May 26, 2024

Health Benefit

Healthy is Rich, Today's Best Investment

Using data analytics to predict chronic conditions

4 min read

Photo by Towfiqu barbhuiya on Unsplash

Opinions expressed by Digital Journal contributors are their own.

The U.S. healthcare system is dealing with an epidemic of chronic diseases. 90% of the country’s $4.5 trillion healthcare budget is spent on tackling these illnesses. Amongst these issues, diabetes is one of the most prevalent. Affecting nearly 38 million Americans, Diabetes is a chronic health condition where the body doesn’t make enough insulin or can’t use it as well as it should. When there isn’t enough insulin, too much blood sugar stays in your bloodstream which can cause serious health problems over time such as heart and kidney diseases. 

Making matters worse, the number of individuals developing diabetes is rapidly climbing every year. However, there is hope. Bharath Srinivasaiah, an expert in healthcare business analytics, has been at the forefront of addressing this challenge with predictive data analytics. It’s a solution that can help curtail the suffering, medical costs, and productivity losses associated with these chronic diseases – costs which are estimated at almost $400 billion per year. 

Data analytics in healthcare

Data analytics is the process of examining datasets to come up with conclusions about the data they contain. Predictive analytics uses statistical models and forecasting techniques to identify trends and determine if they are likely to recur. Applied to healthcare, data analytics can be used to identify patients who are at risk of developing chronic conditions so that providers can intervene early on. Early intervention can prevent diseases from advancing and also reduce the costs that would’ve been involved with managing the ensuing chronic conditions. 

Bharath Srinivasaiah: Healthcare innovator

Photo courtesy of Bharath Srinivasaiah

Originally from Bengaluru, nicknamed the Silicon Valley of India, Bharath is a seasoned Solutions Architect with nearly 15 years of specialized experience in Healthcare Business Analytics and Reporting. His primary role is in designing and implementing robust data solutions including database architecture, ETL design, and advanced data analytics. During Covid-19, Bharath worked on tools like the C19 Navigator and C19 Enterprise Dashboard which allowed healthcare providers to highlight future hotspots for infections, track community mobility, and monitor testing and utilization figures in each area. In recognition of his tremendous contributions, Bharath is the recipient of over 20 awards including the “Global Recognition Award” and “Young IT Leader of the Year 2024”. He has also published 3 academic papers including “Utilizing Data Analytics to Predict Chronic Condition a Focus on Diabetes” in the IJSR journal. His paper in the IJSR journal is the basis for this article. 

The predictive model for chronic diseases

Bharath’s solution for curtailing chronic diseases like diabetes involves a predictive data model employing logistic regression. Logistic regression is a statistical model used to predict outcomes by analyzing the relationships between one or more independent variables. The outcome variable is binary, representing diabetic and non-diabetic conditions. The input variables used in the model included the number of pregnancies, glucose levels, blood pressure, skin thickness, insulin level, body mass index (BMI), diabetes pedigree function, and age. The data, which included medical information and laboratory analysis, was acquired from the National Institute of Diabetes and Digestive and Kidney diseases. Bharath’s model then uses various visualization methods to identify data patterns, outliers, and correlations between variables. 

Bharath’s model was evaluated using a confusion matrix (similar to a report card showing not only how many answers you got right or wrong but also how you got them right or wrong), and achieved an impressive accuracy of 81.57%. His model allows for early identification of at-risk individuals which allows for early preventive measures to be taken. This helps to avert the development of diabetes and its associated complications. With diabetes expected to increase to 14% of the adult population by 2030, Bharath’s predictive model is instrumental in combating the disease. 

The future in fighting disease

Using predictive data analytics as demonstrated by Bharath isn’t just a potential option for addressing chronic diseases like diabetes but a necessity. By leveraging data driven insights, healthcare organizations can improve the quality of care, improve patient outcomes, reduce prevalence of these diseases, and reduce healthcare costs. 

Moving forward, Bharath’s desire to contribute further to the healthcare industry in combating chronic diseases is pushing him to work with advancements in Artificial Intelligence and Machine Learning. As the technologies he uses improve, Bharath will only get better at solving the challenges that our healthcare world needs.


Leave a Reply

Your email address will not be published. Required fields are marked *

Copyright © All rights reserved. | Newsphere by AF themes.