department of Health & Human Services. The partnership committed more than $1 billion of investment to co-fund vaccine research, development, and clinical testing.

In order to develop and test an investigational vaccine candidate against the virus, it is imperative to reach the most at-risk populations and geographic areas. The data science team at Janssen combined location data with several other types of data into an advanced analytical model to derive insights into and develop a deeper understanding of the disease. The clinical study design was supplemented with a rigorous, agile view of critical pandemic data, updated on an hourly basis.

Location-based tracking and forecasting

Janssen developed a global COVID-19 surveillance dashboard that tracks how the disease is impacting certain areas by pulling in data at a country, state and even county level. The dashboard helps paint a real-time picture of how the virus is moving around the world on an hourly basis, and guides the company where it should test its investigational COVID-19 vaccine candidate.

The dashboard does not only track the pandemic, it is also used to forecast hotspots where the disease might spread. The hotspot prediction was critical in the clinical studies of the lead vaccine candidate. The goal was to recruit vaccine trial participants who were most likely to be exposed to the virus in areas with high disease transmission rates. This would give researchers more data to quickly demonstrate efficacy.

Janssen, in partnership with the Massachusetts Institute of Technology, developed Machine Learning-based prediction models that could show where COVID-19 was likely to spike next. This approach used a combination of Machine Learning and epidemiological infectious disease models known as SEIR (which divide populations into four categories: Susceptible, Exposed, Infectious, and Recovered).

The predictive models integrated the global surveillance dashboard data with information about local policies and behaviors, such as how people are traveling and whether they’re being compliant with mask-wearing. The information was then provided as holistic guidance to the Janssen clinical teams as they selected the trial locations for their vaccine candidate.

Identifying high-risk patients with Location Analytics

Location is the most important parameter used to recruit populations for the vaccine trial. The second parameter is biological, such as age and ethnic groups, and the third is environmental, for example, plant workers who have high occupational risk of exposure.

Since the virus doesn’t impact everyone in the same way, Janssen built models to better understand what might make someone more prone to severe illness from the virus, as well as how different treatment courses may affect patient outcomes. The main objective of creating the model was to ensure that the patients who are most likely to be exposed to the virus—and most at risk for severe outcomes—are prioritized for enrollment in the company’s vaccine trial.

Since New York was hit hard and early in the pandemic, the company analyzed data from hospitals in New York City to give data scientists a broad view of the disease in a large population of patients. On top of this, longitudinal data from a global COVID-19 registry where the disease is most prevalent, was also added.

Keeping employees safe and healthy

Janssen used location data to make informed decisions about which labs could remain open, how many people could be onsite at a time, and how different company facilities were configured and sanitized.

The company developed a location-based model to track site density in real time and provide feedback on how employees should vary their work hours. This helped reduce the number of people working at a given location at a given time.

The company also built an analytical model called COVID Lens that takes into account factors like COVID-19 test sensitivity and specificity, the prevalence of the virus in the community, and the number of people working onsite. This helps determine if and how many employees should be tested for the Novel Coronavirus, and how frequently testing should occur.

How Johnson & Johnson Used Location Data to Fight the Pandemic

Location data was used to track the pandemic and forecast hotspots in order to select the best trial locations for the company’s vaccine candidate.
By analyzing location data, along with biological and environmental data, patients who were most likely to be exposed to the virus, and most at risk for severe outcomes, could be identified and prioritized for enrollment in the company’s vaccine trial.
A location-based model was used to track site density in real time, keeping employees safe by making informed decisions about which labs could remain open and how many people could be onsite at a time.