The Vanishing Art of Medical Underwriting

The quote “Underwriting is more of an art than a science” has been attributed to so many people that it could fill Bartlett's Book of Familiar Quotations many times over. However, in this current time and place, the opposite is more likely true. More than ever, underwriting has become primarily a science, and the fields of predictive analytics and predictive modeling along with the accumulation of big data has moved to the forefront of the entire process. The art of underwriting is getting lost in its wake.

Perhaps the key to understanding much of the mechanics of medical underwriting is to understand what constitutes the art and the science of the discipline. Certainly, one can't exist without the other if underwriting is to be both accurate and predictable, and to say that either process can exist in a vacuum would be a great disservice to both. Medical underwriting is essentially the obtaining and evaluating of medical and health information to make an accurate evaluation of mortality risk. Accurate is a key word here in that no model can successfully predict life expectancy in any one individual precisely but can approximate it much more closely when larger sets of numbers are involved.

To concentrate more on the art of the process in this setting, it first helps to know the evolution in the discipline over the recent years and the reasons for the transformation into a more scientific evaluation. Risk assessment does not consist of solely medical information in a vacuum –quite the opposite. A combination of factors, including income, wealth, avocation, hobbies, history, age, gender, driving record, even domicile (among many others) are combined into as complete a picture of an insurance applicant as can be drawn or inferred.

The medical part of the equation has historically centered around the health history, represented by the attending physician statements of the individual. Doctor's notes, medical exams, blood work, imaging tests and medication lists provide a flow of information representing past as well as current health –a trend as well as a current status report. Supplemental information such as bloodwork, EKGs, drug testing, referral and specialist information, and often a current medical or paramedical examination fill in the background of the entire picture. The information then becomes a composite of where on the risk spectrum an individual lies: from excellent health that may qualify for better than average rates to a series of mortality compromising conditions that require added pricing or even a declination of coverage.

The process over 20-30 years before the end of the century remained rather consistent up to the new millennium, when both the increasing sophistication of technology coupled with the ability and demand of those whom the industry was targeting as buyers made necessary significant changes in the process. The business model of insurance purchasing from a single agent or company became diminished with the ability to compare rates from multiple carriers on a spreadsheet literally in seconds. Millennials became targeted in the “go where the money is” scenario, and their wants and needs were different than the industry was historically built for. Ditto for the understanding and traditional ways of doing business.

Such business was now conducted not so much in person but by telephone. More exactly by Smartphone. Virtually anything could be obtained in a flash on the internet that previously required personal time and presentation. Time was money, and the purchase of insurance became no more than a commodity to most, particularly to the more technology savvy, younger age individuals where insurance medical requirements were more minimal. Faster, cheaper, and more convenient were the bywords in a do-it-yourself world.

Insurance purchasing has become yet another part of what might be called the “Amazonization” of today's world (certainly if Amazonization is not a dictionary word, it will be in Webster's in the years ahead). Can it be cheaper, faster to purchase, and obtainable at the click of a mouse with almost instant delivery? Can the internet provide enough information to be able to purchase things without an in-person demonstration or explanation? Can I use bricks and mortar stores, or information obtained from a knowledgeable source and then go to an Amazon like model to purchase it more cheaply, at the time and expense of anyone who has personally given me the information to do so?

Hand in hand with the result of the purchase is the facility, ease, and convenience with which the product can be obtained. To the future consumer, and in particular with regard to life insurance, it means in the now, or as close to now as possible. How soon can I get my approval? How fast can I get the policy in my hands? Do I need to give blood, do any testing, submit to any kind of an examination, or even meet with anyone if the illustration can be emailed to me? To what degree can information I provide to you give you enough assurance that you can be quick, accurate, and perhaps as important as anything, cheaper than the company around the corner?

So it is with this in mind that it becomes much more obvious where the science in the process displaces the art. If I get enough information from a less intensive process, can I price it so that I still make a profit? If I flip a coin enough times, when I get into the millions, isn't the percentage going to even out at 50-50 anyway, especially the larger the database gets? Don't I just need enough data to be the cheapest and most efficient on the block, especially if I am sensitive to demographic changes and scientific advances? The answer may be yes, but in fact the process may play out not quite as expected as the loss of the art evolves.

Attending Physician statements have been the cornerstone of insurance underwriting. They provide an accurate chronology of an individual's health, and a moving timeline of the advancement of any health or disease encountered over time. The science of underwriting is evolving toward whether we need most of these to make a mortality assessment, how many can be bypassed, and can a medication list replace them; and as such, the process becomes cheaper (lowered acquisition costs) and faster (no wait for their receipt). However, the art of knowing what's in those statements is historically what has made medical underwriting more accurate and able to distinguish amongst the risks that will perform better from a mortality standpoint over time.

