Data Analyst Project For Beginner : Phase2 Clinic Trial

Introduction

In the realm of healthcare and pharmaceuticals, the analysis of clinical trial data plays a pivotal role in evaluating new treatments and advancing medical knowledge. This article delves into a comprehensive project focused on the data wrangling and analysis of a Phase II clinical trial dataset for Auralin, an oral alternative to the injectable insulin Novodra.

Understanding the Project

Overview

The project centers on cleaning and analyzing complex clinical trial data related to Auralin and Novodra. This involves examining patient demographics, treatment specifics, and adverse reactions recorded during the trial. Such data, resembling real-world medical complexities, necessitates meticulous handling to derive meaningful insights.

A/B Testing in Clinical Trials

A/B testing, a statistical method for comparing two versions of a variable, is employed to assess the efficacy and safety of Auralin versus Novodra in this clinical trial setting.

Parameters Analyzed

1. Patients Table: Contains demographic and health information of patients.
   • patient_id, assigned_sex, given_name, surname, address, city, state, zip_code, country, contact, birthdate, weight, height, bmi
2. Treatments Table: Details of the treatments given to patients.
   • given_name, surname, auralin, novodra
3. Adverse Reactions Table: Records of side effects experienced by patients.
   • patient_id, reaction, severity

Hypotheses to Test

The project explores two key hypotheses:

• Efficacy Hypothesis: Auralin effectively controls HbA1c levels in diabetic patients more than Novodra.
• Safety Hypothesis: Auralin demonstrates fewer or less severe adverse reactions compared to Novodra.

Methodological Approach

Preliminary Insights using Google Sheets

• Data Import: Importing the raw dataset into Google Sheets for initial exploration.
• Exploratory Analysis: Conducting preliminary data exploration to identify basic statistics, trends, and potential data quality issues such as missing values or outliers.

Data Cleaning and Analysis with Python

• Python Libraries: Leveraging pandas and NumPy for data cleaning and transformation tasks.
• Data Cleaning Steps: Addressing missing data points, handling duplicates, and ensuring consistency in data types.
• Exploratory Data Analysis (EDA): Utilizing statistical and visualization tools to uncover patterns in the dataset.
• Statistical Testing: Performing t-tests to compare HbA1c level control between Auralin and Novodra, and Chi-square tests to assess differences in adverse reaction rates.

Visualization and Reporting with Power BI

• SQL Database Integration: Loading cleaned data into an SQL database for structured querying.
• Power BI Dashboards: Creating interactive visualizations to:
  • Display demographic profiles of patients.
  • Compare changes in HbA1c levels pre- and post-treatment.
  • Illustrate incidences and severity of adverse reactions.
• Insight Communication: Using Power BI to present findings effectively, supporting conclusions drawn from the data analysis.

Conclusion

Analyzing Phase II clinical trial data for Auralin and Novodra exemplifies the intersection of data science with healthcare innovation. By meticulously cleaning, analyzing, and visualizing complex medical data, this project aims to unearth crucial insights that could potentially influence future treatment strategies and patient care protocols. Through the structured application of data analytics tools and methodologies, this endeavor underscores the importance of data-driven decision-making in advancing medical research and improving patient outcomes.

Frequently Asked Questions