Junkfood Science: Say it isn’t so — Part one

November 25, 2006

Say it isn’t so — Part one

It is predictable. A flurry of press releases and articles appear everytime a major study is released that debunks “pop science” — what everyone “knows” to be true — and threatens the research or agendas of special interests. These stories attempt to confuse us, spin the science, or restate the beliefs more emphatically. Many times, they even claim to have new research supporting their position, when they don’t!

Earlier this week, the long-awaited results of the 13-year clinical trial, the Women's Antioxidant Cardiovascular Study (WACS), were released. It followed nearly 16 other major clinical trials released earlier — including the Heart Outcomes Prevention Evaluation (HOPE), Atherosclerosis Folic Acid Supplementation Trial (ASFAST), and the Norwegian Vitamin Trial (NORVIT) — all firmly demonstrating no benefit of folic acid or vitamin supplementation in heart disease, and that lowering homocysteine levels does not translate to real-life reductions in heart disease. And more importantly, has no effect on the most important end point of all: death.

As a Medscape review noted, the clinically measureable evidence is overwhelmingly consistent. According to Dr. Joseph Loscalzo of Brigham and Women’s Hospital, Harvard Medical School, Boston, “the consistency of the results of HOPE-2, NORVIT, and VISP in three similar, although not identical populations, ‘leads to the unequivocal conclusion that there is no clinical benefit of the use of folic acid and vitamin B12 (with or without the addition of vitamin B6) in patients with established vascular disease....[the] incorrect view that folic acid can decrease homocysteine levels and thus reduce the risk of atherosclerosis effectively may be an unintended consequence of oversimplifying a complicated metabolic network.’” [We'll look at even stronger, more comprehensive reviews later.] Bottom line: Believing that vitamins can prevent heart disease doesn't make it so.


On Friday, the press reported the opposite of what we heard last week:

Folic acid can cut heart attack risk: experts

Can taking folic acid supplements reduce the risk of heart disease and stroke? British researchers believe it can. After analyzing evidence from earlier studies, a team of scientists in Britain said on Friday there is enough research that shows folic acid lowers levels of the amino acid homocysteine and reduces the odds of cardiovacular [sic] disease.

David Wald is quoted as saying “The evidence is very persuasive that lowering homocysteine with folic acid will lower your risk of heart attack and stroke by about 10-20 percent.”


The source of this claim is actually a paper published in November 2002 in the British Medical Journal by David Wald. It was a meta-analysis, looking for correlations among data from a multitude of diverse population studies. Its analysis claimed to have found “significant associations between homocysteine” and the risks for heart disease and stroke.

He reported the risks as odds ratios, which were 42% and 65%, respectively. But these are not tenable numbers for these types of studies. (See: Have your steak and enjoy it too!)

Yet this researcher went on to make another fundamental error: that of thinking associations can ever show cause. He even titled his paper “Homocysteine and cardiovascular disease: evidence on causality from a meta-analysis.” He concluded the “significant” results were “strong evidence that the association between homocysteine and cardiovascular disease is causal.”


His belief in homocysteine as a risk factor for heart disease, and that alone, was the basis for his conclusion that “lowering homocysteine concentrations by 3 µmol/l from current levels (achievable by increasing folic acid intake) would reduce the risk of ischaemic heart disease by 16%... and stroke by 24%.”

Given he found untenable associations, not surprisingly actual clinical intervention trials have consistently proved them to be false.

Wald has continued to be the most outspoken proponent for folic acid in the medical literature. In a May 2003 article in BMJ with the dramatic title “Inertia on folic acid has caused thousands of unnecessary deaths,” Wald is quoted as saying: “Until public health agencies fortify our diet with sufficient folic acid, people should consider taking an 800 µg supplement each day, especially if they have coronary heart disease.”

In June of 2003, Wald and a colleague of the Wolfson Institute of Preventive Medicine in London, Malcolm Law (professor at the University of London and University of Auckland in New Zealand), authored another paper in the BMJ: “A strategy to reduce heart disease by more than 80%.”Using statistical analysis of meta-analyses to support potential effects of their new polypill, they estimated that, if taken by everyone aged 55 and older with heart disease, it could prevent nearly nine out of ten heart attacks and 4 out of 5 strokes. This cocktail of six drugs in low dosages — three blood pressure drugs, a statin, aspirin and folic acid — would address what they believed to be cardiac “risk factors.”

They even made the sensational statement that their new pill would have “a greater impact on the prevention of disease in the Western world than any other single intervention.”

In the October 2003 issue of BMJ, Wald and colleagues respondedto those skeptical of their polypill claims, saying (with no supportive evidence) that the reductions in risk factors will be greater when the drugs are used together in low doses than has been shown individually.

They used the popularity of the idea among the public to support their idea. [How many were aware that these authors all have interests in the polypill, with patents granted and pending? When the zeal of claims outpace the strength of the evidence, it is only right to note potential conflicts of interest.]

Responding to the negative feedback received from the medical community, the authors said that their critics “had not given reason or evidence against the concept of the polypill.” [Scientists and quack watchers will recognize this classic argument. In science, the burden of proof falls upon those making sensational claims to prove their claims, not on others to prove a negative.]

So today, the BMJ published an “Analysis and Comment” from David Wald and associates — not a new study with new clinical evidence or any new evidence at all, but another review with the same arguments as earlier papers. Attempting to discredit the clinical evidence was clear in the article’s title: “Folic acid, homocysteine, and cardiovascular disease—judging causality in the face of inconclusive trial evidence.” They supported their conjectures using that same earlier observational population study, reports of higher homocysteine levels in people with a rare genetic disorder and a certain genetic mutation and correlations with heart disease, and statistical interpretations of meta-analyses. While admitting these sources were not randomized controlled clinical trials and suffered from publication bias and other errors, those concerns were dismissed because of the “significance” of the effect of homocysteine on heart disease — not based on actual clinical outcomes, but again on “risks” as defined by homocysteine levels. The 7 (out of many more, some showing adverse effects) randomized clinical trials they selected to review, all showed no measureable effect on relative risks. Yet those studies were also negated, essentially because they didn’t concur with their statistical interpretations and didn't look at as large of numbers as the meta-analyses. [The size of a data base used in a data dredge does not determine soundness.] They asserted that in the clinical trials “the new maximal reduction in ischemic heart disease events” had simply not yet been “appreciated.”

But this is a minor example of science and media spinning information to support a certain belief or marketing interest…..

Bookmark and Share