Interdisciplinary research in mortality/longevity/morbidity

Actuaries have been active in the study of mortality since the earliest days of the profession and it remains an area where actuaries can contribute to the quality of public debate and lead in the development of new thinking.

The ageing population and increasing longevity are welcome evidence of social and health improvements in our society but raise new issues that are of concern to us all. This continues to be an area where the Actuarial Profession can act in the public interest by working with other disciplines, for example in the medical, social sciences and demographic fields. Combining the expertise of actuaries with that from other disciplines will substantially improve our ability to understand mortality and morbidity risk. In 2010, the Profession wishes to move the debate forward by funding external interdisciplinary research in this area up to a total £100,000. We intend this work to build on the platform created by the Mortality Research Steering Committee and to catalyse future research across disciplines that we hope will have a major and far-reaching impact.

Background

With lower inflation and interest rates in recent times, the importance of mortality risk to actuarial practice has increased, just as the difficulties in predicting future mortality have become more apparent. Society as a whole, government, and industry, are all having to deal with an ageing population and increased life expectancy. In 2006 the Actuarial Profession recognised that collaboration with other disciplines offered an opportunity to better understand past, present and future trends and it set-up a multi-disciplinary Mortality Research Steering Group.

The group quickly established the potential benefits of a multidisciplinary approach, particularly in relation to the availability, reliability and granularity of data. An initial scoping study was followed by events which culminated in the successful Joining Forces conference on mortality and longevity held at the Royal College of Physicians in Edinburgh on 21-22 October 2009. This conference brought together leading academics, researchers and practitioners in actuarial and medical science to consider, debate and share knowledge and has created a solid platform for research in this area. The conference reinforced the view that actuaries must link up with and understand the work of others as an input to their work and, crucially, to influence the direction of future research.

The call for research

The Profession would welcome proposals in the three themes underpinning the Joining Forces conference:

I.   Understanding the drivers for change in mortality and longevity, for example:

• What exactly causes death rates to continue to fall?
• What is the scope for current trends to continue?
• Understanding the closing gender gap in mortalit
• What will be the impact of future medical advances?
• What would be the impact of future behavioural/social/environmental changes?

II.   How will successive cohorts differ and why

• Are those now reaching old age in some way different to previous generations in terms of their intrinsic biology and exposure to risk?
• How will younger generations differ, and why?

III.   How far can the approach to drilling down to individualised risk help in exploring the connections between populations and individuals including the impact on mortality of biological and life style risk factors such as nutrition, exercise, and alcohol consumption?

Areas with immediate practical impact:

• The development of mortality/morbidity models – this could include new methods or modelling methodologies from other areas of statistics/mathematics/science, incorporating cause of death, the potential for using stochastic models
• Analysis of alternative datasets e.g. pensioners, different socio-economic groups
• Modelling and/or projecting different populations (e.g. different geographic regions, different socio-economic groups, migrant populations).

The public interest

Research into mortality and/or morbidity that will help move forward the current debates around:

• Long term care/the retirement age/healthy ageing
• The social and financial consequences of changes in mortality/longevity and morbidity
• The impact of mortality on the economics of pensions: public spending and education.

Awards

Three awards were subsequently made:

• University of Southampton and Barnett Waddingham LLP - Bayesian Modelling of Mortality Projection Uncertainty
• Heriot-Watt University - Mortality Models for Multiple Populations using Covariates
• King’s College London - Genetic risk profiling for common diseases

Dr Gerard Kennedy ASA of the University of Southampton of the research team of himself, Professor Jon Forster, also of the University of Southampton and Mr Neil Robjohns FIA of Barnett Waddingham LLP said: “Mortality projection is of vital importance to the actuarial profession, but is subject to uncertainty, and it is critical that such uncertainty be accurately quantified. Currently, uncertainty quantification, if performed at all, is done within the context of a single mortality projection model. However, there is no universally agreed such model, and fully coherent uncertainty quantification should also account for model uncertainty. Bayesian statistical methodology provides a fully coherent framework for estimation in the presence of model uncertainty, but practical implementation within the context of mortality projection is under-explored, and will form the main focus of our research. Potential benefits of this research would be improved pricing and assessment of risk exposures and capital requirements in respect of longevity risk transactions, greater understanding of and confidence in the quantification of both mortality projection uncertainty and the tail of longevity risk, and improved management of such risk.”

Dr Torsten Kleinow of Heriot-Watt University said: "Our project willdevelop new mortality models for multiple populations. These models will be based on covariates; in particular smoking prevalence. Our aims are to refine predictions of mortality rates, and explain differences between mortality rates for different cohorts and populations."

Professor Cathryn Lewis of King’s College London said: “Genetic studies have identified genes contributing to the common, complex disorders that confer a major public health burden, such as heart disease and diabetes. This raises the prospect that individual-level genetic screening can be used to identify those at increased risk of such diseases, which will have implications for actuarial practice.  This research proposal will develop statistical models for genetic risk profiling, providing a framework for investigating genetic risks at a population level.”