| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||
COMMENTARY |
School of Public Health, University of Illinois at Chicago.
Address correspondence to S. Jay Olshansky, PhD, School of Public Health, University of Illinois at Chicago, 1603 West Taylor St., Rm. 885, Chicago, IL 60612. E-mail: sjayo{at}uic.edu
Authors Jean-Marie Robine and Jean-Pierre Michel (1) suggest that the time has come for the development of a new general theory of population aging. A variety of theories or models of population change have been proposed in the past, including the one most relevant to this manuscript—the Epidemiologic Transition Theory devised by Omran (2), in which human populations appear to move in transitional phases from high to low rates of fertility and mortality. This model was later extended to include a fourth (3) or hybristic (4) stage during which declining death rates at older ages were explicitly accounted for as relevant new attributes of population dynamics.
These general models of population change are interesting and useful because they enable those of us living in the time in which such transitions are occurring to label entire nations and population subgroups as falling within one of the various stages. In turn, this theoretically enables public health practitioners to target for intervention, population subgroups that are lagging behind. It is also a useful teaching tool that enables researchers to communicate to students entering the various fields that address population dynamics, to understand general attributions of population change both ongoing and observed in the past.
Robine and Michel (1) are now calling for the development of a new, more comprehensive model of population aging that is sensitive to trends in both longevity and health status. Their rationale is straightforward. Population transition theories and, in particular, the Epidemiologic Transition Theory tend to be overly simplified generalizations of trends in the vital rates of a country. In reality, most nations have population subgroups in various proposed phases of the epidemiologic transition at the same time. Add to this complexity the fact that death rates continue to decline (even at older ages), life expectancy is on the rise, advances in the biomedical sciences are progressing at an accelerating pace, and the health status of older populations across the globe are experiencing a complex mixture of increased frailty accompanied by reductions in some measures of disability at some ages, and you have a recipe for the unique contemporary phenomenon of population aging that falls outside the boundaries of understanding offered by existing models of population change. It has even been suggested recently that the Epidemiologic Transition Theory and its extensions represent a poor model of population change because they fail to take into account the ongoing reemergence of communicable diseases (5).
In developing the rationale for a new general theory of population aging, the authors have embellished elements of the population trends to support their reasoning. This was not necessary. For example, although some researchers have argued that the maximum verified age of humans has been on the rise in the past 150 years (6), the fact is that the chronological ages, and more importantly, the exact duration of life, of most people on earth cannot currently be measured because of the lack of reliable birth records and other trustworthy makers of time during the life span. Thus, while it is certainly true that there are more centenarians now than in the recent past, it is not possible to know with certainty the frequency with which centenarians occurred in our distant past, nor is it possible to know whether the maximum age at death for humans has ever increased. The fact is that while the extreme elderly are on the rise, and they certainly have a growing presence in low-mortality populations, the existence of more people in this extreme region of the life span is insufficient as a rationale for developing a new theory of population change. The same logic could (and should) just as easily have been applied to the population aged 85 and older, where the number of people entering this region of the life span is so large, so rapid, and so well documented, as to represent sufficient justification for the argument developed by Robine and Michel (1).
The tail of the human survival distribution is interesting for the moment, but it is certainly too early to determine whether the dynamics of survival and mortality change in this extreme region is of sufficient interest to influence future population-level statistics. Moreover, Robine and Michel (1) suggest that future increases in life expectancy at birth will result from declining mortality at the highest ages, which can only mean the extreme tail of the survival distribution (i.e., ages 100 and older), and that sustained increases in life expectancy resulting from declines in old-age mortality was unexpected just a few years ago. Neither point is accurate. The importance of declining old age mortality as a relevant factor in epidemiologic transition theory was first described in articles by Greunberg (7) and Kramer (8), and which later served as the primary rationale for supporting an extension of epidemiologic transition theory (3,4). As for the importance of declining old-age mortality on life expectancy at birth, the effects of mortality dynamics among those aged 50 to 100 is far more important than declines in death rates at the extreme ages because most people die before the age of 100.
In the end, what Robine and Michel (1) offer is not what I had expected, which was a proposed new theory of population aging that takes into account the unique combination of life extension into extreme regions of the life span and the extraordinarily complex dynamics of temporal transitions in frailty and disability across the globe. Moreover, Robine and Michel (1) ignore the obvious importance of ongoing global trends in obesity and communicable diseases—both of which would have a significant negative impact on the rise in life expectancy if left unchecked. What has been offered instead appears in the final sentences in the form of a suggestion that scientists harmonize the various measures of health expectancy (which investigators who study health expectancy have been attempting to do for some 20 years), and that the time has come to develop a new periodic aging survey to monitor global trends in aging. Neither proposal represents a new theory of population aging, but rather, the implementation of REVES (Réseau sur l'Esperance de Vie en Santé)-initiated efforts and ideas that date back 20 years. To that end, I completely agree with Robine and Michel (1)—their proposal would yield valuable new data that would enable investigators to determine whether the ongoing extension of life is leading to improved health or increased frailty, and whether we can expect these trends in health and longevity to continue. Perhaps from these new data, scientists will eventually be sufficiently stimulated to devise an innovative and useful theory of population change that the authors initially call for.
References
| ||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
