Health Triggers and Preventive Medication

Our body send signals related to some of the challenging diseases well before the situation become critical and quite often we ignore them.

As mentioned in the article on “These 4 Things Happen Right Before a Heart Attack” 1, body send some signals and some of these signals are related to Heart Attack.

Data Science and Analytics can help in linking the triggers which are strongly related to the diseases.  And educating patients and customers about these signals can save lives. Some of the key activities are first defining the events or triggers.  Once we define the events or triggers, then important things would be to link them with the different medical complications.   Some of the statistical techniques and analytical frameworks can help in finding out the strength of relationship.  The event or trigger based insights can help  the medical practitioners to recommend require investigations or actions to the patients and reduce the health risk for the patients.

One of the other angle could potentially be identified would be to estimate time between an event and serious complications

If non significant events are used for recommendations and the patients would not take next recommendation (may be serious one)  positively and also inaccurate triggers will cause waste of money & time for medical practitioners.

The advantage of identifying the triggers are

  • Understanding the factors which are linked to a disease and why?
  • Taking measures for Preventive Medication for the clients/patients
  • Recommending life style and other changes for the clients
  • Reducing medical expenses and  complications for the clients

Reference

  1. http://www.newsmaxhealth.com/MKTNewsIntl/heart-attack-four-things/2013/08/06/id/518985#ixzz2x2yjanCC

Relevance of Credit Scoring and Analytics

The banks and financial institutions have moved away from manual credit authorization and credit approval processes along back. The reasons of such migration are

  • Standardisation: Adopting a standard process for all the applicants instead of local branch manager taking decision which can have certain bias and discriminations.
  • Reduction of Risk: Measuring and reducing is one of the important competitive advantages for financial institutions. The centralised decision has helped in measuring and managing credit risk not only at the time of acquisition but across customer life cycle.
  • Technology: Due to advancement of technology and adoption of data warehouse based systems facilitated the banks and financial services to migrate to centralised & automated credit approval process.
  • Regulations: Basel and other regulation has played a role in banks and financial institutions adopting prudent approach of credit approval.
  • Data Driven Intelligence: Due to regulations, technology and data availability, the financial institutions have leveraged Data Driven Intelligence for credit approval process. This intelligence is called Credit Scoring which is built using statistical and machine learning techniques.

A person loan, credit card or Mortgage loan can rejected for various reasons but one of the main reasons is Bad Credit Score.

Credit Score for a person is calculated based on applicants person characteristics (Demographic Data), Payment History and Financial Conditions (such as Debt level, # of Credit Cards etc) Data.

Increased number of Banks and Financial Institutions are employing external data such as social media and web data.

These additional sources of data help them in improving accuracy of applicant risk level. Example, job losses  for employees from a company could lead to increase in risk level for the applicant from the company. The financial institution can be more prudent in underwriting credit to the applicants. The points to note is that not all applicants from the company will default and this could be one of the factor in credit approval.