Data analysis in insurance business

When we discuss about risk, the focus is generally on the insurer to takes insurance to mitigate their risks. It is imperative that insurers must also protect themselves. In the process the insurers collect variety of data from the personal details to the credit worthiness of their customers. For effective risk management, it is important to analyse the data available with the insurer and also for the auditor who audits an insurance company. The audit risk arises as many insurance policies are based on estimation when underwritten. Data analytics provides the auditor with an insight to identify and assess the risk of each applicant. In this article, we would discuss some data analytics procedures which can be applied while taking up Insurance audit assignments.

Data Analysis across the files

1. Existence of policy in the same period as the claim
2. Checking details between claims and policy
3. New policies experience claims shortly after inception
4. Identification of unadjusted credits in cash deposit account

Premium

Audit objectives

1. Accuracy of premium collected
2. Authorizing of discounts offered
3. Justifying policies cancelled shortly after inception
4. Accuracy of penalty levied in case of repetitive past claim history

Data Analysis for the above objectives

1. Stratify the premium by size and product type
2. Summarize new business by agent or area
3. Summarize debts by agents
4. Exception test: Extract old premium due on policies that have not lapsed
5. Identify policies that were cancelled soon after inception
6. Extract negative debts
7. Identify unallocated cash
8. Extract policies with blank or invalid policy numbers
9. Identify policy premium that have only been part-paid
10. Direct debits that have not been collected

Claims

Audit Objectives

1. Establishing the validity of the claim
2. Validating close proximity claims
3. Justifying operational delays in claim settlement
4. Accuracy of agency commission
5. Settlement of claims on policies underwriting risk of stolen motor vehicle

Data Analysis for the above objectives

1. Analyse by month & type and calculate average values
2. Group claims by product type
3. Matrix analyse date of the incident to date claim reported as part of assessing claims incurred but not recorded.
4. Exception tests: Identify claims with no movement during the last 6 months (some claims get reported but queries on the claim are not responded to leaving old expired claims on the system)
5. Identify negative balances (recoveries)
6. Identify the duplicate claims (same policy number - same policy amount - same date of incident)
7. Check for multiple claims on the same policy

Review of the Fraud areas - Emphasis on the Health Care fraud

Audit objectives

1. Excessive procedure billing of same diagnosis, same procedures
2. Identify excessive number of procedures per day or place of service per day / per patient
3. Identify multiple billings for same procedures, same date of service
4. Analyse for mismatched services to diagnosis codes
5. Review diagnosis fees in excess of masters
6. Review procedure fees in excess of masters
7. Identify and report diagnosis and/or treatment that is inconsistent with patient’s age or gender

Data to be obtained from the client for the above analysis

1. Premium Report
2. Cancelled policy report
3. Cover note report
4. Premium master for various policies
5. Discount master
6. Policy number report for rolling policies
7. Agent commission report
8. Commission master
9. Claim report
10. Incurred loss summary report
11. Incurred loss report - Agent wise