Doctors put a lot more into their notes than a lay reviewer gathers from their literal interpretation. The wording and inflections tell you whether you have a compliant patient or not, one where the trend of care is your friend rather than a downward progression, and the doctor's assessment of how he or she expects the next step to go. I have been out of active practice as an internist and endocrinologist longer than I care to admit, but I could look over my old notes and tell you just from how they were written about my evaluation and prognosis in a medical case. This is lost when a physician/medical director or a well-trained underwriter does not see the actual progress note and doesn't get to put it in perspective of the whole story that is being told. “Lost in translation” is probably the most apt description –many companies use third party summarization services based in foreign countries where the facts are there, but the underlying meaning isn't. I took Spanish for many years in school and am fluent enough, yet my translation of conversations and events often fall far short of what a trained or native Spanish speaker gets out of them.

The use of underwriting manuals has become a tried and true means of accumulating data on impairments over time to estimate mortality of a disease entity. Enough people with disease “X” at stage “Y” treated over time allows a good estimate over large numbers of a given population. Yet, no disease ever exists with a consistent outcome. Everyone who has the same category of disease at a similar stage after a certain time does not live the identical amount of years. Other concurrent diseases may have an impact or play a role, and it is impossible for a manual to cover every set of circumstances. Maybe it covers disease A with situation B, but they are not the only health situations coming into play. The art of underwriting is to use clinical experience and data interpretation to combine all risk factors into a more comprehensive and more accurate assessment of mortality risk. Perhaps 10,000 cases would bear out the generalization a manual suggests, but would an insurer be better prepared and more competitive if it was able to reward the more favorable cases with pricing enhancements and separate more dire circumstances and more accurately price for them? The art of underwriting is almost a necessity in these circumstances.

Lowered acquisition costs certainly enable companies to pass those savings to customers, but is cheaper always better? A trusted line in any purchase is that “you get what you pay for.” Is getting added bloodwork a headache when the art of underwriting allows the insurer to find better cases and avoid salient risks? Does a blood test like BNP (certainly easier to obtain than an EKG as it is reflexed off an already drawn sample) replace that EKG in terms of predictive value – does it tell you about arrhythmia, conduction defect, old heart attack, rate and axis, hypertrophy, etc? Is passing on an exam or test justifiable because over innumerable cases the odds and probabilities will average out? The art in handling these circumstances is in evaluating what makes each case different. Perhaps it does make a difference in insuring the better client whose premium payment will persist far longer over a greater lifespan or excluding one whose placement today will only result in a claim tomorrow.

A lot of advancements in today's underwriting have to do with a healthy life style on the part of an insured, and how much it will play into future longevity. Do you eat fruits and grains? Do your wearables illustrate a pattern of exercise and fitness? Can you run a 9-minute mile? But mostly, do I trust these things are facts because you say they are? Gathering this information in actuality can be as labor intensive as any traditional underwriting requirement to obtain. Taking this information on faith will go nowhere in improving the accuracy of the underwriting process. Trust a doctor's note. The information a doctor gets in the face-to-face interaction with a vulnerable and trusted relationship goes a lot farther than what is written on an application's form. It is truly an art to read them informatively.

The art of underwriting also goes hand in hand with relationship building and purchasing. Given the ability to spreadsheet anything based on age, gender, and amount, years of coverage, and their assumptions on what the value of money will be sometime in the future, insurers compete on product enhancements, banding, pricing, guarantees, and their assumptions on what they feel the value of money will be in the future. Medical underwriting incorporates a relationship between insurer and agent/broker where personal knowledge goes a long way in building a committed client. Most newer underwriting models and outcomes have begun to build in a “no appeals” type of issue for a product –similar perhaps to “bottom line pricing” in automobile purchase. In commodity purchase though, a relationship can make a big difference in obtaining a successful outcome.

There are many cases where medical facts, dates, and even a doctor's observations about a patient can go a long way in helping to paint a better picture of risk in an applicant. In the art of underwriting, the medical director or doctor goes the extra mile to find out if any of those factors exist, and how much they can benefit an applicant's assessment. The broker or agent is key to facilitating this understanding, and when there is a relationship between underwriter/doctor and field force more pieces of a puzzle can be obtained. Additionally, it builds a stronger client in the producer, who has incentive to do business where he feels he is heard and makes a difference. The alternative is the eventuality that there are fewer insurers offering standard products with almost identical features because there is nothing to differentiate them in the process except a catchy logo, and there is no incentive for a producer to take business anywhere else but to where the bottom line prevails.

With the advent of more mortality experience over larger blocks of time, and the ability to model more variables into designing both saleable and practical products, the science of underwriting is not only here to stay, but likely to expand in scope. Underwriting today is based on convenience to the consumer, ease of application and delivery, lowered costs in a world of competitive pricing, and companies focusing their resources toward those most likely to purchase its products. Insurers are looking to use an increasing amount of data that is now easier than ever to access to help make decisions. In this context, the individuality, clinical experience, direct evaluation of the medical records, relationship building, intuition, and the time it takes to make these happen on the part of the medical director and the underwriter are falling victim to the new reality of the purchase of insurance. The art of medical underwriting may never be the same.