USC-Oxford Study on Diabetes

USC-Oxford Study Alleging HFCS-Diabetes Link Is Flawed and Misleading

A study released November 2012 by researchers from USC and Oxford uses a severely flawed statistical methodology and ignores well established medical facts to “suggest” a unique link between high fructose corn syrup (HFCS) and Type 2 diabetes.  A previous study critical of HFCS from the lead author, Dr. Michael I. Goran, also met with severe criticism.  Most importantly, Dr. Goran’s newest attack on HFCS fails to account for widespread agreement among scientists and medical doctors that HFCS and sucrose (table sugar) are nutritionally equivalent.  In early December, the Food & Drug Administration confirmed that it is unaware of any evidence of a difference in the safety of HFCS in foods and sucrose (table sugar) in foods. (See FDA).

Independent Experts Immediately Challenge the Study's Design and Conclusions

In news reports, leading research scientists and experts challenged or raised questions about how the study was conducted and its leap to suggest causation between HFCS and diabetes.

  • “Dr. Robert Lustig, a professor at UC San Francisco and director of the Weight Assessment for Teen and Child Health Program there, noted that the researchers did not show that higher consumption of high fructose corn syrup caused the increase in diabetes, only that there was a link between the two.” – LA Times
  • “Canadian researcher John Sievenpiper at St. Michael’s Hospital backs the [HFCS] industry’s case.  ‘When you isolate the effects of fructose you don’t see that it behaves differently than other forms of calories,’ he said.” – CTV
  • “Lona Sandon is a registered dietician and assistant professor of clinical nutrition at the University of Texas Southwestern Medical Center at Dallas. She said that, ‘no matter what the added sugar is -- honey, agave, sugar, or high fructose corn syrup -- we should not be consuming it. In fact, honey and agave syrup have higher levels of fructose than high fructose corn syrup, but people think they're somehow better for us because they're 'natural.'” – Health Day (as reported in US News & World Report)

James M. Rippe, M.D., Professor, BioMedical Sciences, University of Central Florida (and consultant to CRA) added:  "Diabetes is a complex disease with many underlying factors.  It is highly unlikely that one component of the diet is uniquely related to diabetes. There are well-established links between obesity and diabetes. That is where we should be focusing our attention rather than vilifying one component of the diet."
The Study's Chief Author Admits to Flaws and that HFCS and Sugar are "Almost Identical"

In media interviews, Dr. Goran conceded that his study has limitations and criticisms by independent experts are valid.  Dr. Goran acknowledged to Food Navigator that “HFCS and sucrose are almost identical.” On WOR-710 AM, Dr. Goran  admitted to a central flaw in his research: an inability to accurately and reliably determine how much HFCS is being consumed.

Five Major Reasons the Goran Study is Flawed and Unreliable

  1. HFCS and sugar are nutritionally equivalent. There is broad agreement among scientists that HFCS and table sugar are nutritionally and metabolically equivalent, and the American Medical Association has concluded that HFCS is not a unique cause of obesity.
  2. Glucose is wrongly compared to fructose. The studies cited by Dr. Goran to justify his belief that fructose is dangerous are largely studies that compare pure fructose to pure glucose, neither of which is consumed in isolation in a normal diet.  HFCS and table sugar contain about equal amounts of glucose and fructose–not fructose or glucose alone. When the same measurements cited by Dr. Goran are made comparing HFCS to sucrose, all of the differences disappear.
  3. The study is based on an “ecological fallacy”. It associates a characteristic of a large group (diabetes) to a single factor (high fructose corn syrup availability), with no clear evidence of causation. Comparing levels of HFCS “availability” to a disease with multiple contributing factors (diabetes) is an extremely weak association, particularly given the study’s grandiose claim of a global significance. In fact, Dr. Goran and his co-authors are forced to admit in their own paper that their approach “might introduce errors.”
  4. Goran’s previous HFCS study was shown to be deeply flawed. For a prior study purporting to find higher than expected fructose in HFCS sweetened drinks, Dr. Goran used the wrong laboratory test methodology. This flawed approach falsely inflated the levels of fructose and glucose in HFCS and in HFCS-sweetened products. Dr. Goran’s previous study has been roundly criticized.
  5. Isolating one aspect of a national diet and implying that it is a singular cause of diabetes is unwise, unscientific and highly speculativeDiabetes, like obesity, is a complicated condition with numerous contributing factors. Therefore, most of the scientific community would not support this kind of approach, which links the availability of a single component of the diet to an increased risk of diabetes.  There are many countries with high HFCS availability but low indicators of diabetes. Conversely, there are also many nations with low HFCS availability but high diabetes rates.  For example, drawing from the study’s own data, Japan consumes more high fructose corn syrup every year than Mexico.  Yet the rates of diabetes in Japan are about half of what they are in Mexico.  Hungary and Slovakia, which have the second and third highest rates of HFCS consumption, have lower diabetes rates than eight out of the 14 countries that have no HFCS availability at all.  There are epidemics of obesity and diabetes in many countries where high fructose corn syrup is not even used.  In fact, sucrose comprises more than 90% of annual sweetener consumption worldwide.