Many of us believe that vegetarianism gives us the healthiest diet.
But how does it stand up to scientific scrutiny as the optimum diet for everyone?
We, like many of our readers, support the many issues surrounding a vegetarian lifestyle. There are many important health and moral issues here, and there is no doubt that an increasing number of studies being performed prove the extraordinary health supporting nature of plant foods. However, what never gets asked is the simple question: is there proof that vegetarian is the optimum diet for everyone? What Doctors Don't Tell You asked contributing editor Pat Thomas to scour the scientific literature on this issue in effect, to subject vegetarianism to the same detailed scrutiny under which we put aspects of conventional or alternative medicine.
We present the results with the caveat that it represents what we can say for certainty about the effect on health and longevity of a vegetarian diet versus a meat eating diet. Our article does not examine the copious moral issues or the equally important issue of the decreasing quality and increasing contamination of our foodstuffs with hormones and chemicals. But see Viewpoint, p 6 Lynne McTaggart.
There are many positive aspects of adopting a vegetarian diet. For some who are at a specially high risk of heart disease or cancer, it may strengthen and protect the body.
What we do know is that diets rich in brightly coloured fruits and vegetables, and polyunsaturated (and particularly monounsaturated) oils, such as the Mediterranean diet, have been shown, through good research, to reduce the risk of heart problems and stroke. They may even reduce the risk of certain cancers and prolong life. But does the current data support the hypothesis that a wholly vegetarian regime constitutes an "optimum" diet?
Research into vegetarianism can be flawed. Firstly it usually involves self selected populations. Secondly its focus tends to be very narrow. The hypothesis of vegetarian versus non vegetarian may lead to a conclusion that a vegetarian diet is healthier in much the same way that a study into thalidomide versus no thalidomide would have proved that controls experienced greater morning sickness than thalidomide users. Not surprisingly, research which concludes that vegetarian regimes are "best" usually contains the proviso that vegetarians are by their own admission health conscious individuals who may be involved in other regular practices, such as exercise, which have an equally beneficial effect on their health.
No firm conclusions can be made about death from heart disease among vegetarians and meat eaters, though many have tried. Indeed, in many of the studies into diet and heart disease, it is the inclusion of fruits and vegetables, rich in antioxidants such as vitamins A, C and E, rather than the exclusion of meat, which has been the decisive factor (BMJ, 1996; 312: 1479; BMJ, 1996; 312: 478-81; JAMA, 1995; 274: 1197).
In one widely quoted study by Thorogood (BMJ, 1994; 308: 1667-70), 6,115 self selected non meat eaters were compared with 5,015 meat eaters over 12 years. Meat eaters were nearly twice as likely to die from heart diseases and more likely to die from cancer than non meat eaters. Nevertheless, the study design offered an absurdly elastic definition of what constitutes a vegetarian. Of the 6,115 vegetarians (referred to in the study as non meat eaters), 5,728 were defined as those "who did not eat meat or fish, or ate these less than once a week, but did eat eggs or dairy products or both, or vegans". A further 387 ate fish at least once a week, either abstained from meat altogether or ate it less than once a week and may or may not have eaten eggs or dairy products.
By the study's definition, a person who ate meat once every two weeks or once a month would be a vegetarian. Likewise, someone who ate fish would be a vegetarian.
Other factors such as the "healthy volunteer effect" came into play in this criticised study. Recruitment of individuals was through health food shops, the Vegetarian Society and the media. The vegetarians were then asked to nominate a meat eating relative or friend to compare themselves with. As one commentator points out, people enthusiastic about their vegetarian lifestyle may have volunteered in order to prove a point (BMJ, 1994; 308: 1671). Equally, their meat eating friends and relatives may not have been as health conscious as themselves. Here the problems of self selection become apparent. The authors did what they could to compensate for these problems but admitted that direct comparison between meat eaters and vegetarians was not really possible. They also concluded that there is no data to justify the advice to "exclude meat from the diet since there are several attributes of a vegetarian diet, apart from not eating meat, which might reduce the risk".
