How to Create World-Class Teleinterview Questionnaires
Hank George, FALU, CLU, FLMI
June, 2007 • Teleunderwrtiting Essay 5
“Garbage In, Garbage Out”
As I said in earlier essays, this sharp phrase is the most important concept in all of teleunderwriting.
If your teleinterviewers aren’t doing their job, their output won’t allow your underwriters to take as many actions as possible without getting M.D. records they really don’t need.
The same goes for your teleinterview questionnaires.
If you don’t ask the right questions, then no matter how well the callers do their job, the net result will be fewer rapid actions, more medical records, unnecessary delays, added costs, and so on.
Two Basic Kinds of Questionnaires
Since the teleinterview process is designed to replace the traditional gathering of risk-related information from the proposed insured, you need to have 2 distinct types of questionnaires.
The first one reprises the risk questions on the Parts I + II of your insurance application.
This is the easy one to design, because it simply asks for a YES/NO answer. The applicant either does or does not use tobacco, did or did not have hepatitis and so on.
The main consideration here is to get away from the notorious lists of impairments found on traditional applications, as in:
G. diseases or disorders of the heart, high blood pressure, chest pain…etc.
Virtually every mortality and morbidity insurance application we have seen asks about long lists of impairments related more or less to a given organ or bodily system.
The problem with reading lists is that it can confuse the informant. By the time you get to the sixth item, he may have forgotten the first one. He may have three YES answers among the eight you rattle off and become confused. Or, he may not understand the meaning of one of the terms and gloss over this as he listens to the whole list.
Ipso facto: inadvertent nondisclosure, the mechanism underlying 9 out of 10 episodes of nondisclosure and often mistaken later for antiselection (intentional nondisclosure).
It is our view that the baseline teleinterview questionnaire designed to recount the Part I + II risk questions should ask about each impairment separately, or, if that is too lengthy a process to tolerate, then certainly no more than two closely- related conditions in each question asked by the interviewer.
This gives the interviewee a chance to ask for amplification as to the meaning of the term or intention of the question, thus forestalling inadvertent disclosure of misinformation.
Rather than repeat “In the last 5 years…” with every condition, you might begin with: “I am going to ask you about a series of medical conditions. In each case, please tell me if you have had or been told you have had this condition in the last 5 years” and then begin the list. You might reiterate this phrase when you resume reading the list after completing an impairment-specific drilldown for any ensuing YES answer; just to make sure the listener understands where you are at in the core interview process.
Writing this basic interview is not very difficult because you are tethered, so to speak, to asking all of the risk questions on the application in the order they appear and without adding any elements to the process. The more challenging part of this exercise comes next: the drilldown questionnaires.
Drilldown Questionnaires: Preliminary Comments
Since we’ve been in business (May, 2003), we’ve designed four sets of customized drilldown questionnaires for specific companies. We’ve also edited several sets initially drafted by companies, making our recommendations. And we’ve fashioned a comprehensive set of 90+ generic drilldown questionnaires which have been acquired at an incredibly low cost by at least 10 insurers.
I mention the foregoing to assure you that we have considerable experience in this undertaking.
In the course of our work as consultants and educators, we’ve seen a fair number of examples of drilldown questionnaires designed in-house by insurers. In some cases, this laborious process was said to have taken as long as a year, involving veteran underwriters, medical officers and sometimes persons with other expertise.
We estimate the total cost (mainly high-salaried labor) of designing a set of, say, 75-95 original questionnaires (yes, you may very well need THAT many) in the range of $100,000-150,000 US. If anything, this is a conservative estimate, based on what we have seen and been told anecdotally at study group meetings.
How would we characterize the quality of drilldown sets that have been concocted in-house?
There have been a few that were very impressive.
A few…
In one case, we were asked to review a set of over 100 questionnaires prepared by a rather veteran team of M.D.s and underwriters at a major US company. We recommended merely 360 changes and we were told most of them were ultimately embraced.
We have also seen some gosh-awful sets. It would be compassionate to describe the phraseology of some of them as “wooden.” They almost begged for a cryptic response (the antithesis, of course, of the intent of the process!).
I recall one set which was haphazardly extracted directly from language used on hard copy questionnaires (which, for decades, had been completed by applicants in long-hand in the presence of agents). Suffice to say it was impossible to find any basis for complimenting the authors of these miserable documents!
