OPIOID DRUG ABUSE
Fatal drug overdoses account for a significant portion of accidental deaths in adolescents and young adults in the United States. The majority of fatal drug overdoses involve opioids, a class of inhibitor molecules that disrupt the transmission of pain signals through the nervous system. Medically, they are used as powerful pain relievers; however, they also produce feelings of euphoria. This makes them highly addictive and prone to abuse. The same mechanism that suppresses pain also suppresses respiration, which can lead to death in the case of an overdose.
STUDY OBJECTIVES
DATABASE DESCRIPTION
This dataset prescriber_info.csv contains summaries of prescription records for 250 common opioid and non-opioid drugs written by 25,000 unique licensed medical professionals in 2014 in the United States for citizens covered under Class D Medicare as well as some metadata about the doctors themselves. This is a small subset of data that was sourced from cms.gov. The full dataset contains almost 24 million prescription instances in long format.
DATA SETS
Prescribers
Summaries of prescription records for 250 common opioid and non-opioid drugs written by 25,000 unique licensed medical professionals.
Economics
Social and economic factors of US States and county, i.e., Total population, Gender, Race, Profession, Per capita income etc.
Opioids
All opioid drugs included in the data and is of dimension 113 observations and 2 columns.
​
​
Education
Information about educational aspects of each States in US, percentage of population undergone/completed different educational levels
DATA SETS
The goal of this experiment is to demonstrate the possibility that predictive analytics with machine learning can be used to predict the likelihood that a given doctor is a significant prescriber of opiates, Predict the Opioid-related death ratio in the US States based on the socio-economic characteristics, Predicting the number of prescriptions based on specialty, credential, gender, etc. Predict which prescribers tended to prescribe opioid drugs, Predicting opioid-related Accidental Deaths in Connecticut State, etc. We did some cleaning of the data, and build a predictive model using some statistical models like, Ridge regression, Lasso Penalized ZIP Model, and logistic regression that predicts with >80% accuracy that an arbitrary entry is a significant prescriber of opioids. We also did some analysis and visualization of my results combined with those pulled from other sources.
GENERAL CONCLUSION
When we consider the socio-economic factors mostly people having high school graduation or Batchelor's degree have more exposure towards opioid drug usage. When we look at people by profession, people engaged in construction work are more likely to use of opioid drugs. Some of the previous studies from New York University, USC News, and other Researchers also stated the same claim.
​
While looking at the Speciality of Doctors, mostly surgeons are tended to prescribe opioid drugs. While considering the accidental deaths associated with drug overdoses, the opioid such as OpiateNOS, Hydromorphone, and Fentanyl have more impact on accidental deaths in Connecticut state. In terms of gender, Male has a slight high chance of accidental death due to drug overdoses compared to female.