Skip to Main Content

Since colorectal cancer is on the rise in people younger than age 50, national guidelines have recently pushed the starting age for colorectal cancer screening down to 45. We need to spread the word.

Yet we specialists continue to provide outdated and disproven recommendations on colonoscopy bowel preparation that make the process more difficult for the millions of patients who undergo colonoscopy every year: We tell them to adopt a clear liquid diet the day before their procedure.


Patients rate the preparation as the most difficult aspect of a colonoscopy. This is due in part to the unavoidable, unpleasant process of drinking a purgative so that we can get a clear view of the colon. But there’s a second, unnecessary part: Most colonoscopy instructions direct the patient to stick to a clear liquids-only diet for much of the day before, or even for the entire day. On the surface, the rationale seems sound: Sticking to liquids should make it easier for the laxative to do its work. Some colonoscopies are unsuccessful or need to be repeated early because of an inadequate preparation, so it makes sense to stick to liquids and reduce the amount of solid waste.

It is true that diet matters for colon cleansing. Certain foods can greatly obscure our views of the colon: corn, nuts, seeds, and many kinds of vegetables. But that leaves many solid foods that people can eat and digest without affecting the quality of their colonoscopy. A low-residue diet, one that is heavy on starch and protein and free of vegetables, is just as effective, and much less difficult to endure, than a diet restricted to clear liquids all day. (I have seen corn and seeds many times during a colonoscopy, but I have never seen a piece of chicken!)

A pre-colonoscopy diet that includes solid foods is safe and effective, and there are abundant data to prove it. A large, randomized trial published in 2013 compared the results of two diet instructions the day before colonoscopy: clear liquids only, or a diet that permits solid foods such as a bagel with cream cheese, a turkey sandwich, and mac and cheese. The solid diet group had just as good clean-outs as the spartan clear-liquids-only group. There were some differences, though: Those who were allowed to have solid food reported higher satisfaction with the preparation experience and were actually less likely to cancel their appointments.


In the ensuing years, study after study found the same outcome. One systematic review identified 13 randomized trials comparing a low-residue diet to a clear liquid one. Not a single trial found that a clear liquid diet resulted in better cleansing, and the low-residue diet was consistently rated by patients as more tolerable than the clear liquid diet.

We have known for more than 10 years that there is an easier diet to prescribe before a colonoscopy. Yet gastroenterologists stubbornly cling to the clear-liquid diet. Academic medical centers, which should be the vanguard of evidence-based and patient-centered care, fail our patients in this matter. When I ask Google “Can you eat solid food the day before your colonoscopy?” the first hit gives me a firm “No” from instructions provided by the University of California Los Angeles: “You must not eat any solid foods the day before your colonoscopy. You may only eat a clear liquid diet.” Harvard-affiliated Massachusetts General Hospital’s preparation materials admonish, “Follow a clear liquid diet only!” The Cleveland Clinic agrees: “Only drink clear liquids the ENTIRE DAY before your colonoscopy. Do NOT eat any solid foods.”

Patient advocacy groups do not fare better, likely because they are receiving outdated advice on this subject from gastroenterologists: Fight Colon Cancer includes much valuable information on the importance of screening, but in its preparation materials it advises, “The day before a colonoscopy, patients must stick to a liquid diet.”

The clear-liquid diet is unnecessary and harder on the patient. As the pool of people eligible for screening has enlarged to include those age 45 and up, it behooves us to minimize all impediments to this procedure. So why won’t gastroenterologists embrace this development and let their patients eat?

Part of the answer relates to long-known obstacles to diffusion of innovation. Gastroenterologists may not know about these studies, and they are not learning from patients or colleagues that their preparation dietary instructions are associated with worse patient satisfaction and possibly increased cancellation rates. But knowledge of the literature may not be enough to convince our specialty to make this change. When I ask my colleagues why they continue to prescribe a clear-liquid diet, the most common answer I get is some version of this: “My patients may have difficulty understanding the nuances of the low-residue diet.” This strikes me as pessimistic and even paternalistic. If it were really so hard to stick to this diet, we would not have seen the clinical trials show such a strong and consistent result.

There is no major force pushing gastroenterologists to catch up with the evidence; the pharmaceutical industry and government regulatory authorities, the common spurs of change in medical practice, are not interested in this matter. Medical inertia can be powerful. But change may yet occur, first slowly, and then rapidly once a critical mass is attained. This will happen one gastroenterologist or one endoscopy unit at a time.

I urge my colleagues to look at their preparation instructions and to consider catching up with the evidence. We can advise patients that the diet on the day before a colonoscopy may include the following foods: cheese, eggs, milk, white bread, olive oil, butter, beef, chicken, fish, white rice, vanilla ice cream, and plain yogurt. They must avoid the following: popcorn, seeds, nuts, corn, multigrain bread, salad, raw and dried fruits, berries, kiwi, broccoli, cauliflower, Brussels sprouts, and cabbage. These instructions may take longer to outline than a clear liquid diet, but they will allow our patients to prepare for a colonoscopy in a way that is practical, humane and effective.

It is time to follow the science and let our patients eat.

Benjamin Lebwohl is a gastroenterologist and director of clinical research at the Celiac Disease Center at Columbia University.

Have an opinion on this essay? Submit a letter to the editor here.

STAT encourages you to share your voice. We welcome your commentary, criticism, and expertise on our subscriber-only platform, STAT+ Connect

To submit a correction request, please visit our Contact Us page.