Do Vegetarians Live Longer?

(These notes are from a talk given by Paul Appleby, Secretary of Oxford Vegetarians, to members of Oxford Green Party Students at the Friends Meeting House, Oxford, on Friday 1 March 2002, and at an Oxford Vegetarians speaker meeting at the same venue on Wednesday 20 March 2002.)

It is often claimed that vegetarians enjoy greater longevity than non-vegetarians. Is there any truth in this claim? If so, for how much longer do vegetarians live, on average, than non-vegetarians, and are the benefits attributable to diet alone, or to other lifestyle factors?

When studying longevity, one approach is to simply compare the average (or mean) age at death of one population with that of another. This approach works well when making international comparisons, but is not suitable for 'within-population' comparisons because governments do not generally record the diets of their citizens. A more common approach is to compare the death rates (mortality) of population groups. This approach involves studying a large group of people in what is called a prospective or cohort study. At the beginning of the study participants complete a questionnaire describing their diet and other aspects of their lifestyle. Participants are then 'followed-up' over a period of time and details of any deaths and their causes are recorded.

Statistical methods are then used to compare death rates (mortality) between the cohort as a whole and some reference population, or between different groups within the cohort, for example, between vegetarians and non-vegetarians. Relative mortality can then be expressed as a death rate ratio (DRR) for one group (e.g. vegetarians) compared with the reference group (e.g. non-vegetarians), together with its 95% confidence interval (95% CI - a measure of how accurately the DRR is estimated).

Several prospective studies have recruited large numbers of Western vegetarians (and comparable non-vegetarians), enabling estimates of the relative mortality of vegetarians to be made:

• Adventist Mortality (California, 1959-60)
• Health Food Shoppers (UK, 1973-79)
• Adventist Health (California, 1976-80)
• Heidelberg (Germany, 1978-81)
• Oxford Vegetarian (UK, 1980-84)
• EPIC-Oxford (UK, 1993-2001)

Mortality for the vegetarians in these studies may be compared with death rates in the population from which they were recruited by calculating a Standardised Mortality Ratio (SMR). For example, SMRs for vegetarians in the Health Food Shoppers study and the Oxford Vegetarian study are shown below:

These results show that the vegetarians in these studies are dying at about half the rate of their contemporaries in England & Wales (the reference population) as a whole.

However, SMRs can be misleading because of the "healthy volunteer effect" (sick people do not usually volunteer for long-term studies of health) and because the cohort may not be representative of the reference population in many respects.  For example, vegetarians are less likely to smoke and are more likely to be from the middle classes than the 'average' Briton. Therefore, it is not surprising to find that the 'health-conscious' non-vegetarians in these studies have similarly low mortality, as we see when we add their SMRs to the table:

A better approach is to compare death rates for vegetarians and non-vegetarians within studies. Within-study comparisons have been made in several of the studies listed, with varying results.

Individual data from the first five studies listed above were pooled in a collaborative analysis of mortality in vegetarians and non-vegetarians. Data were available for 76,000 persons of whom nearly 28,000 were vegetarians. There were 8,330 deaths between the ages of 16 to 89 after an average of 10.6 years follow-up, allowing accurate estimates to be made of the DRR for several common causes of death, as well as for all causes combined.

Death rate ratios for vegetarians compared with non-vegetarians, adjusted for age, sex and smoking, were calculated for each of the 5 studies and then combined to give an 'all studies' DRR. The results for all-cause mortality were as follows:

The results of the collaborative analysis suggest that vegetarians may have a lower overall mortality than comparable non-vegetarians (by about 5%), perhaps enough to confer a slightly greater life expectancy.

In the same analysis, vegetarians were found to have a significantly lower mortality for ischaemic heart disease (heart attack), but not for other common causes of death, as shown below:

A recent re-analysis of mortality data from the two British studies found no differences in overall death rates between vegetarian and non-vegetarian participants. The researchers concluded that the low mortality of British vegetarians compared with the general population "may be attributed to non-dietary lifestyle factors such as a low prevalence of smoking and a generally high socio-economic status, or to aspects of the diet other than the avoidance of meat and fish".

A more positive outcome for vegetarians was found in a recent analysis of data from the Adventist Health Study which predicted life expectancy in Seventh-day Adventists following different behaviour patterns. The researchers found that a combination of different lifestyle choices could influence life expectancy by as much as 10 years. Among the lifestyle choices investigated, a vegetarian diet was estimated to confer an extra 1½ to 2 years of life. The researchers concluded that "the life expectancies of California Adventist men and women are higher than those of any other well-described natural population" at 78.5 years for men and 82.3 years for women. The estimated life expectancies of vegetarian California Adventists were 80.2 years for men and 84.8 years for women. Other beneficial lifestyle choices included high nut consumption and a high level of exercise.

In conclusion

• vegetarians have low mortality compared with the general population
• much of this benefit is attributable to non-dietary lifestyle factors such as the avoidance of smoking and a high socio-economic status
• vegetarians have similar mortality to comparable non-vegetarians, although a vegetarian diet may confer an additional 1-2 years of life (at least among US Adventists).

Of course, life expectancy is not the only measure of health status, and other studies have suggested that vegetarians may enjoy a number of health benefits including being generally slimmer and having lower blood cholesterol levels than non-vegetarians.

So, should you be a vegetarian for health reasons? When asked this question, the Nobel-prize winning author Isaac Bashevis Singer replied: "Yes, for the health of the chicken! "

References:

• Key et al. Mortality in vegetarians and non-vegetarians:detailed findings from a collaborative analysis of 5 prospective studies. Am J Clin Nutr 1999; 70(suppl): 516S-24S.
• Key et al. Health benefits of a vegetarian diet. Proc Nutr Soc 1999; 58: 271-5.
• Fraser & Shavlik. Ten years of life - is it a matter of choice? Arch Intern Med 2001; 161: 1645-52.
• Appleby et al. Mortality in British vegetarians. Public Health Nutr 2002; 5: 29-36.

Paul Appleby

[ OxVeg Front Page ]

Oxford Vegetarians Web site is hosted by the International Vegetarian Union.

This page created 16 March 2002 by Paul Appleby.