RCTs, health claims, and common sense deficiency syndrome (CSDS)

The limitations of applying a pharmaceutical-style model to foods are well known in the nutrition science community, and have been explored at length on NutraIngredients-USA. Click here​ and here​.

Insistence on RCTs will lead to stagnation​

However, regulators continue to regard RCTs as the gold standard of scientific evidence, said Dr Robert Heaney, professor of Medicine at Creighton University, who was speaking at the recent IFT Wellness 2012 conference.​

Heaney, who was followed on the podium by Dr Jeffrey Blumberg, Professor, Friedman School of Nutrition Science and Policy at Tufts University, and Duffy MacKay, VP, scientific and regulatory affairs at the Council for Responsible Nutrition, said policy makers must now accept that intervention trials designed to test the efficacy of powerful drugs on sick people are not always suitable for determining the benefits of nutrients.

He added: “Progress can’t be made until the nutritional policy establishment accepts this fact and agrees to alternatives to RCTs.​

“To continue to insist upon RCT data for validation​ [of the health benefits of nutrients] will guarantee not certainty but stagnation… The research question ​[for nutrients as opposed to drugs] is not ‘whether’ but ‘how much?’ and ‘for what endpoints?’ But the RCT is poorly suited to answer such questions.​

“We should move to the use of physiological criteria of adequacy or adopt a least harm strategy.”​

You can’t test a hypothesis of harm​

If the problem is deficiency, as is frequently the case with nutrients - unlike drugs – he pointed out, supplementing the diet of healthy people that already have sufficient intakes of the nutrient cannot be expected to deliver spectacular results. But that doesn’t mean that the nutrients in question confer no benefit.

If you go without vitamin C for an extended period, you’ll get scurvy. But it would be unethical – and unfeasible – to stage a lengthy trial to prove this in which a control group is denied vitamin C for six months.

“How do you do a RCT to prove you need vitamin D in pregnancy?” ​said Heaney.  

“You can’t split them in two and feed half high levels and not give any to the other half. That’s not ethical. You can’t test a hypothesis of harm.”​

Poorly designed trials continue to be included in systematic reviews​

And yet meta-analyses continued to include results from intervention trials deemed to have negative results, when in fact, they were simply null results, because the healthy participants volunteering for such trials already had adequate basal levels of the nutrient in question, he said.

“Yet studies that are badly designed ​continue to be included in systematic reviews.”​

A multi-system assessment of total body health?​

Dr Blumberg said the failure to find a solution to this problem is directly “impeding the application of nutrition research to public health issues”.​

Additional surrogate endpoints for disease and wellness are urgently needed, he said: “You need biomarkers, but can we look at things like a multi-system assessment of total body health, or a global health index?​

“The challenge is that nutrients act in pan-systematic ways that affect overarching processes. Inflammatory stress, metabolic stress, and so on.​

“We need to look at omics approaches where you are moving beyond looking at one or two specific pathways and biomarkers and looking at 100s if not thousands.”​

You can’t have a true placebo with food​

The problem, said the CRN’s MacKay, is that drugs are typically designed to deliver a big effect in a short time on a single pathway, enzyme or or receptor, whereas nutrients are designed to be consumed in smaller doses and deliver their beneficial effects across multiple pathways, cells and tissues over years or even decades, often in combination with other nutrients.

Meanwhile, control groups in nutrition and drug trials were not the same, he said. “You can’t have a true placebo with food – you can’t have people in a nutrient free state.”​

Currently, there are still only a handful of validated biomarkers that regulators will consider in nutrition research, such as blood cholesterol, blood sugar, blood pressure and bone density, he said.

“The FDA health claims are all pretty old now, and it’s proving to be more and more difficult to get claims, so the industry is kind of giving up. People won’t make the investment unless the ROI is there.”​

We need to think outside the box​

He added: “We need to think outside the box or this field will stagnate.”​

One helpful approach would be to look towards current strategies in obesity research which use all forms of useful evidence—including expert opinion— and factored in risks, he said.

"If these ingredients are not risky why are treating them like drugs?”​

He added: “In the real world, we need a more nuanced approach that honors multiple disciplines and embraces complexity.​

“If your doctor relied on RCTs to give you medical advice, he would not advise you to not smoke... not drink enough water... and not use a parachute when jumping out of an airplane.​

“Over-reliance on RCTs leads to CSDS: Common Sense Deficiency Syndrome.”​