Diabetes The increasing prevalence of diabetes in the 21st century is a problem. Patients have symptoms like unusual thirst frequent urination extreme fatigue Diabetes can also lead to more serious complications like stroke, blindness, loss of limbs, kidney failure, and even heart attack.
Discovery of Insulin In the 1920s, insulin was discovered by Frederick Banting. Most of the food we eat is turned to glucose, or sugar, for our bodies to use for energy. The pancreas, an organ near the stomach, makes a hormone called insulin, to help glucose get into the cells of our bodies. When you have diabetes, your body either doesn't make enough insulin or can't use its own insulin as well as it should. And this causes sugars to build-up in the blood. With Banting's discovery of insulin, pharmaceutical companies began large-scale production of insulin.
Although it doesn't cure diabetes, it's one of the biggest discoveries in medicine. When it came, it was like a miracle. Challenges with Insulin The default method of administration is by a needle, multiple times a day. Insulin pumps are a more recent invention. These are insulin delivering devices that are semi-permanently connected to a diabetic's body. The Future: Oral Insulin? Wouldn't it be great if diabetics could take insulin orally? This is an active area of research, but historically the roadblock is getting insulin through the stomach's thick lining.
Auralin and Novodra Trials We will be looking at the phase two clinical trial data of 350 patients for a new innovative oral insulin called Auralin - a proprietary capsule that can solve this stomach lining problem. Phase two trials are intended to: Test the efficacy and the dose response of a drug Identify adverse reactions In this trial, half of the patients are being treated with Auralin, and the other 175 being treated with a popular injectable insulin called Novodra. By comparing key metrics between these two drugs, we can determine if Auralin is effective.
Healthcare data is notorious for its errors and disorganization, and its clinical trial data is no exception. For example, human errors during the patient registration process mean we can have duplicate data missing data inaccurate data You're going to take the first step in fixing these issues by assessing this data sets quality and tidiness, and then cleaning all of these issues using Python and Pandas. Our goal is to create a trustworthy analysis.
This Data Isn't "Real" The Auralin and Novodra are not real insulin products. This clinical trial data was fabricated for the sake of this Project. When assessing this data, the issues that you'll detect (and later clean) are meant to simulate real-world data quality and tidiness issues. That said: This dataset was constructed with the consult of real doctors to ensure plausibility. This clinical trial data for an alternative insulin was inspired and closely mimics this real clinical trial for an inhaled insulin called Afrezza.
-
The data quality issues in this dataset mimic real, common data quality issues in healthcare data. These issues impact the quality of care, patient registration, and revenue. The patients in this dataset were created using this fake name generator and do not include real names, addresses, phone numbers, emails, etc.
-
image is preview of the dataset that is intended to motivate. If you're not comfortable with the meanings of each column in each table.