Some commentators believed that Thorogood's study did not last long enough to conclude properly that the death rate was significantly different in the two groups. Had he extended the study another five years, they argued, the "healthy volunteer effect" would have worn off and figures would be more reliable (BMJ, 1994; 309: 955). Certainly, when other researchers have compared death from all causes in vegetarians and non vegetarians, the death rate did not alter appreciably after the exclusion of early deaths (during the first years of observation) (Am J Clin Nutr, 1982; 36: 873-7).
Perhaps in response to this, Thorogood and associates published a reanalysis of the same group three years later. They noted that the difference in death rate due to cancer and heart disease between non meat eaters and meat eaters had narrowed significantly in just two years (from 30 per cent to 15 per cent). Across the board, the only significant factor in lowering mortality from all causes was daily consumption of fruit. Health consciousness, not vegetarianism, was the decisive factor (BMJ, 1996; 313: 775-9).
Their main findings were consistent with other studies, many of which show that an increased intake of foods rich in vitamin C and beta carotene are the significant factors in reducing death from heart disease and stroke (Am J Epidemiol, 1995; 142: 1269-78; Epidemiol, 1992; 3: 194-202; BMJ, 1990; 300: 771-3; Am J Epidemiol, 1994; 139: 1180-9; Lancet, 1983; i: 1191-3; JAMA, 1995; 273: 1113-7).
As for lowering cholesterol levels, studies show that adding fish to the diet can also bring significant improvements. In one study conducted in 1960, 852 healthy middle aged men were selected, and dietary histories were obtained with special attention to fish consumption. Some 20 per cent ate no fish; the rest ate on average three quarters of an ounce per day. The men were followed for 20 years, during which time 78 died of coronary disease. The conclusion of the research was that the more fish the men ate, the lower their risk (N Eng J Med, 1985; 312: 1205-9). In this study, eating just an ounce of fish a day reduced the risk of heart disease by 50 per cent. The influence of the long chain fatty acid EPA on clotting tendencies was one of the likely reasons for the result, the authors said.
In the same issue of the Journal, a smaller Dutch study of fish consumption placed 20 patients on three successive controlled diets, differing only in the kind of fat included. The first diet was low in fat and cholesterol and contained no fish oil; the second contained fish oil (in fish and in supplements) amounting to 20 to 30 per cent of daily calories. The third diet was identical to the second diet, but substituted an equal amount of polyunsaturated vegetable oils for the fish oil. In every patient on the fish oil diet, blood fat levels dropped, as did cholesterol levels (N Eng J Med, 1985, 312: 1210-16). Surprisingly, when the patients were taken off the fish oil rich diet and placed on a diet rich in polyunsaturated vegetable oils, levels of both triglycerides and cholesterol rose considerably. While vegetable oils are known to reduce these blood fats considerably, they proved less effective than fish oils.
Because earlier studies had demonstrated the benefits of adding a little polyunsaturated oil to the diet, we have taken on board the idea that a lot will do us a lot more good. But more recent studies have shown that while increasing consumption of vegetable oils and foods rich in vegetable oils may lower the overall cholesterol level of the blood, they do not lower the level of more harmful LDL cholesterol.
Often when talking about "good" diets we fail to make the distinction that the fat which is on the meat when purchased does not necessarily end up on the plate or in the mouth. Studies conclude that it is beef fat, not beef itself which is responsible for elevated cholesterol levels and a greater risk of heart disease (Am J Clin Nutri, 1990; 19: 491-4).
In the Mediterranean diet, meat is used as a "condiment". Studies confirm that for meat eaters, small amounts are all that is needed to provide optimum nourishment. For instance, in a study of 504 young adult meat eaters, those with a low to moderate intake of meat (less than the 25th percentile) came closest to meeting dietary recommendations with 11 per cent of energy from protein, 55 per cent from carbohydrate, 32 per cent from fat, 11 per cent from saturated fatty acids and 264mg dietary cholesterol (J Am Diet Assoc, 1995; 95: 887-92).
Lowering Your Cancer Risk
It is widely accepted that between 30 and 40 per cent of cancers are caused by dietary factors. But not all dietary factors are factors of excess. Although cancers of the colon, rectum, breast and prostate are conditions linked to dietary excess, cancers of the oesophagus, gastric tract, throat and liver are often associated with dietary deficiencies.
