Balabodhi Y

Business Impact Analysis using SQL Server and data visualising on PowerBI.

• Develop, deploy and manage PowerBI, handling over 20 million Clinical data and 10 million Patient data (demographics) for revenue analysis, KPIs and business decisions, resulting in a 100% increase in usage by analytics teams and stakeholders
• Designed the dashboards and data products to effectively communicate the results of data analysis for a clear understanding of the analysis.
• Revenue Growth: Achieved 5-10% monthly revenue growth and improved data accuracy by 40%.
• Real-time data update of the Dashboards through backend automation of the analysis store procedure.
• Improved Accuracy: Improved data accuracy and reduced data processing time by nearly 40% by optimizing and automating data extraction, cleansing, and analysis processes.
• Nearly a 5 to 10% increase in overall revenue monthly than the normal increase rate, by monthly analysis of clinical and patient visits.

Turn-Around-Time (TAT) Analysis:

• Initiative: Independently proposed, developed, and deployed a TAT analysis solution.
• Data Analysis: Identified data errors and discrepancies affecting TAT.
• Performance Improvement: Reduced average TAT by 10%, and achieved 90% accuracy in identifying issues.

Data analytics for patient retention.

• Data Understanding: Effective Exploratory Data Analysis on the patient data to uncover hidden patterns, findings of clinical outliers, and the key relationships within the data.

Clustering Analysis:

• Patient Segmentation: Employed K-means clustering to group patients into distinct therapeutic categories based on their characteristics.
• Achieved a silhouette coefficient of 0.85 for patient segmentation, indicating highly cohesive and well-separated clusters.
• Sub-Grouping: Utilized clustering techniques to further segment patients within each therapeutic category, tailoring interventions.

Adherence Analysis:

• Patient Behavior: Calculated patient loyalty, regularity, and adherence to medication and treatment regimens.
• Predictive Modelling: Employed logistic regression (binary or ordinal) to predict adherence probability and identify at-risk patients
• Impact: Achieved a 25% increase in patient retention through targeted adherence interventions based on predictive modeling and clustering analysis.