American Journal of Epidemiology Advance Access published online on August 11, 2008
American Journal of Epidemiology, doi:10.1093/aje/kwn242
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Letter to the Editor |
THE AUTHORS REPLY
1 Department of Hygiene and Epidemiology, University of Athens, Medical School, Athens, Greece
2 Hellenic Health Foundation, Athens, Greece
3 Department of Epidemiology, Harvard School of Public Health, Boston, Massachusetts
(e-mail: cbamia{at}nut.uoa.gr)
We thank Dr. Alexopoulos (1) for his interest in our work (2). In his letter, Dr. Alexopoulos expresses his concern regarding the appropriateness of using the ages of 60 or 65 years as cutoffs for defining retirement as early or usual, given the pension schemes that existed in Greece until recently. It is for this reason, however, that, in our main analysis, we avoided using a cutoff for age at retirement, but we assessed the impact of retirement in a comparison of retirees with still-working persons of the same age as well as by estimating the mortality ratio associated with a 5-year increase in age at retirement among retirees. The ages of 65 and 60 years were used to assess the possibility of effect modification only. These analyses revealed no evidence of such an effect. Dr. Alexopoulos refers to stratification, but gender was the only factor used for stratification, and it was not related to the issues he refers to.
As we state in our paper (2), we did not have usable information on the prevailing conditions in occupations of the subjects included in this cohort, and it may be true that some of those who retired early may have been working in hazardous jobs, thus being at higher mortality risk. The overall mortality ratios associated with common unsatisfactory working environments, however, are generally not large. Thus, it is unlikely that occupation would have explained more than a small fraction of the excess mortality associated with retirement (3). We could not adjust for socioeconomic status, but education is a satisfactory indicator of socioeconomic position in this cohort of adults and elderly Greeks (4, 5).
It is also possible that some early retirees may have been suffering from diseases other than those excluded, such as mental, neurologic, orthopedic, and respiratory disorders, that is, diseases with a frequently prolonged clinical course. These subjects could have introduced a bias in the estimation of mortality ratio associated with early retirement, but few deaths were attributed to these causes.
We are not quite sure about the statement concerning healthy-worker effect. The extent of confounding is maximal when the prevalence of exposure is about 50%, not when it is overwhelmingly high (6).
Finally, access to health care is not equitable in Greece, but this inequity applies to both working people and retirees.
We do not discourage earlier retirement. We simply exploited data from a valuable database to obtain insight into the processes that affect longevity, as others have also done (7–12).
ACKNOWLEDGMENTS
Conflict of interest: none declared.
References
- Alexopoulos EC. Re: "Age at retirement and mortality in a general population sample: The Greek EPIC study" [letter]. Am J Epidemiol (2008) 168(00):000–000.
- Bamia C, Trichopoulou A, Trichopoulos D. Age at retirement and mortality in a general population sample: the Greek EPIC study. Am J Epidemiol (2008) 167(5):561–569.
[Abstract/Free Full Text] - Clayton D, Hills M. Statistical Models in Epidemiology (1998) New York, NY: Oxford University Press.
- Trichopoulos D. Epidemiology. Principles, Methods and Applications [in Greek] (1982) Athens, Greece: Parisianos. 1–555.
- Trichopoulou A, Costacou T, Bamia C, et al. Adherence to a Mediterranean diet and survival in a Greek population. N Engl J Med (2003) 348(26):2599–2608.
[Abstract/Free Full Text] - Walker AM. Observation and Inference. An Introduction to the Methods of Epidemiology (1991) Chestnut Hill, MA: Epidemiology Resources Inc.
- Haynes SG, McMichael AJ, Tyroler HA. Survival after early and normal retirement. J Gerontol (1978) 33(2):269–278.
[Abstract/Free Full Text] - Baker D, Packard M, Rader AD, et al. Mortality and early retirement. Soc Secur Bull. (1982) 45(12):3–10.[Medline]
- Morris JK, Cook DG, Shaper AG. Loss of employment and mortality. BMJ (1994) 308(6937):1135–1139.
[Abstract/Free Full Text] - Quaade T, Engholm G, Johansen AM, et al. Mortality in relation to early retirement in Denmark: a population-based study. Scand J Public Health (2002) 30(3):216–222.[CrossRef][Web of Science][Medline]
- Hammerman-Rozenberg R, Maaravi Y, Cohen A, et al. Working late: the impact of work after 70 on longevity, health and function. Aging Clin Exp Res. (2005) 17(6):508–513.[Web of Science][Medline]
- Tsai SP, Wendt JK, Donnelly RP, et al. Age at retirement and long term survival of an industrial population: prospective cohort study. BMJ (2005) 331(7523):995.
[Abstract/Free Full Text]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||