Main results from the Oxford Vegetarian Study

The Oxford Vegetarian Study is a nationwide long-term study of the health of 6000 persons who do not eat meat (mostly vegetarians, but including some persons who eat fish) and 5000 meat-eating controls. Vegetarian subjects were recruited with the help of The Vegetarian Society, through vegetarian and health food magazines, and via the media. The control group consists of meat-eating friends and relatives of the non-meat eaters, matched on age and gender. Recruitment to the study took place between September 1980 and January 1984.
Volunteers were sent an initial questionnaire requesting information about their diet, lifestyle (smoking, drinking, work and exercise) and previous medical history. Later, volunteers under the age of 70 were asked to provide a blood sample, and in 1985-86 all participants were sent a dietary record form which they were asked to complete on two weekdays and both weekend days. As a result, blood cholesterol measurements are available for nearly 3800 participants and detailed dietary intake data are available for over 5000 participants.
Diet and serum cholesterol concentration
Total, low density lipoprotein (LDL) and high density lipoprotein (HDL) cholesterol levels were compared in each of four diet groups (vegans, vegetarians, fish eaters and meat eaters) in a paper published in 1987. Total and LDL cholesterol concentrations were both significantly lower in the vegans than in the meat eaters with the vegetarians and fish eaters having intermediate and similar values. HDL cholesterol concentration was similar in all four diet groups. The differences suggested that the incidence of coronary heart disease may be 24% lower in lifelong vegetarians and 57% lower in lifelong vegans than in meat eaters. A subsequent analysis of the diets of a sample of 208 participants (52 from each diet group) showed that the nature rather than the quantity of dietary fat is an important determinant of cholesterol concentration, with health-conscious individuals selecting a fat-modified rather than a low fat diet.
Plasma lipids and lipoprotein cholesterol concentrations in people with different diets in Britain. Thorogood M, Carter R, Benfield L, McPherson K, Mann JI. British Medical Journal 1987;295:351-353.
Dietary intake and plasma lipid levels: lessons from a study of the diet of health conscious groups. Thorogood M, Roe L, McPherson K, Mann J. British Medical Journal 1990;300:1297-1301.
Diet and mortality
A comparison of the mortality (death rate) of meat eaters and non-meat eaters after 12 years of follow-up was published in the British Medical Journal on 25 June 1994. The results show that after adjusting for smoking habits, body mass index (a measure of obesity) and social class (three factors known to influence mortality), the non-meat eaters experienced lower death rates for all causes of death combined, ischaemic heart disease and all cancers combined, the differences being statistically significant (meaning that they were unlikely to have arisen by chance) for all causes and all cancers combined. Only time will tell whether these differences are real or a manifestation of the 'healthy volunteer effect' (the vegetarian volunteers may have been better motivated and therefore healthier than their non-vegetarian controls). If real, it will be interesting to see whether the lower mortality for all cancers combined applies 'across the board' or only to certain cancers which other studies have suggested vegetarians may be less likely to suffer from. Eventually, it may even be possible to determine which features of the vegetarian diet account for the lower mortality.
Risk of death from cancer and ischaemic heart disease in meat and non-meat eaters. Thorogood M, Mann J, Appleby P, McPherson K. British Medical Journal 1994;308:1667-1671.
Emergency appendicectomy and meat eating
Rates of reported emergency appendicectomies (indicating acute appendicitis) were compared according to participants' history of meat consumption in a paper published in the Journal of Epidemiology and Community Health. Participants were grouped according to whether they had always eaten meat, never eaten meat or stopped eating meat. The percentage who reported an emergency appendicectomy was higher among the lifelong meat eaters (10.7%) than either the lifelong non-meat eaters (7.8%) or the former meat eaters (8.0%); and the operations were performed at an earlier age in the first group (average age at operation 18.9, 26.0 and 19.6 years respectively). The overall age-adjusted emergency appendicectomy rate ratio comparing participants who did not eat meat with those who did was 0.47, suggesting that vegetarians have a 50% lower risk of requiring an emergency appendicectomy than non-vegetarians.
Emergency appendicectomy and meat consumption in the UK. Appleby P, Thorogood M, McPherson K, Mann J. Journal of Epidemiology and Community Health 1995;49:594-596.