Garbage In, Garbage Out
You cannot go into this critically important undertaking with the intention of “whipping something out by Friday.”
That is, not unless you’ve got a cozy rock to hide under when your much-ballyhooed teleunderwriting project comes crashing down around your ears!
Lousy drilldowns are one of the leading causes of teleinterview program implosion!
How about the drilldown questionnaires you can sometimes get for “free” from outsourced teleinterview providers?
We have seen a fair number of these. Some are clearly superior to lesser sets “crafted” in-house. However, none can stand toe to toe with the cream of the crop.
But, then, you’d hardly expect service firms to expend an effort equivalent to six figures of direct labor cost to develop a resource which they have every reasonable expectation the customer will provide!
No two sets of risk history application questions we have seen are 100% alike. Thus, no one set of generic drilldowns will be spot on target as regards any given insurers’ needs. Moreover, some insurers want minimalist drilldown questioning (due to the ominously mistaken belief that the primary concern is call duration), whereas others will covet more in-depth questioning.
You can opt to use what, if anything, you get for free. Just remember what “for free” means in our socioeconomic system!
Have we made a case for designing your own (or, if you are so disposed, using ours, which are conveniently formatted for customization by the user)?
If we have, then read on and we will share our approach to designing “world class” drilldowns that gather as much salient information as possible, as quickly as possible.
The Key Elements of Most Drilldowns
Why only “most”?
Because you need fewer questions to amplify some application risk questionnaires than others!
How much do you REALLY need to know when someone says he is HIV-positive or he has been diagnosed with amyotrophic lateral sclerosis?
Why ask for more than you need to take the inevitable (for most carriers, anyway) action?
Most drilldown questionnaires for medical impairments will contain, more or less, the same core elements. In addition to these, there will be some impairment-specific questions embedded in some but not all of the question sequences.
A predictable example of the specific questioning occurs with diabetes. Obviously, you are going to want to inquire about microvascular disease, how often blood sugar is checked and the most recent hemoglobin A1-c (you might be surprised how often they know the number…and you’ll never know if they know unless you ask for the information, right?).
Here is a list of the core elements, some or all of which one might cover in a garden-variety medical impairment drilldown:
- When did the condition first occur?
- What were the manifestations that caused the person to seek medical care?
- What tests did the doctor do?
- What was the proposed insured told of the results of these tests?
- What was he told was the cause of his symptoms?
- What medication was given (and more if any is acknowledged)?
- What other forms of treatment, if any?
- Was the individual hospitalized or did he go to an emergency room (and there is still more to ask in either case if the answer is YES)?
- Is the person using any alternative or complimentary remedies…and if so, which ones?
- When was the last time the person had symptoms and what were those Sx?
- How does the condition impact the proposed insured’s daily activities…with modifications of language depending on the impairment?
- Has the person been advised to have any tests that have not been done (with elaboration if YES)?
- Has he been told to have any form of treatment not yet undertaken or completed (with elaboration if YES)?
- Is there anything else the proposed insured would like to tell us about the condition at hand?
14 issues to address (plus answer-dependent further drilling down of some).
Not all that onerous, eh?
We like to hear from the insured about the “manifestations” (symptoms) they first experienced. This is critically important in trying to frame both the likely diagnosis and the severity of the disorder (if, indeed, there IS a disorder).
Test information is crucial. Knowing the tests chosen by a primary care physician (even more so a specialist) is a huge clue as to the differential diagnosis (list of possible causes, in order of probability, by the physician, which guides him in choosing which tests to do and in what sequence).
Notice we didn’t ask for a diagnosis. We prefer to ask “what did your doctor say was the cause”, as this has been shown to put the interviewee more at ease. They don’t have to worry about getting a complicated medical term correct.
How an individual is treated is the #1 clue to both diagnosis and impairment severity.
If the proposed insured is taking medication, we push him to grab the pill bottle(s) and read us the drug names. We also want the dosage and times taken per day.
Then, we ask about whether Rx has been changed, or new drugs added, or a drug removed from his treatment regimen, over the last year or two.
Knowing the history of his treatment, at least in the relatively recent past, can be a virtual roadmap to pinning down the success of treatment and whether the impairment has improved, stayed the same or gotten worse.
The whole reason our company created Hank’s Underwriting Guide to Rx was to make the process of analyzing the implications of medical therapy as easy as possible. Most underwriters weren’t pharmacists in a prior life! The better you understand all of the possible ramifications of the Rx, the better your analysis of the risk will be.
