石榴直播

KENNESAW, Ga. | Jan 18, 2023

With the most severe flu season in a decade in full swing, businesses are anxious about the health of their employees, both for the employees鈥� wellbeing and the company鈥檚 bottom line. Faculty and student researchers at 石榴直播 recently partnered with health technology company radaRx to determine the effectiveness of new technology designed to help businesses monitor employee health and prevent infections.

鈥淩adaRx is on an ambitious journey to improve business resilience from common workplace infections and needed research-focused methods to validate the effectiveness of our approach,鈥� said Saurabh Srivastava, founder and CEO of radaRx. 鈥淭he 石榴直播 team led by Dr Sweta Sneha brought unique cross-functional skills, including syndromic analysis, healthcare policy, and data science to collaborate with the radaRx team on this complex initiative. The research partnership has been invaluable for us.鈥�

How Serious is Health-Related Absenteeism?

An estimated and . The total impact of infectious diseases on the US economy is estimated at $200B based on 158 million full-time employees with a median income of $41.5K per employee. 

Currently, most organizations make their employee health policy decisions for common illnesses like flu or gastro-intestinal issues based on data from public health agencies. However, this data, which relies on reporting from doctors and hospitals, has several shortcomings:

• No workplace specific data: public health agencies report infection data at a county or state level, which is not specific enough for their work locations to help organizations drive their preventive measures.
  
  
• Underestimated infection: the data is based on patients whose symptoms are severe enough that they seek medical care. It excludes people with milder symptoms or those who choose not to seek medical care.
  
  
• Delayed reporting: the reporting lags the spread of infections by several weeks as the data moves from through the healthcare system to the public health agencies for reporting.

These shortcomings make it difficult for organizations to predict infection risk at work locations and to take appropriate and timely actions to mitigate the spread of infection.

Enter Workplace Syndromic Surveillance

With the goal to address these gaps and improve workplace safety against infectious diseases, radaRx developed an 鈥渁lways-on鈥� system for assessing, predicting, and mitigating workplace risks associated with infectious diseases. 

Rather than tracking data from doctors and hospitals, radaRx focuses on tracking employees鈥� symptoms before a diagnosis. Here is how this system, called syndromic surveillance, works:

1. Workplace Events Monitoring - When employees have symptoms of illness (cough, runny nose, headache, etc.), they can use the radaRx app to report their symptoms and inform managers of their absence.
   
   
2. Syndromic Identification - Based on the self-reported infection cases, the app identifies the syndromes.
   
   
3. Risk Modeling 鈥� The syndromes lead to assessment of infection risk at the workplace. The app updates and projects the risk of infections to managers and employees daily.
   
   
4. Projections 鈥� The app sends alerts when the risk of infection exceeds a given threshold set by radaRx.
   
   
5. Safety Policy 鈥� The alerts provide strategies to mitigate risks of spread and preventive measures such as wearing face covering, social distancing or remote work depending on the level of risk and types of syndromes. 

How Effective is Workplace Syndromic Surveillance?

Syndromic surveillance has been used by public health professionals to rapidly detect spikes in illness. Using infectious disease modeling, the radaRx solution identifies illnesses, determines the growth factor and risk, and forecasts future infections and absenteeism. 

The research team at 石榴直播 simulated a series of outbreaks of different illnesses affecting different company sizes using the popular SIR (Susceptible, Infected, Recovered) epidemiology model. 

The following example considers a flu-like illness outbreak at an organization with 500 employees. 

The image shows the simulated daily infected cases for the organizations with and without workplace syndromic surveillance. The outbreak begins around day 30. The black line represents the daily infected cases without surveillance and the red line shows the daily infected with workplace syndromic surveillance. 

When the risk monitoring system detects an outbreak from the reported infections (after around 5 days), it alerts the managers and recommends that employees wear masks. If employees follow the recommendations, daily infectious cases are significantly reduced over time.

The simulation study demonstrates the potential benefits of workplace syndromic surveillance. It significantly reduces the peak of daily infections and the total number of infections in each outbreak cycle. For a medium-sized company with 500 employees, the workplace syndromic surveillance program reduces the infection peak and the overall number of infections by 50 percent. For larger companies, the reduction could be as much as 75 percent. 

The Diagnosis

The Kennesaw State study shows businesses can see significant benefits from having a workplace syndromic surveillance system. While the specific response would of course vary based on the business and the outbreak, i.e., respiratory, food-borne, water-borne etc., syndromic monitoring can combine with guidance from the CDC to create a playbook to follow during an outbreak that could lead to a healthier workforce and lower costs. 

Collaborations with organizations like radaRX allow academic programs like Kennesaw State鈥檚 Master of Science in Healthcare Management and Informatics program solve real-world problems by working with real-world data. Industry partnerships allow us to perform the valuable job of turning theory into practice. 

Authors
, Executive Director, Master of Science in Healthcare Management and Informatics Program
Yifan Zhang, Assistant Professor of Quantitative Analysis
, Assistant Professor of Economics
, Ph.D. Research Assistant, Analytics and Data Science