By Tom Edwalds
The fifth triennial Living to 100 Symposium was held in Orlando, Fla., from January 8 – 10, 2014. Sponsored by the Society of Actuaries with seven co-sponsors and 39 other participating organizations, the symposium attracted nearly 200 attendees from 15 countries, including experts in demography, gerontology, medicine, biology, economics, and financial planning, as well as actuaries. Presentations included 32 research papers with formal discussions, eight panel discussions, and four featured speakers. Some of the key topics covered include societal adaptations to longer average lifetimes, advances in medicine, forecasting future changes in life expectancy and life spans, the effect of genetics on longevity, and the shape of the mortality curve at advanced ages.
According to Anna Rappaport, societal adaptations to longer average lifetimes have been slow in coming and have a long way to go. The swiftest change has been the switch from defined benefit to defined contribution private pension plans, offloading longevity risk from employers to employees. The average age at retirement has stopped decreasing and has recently started increasing. Retirement is no longer abrupt, with half of retirees working either in bridge jobs or to supplement retirement income. However, research done by the Society of Actuaries Post-Retirement Needs and Risks Committee shows that most people at or near retirement age are ill-prepared for retirement. Many retire involuntarily, and even those who voluntarily retire have woefully inadequate (or even non-existent) planning horizons.
Dr. Anthony Atala of the Wake Forest Institute for Regenerative Medicine described the progress that has been made in this field and showed videos of organs being “printed” and of patients who had received successful transplants of printed organs and tissues. Almost all organs and tissue types have been successfully constructed ex-vitro, except for the brain, liver, and kidneys.
A wide range of opinions was expressed concerning future changes in life expectancy and life span. Sam Gutterman of PwC pointed out that the current obesity epidemic has yet to reach the ages of greatest mortality impact and that smoking prevalence has stabilized, so future improvement in life expectancy could be significantly less than it has been recently. Jim Vaupel of the Max Planck Institute for Demographic Research pointed out that progress in life expectancy at birth based on the country with the lowest mortality has followed a remarkably linear pattern, increasing by about three months per year. He believes that this trend will continue for many years, with future improvements generated by reductions in mortality due to cancer and dementia and by progress in regenerative medicine and gene therapies.
A truly optimistic view was presented by Aubrey de Grey of the SENS (Strategies for Engineered Negligible Senescence) Research Foundation. He categorized all of the deterioration due to old age into seven categories of biological damage, and asserted that therapies already exist (at least in laboratories) to remove, repair, reinforce, or replace all such damage. He envisions achieving “longevity escape velocity” as these therapies become enhanced and refined and become standard in clinical practice, leading to reductions in mortality rates at all ages down to the level of accidental deaths. This would mean a ten-fold increase in life expectancies.
The effect of genetics on longevity was discussed by several speakers. Nir Barzilai of the Institute for Aging Research at the Albert Einstein College of Medicine presented his findings from his study of a group of nonagenarian and centenarian Ashkenazi Jews along with their offspring and demographically matched controls. He has successfully isolated genes that increase HDL (good) cholesterol, decrease LDL (bad) cholesterol, improve insulin action, prevent Alzheimer’s, and dictate longer telomeres, all of which are associated with lower mortality and increased longevity.
Leonid Gavrilov of the National Opinion Research Center at the University of Chicago investigated predictors of exceptional longevity by linking computerized genealogies of confirmed centenarians and matched controls from the 1890-91 birth cohort to the 1900 and 1930 censuses and found that having parents who lived to be 80 or more was the strongest predictor of survival to age 100. A surprising finding was that having a radio in the household in 1930 was a strong predictor of longevity for women, but not for men.
The shape of the mortality curve at advanced ages was also a topic of debate. Natalia Gavrilova of the National Opinion Research Center at the University of Chicago showed that the Gompertz model (no mortality deceleration) fit the estimates of the force of mortality better than the Kannisto model when the Sacher estimate was used on single-birth-year cohort data from the Social Security Death Master File. Nadine Ouellette of Institut National d’Etudes Demographique investigated estimates of the force of mortality from exceptionally clean data on Quebecoise women centenarians born in the last quarter of the 19th century and showed that fitted P-splines exhibited downward curvature (mortality deceleration). Meanwhile, Bob Howard constructed data displaying obvious mortality deceleration by contaminating data constructed deterministically from a Gompertz model with a small proportion of misstated ages at death, illustrating the possibility that observed mortality deceleration could be an artifact of data errors.
This is a small sample of the ideas and information presented at the symposium. The Society of Actuaries will publish a monograph of the papers presented at this symposium. For more information, visit http://livingto100.soa.org.
Tom Edwalds, FSA, ACAS, MAAA, is assistant VP, mortality research at Munich American Reassurance Co in Chicago, Ill. He can be reached at email@example.com.