Smart Score®: A Transparent Algorithm to Improve Mortality
Life insurance underwriting is, arguably, the most successful implementation of predictive analytics ever developed. One need only look at the beautiful, enormous buildings and skyscrapers that house many of America’s largest insurers to demonstrate that fact. But even the best algorithm can be improved.
Why use Smart Score®?
The most common method used to determine Preferred class eligibility is the cut-off system, where specific thresholds of BMI, blood pressure and cholesterol values are used to make the decision. For example, a cholesterol of 240 mg/dl may be the knockout threshold between Preferred classes. These thresholds vary from company to company, and likely came about because of the Framingham score which uses many of the same factors and is derived from a large cohort study started in 1948. The knockout system is popular because it works and is clear and easy to understand. It also has its challenges; all underwriters have had unpleasant conversations with field agents about why their client did not get a better class despite being just 1, 2, 5 or 10 pounds over the limit.
The Smart Score® approach is different: it is a risk score like Framingham but based on many more factors. Specifically, CRL has data on more than 15 million applicants, over 500,000 of whom have died since the start of data collection. Furthermore, each individual has up to 50 different laboratory values and body measurements. This makes for a very rich data set to determine which factors are most important and how they balance each other. In studies comparing performance of Smart Score® versus real life knockout systems, Smart Score® comes out ahead every time. This stands to reason because Smart Score® considers about 3 dozen different factors instead of the usual 3 or 4 used by knockout systems.
Of course, this increase in accuracy comes with an increase in complexity and thus conversations with the field may be more difficult. That is why Smart Score® breaks down the individual contribution of each factor to the overall score in a simple debit/credit scale familiar to life insurance professionals. Even in situations where an insurer prefers to continue to use knockouts, Smart Score® is an ideal way to help make those “wiggle” decisions on borderline cases.
Is Smart Score® dependent on age or sex?
In a word, no. Although many laboratory tests vary with age and sex, Smart Score® is corrected so that these differences are minimized. Further, the entire distribution of scores is held constant when updates are made.
Is the evidence of the effectiveness of Smart Score® available?
CRL has published numerous articles in peer-reviewed literature as well as in industry journals which provide detailed evidence of the mortality risk associated with individual laboratory measurements and the overall performance of Smart Score®.
Test Smart Score® Today
If you would like to see what your data looks like with Smart Score®, we can run an example for your business in just a few minutes.
To learn more or start using Smart Score®, contact your sales representative or email ILSsales@crlcorp.com.
Steven J. Rigatti, MD is a consulting medical director with Clinical Reference Laboratory, with 12 years’ experience in the life insurance industry. He is the current chair of the Mortality Committee of the American Academy of Life Insurance Medicine.
Michael Fulks, MD, Consulting Medical Director, is board-certified in internal and insurance medicine. After leaving practice, he served as a medical director, creating or editing several underwriting manuals and preferred programs. For the past 13 years, Dr. Fulks has consulted for CRL, participating in its mortality research on individual tests and all laboratory test results, BP and build in combination. He is also involved in the development and implementation of automated screening tools for non-laboratory data.