SCORviews hero image

Using Expert Judgement to Assess Mortality Trends
April  2019

During the Global Mortality Improvement initiative, SCOR assessed trends in mortality improvement and quantification in the basis risk (the mis-estimation of the difference in mortality experience between general and insured populations) between insured, reinsured and population mortality with the goal to improve mortality forecasting for mortality products. 

To complement our primary objectives, SCOR performed two expert judgement studies to collect opinions from both insurance professionals and the medical community. The objective of both studies was to advance the collective knowledge of both SCOR and the study participants on the topic of mortality risk, including causes of death and potential changes in lifespan in the future due to medical advances, behavior changes, public policy, socioeconomic and other factors. 

Delphi Study
The first of these studies was a Delphi Study. The Delphi Study method is a structured communication technique developed as a systematic, interactive forecasting method which relies on a panel of experts. The aim is to gain a level of consensus by surveying the experts in multiple rounds of questions, allowing the experts to review responses from the other experts at the end of each round.

Participants in the Delphi Study are subject matter experts in the area of mortality improvement who shared an understanding of these characteristics:
  • Medical, cultural, political, social, legal and demographic impacts on mortality
  • Medical advances, including their implications on causes of death
  • Mortality improvement and mortality experience analysis
  • Mortality and mortality improvement assumption setting, including consideration of risk
  • Approaches to using historical mortality data to estimate future mortality
  • Risk classification

They answered questionnaires in two rounds after which we created a final research report.

Overall, responses in the Delphi Study showed varied opinions regarding future mortality improvement levels compared to recent historical experience. In general, the panelists expected mortality improvement levels to be similar in the short term (5 – 15 years). For mortality improvement by socioeconomic class, panelists expect it will expand in the short term and possibly converge over the longer term (16 to 40 years). 

The largest positive drivers that contributed to improved mortality in recent history are medical advances and fewer smokers. The largest negative drivers are obesity (with the resulting impacts of related issues) and opioids/drug epidemic. There is consensus that mortality improvement for the insured population is different than the general population, mainly due to levels of education and income, access to healthcare and healthier lifestyles. In general, the expectation is that mortality improvement is and will continue to be higher for the insured population. 

Mortality improvement will continue to differ for impaired lives; however, with future medical advancements the gap may begin to close to normal life expectancy. On average, panelists believe that the maximum duration (years from life insurance policy issue) at which mortality improvement rates should be applied is 30 years and the maximum attained age at which mortality improvement rates should be applied is 99.

Medical Expert Judgement Study 
The second study executed was focused on Medical Expert Judgement. This study consisted of a single survey followed by phone interviews with 16 physicians/medical experts (internally and externally). 

As one might expect, our medical experts consider obesity as the major risk to population mortality over the next decade, with most participants saying that it would have a high negative impact. Interestingly, there was less consensus 2.5 decades out, as a similar number thought that the impact of obesity would be high, medium or low. Opioids and illegal drugs were thought to have a medium negative impact in the near and medium term but were projected to have less impact in the longer time frames. 

Medical innovations’ impact may have been diluted by the fact that many different specific innovations were considered. Even so, the panel’s consensus was that there would be a medium sized positive effect in the short term, and innovations in the prevention of disease were expected to hold sway in the longer term. 

Another consensus arose around the causes of deaths that would increase in the future with Alzheimer’s/dementia and heart failure. However, the modal age at death is expected to be slightly older for all causes of death that were surveyed. 

To summarize, both expert judgement studies performed during the GMI project were major factors in helping SCOR validate assumptions, while providing avenues to incorporate the perspectives of leading industry experts. In addition, the processes associated with the Delphi Study and the Medical Expert Judgement Study provided valuable insights and learnings that will be applied to future mortality improvement studies throughout SCOR.