Recent Newspaper & Online Columns by Kate Scannell MD

Carbs vs. fats: Studies cause heartburn

September 7, 2014

Tags: diet, carbs, fats, surrogate endpoints, surrogate markers, diet studies, low fat, low carbohydrates

By Dr. Kate Scannell, Syndicated columnist
First published in print: 09/07/2014

Reading and writing about new diet studies is bad for my weight and health. I often develop headaches trying to make sense of them because they're usually so muddled and imprecise. That means frequent reaching for food to prepare my stomach for anti-inflammatory pain relievers.

Nonetheless, with bagel in hand, I'm now prepared to discuss the two megawatt diet studies released last week.

The first, published in the Annals of Internal Medicine, has been rockin' the headlines. It's been interpreted in various ways by popular media, mostly as evidence that a low-carb diet trumps a low-fat diet for people wanting to lose weight and reduce their risk of heart disease.

Others claimed the study redeemed dietary fat's tarnished reputation because, finally, "the evidence is in" -- we should be focusing more on lowering our carbs, not our fats! Indeed, it may be advisable to "embrace" more dietary fat (a slippery proposition) as we simultaneously lower carb intake. But the most serious misinterpretation claims the study proves that low-carb diets are superior for healthier hearts.

Argh. It's hard to know where to begin a critique, so I'll just swallow another anti-inflammatory and start at the bottom line: Readers beware!

Turning to the basics, this study randomly assigned adults to either a "low-carb" or "low-fat" diet, hoping to make some legitimate claims about which was superior for weight loss and cardiac health. But a fundamental problem here involves how researchers defined "low" for each group. A "low-carb diet" meant incorporating less than 40 grams of carbohydrate intake per day, while a "low-fat diet" meant less than 30 percent of daily caloric intake coming from fat.

However, this actually meant that people joining the low-carb group were required to drop more than 200 of their daily 242-gram carbohydrate intake at baseline -- a whopping change in their eating behavior, representing 40 percent of all meal content and about 800 calories per day. Meanwhile, people joining the low-fat group had to reduce the percent of their daily calories derived from fat by a mere 4.7 percent (from a baseline 34.7 percent) -- representing a relatively minor dietary intervention and approximating only 100 calories per day. The point is, it's hard to make a meaningful comparison.

We also shouldn't be too surprised that the low-carb intervention -- requiring an 8-fold greater caloric restriction -- resulted in greater weight loss. While it's clear that people in this group didn't strictly adhere to the diet, nonetheless, compared to people in the low-fat group, they consistently consumed 80-to-160 fewer daily calories throughout the yearlong study.

Another fundamental problem involves bundling all fats and carbohydrates together and making blunt cuts to either to define -- or recommend -- a dietary intervention. That's because, as Clint Eastwood and Sergio Leone taught us, we must always contend with the good, the bad and the ugly -- whether in our search for gold or our view of fats and carbs. For example, indiscriminately eliminating good dietary fats -- like omega-3s or monosaturated fats -- may unfavorably affect our cardiovascular health. And aiming to equate the carbs in a cupcake with those in a celery stalk is nothing short of a long-shot.

My main concern regards the popular misperception that the study linked low-carb diets to healthier hearts. Yet no such evidence was actually provided. In fact, to know if that were true would have required the authors to monitor and document the incidence of strokes and heart attacks in many more study participants, over many more years of observation.

Instead, we're provided with information about "surrogate endpoints" for cardiovascular disease -- the "clinical endpoint" that actually matters to us. That is, we're told that people in the low-carb group tended to show better changes in weight, fat mass, HDL ("good") cholesterol, and triglycerides -- surrogate markers that presumptively stand-in for future cardiovascular disease.

The authors openly acknowledge this limitation of their study. But using surrogate endpoints to substitute for actual clinical endpoints deserves greater scrutiny because it's become so prevalent in research that shapes routine medical practice.

While I'm not from Missouri, I'm always in a "show me" state of mind when it comes to research using surrogate markers to stake serious health claims. I prefer a tighter chain of evidence that includes a close view of the actual end-point. The X-ray may look better after antibiotics, but I still want to know whether the patient actually survives her pneumonia or sustains long-term respiratory problems. The PSA may be lower after chemo, the tumor may be smaller -- still, I want to know whether the patient's fared any better with the treatment.

Clearly, substituting surrogate endpoints for clinical endpoints can be problematic. The purported chain of events linking them may be tenuous, fragile or purely speculative and subject to future revision. It can be weakened or twisted when parallel events in a person's life happen to affect the same endpoint.

Still, researchers often choose surrogate endpoints when it's hard to access meaningful clinical endpoints. That doesn't mean that such research is invalid, but it does mean that you shouldn't miss the point when making evidence-based decisions about your own medical care.

Oh, yes -- about that second study from last week. It mined the data from 48 prior studies on brand-named diets, like the Ornish (low-fat) and Atkins (low-carb) diets. Published in the Journal of the American Medical Association, it concluded that people who followed either general approach -- low-carb or low-fat -- lost a similar amount of weight in a year's time.

In the end, with all the available evidence on the scale, we still can't definitively determine whether any particular diet is superior in terms of lifetime sustainability and health outcomes. Nonetheless, in the uncertain meanwhile, a girl's gotta eat, so I'm going to weigh-in, favoring a balanced approach akin to the Weight Watchers model.

For me, that means a low-carb, low-fat, calorie-restricted diet that prioritizes wholesome non-processed foods, plant-derived nutrients, and heart-healthy fats. And that's my endpoint for the moment.

Kate Scannell is a Bay Area physician and the author, most recently, of the novel "Flood Stage."
© 2014, Kate Scannell

Comments have been disabled for this post