02/09/2011
By Ilonka Oudenampsen
Predicting life expectancy has proven very difficult and imprecise, but a new report by Swiss Re calls for a different approach to predicting future life expectancy by looking at the pattern of diseases that often cause premature death.
Traditional approaches for forecasting life expectancy combine current rates of mortality improvement with a long-term future assumption or rely on stochastic computer-based models, which are based on historical mortality experience. However, in A window into the future: Understanding and predicting longevity, the reinsurance company states that in order to improve the method of predicting longevity, many different professionals need to be working together.
Author of the report Daniel Ryan explained Swiss Re had looked at 30 different diseases that are significantly associated with increased chance of death. If these diseases can be better monitored and their progress delayed, predicting future life expectancy might become more accurate.
However, he acknowledges that this method does not include the emergence of new diseases. “The best approach for modelling new diseases is to develop scenarios where new diseases are discovered or are found to be present in the population, and hence make explicit allowance for this possibility. We can quantify the likelihood of these scenarios based on how frequently new diseases have arisen in the past, and consider whether new diseases are more or less likely to occur now,” he said.
The report explained that most deaths follow an escalating pattern of disease as a result of harmful risk factors, with different diseases impacting one another in old age. “This co-morbidity means that analysing diseases in isolation is not sufficient. Any model attempting to predict future mortality needs to track the interaction between different key diseases.
“In order to build this type of mortality model, we need to develop a multi-disciplinary approach, using large patient databases that provide detailed information on individual diseases.”
This would involve different professions combining their expertise and actuaries, medical experts, epidemiologists, pharmacologists, demographers, gerontologists and governments will need to work together to better understand and model future life expectancy.
Predicting future life expectancy will remain difficult, but with this model it will be possible to more accurately predict the life expectancy of those with the most common diseases that often lead to a premature death.
This new model is also good news for pension schemes, Ryan said. “That is clearly to the advantage of pension schemes facing these risks at the moment, but also to the advantage of the members of the pension scheme, as their pensions are being protected in the future. We would have a better understanding as to how long people might have to work to make adequate pension provision to support themselves in retirement and how likely it would be that additional years of life would be spent in good health such that individuals would be able to continue to work and delay retirement.”
Other stories you may find of interest:
Multi stakeholder approach key to tackling longevity issues
With over $17trn of pension assets exposed to longevity risk across Europe, UK and the US, governments and key stakeholders have been called upon to take action to help tackle risks
Market volatility biggest perceived risk
Market volatility is the biggest perceived risk in the next 12 months according to institutional investors questioned for the latest biannual RiskMonitor survey from Allianz Global Investors. This was followed by sovereign debt and falling equity markets
RMS launches model to analyse Dutch longevity risk
Risk Management Solutions (RMS) has conducted a mortality risk analysis of the Dutch population, incorporating estimates of lifestyle trend changes, medical advances and future health care environments specific to the Netherlands
