Problem

The pharmaceutical industry is constantly seeking ways to improve patient care and targeting strategies. One way to achieve this is through patient segmentation, which involves dividing patients into groups based on specific characteristics. However, there are limitations in the data available for patient segmentation, such as a shortage of time dimension and limited sample size.

In September 2023, Purplelab Inc. was tasked with conducting patient segmentation in multiple sclerosis (MS) for the pharmaceutical industry. The team needed to conduct comprehensive research and exploratory data analysis on 30 patients' pharmacy and medical claims data, focusing on ICD-10 (diagnosis, treatments, and patient behaviors) and NDC11 (drug use) for 27 associated data files with the primary key. They also needed to apply advanced data modeling techniques to personalize target patient group segmentation through R and PowerBI.

Solution

To address the limitations in the available data, our team conducted comprehensive research and exploratory data analysis on 30 patients' pharmacy and medical claims data. We focused on ICD-10 (diagnosis, treatments, and patient behaviors) and NDC11 (drug use) for 27 associated data files with the primary key.

We then applied advanced data modeling techniques such as Random Forest, SVM, K-Means clustering, Elbow clustering, silhouette clustering to personalize target patient group segmentation through R and PowerBI. This allowed us to present actionable segmentation insights to industry professionals that contributed to improved pharmaceutical patient care and targeting strategies.

Impact

Improved patient care and targeting strategies
Customized treatment plans for patients
Patient-centric marketing for pharmaceutical companies

"This project had an incredible impact on the pharmaceutical industry, allowing us to better understand patient needs and provide more personalized care."

Advanced MS Patient Segmentation Boosts Care