Our previous article discussed how risks associated with mental illnesses may not always be what one might think in both magnitude and causes of mortality, especially if environmental factors such as socioeconomic status and support networks are taken into account.
So how can we underwrite these cases better? How can we reach fair decisions quickly and efficiently? Let’s deal with life and critical illness covers first.
For a start, many cases of anxiety and mild depression can be regarded as standard risks, so long as the underwriter is happy that the case is straightforward with no adverse features. Even cases of moderately severe depression that are well controlled by maintenance therapy may be considered as qualifying for standard rates.
However, to make such judgments a certain minimum amount of information is required and, for more complex risks, good information is key to reaching an equitable decision. And underwriters need to look beyond the basics of diagnosis, treatment, response and relapse. They need to probe into those important environmental factors, for example, the persistence of issues that contributed to the illness in the first place and the nature of relationships with family, friends and colleagues which can be strong influences on the course of the illness. Physical co-morbidities too affect prognosis.
The information comprising the bigger picture may be available from a GPR but getting it via that route relies on good supplementary questioning and not just a bit of luck. If we need to know the detail, why not ask the applicant? And a good medium for doing so is the tele-interview. This would not be a mechanical navigation of a series of questions but a sympathetic – but still objective – exploration of the history, from lead-up, to onset and diagnosis, through treatment and subsequent course to current situation. Tele-interviews are not cheap and they do take time to arrange and conduct, but if an insurer is serious about fair underwriting the customer is owed the time and expense involved.
Any argument for more detailed information and the personal nature of a tele-interview raises a question mark about the suitability of automated underwriting for mental health risks. To say that underwriting engines have no place in mental health underwriting would be going too far: not every case is complex or needs a deeper exploration. But how to triage the cases, to identify the ones that ably can be processed by the technology and those that merit special handling? That will require some new thinking by those that write the rules and devise the questions embodying the underwriting philosophy.
In the process of doing that they might consider how an applicant may feel, after disclosing something that to him or her is rather innocuous, such as work stress or bereavement, to be promptly asked ‘Have you ever attempted suicide, taken an overdose, self-harmed or had suicidal thoughts?’ This is heavy-handed and lacking in empathy. The underwriting rules within engines need to explore more carefully what they are dealing with and avoid what might be close to a ‘one size fits all’ approach’.
And thinking of the scope of those questions and the way in which they are phrased, are development underwriters and even their regular medical advisers well enough equipped to pose the right ones in the right order? Do they have the experience? Is their approach based on conventional underwriting wisdom that is past its sell-by date? Maybe they need to work with some specialist medical professionals who can provide expertise and new insights, enabling a quicker and clearer understanding of the issues at stake.
Finally, it may be that underwriters are approaching these risks in the wrong way. I remember an industry meeting at which some case studies – derived from actual cases involving mental health issues – were discussed. The histories were complex but nevertheless I was alarmed at the way some of the underwriters read a story into the relatively undetailed information that had been obtained via a GPRs and applicant questionnaires.
This is not the way to reach underwriting decisions, which need to be supported by the facts of the case concerned. While it is true that underwriters may differ in their opinions, the facts of an individual case should be clear and there must be sufficient information on which to base a decision. While a little ‘reading between the lines’ is on occasion justified, creating fiction is not. Underwriters, as guardians of profitability, need to do better and customers deserve better.
While much of the foregoing relates to mortality, it applies to disability products too – and in some respects more so. Morbidity data with the volume and detail required to provide a good evidence base for underwriting are, for a variety of reasons, rare. Nevertheless, it is well known that a high proportion of income protection claims are due to mental health problems. So how to deal with them at the underwriting stage? In the absence of good data, any rating philosophy has to be based on guesswork, which should be viewed as unacceptable to any of the stakeholders in these matters.
The only realistic solution, then, is to use an exclusion if the risk is not acceptable at the normal rate of premium. And because mental problems may manifest themselves in different ways over the course of time, leading to new diagnoses, it is not sufficient to exclude what has been diagnosed thus far; instead a wide-ranging exclusion embracing a number of mental conditions needs to be applied. As one who believes in offering cover as opposed to denying it, such broad exclusions make me a little uncomfortable but, given the nature of the risk and the potential for claim, it is difficult to see an alternative solution.
By modifying their approach to risk appraisal and their evidence requirements, insurers can create a win-win situation: they get better information and reach better decisions, and customers get fairer outcomes and feel their risk has been both better understood and priced fairly.
This article first appeared in Cover – www.covermagazine.co.uk.