In July we will get into the matter of using treatment information disclosed on teleinterviews – medications, OTC remedies and other interventions like surgery – and how they are the most important clues of all to insurability.
Hospitalization can be inevitable (hip fracture) or a huge red flag (depression). Once we get a YES, we want certain key details including in-patient testing, treatment, involvement of specialists and the like.
By the way, we also add a question about being referred to a specialist on some drilldowns but not others. The determining factor is the nature of the impairment.
Most veteran underwriters know that visits to an emergency room are often even more revealing, in terms of nature and severity of impairment, than in-patient episodes. Thus, we ask about this whenever the impairment in question might be more clearly understood if an E.R. visit occurred.
Why do we ask about alternative and complementary medicine? Because we seem to know what too many underwriters don’t…that A&C remedies are HUGELY important at a time when the upper 1/3rd of the socioeconomic spectrum spends more out-of-pocket money on this stuff than they do on conventional care!
Certain herbs, for example, are RED FLAGS for specific impairments. Many times, their use can point us to one probable diagnosis instead of another. The same can be said for other interventions such as chiropractic, massage therapy, acupuncture and so on.
Leave it off if you wish. It is your mistake to make, and you (sadly) won’t be alone in making it.
Needless to say, the way in which any impairment impacts all aspects of the patient’s life tells us a great deal about morbidity implications and, indirectly, also gives us many clues as to diagnosis and disease severity. Let’s say a person with chest pain is told not to ever shovel snow again, or a person with “mild memory problems” is advised to discontinue driving. See any implications in these scenarios?
Few YES answers are more revealing than when we ask about future tests or treatments. If they say YES, we ask what; then, ask if it is scheduled. If YES, when? If NO, why not? Short, sweet and oh, so revealing in a few cases per hundred.
Lastly, we always give the proposed insured a chance to “say their piece.”
If you’ve never considered a question like this, you’ll be surprised at the results. We’ve seen cases where the most important information on the whole drilldown was shared in answer to this question.
What Else Matters on Drilldown Questionnaires?
The tone of the questions…we prefer soft, not harsh. We like “what did your doctor say…” over “what was…” and so on.
The whole idea is to make the informant as comfortable as possible.
When the impairment is one that is especially sensitive (cancer, psychiatric, substance abuse, sexual), we often say “your condition” or something equally gentle, shunning any word we can may trigger emotion or embarrassment.
Emotion and embarrassment fuel nondisclosure.
Use the teleinterview to accomplish some things you cannot get done with your application as it is presently configured.
Such as?
Well, how about building in enough questioning to get a reasonable approximation of “pack years” of cigarette smoking?
Current smoking pales by comparison with pack- years of consumption. But most companies settle for the lesser answer. Why? Beats me…
Want to get more radical?
When they say they quit smoking after 30 years, ask’em why. Even more so when they say they threw out their last empty Jack Daniels bottle in January, 2005!
People don’t wake up one morning, light up and say “egad, what a filthy habit!” There is a reason behind the decision to quit. Maybe it’s a benign one. Maybe it was an episode of hemoptysis.
If this is true for smoking it is doubly true for alcohol use.
No, make that quadruply…because people don’t just stop a lifetime of using alcohol because some doctor said drinking is bad for you (the way they should to folks who smoke!).
That is, unless the applicant’s drinking was Drinking with a capital “D” (if you get my meaning!).
Hopefully, this essay has given you some perspectives on how to approach either designing your questionnaires or making timely edits to the set you already use. Just keep in mind how important your teleinterview questions are. They lay the foundation for how you underwrite the case. They are the fulcrum on which the YES/ NO decision about securing medical records is balanced.
GARBAGE IN, GARBAGE OUT
With the best possible teleinterview questionnaires, you can make it:
QUALITY IN = LOWER COSTS, FEWER REQUIREMENTS, FASTER APPROVAL, HAPPIER PRODUCERS, HAPPIER CLEINTS = SUCCESS
DISCLAIMER This essay was written for informational purposes only. Hank George and Hank George, Inc. do not recommend or endorse any specific business practice or procedure discussed herein. All business considerations concerning matters covered herein should undergo proper and sufficient scrutiny by appropriate management personnel of the companies involved prior to implementation on any basis. © 2009-10 Hank George, Inc.
