Secular Trends in Life Expectancy by Diet Status Among Never Smoking Seventh-day Adventists

To evaluate how life expectancy changed by age and secular time from 1960 through 1988, two cohort studies, the Adventist Mortality Study (AMS, 1960-1985, n = 27,530) and the Adventist Health Study (AHS, 1976-1988, n = 34,192) were combined. The life expectancy of omnivores was compared to vegetarians separately in never-smoking males and females for all-cause mortality, and where the cause of death was cardiovascular disease or all site-cancer. Life expectancies were calculated for subjects who had survived to ages 30, 50, 70, and 80 during six calendar periods: 1960- 1964, 1965-1969, 1970-1974, 1975-1979, 1980-1984, and 1985-1988. Ten-year calendar periods were used for the specific causes of death: heart disease and all-sites cancer, because of the small number of events. The basis for the life table calculations was the age-specific mortality rates, which were calculated using the Multiple Decrement Life Table Analysis Program (MDLTAP) taking into account competing causes of death and allowing for the control of confounders. Both univariate and multivariate analyses were performed controlling for the potential confounders: education, body mass index (BMI), and exercise. Secular trends in life expectancies were examined using weighted linear regression, where inverse variances obtained from the MDLTAP program were used as weights for the respective calendar periods.

The life expectancy for both males and females increased with secular time from 1960 to 1988 for all ages, and the linear trend was significant at most ages. The trend tended to be steeper in omnivores compared to vegetarians and in females as compared to males. The life expectancy of vegetarian females who had survived to 30, 50, and 70 years of age was about two years higher than that of omnivores for the calendar periods 1965-1969, 1975-1979, and 1980-1984. The life expectancy of vegetarian males up to 80 years of age was 1-4 years higher than that of omnivores for all calendar periods.

The life expectancy at all ages of both females and males who eventually died from heart disease or cancer increased with secular time from 1960-1988, irrespective of their type of diet. This increase was 2-10 years in females and 2-7 years in males. The life expectancy at all ages for vegetarian females who died from heart disease or cancer was 1-2 years higher than that of omnivores in the calendar period 1970-1979. The difference in life expectancy that was attributed to a vegetarian status in males who died of heart disease or cancer was 1-4 years. This difference was statistically significant at most ages.

In conclusion, the life expectancy for both males and females increased with secular time from 1960 to 1988, irrespective of type of diet and cause of death. The life expectancy of vegetarians at all ages was 1-4 years higher than that of omnivores. Females had a higher life expectancy than males. The gender differences in life expectancy were less in vegetarians compared to omnivores. A vegetarian diet does contribute to greater longevity, and this lifestyle decreases the gender gap in life expectancy.