When people with multiple sclerosis ask their doctors about whether diet has a role in the management of MS, they are often told there is no evidence to support it. This is a remarkable statement. A comprehensive research effort from medical scientists for over half a century has produced good evidence supporting the role of diet in managing MS, from laboratory and animal research, human case reports, epidemiological, case-control, and uncontrolled intervention studies and finally randomized controlled trials. Interestingly there has been virtually no contradictory evidence published. There is insufficient room here to document all this research, but interested readers are directed to my book Overcoming Multiple Sclerosis, in which I outline all the major studies in the field.
The major contribution in my view is the seminal work of pioneer neurologist Dr Roy Swank. He initially made the observation that the incidence of MS seemed to follow the consumption of saturated fat, particularly dairy products, and was lower wherever fish consumption was high. Subsequently in 1949, he enrolled 150 MS patients in a meticulous 34 year study, managing these patients with a very low saturated fat diet. The study was supported by grants from the MS Society of Canada, the Montreal Neurological Instititute, the Department of Health and Welfare of Canada, the MS Society of Portland and other grants. 
The study was published in one of the world's premier medical journals, the Lancet in 1990. Of 144 patients, 72 stuck to the diet (good dieters), that is they consumed less than 20g/day of saturated fat. The results were dramatic. Regardless of level of disability at entry to the trial, good dieters did not deteriorate significantly. The results were best for those who started with minimum disability, with 95 per cent surviving and still physically active 34 years later, excluding those who died from non-MS diseases. The benefits occurred in all groups; even people with significant disability markedly slowed the progression of the disease if they stuck to the diet.
The picture for poor dieters was terrible, in line with results from other series of MS patients whose conditions have been monitored over time. Only 7 per cent of patients who did not stick to the diet remained active. The death rate amongst the poor dieters was extremely high. 58 of the 72 were dead by 34 years, 45 from MS-related causes. There was a dramatic fall in relapse rate after commencement of the diet, from around 1 relapse per year pre-diet to about 0.05 per year once stable on the diet.
So why do many doctors say there is no evidence for diet in MS in the face of the large body of literature, but particularly this remarkable piece of research? Many will say that Swank's study does not meet modern standards for what is considered evidence. But it is important to remember that in medicine there is actually no such thing as proof. We analyse data, trying to remove as many sources of bias as possible in the study design, then make inferences from the data. These days, for the highest form of medical evidence, we usually use a design where we randomly allocate patients to treatment groups, then apply a treatment with both patients and doctors being unaware of which group gets which particular treatment. Often one of the ‘treatments' is an inactive placebo. This study design is called a randomized controlled trial (RCT). But it is important to remember that there are many study designs apart from RCTs, and we often use other levels of evidence to make recommendations to patients.