Background: It is unclear why relatively fewer men live to very old ages compared with women. Several prospective studies have examined survival to the average life span of 75 to 85 years of age, but studies of even older persons have been largely retrospective and included relatively few men. They are also complicated because of the high prevalence of disease and disability in these oldest populations. Studies of twins have estimated that genetic traits account for about 25 percent of the variability in life span, with about 75 percent influenced by potentially modifiable factors. It is unclear what factors earlier in life might correlate with survival to older ages or with good health and function at an older age. Yates and colleagues attempted to answer two questions, including which potentially modifiable factors are associated with survival to an older age, and if those with exceptional longevity have reduced incidence of age-associated disease.

The Study: Prospective data was gathered on men in the Physicians' Health Study, a cohort of male physicians started in the early 1980s to assess the use of aspirin and beta-carotene in the prevention of cardiovascular disease and cancer. Men were at least 65 years of age at enrollment and had already exceeded the average life expectancy for men born between 1900 and 1915 (46 to 52 years). At enrollment, participants were generally healthy without a history of cancer, myocar-dial infarction, cerebrovascular disease, or other serious diseases.

Baseline data collection included information on height, weight, blood pressure, cholesterol level, diabetes, smoking status, alcohol consumption, and frequency of participation in vigorous exercise (defined as enough to cause sweating). Information was updated annually, including data on changes in health and lifestyle habits. The primary outcome of the current study was survival to 90 years of age. Secondary outcomes included development of major age-related diseases, including cardiovascular disease and cancer (excluding nonmelanoma skin cancer). Functional status was assessed by the Medical Outcomes Study 36-Item Short-Form 36 Health Survey (SF-36), which was included in the 16-year follow-up questionnaire. Men who lived to 90 years of age (survivors) were compared with nonsurvi-vors in baseline characteristics, follow-up data, and incidence of outcomes.

Results: Of the 2,357 men followed over 25 years, 970 (41 percent) survived to at least 90 years of age. Of these survivors, slightly more than one half (53 percent) had died by the study's conclusion, with a mean age at death of 93 years. The mean age at death of nonsurvivors was 83 years.

Several risk factors at study enrollment (current smoking, diabetes, hypertension, and obesity) were independently associated with greater risk of dying before 90 years of age (hazard ratio of 2.10, 1.86, 1.28, and 1.44, respectively). Regular vigorous exercise reduced mortality before 90 years of age by 20 to 30 percent, even in those who exercised as infrequently as one to four times per month. When examined in aggregate, the presence of one, two, three, four, or more baseline risk factors (i.e., smoking, diabetes, body mass index greater than 25 m per kg², hypertension, or sedentary lifestyle) reduced the age-adjusted odds of survival to 90 years of age by 33, 49, 75, and 88 percent, respectively. No significant survival associations were noted with alcohol intake or cholesterol level.

Survivors had significantly better self-rated physical and mental health than nonsurvi-vors, and there was no difference between the two groups in level of social contact. Survivors were less likely to have cancer and cardiovascular disease, and when present, they occurred on average three to five years later than in nonsurvivors. In age-stratified analysis, similarly significant findings were noted between survivors and nonsurvivors.

Conclusion: The authors conclude that the likelihood of men living to 90 years of age is significantly reduced by smoking, diabetes, obesity, and hypertension, whereas regular vigorous exercise substantially improves it. Men who lived to 90 years or older also had better physical and mental functioning than men who died before this mark. This supports the idea that longevity is achieved by escaping or delaying disease processes, and that this can be influenced by behavioral factors.

editor's note: This study gives us more ammunition to use when discussing preventive health medicine with our male patients. For instance, men might be more motivated to address being overweight once they realize that it reduces their chances of living to 90 years of age by one third, or that addressing modifiable issues could improve their chances of being in better mental and physical health as they age. Also, because even exercising one to four times per month reduces the likelihood of early death by approximately 20 percent, many patients might see this as an achievable initial goal.

This study only recruited men 65 years or older; therefore, incorporating lifestyle changes later in life could still positively affect our patients' lives. The authors calculated that a 70-year-old man who had none of the five risk factors identified above had a 54 percent greater likelihood of living another 20 years. This study's main drawback is that most participants were well-educated, white physicians, but it still provides a good starting point until results from a more heterogeneous population are available.—k.t.m.