Abstract
Today’s competitive underwriting environment requires faster decisions with less friction. Contrasting this is the need to place risks in the appropriate risk and premium categories. Underwriting requirements have shifted due to these competing priorities, and underwriting decisions are made now faster than ever before. To ensure accurate underwriting decisions, verifying an applicant’s personal identifiable information and criminal records can improve mortality and morbidity results. These can be done quickly and in an automated environment for more confident, faster decisions.
Proper underwriting classification is a must
Effective underwriting requires selecting acceptable risks and then ensuring that those risks are assigned to the appropriate risk cohorts. This requires the underwriter (or computer system) to review the information submitted to the company by way of an application and then apply established rules that then determine whether the applicant is acceptable, and if so, the premium classification to be assigned. The rules attempt to match like-risks so that an accurate mortality or morbidity rate, usually expressed as the number of cases expected per 100,000 exposures, can be applied so that claims can be predicted and appropriate premiums for that risk cohort charged. Misclassification of accepted risks can have significant and long-term financial ramifications.
Most people are honest, some are careless, others lie
In an ideal world, the information in the application is accurate, and the company can rely on the information provided. Unfortunately, the reality is that some people are careless when completing an application and provide inaccurate information. Others may intentionally fail to disclose or misrepresent information that could adversely affect the insurer’s decision to issue coverage (or the amount to charge for the coverage). Either way, the insurer is at risk of understating the rate of claims to be expected. Compounding this further, insurers are usually limited to two years from the application date or policy effective date to find materially misrepresented information in the absence of fraud.[1]
This can easily result in tens of millions of dollars being paid out in claims that were otherwise unwarranted or unexpected. Additionally, insurers are constantly being tested by fraudsters to determine which companies have gaps that may be exploited. Once a gap is found within a company’s underwriting controls and processes, that company risks being targeted for higher fraudulent activity. This makes having effective and efficient controls to find “misinformation” in place essential.
Validating data must be quick and frictionless
Speed-to-decision and straight-through-processing are priorities competing with the need to verifying the information on an application. In today’s environment, consumers expect fast decisions, and insurers who are perceived as taking too long or creating too much friction risk losing a sale or worse – an agent.
Any third-party data requests that rely on the insured’s name, date of birth, or Tax ID number (TIN) to find information may miss significant records and result in a false negative finding. Without these hidden records, the insurer may conclude that the insured is an acceptable risk when, in fact, the applicant may not be. The key to accurate frictionless and straight-through processing is validating the applicant’s personally identifiable information (“PII”) up front and automation.
Often, misrepresentations of the PII appear to be typographical errors or “finger fumbles;” however, applicants may intentionally make it appear as such to hide information. Simple “mistakes” like variations in the spelling of a first or last name, a difference of a day on a date of birth, or a digit off on a TIN may appear to be innocent mistakes, but the impact can be significant. For any HIPAA-compliant data request, the identifiers usually must match exactly. For example, the spelling of the name “Jackie” v. “Jacqui” or a date of birth “September 21” v. “September 12” will likely be enough to throw off record requests.
Validating PII is quick and frictionless. If this check is automated, validations can take place within seconds. Mistakes can be clarified, and then requirements ordered. This is a prerequisite with Diligence when ordering any requirement on behalf of our clients, and the results are nearly instantaneous.
In recent years, advances in the coverage of EHRs and the like along with AI to summarize records are enabling insurers to more quickly assess medical information, assuming the PII was accurate to begin with. AI can also incorporate underwriting manuals and treaties to provide insight into the medical information obtained. But there is other information beyond medical information that can significantly impact mortality and morbidity rates.
Automated criminal record searches
Knowledge of criminal records is an important part of underwriting. While most applicants do not have a criminal history, some do. Criminal activity denotes risky behavior especially if the activity involves violence or drugs. Even “non-violent” activities such as so-called “white collar crimes” exhibit someone willing to take risks. Depending on the nature of the criminal activity and the time that has since elapsed, a criminal history could result in a significant rating or declination of coverage.
Studies have shown that former prisoners have a higher mortality rate than the general population thus pose an increased risk for insurers. The American Journal of Public Health published a study in 2012 which concluded:
“We have systematically reviewed research examining the mortality of released prisoners, and found that released prisoners are at substantially increased risk of death from all causes, and from drugs, suicide, and homicide in particular.”[2]
Drug use and mental illness are more common amongst former prisoners. Imprisoned people may also be exposed to HIV transmission more so than the general population without the same access to health resources.[3] Criminals may also exhibit higher mortality even if not incarcerated. Skinner concluded in his thesis on mortality and offenders that there is a correlation between “community offenders,” health, and premature mortality.[4] This stands to reason as the actions of those involved in committing an offense regardless of whether violence was involved are exhibiting risky behavior.