A report published by the World Cancer Research Fund (WCRF) entitled Food, Nutrition and the Prevention of Cancer: A Global Report, recommended that reducing an individual's intake of red meat to 80g per day or 150g three times a week would significantly reduce the incidence of colorectal cancer. These figures have been widely accepted as a healthy level of meat intake, but whether they can actually cut the risk of colorectal cancer has yet to be proven.
Comparing vegetarian and non vegetarian populations, there is evidence that the incidence of colorectal cancer among non vegetarian Mormons is lower than that of vegetarian Seventh Day Adventists (N Eng J Med, 1976; 294: 132). While the high fibre content of the vegetarian diet is thought to exert a protective effect, an analysis of 13 case controlled studies of the relationship between fibre intake and colorectal cancer showed that low fibre intake was not a significant risk factor in the development of the disease (Epidemiol, 1994; 5: 66-79).
Although reviews in America and Australia have showed a positive correlation with meat intake and the development of colorectal cancer (Eur J Cancer Prev, 1991; 2: 13-20), reviews of European populations show no such association (Eur J Cancer Prev, 1997; 6: 415-17).
The most recent epidemiological study into the relationship between bowel cancer and meat eating also showed no such relationship (Pub Health Nutri, 1998; 12: 31-44). The analysis pooled data from five studies looking at mortality from all causes among 76,000 men and women, 28,000 of whom were vegetarians. After 11 years of follow up, there were 8,330 deaths from all causes. While vegetarians who had followed their regime for more than five years were 35 per cent less likely to die from the condition, they were no less likely to develop bowel cancer in the first place. The research also showed that the risk of developing cancers of the lung and prostate was not significantly different between meat eaters and non meat eaters. Thus, while a plant based regime may help you fight off disease more efficiently once it strikes, it may not stop you from getting it in the first place.
Currently the UK has the lowest intake of red meat in the European Union but one of the highest incidences of colorectal cancer. The four countries with the lowest mortality from colorectal cancer, Greece, Italy, Spain and Portugal, all have a higher intake of red meat. But they also have a much higher intake of protective fruits, vegetables and cereals. This suggests that increasing protective dietary factors may carry more weight than cutting out purported "causal" factors such as red meat (Nutri Bull, 1998; 23: 79-83).
Many of the longest living people in the world are not necessarily vegetarian. Of 1,200 people in one survey who reached the century mark between 1932 and 1952, only four were vegetarians (See
O Segerberg, Living to be 100: 1200 Who Did and How They Did It, New York: Charles Schribner & Sons, 1982). Compare meat eating Mormons to vegetarian Seventh Day Adventists, and you will find similar mortality rates (Arch Anat Cytol Pathol, 1992; 40: 73-8). Both groups have a life expectancy which is two to four years greater than the average population. Religious groups, which have lower mortality rates than the general population, may also have lifestyles that include other healthful practices, such as a firm spirituality; they may also be more emotionally or spiritually stable.
In his book Native Nutrition (Vermont: Healing Arts Press, 1987) Ronald F Schmid details the lifestyles of three long living cultures: the Vilicamba of the Andes Mountains, in southern Ecuador; the Hunza people who live on the borders of China and Afghanistan; and the residents of Georgian Russia. What these cultures all have in common is an active and emotionally rewarding lifestyle and a hunter gatherer diet which consists of natural, seasonal and locally available foods, including meat. In these cultures, the chronic diseases which plague our civilised culture are unknown.
In terms of longevity, exercise has been shown to be at least of equal importance as diet in maintaining long term health. Studies of Harvard alumni over 30 years show that the more physically active an individual was, the lower the risk of early mortality from all causes, including heart disease and cancer (N Eng J Med, 1986; 314: 605-13).
Diet is inarguably important to our health. Many people thrive on vegetarian diets and would enthusiastically attribute their good health status to this regime. But published evidence does not yet suggest that a solely plant based diet is the optimum diet for everyone. And some vegetarians may be as prone as anyone else to making poor nutritional choices. All that we can say with any certainty at the moment is that there are, it seems, many pathways to good health. It may be that there is no single optimum diet which works for everyone.