Searching For The Best Colonoscopy In Town


All Martha Bebinger wants is a good colonoscopy. Don’t we all? But look at the lengths she has to go to in order to find out which doctors are good, who finds the polyps and what the actual cost of the procedure is. Her search shows that while everyone in health care talks about the importance of transparency, it’s not here yet. Here’s the top of Martha’s story:

Someday soon, you’ll need a routine medical test, perhaps an ultrasound or a mammogram, and you’ll obviously want the best. But the quality of health care tests and procedures can vary a lot depending on the doctor or hospital, and it’s not easy for patients to find information about quality.

Still, we’re all supposed to be choosing our care more carefully these days. So when I got a scribbled note from my doctor saying that it was time for a colonoscopy, I set out to find the best one in Boston.


I used to think a colonscopy is a colonoscopy, just one of those tests we have to get when we hit middle age. Then I met James Tracey, a gastroenterologist (GI) who does more than a thousand colonoscopies a year. He starts telling me about this running competition he has with other GI docs, one in particular at Hawthorn Medical Associates in North Dartmouth. It’s about who has the highest rate for finding polyps in patients during a colonoscopy.

“When he and I walk down the hall, it wouldn’t be uncommon for him to mention his percentage just to get me going,” Tracey says. “And of course that redoubles my effort that I’m not going to have that edge over me! I’m going to make my numbers as good as his.”

Tracey tells me some doctors in his practice find polyps in twice as many patients as others. Hold on, I think, this seems pretty important. I’m getting this not-so-pleasant test to look for adenomas, those polyps that can develop into colon cancer. So I ask Tracey, why is there such a big difference? He reminds me the colon is a five- to six-foot-long tube packed into the abdomen.

“Sixty to 70 percent of the colon you can see clearly, no matter what you do,” Tracey explains. “But 30 percent of the colon, you need to work at it. You need to irrigate the colon out if there’s any fluid or debris. You need to reexamine an area several times if the anatomy is distorted so the colonoscope cannot clearly see.”

My takeaway is that some doctors look more closely for polyps than others. And, of course, patients play a big role here. If we don’t follow the prep instructions, such as fasting, and come in for this test with a clean colon, the doctor is not going to have a clear view of potential problems. OK, enough description. So I make a list, based on conversations with doctors, of ways that one colonoscopy can be better than another. I email the list to some large hospitals and physician groups in Greater Boston and ask them how they measure up.

At this point, with or without solid quality data, patients must take a more activist role in their own care, Martha says:

For now, if you need a colonoscopy, be sure you ask at least one question: How often does your doctor find adenomas, the polyps that can develop into cancer? That answer is important.

If you really want to find the best “value” colonoscopy, try comparing the quality chart [top] and the cost chart… Both are rough. The quality information has not been cleaned up or adjusted for the sickness of patients. And on the cost chart, some of the prices you see are based on what my insurer, Blue Cross, would pay, and some aren’t. This isn’t close to what we patients need to make smart choices, but it’s a start.

Please follow our community rules when engaging in comment discussion on this site.
  • ISO_Data

    Thank you for an excellent piece of reporting! I called around to a couple of practices and asked them the information you shared on your table, with the additional question of pain following the procedure. So far, NO practice has be wiling to disclose their stats and Harvard Vanguard said that they would only do so at an appointment. Can you please advise how to best garner this information?

  • MMRosenthal

    James Piper, MD at Harvard Vanguard, has guided me through this test twice in the last 11 years (Other medical conditions make prep very difficult for me.) and he is EXCELLENT. Kind, patient and professional. And, unlike many surgeons, he exudes a warmth and tenderness that is very reassuring. I don’t think I would have undergone either procedure with any other doctor.

    (BTW, prep is much easier now, than 20 years ago. With all the dye-free foods available at Whole Foods and Trader Joe, patients can do a good prep, without starving during the days preceding the procedure.)

  • anon

    There is no question that data like this is important to evaluate, and I’d rather see it, than not see it, as a patient. But your point is correct that factors other than ‘doctor quality’ may dramatically affect the results. Polyp finding is probably dramatically lower in patients who can’t prep or have difficulty following instructions. I bet #s are much worse in places that serve lower socioeconomic profiles. Should doctors be punished for working in richer vs. poorer communities? In cities vs. suburbs? Also, I see MGH and BI have higher perforation rates- more complex patients, more complex procedures there, more rigorous data recording? Complex stuff!

    • Martha Bebinger

      Hi anon – there is no question that such data should be adjusted for the factors you mention, and possibly others, before it is published widely. I hope we can find a way to do that soon so that patients have a clear, fair picture of quality differences among doctors.

  • Reasonable?

    This is really great reporting. Martha managed to accomplish what many health policy experts have struggled with for years. This is a great model of what can be done.

    In medical science there is common set of calculations used to simplify and clarify stats in areas like this. Increasingly consumers will have to learn this vocabulary to become savvy shoppers. It’s called Detection Theory.

    a. Sensitivity=> How many polyps are detected (cancerous or not).

    b. Specificity=> how many polyps that are detected are actually cancerous/precancerous eg not benign..

    Does the detection rate above reflect sensitivity or specificity?

    Both numbers are important.
    The best organizations have high percentages on both measures.

    Sensitivity and specificity can be used to calculate the positve and negative predictive value of a procedure. Those would probably be the most helpful for consumers along with cost and perforation rate which are reported above.

    Medical organizations should be challenged to collect better data to elucidate these stats so consumers can make better decisions in a large number o healthf domains.

    Strong work by Martha and team!

    • Reasonable?

      Correction on the Sensitivity Specificity definitions (they never stick in my brain).

    • Martha Bebinger

      Hi Reasonable – this is the adenoma detection rate. The rate in the chart does not include hyperplastic polyps because I was told they do not develop into colon cancer. I did not ask for the percentage of adenomas that are found to be pre-cancerous. Perhaps that should be added to a future quality score card.

      Thanks, as always, for your comments.