The value of criminal record searches
Automated criminal record searches were previously spotty, not FRCA compliant, and difficult to read. All that has recently changed. Criminal records, which can also include validating an applicant’s PII, is simple, affordable, and can be easily automated as part of an underwriter’s workbench. Diligence has found in conducting automated criminal record searches that nearly 5% of applications contain misrepresentations of applicants’ dates of birth and TINs. For active block reviews, the rate is nearly 10% albeit typically block reviews are pre-screened and already suspected to contain misinformation. Consider the longer-term mortality and morbidity results due to false negatives of data inquiries using incorrect PII. In addition to misrepresented PII, we are finding approximately 1.6% of cases reviewed have undisclosed criminal records.
Consider the mortality and morbidity costs for issuing this volume of business that is in the wrong risk category or perhaps should not have been issued at all. Having better information can lead to improved margins for insurers and perhaps better rates for the honest applicants. Not having this information could result in unexpected financial losses due to higher than desired actual to expected results.
Most criminal record searches are returned within seconds because most people do not have a criminal record to find. If there are no potential records or mismatched PII, the automated search immediately sends a response back to the client, and the client can then proceed with other underwriting tasks. There are cases, however, in which potential records are located that must be researched to determine if the record belongs to the applicant. This can take as little as a few hours or a few days in a minority number of cases depending on the jurisdiction. The results, however, can be impactful to an underwriting determination.
Summary
The key to winning in today’s environment and staying competitive is having accurate information quickly. Speed-to-decision and straight-through processing are part of this equation. To obtain accurate information, an applicant’s PII must be validated before ordering any third-party data. Without this, the results of all other inquiries are suspect. There have been significant advances in recent years regarding the coverage of EHRs, medical information, and AI summarizations and insights. We can also add to this that advances in FCRA-Compliant criminal records checks to ensure that insurers have accurate information to assess risks appropriately.
Diligence provides automated solutions for validating an applicant’s PII, criminal records, and more. We are more than a vendor; rather, we are an information partner to assist clients with assessing the validity of information received – whether it be at time of underwriting, an active blocks of business, or at time of claim. For more information on PII validation, criminal record searches, or Diligence’s broad offering to the insurance industry, contact Kevin Glasgow or Paul Marquez at Diligence International Group. Their contact information is Kevin.Glasgow@DIGroup-US.com, (214) 316-0917, and Paul.Marquez@DIGroup-US.com, (469) 892-4481.
About the author
Kevin Glasgow has been in the insurance industry veteran with over 40 years working with both retail insurers and reinsurers in the United States and Canada. His roles have included leading claim teams in the United States and Canada, and as such, he has worked extensively collaborating with underwriters and others to mitigate fraud risks as well as defending companies against fraudulent activities. He has worked as an expert witness in insurance matters, and he is currently working with Diligence International Group which specializes in fraud mitigation and identification. He is also on the Advisory Board with Friendly, an AI company serving the insurance industry. He holds a bachelor’s and master’s degree in business administration, is six-sigma trained, and is the past president of the International Claim Association and the Eastern Claim Conference. His designations include the ARA, FLMI, FLHC, CLU®, and CFE.
[1] Most state regulations state that the maximum contestability period stated in a policy cannot exceed two years. As of this writing, Missouri’s maximum contestability period is 12 months. Depending on whether intent can be proven and the jurisdiction, defenses may be available for fraud outside of the contestable period.
[2] Zlorde and Fazel, All-Cause External Mortality in Released Prisoners: Systematic Review and Meta Analysis, American Journal of Public Health, December, 2012, https://ajph.aphapublications.org/doi/10.2105/AJPH.2012.300764
[3] Flath, N., White, J.J., Tobin, K. et al. Criminal justice involvement, structural vulnerability and social safety net services among people living with HIV in Baltimore. Crime Law Soc Change 81, 127–141 (2024). https://doi.org/10.1007/s10611-023-10112-y
[4] Skinner, G. (2021). Offending, Physical Health and Premature Mortality: Associations Derived from Longitudinal and Meta-Analytic Evidence [Apollo – University of Cambridge Repository]. https://doi.org/10.17863/CAM.78151