【律商风险】2024双重视角下的保险业死亡率风险管理白皮书医疗与非医疗数据并用构建全面风险管理体系_19页_9mb

2024 LexisNexis® Risk Solutions Insurance Mortality Risk Management Study Summary

Core Content

This report discusses how combining medical and non-medical data can enhance mortality risk assessment and improve underwriting decisions in the life insurance industry. It highlights the potential for accelerated underwriting by identifying segments of applicants with medical conditions who may still have average or better-than-average mortality risk, thus allowing for more efficient and accurate risk evaluation.

Main Points

• Accelerated underwriting is a method used to process life insurance applications faster and improve the customer experience without compromising risk assessment.
• Traditional underwriting processes often overlook opportunities to automate decisions for certain medical conditions due to limited data integration.
• The 2024 LexisNexis Risk Solutions Insurance Mortality Risk Management Study analyzed 50 million de-identified lives and 1.7+ billion medical claim lines to identify hidden risk segments.
• Standardized Mortality Ratio (SMR) is used to compare mortality risk across populations and conditions, where:
  • 100% SMR = average mortality risk
  • <100% SMR = lower mortality risk
  • >100% SMR = higher mortality risk
• LexisNexis Risk Classifier Score combines non-medical data (e.g., driving history, credit records, public records) to provide a comprehensive risk assessment.

Key Insights

Example 1: Type 2 Diabetes

• 10% of applicants with type 2 diabetes have average or better-than-average mortality risk.
• These individuals could be good candidates for accelerated underwriting.
• 30% of the population have twice or more the average mortality risk, warranting manual review.

Example 2: Asthma

• 60% of applicants with asthma have average or better-than-average mortality risk.
• 20% have elevated mortality risk and may need closer scrutiny.
• This shows that non-medical data can help segment applicants more effectively.

Example 3: Sleep Apnea

• 50% of applicants with sleep apnea have average or better-than-average mortality risk.
• 20% have higher-than-average mortality risk and may require manual underwriting.
• This highlights the value of combined data analysis in identifying hidden opportunities.

Example 4: Medium-High Medical Risk

• 40% of applicants with medium-high medical risk can be accelerated if non-medical data is integrated.
• 20% still require manual review due to higher risk.
• This demonstrates the importance of simultaneous analysis of both data types to optimize underwriting.

Validation of Combined Model

• The combined model of medical and non-medical data provides more accurate and nuanced risk assessment.
• Shapley Value method is used to determine feature importance, revealing how each attribute contributes to mortality prediction.
• Age, gender, and medical conditions are traditional risk factors, but non-medical data (e.g., driving records, credit) adds new predictive power.
• Non-medical data is especially valuable for applicants under 60, who may lack extensive medical history.

Conclusion

• By integrating medical and non-medical data, life insurance carriers can better segment applicants, improve decision-making, and mitigate mortality slippage.
• This approach enables more informed underwriting and differentiated customer experiences.
• The LexisNexis Risk Classifier Score helps identify both high and low-risk segments, which is crucial for automating decisions and optimizing resources.

Call to Action

• Consider how your organization can incorporate medical data from EHRs into your risk assessment workflow.
• Explore the potential to expand accelerated underwriting by combining medical and non-medical data.
• Share this report with stakeholders to advocate for data integration and improve underwriting efficiency.

About the Authors

• Matthew Stull: Director of Data Science at LexisNexis Risk Solutions, with over 15 years of experience in predictive analytics and model development.
• Jason Cai: Senior Data Scientist with expertise in life, health, and property/casualty insurance, holding a Ph.D. in Operations Research.
• Xueping Meng: Senior Data Scientist with a background in predictive modeling, leveraging medical, credit, public records, and motor vehicle data.
• Lingping Liu: Senior Data Scientist with a focus on healthcare data and predictive model development in the life insurance industry.

Sources

1. Munich Re. "A Decade Into Accelerated Underwriting: The new normal for advanced risk selection." Accessed April 16, 2024.
2. McLaughlin, K. and Gardner, V. "What has accelerated underwriting done for life insurers?" InsuranceNewsNet, October 23, 2023.
3. National Center for Health Statistics. "Leading Causes of Death." Accessed April 16, 2024.

About LexisNexis® Risk Solutions

LexisNexis® Risk Solutions is a global provider of information-based analytics and decision tools for the insurance industry and other sectors. Based in Atlanta, Georgia, it is part of RELX, a global leader in professional and business information solutions.