Are we underwriting what we price for?

Actuarial risk profiles are based on general population mortality statistics and insured population claims experience. Medical underwriting identifies risks that are consistent with those risk profiles and thus delivers an appropriate and correctly priced risk portfolio – but, it can only do that if it is fully capturing all the material risk factors. Our research from the U.K. life protection market shows that lifestyle risks are a major risk factor for the under 40’s, but that these risks are not adequately captured by medical underwriting.

Similarly, population mortality statistics from the U.S. show that the majority of deaths in the under 40’s are trauma related. A similar issue applies - the same underwriting approach is typically used across all age groups, relying largely on medical evidence. However, over the past few years measures have been taken in the U.S. to target lifestyle risks in the underwriting process. In particular, insurance companies can gain access to pharmaceutical and motor records which allow a more targeted approach to be taken while avoiding issues of non-disclosure. Tele-interviewing, which prioritizes lifestyle risks for the under 40’s, is used to supplement this information.

We present the results of our study - which includes input from leading U.K. Chief Medical Officers (CMOs) and medical underwriters in the U.K. & Irish markets - and taking into account developments in the U.S. propose improvements that would help to resolve the discrepancy.

The statistical background

What we found is that in pricing, the risk factors for mortality are commonly split into three broad age-groups, as per the table 1 below.


Table 1: Leading causes of death by age group, from U.K. mortality statistics. Source: The U.K.’s Office for National Statistics.

Statistics produced by the Office for National Statistics in the U.K. (2008¹) show that external causes account for over 60% of male deaths for ages 20-24; this reduces to less than 5% for ages 60-64. Cancer accounts for less than 10% of male deaths for ages 20-24; this increases to over 40% for ages 60-64. Whilst respiratory diseases account for less than 5% of male deaths for ages 20-24, this increases to over 30% for ages 60-64. So in broad terms, life-style is the key risk factor for mortality in the under 40’s but is much less important for older ages.

This picture is backed up by our analysis of the claims experience of a large U.K. insurer. Based on this sample, 46% of deaths in males under 40 relate to road traffic accidents, suicide or other accidents. This compares to 10% for males over age 40.

The above statistics show that the risks to which a younger life is exposed are quite different to those seen in older lives. In contrast, medical underwriting generally follows a “one-size fits all ages” approach that has also not changed much since the 1950’s.

It would appear therefore that the gate-keeping role of medical underwriting to deliver a mix of business broadly consistent with what is priced for does not always happen in practice. Medical underwriting is still generally approached via the rear view mirror: loadings are applied for previous medical history, but despite major advances in lifestyle analysis, little is done to identify and allow for this risk factor in rating.

CMO and Chief Underwriter opinions

Beyond the statistics, we also sought expert CMO and U.K. Chief Underwriter opinions. The responses from a leading cardiologist and oncologist were particularly interesting. A leading cardiologist felt that:

“Alcohol is a major problem at the moment with 25% of all hospital admissions being related to acute or chronic problems from alcohol. In the young it is binge drinking and the ‘alcopops’ saga. What middle England drinks has changed and also the amount (especially in the 30-40 age group) has increased dramatically.”

“There is no doubt that the use of cocaine since the 1980’s has surged due to the falling price. Cocaine causes coronary artery spasm (sustained contraction of the coronary arteries) which can cause heart attack. This is often the cause in young people (in their 30’s etc).”

Whilst a leading oncologist felt that:

“The major changes in cancer risks from lifestyle in < 40 are in oral and oesophageal cancer. Both are showing significant epidemiological change in traditionally low risk age groups. Alcohol and possibly obesity remains a significant risk factor.”

Further CMO input revealed the following concerned observation:

“Smoking is declining in older populations but is on the increase in the young. In particular there is an increase in young women starting smoking which is concerning. COPD/chronic airways disease is related to the total amount you have smoked as is lung cancer. The risk of several other cancers (renal definitely and also oral cancer) is also increased with smoking.”

Chief Underwriter Survey

Feedback from a broad range of Chief Underwriters across the U.K. & Irish life markets supported the view that assessment of the under 40's was a different challenge, and one which we are not currently fully addressing. A representative comment was:

“There are three key areas which are going to cause us problems over the next decade or two – obesity, drug use and alcohol. None of these are easy to assess or predict through traditional underwriting but we must find better ways of drawing out these issues through increased lifestyle and behavioral questioning, perhaps using a very different and (unfortunately) possibly longer application form.”

We asked Chief Underwriters to consider the following questions:

Reacting to the statistics and opinions

At PartnerRe, we therefore consider that it is essential to underwrite and rate distinctly by age-group in order to give the closest focus to the illnesses or accidents that are likely to be experienced in each age group. For younger lives, this will mean looking beyond medical history to the more predictive lifestyle and habit queries.

This could be achieved by making adjustments to the application questions and by introducing screening to test for current lifestyle habits whose debilitating effect will have a significant impact on claims experience. Specifically we suggest:

Redesigning the application form to elicit lifestyle risks

Alcohol Risk

• Type and quantity of alcohol consumed (rather than how many units)
• Has the applicant ever attended A&E for alcohol related reasons?
• Has the applicant ever been convicted of driving whilst under the influence of alcohol?
• How many days per week is alcohol consumed?
• Has the applicant ever been advised by a doctor or counsellor to reduce or modify their alcohol consumption?
• Has the applicant ever missed work due to alcohol?

Accident Risk

• Has the applicant ever been convicted of a drink driving offence?
• Has the applicant’s driving licence been endorsed for speed related offences?
• Mode/s of transport (car/motorcycle)
• Vehicle/s engine size
• Previous injuries including broken bones, fractures and sprains

Suicide Risk

• Include detailed questions around depression, stress and anxiety
• Ask for details of time off work due to illness (number of occasions and days)
• Approximate number of visits to general practitioners in the last 12 months
• Is there a family history of depression or suicide?
• Details on recreational drug use

Introducing additional sources of information

Routine screening for some of the important lifestyle risk factors for the under 40’s should also be considered. In particular, tele-interviewing (as now increasingly used in the U.S. market), nurse screenings and tests should replace GPR’s as the cornerstone of medical underwriting to better assess lifestyle risks. Consideration also needs to be given to how additional information can better be obtained by insurance companies.

Conclusion

Medical underwriting is heavily based on medical evidence and, by extension, prior medical history, but it does not adequately identify the risk factors in the under 40s relating to alcohol, stress and obesity; these factors have yet to manifest in medical history, but will impact future claims. Medical underwriting is therefore not fully synchronized with the statistics and underlying claims experience that actuaries are using to price the business. The two functions need to be re-aligned to ensure there is less cross-subsidisation in insurance rates and risk controls improved to avoid a future negative claims impact.

Possible measures to strengthen and re-align medical underwriting to the underlying risk and its development include updating the application process for younger lives to elicit additional and more detailed information on lifestyle risk, introducing routine screening for some of the more important risk factors for the under 40’s, and by making tele-interviewing, nurse screenings and tests the cornerstone of underwriting over GPR’s.

All told, this should lead to improved, equitable risk selection and a more sustainable and successful product for protection providers.

¹ www.statistics.gov.uk