Effects of diet, lifestyle and physical characteristics on cholesterol concentration
The effects of dietary, lifestyle and physical factors on participants' concentration of total and high density lipoprotein (HDL) cholesterol in the blood was investigated in a paper published in the Journal of Human Nutrition and Dietetics. High total cholesterol concentration is associated with an increased risk of coronary heart disease, whereas HDL cholesterol is thought to have a protective effect against heart disease. After adjusting for age there was a progressive decrease in total cholesterol concentration from meat eaters to vegans for both male and female participants, with vegetarians having intermediate values. In contrast, HDL cholesterol concentrations were similar in all diet groups for both men and women. When the effects of specific dietary and lifestyle factors were considered meat and cheese consumption were found to increase total cholesterol, whereas a high intake of dietary fibre was associated with a reduction in total cholesterol for both men and women. In accordance with results from other studies, body mass index (a measure of obestity) and alcohol consumption were associated with a decrease and increase respectively in HDL cholesterol concentration in both men and women. The results provide further evidence of the cholesterol-lowering effect of a vegetarian diet with a high dietary fibre content and limited use of cheese. Excluding meat from the diet might be expected to result in a 15-25% reduction in the risk of coronary heart disease, with a similar beneficial effect if cheese is also excluded.
Associations between plasma lipid concentrations and dietary, lifestyle and physical factors in the Oxford Vegetarian Study. Appleby PN, Thorogood M, McPherson K, Mann JI. Journal of Human Nutrition and Dietetics 1995;8:305-314.
Diet and ischaemic heart disease
Ischaemic heart disease is the most common cause of death in the UK and in many other developed countries. A previous analysis of data from the study showed that non-meat eaters had a 28% lower risk of death from ischaemic heart disease before age 80 than meat eaters after adjusting for differences in smoking habits, body mass index and social class. However, it was unclear which dietary factors might account for this difference. Therefore, the effects of various dietary factors on mortality from ischaemic heart disease were examined in a recent analysis. Participants were grouped not just according to their diet (meat eater, semi-vegetarian, vegetarian/vegan), but also by their consumption of various foods and alcohol according to the answers provided on the recruitment questionnaire. Participants were also divided into thirds of estimated intake of total fat, saturated fat and dietary cholesterol from land animal sources, and into thirds of estimated dietary fibre intake from their reported consumption of fibre-rich foods. The main analysis was restricted to participants with no history of cardiovascular disease or diabetes at recruitment, of whom 392 died before age 80, including 64 deaths from ischaemic heart disease. After adjusting for differences in age, sex, smoking habits and social class, vegetarians and vegans had a 17% lower risk of death from ischaemic heart disease than meat eaters (defined as participants eating meat at least once a week) although the result was not statistically significant. The most striking results from the analysis were the strong positive and highly significant associations between increasing consumption of animal fats and ischaemic heart disease mortality, the death rates being roughly three times greater among participants in the highest third of intake compared with the lowest third of intake for each of total animal fat, saturated animal fat, and dietary cholesterol. Consumption of eggs and cheese were also positively associated with ischaemic heart disease mortality, but no protective effects were noted for dietary fibre, fish or alcohol consumption, as might have been expected from findings in other studies.
Dietary determinants of ischaemic heart disease in health conscious individuals. Mann JI, Appleby PN, Key TJ, Thorogood M. Heart 1997;78:450-455.
Diet and body mass index
Body mass index (BMI - a measure of an individual's relative weight calculated by dividing their weight in kilograms by the square of their height in metres) was calculated for 1914 male and 3378 female non-smokers aged 20-89 at recruitment to the study. Answers to the food frequency questionnaire completed by all participants were used to classify them as either meat eaters or non-meat eaters, and to estimate intakes of animal fat and dietary fibre. Participants were further classified according to their alcohol consumption, exercise level, social class, past smoking habits and parity (whether or not a woman has had children). Mean (average) BMI was lower in non-meat eaters than in meat eaters in all age groups for both men and women. Overall mean BMIs in kg/m2, adjusted for differences in age between the two groups, were 23.18 and 22.05 for male meat eaters and non-meat eaters respectively and 22.32 and 21.32 for female meat eaters and non-meat eaters respectively, differences of about one kg/m2 which were massively statistically significant, indicating that they were very unlikely to have occurred by chance. In addition to meat consumption, dietary fibre intake, animal fat intake, social class, past smoking habits, alcohol consumption in men only and parity in women were all independently associated with BMI. However, these factors accounted for only about one-third of the difference in mean BMI between meat eaters and non-meat eaters in both men and women. It was concluded that non-meat eaters are thinner than meat eaters, and that this may be partly due to a higher intake of dietary fibre, a lower intake of animal fat, and in men only a lower intake of alcohol.
Low body mass index in non-meat eaters: the possible roles of animal fat, dietary fibre and alcohol. Appleby PN, Thorogood M, Mann JI, Key TJ. International Journal of Obesity 1998;22:454-460.
A summary of the published papers up to and including 1997 may be found in the following reference:
The Oxford Vegetarian Study: an overview. Appleby PN, Thorogood M, Mann JI, Key TJ. American Journal of Clinical Nutrition 1999; 70(suppl): 525S-531S.
(Paul Appleby, July 1994, April 1996 and July 1998)
[Note that there is no formal link between Oxford Vegetarians and the Oxford Vegetarian Study, although some past and present members of Oxford Vegetarians are participants in the study.]