The health experts are united in proclaiming the importance of preventative health in general, and regular colonoscopies in particular. Get one so you can avoid dying of colon cancer! As is sadly all-too-usual, the colon cancer early detection campaign is propaganda to cajole people into putting lots of money into the pockets of doctors and health systems -- with no provable benefit to the patients who do their best to patiently put up with the pointless nonsense.
Do people who get colonoscopies avoid getting colon cancer? Yes, the vast majority of tested people avoid it; but then the vast majority of untested people also avoid getting it. Does getting colonoscopies as recommended help you live longer? All the authorities say it does, but the recently published RCT (random controlled trial) with over 80,000 subjects -- the ONLY such gold-standard trial EVER conducted -- shows that colonoscopies do NOT help the people who get them live longer.
It's no wonder that the medical-industrial complex has united to discredit this trial that threatens their revenue stream. Experts are speaking out, waving their arms wildly and pronouncing with deep-voiced authority that the trial is misleading. I guess all the people and organizations drinking from the fire hose of tens of billions of dollars a year in testing fees have been too busy to conduct a trial of their own to demonstrate that what they do actually helps people.
Colon cancer is a big deal
There is no doubt that colon cancer is something to be avoided, if at all possible. It's the fourth leading cause of cancer deaths, behind breast, prostate and lung cancer. Lots of people are diagnosed with it and die from it, according to the National Cancer Institute:
Getting colon cancer isn't a death sentence -- look at the survival rate above -- but we would all like to avoid getting it.
An ever-increasing number of people are screened for this terrible disease. According to the National Cancer Institute:
This adds up to over 16 million colonoscopies in 2019! The cost? Good numbers are hard to find, but it's probably in the range of $30 to $60 Billion dollars a year for screening.
The Voice of the Experts
Here is the summary recommendation of the U.S. Preventative Services Task Force:
The Grade A is primarily because the screening has "substantial net benefit." (Bold in the original.)
This organization has plenty of prestigious company in making this recommendation. For example, they say:
There is a general consensus that average-risk adults aged 50 to 75 years should be screened. The American Academy of Family Physicians (AAFP),42 American College of Physicians (ACP),43 American Cancer Society (ACS),44 and the US Multi-Society Task Force (which includes the American College of Gastroenterology, American Gastroenterological Association, and American Society for Gastrointestinal Endoscopy)31 all recommend routine colorectal cancer screening in this age group, although specific recommended tests and frequency of screening may vary.
It's not just the big organizations. Doctors and health systems on the front lines pitch the same message. Mount Sinai hospital in New York City recently put up a strong pitch on the subject. They lead their long article with this:
Colonoscopy is one of those important, routine medical procedures that most people would rather avoid. But experts say the test is a highly effective tool for both preventing colorectal cancer and diagnosing it at an early stage.
Along with the American Cancer Society, they recommend that screening start at age 45.
The trouble is, when you read all this carefully, particularly the extensive review that led to the USPST report above, you find a complete lack of RCT's for the effectiveness of colonoscopies. Nothing but modeling and authoritative-sounding guessing.
Experts attack the skeptics
Every once in a while, someone pops up who says that colonoscopies don't do any good. Fortunately for concerned patients, the experts tend to jump on such baseless assertions and tear them apart with facts and sound reasoning. Most of the Mount Sinai article just mentioned is devoted to this. It says:
Those looking for an excuse to put off a colonoscopy might now point to a large study conducted in Europe and published in September 2022 in The New England Journal of Medicine (NEJM) that appeared to question the benefits of colonoscopies.
For some reason, the Mt Sinai article fails to give a link to the study in question, which is here. I guess they feel that readers don't need to waste their time, since Dr. Greenwald's take-down is authoritative. It continues with this:
In this Q&A, Dr. Greenwald, Immediate Past-President of the American College of Gastroenterology, and Co-Chair of New York’s Citywide Colorectal Cancer Control Coalition (C5), discusses the recent study and why the value of colonoscopies remains unchanged.
Here's the man himself. He is a seriously authoritative-looking guy:
The NordiCC Study and the response of the Experts
The NordiCC study is the one described in the recent New England Journal of Medicine article that Dr. Greenwald tore apart in the Mt. Sinai article promoting colonoscopies. The study followed over 84,000 "presumptively healthy men and women 55 to 64 years of age drawn from population registries in Poland, Norway, Sweden, and the Netherlands between 2009 and 2014."
The article seems to support performing colonoscopies. Here's the conclusion of the Abstract at the beginning of the paper:
In this randomized trial, the risk of colorectal cancer at 10 years was lower among participants who were invited to undergo screening colonoscopy than among those who were assigned to no screening.
I guess Dr. Greenwald and others were concerned about the fact that the numbers weren't strong. The Mt. Sinai article said the study "appeared to question the value of colonoscopies." Dr. Greenwald is quoted by Mt Sinai as saying
This study, along with prior studies, shows that colonoscopy decreases your chances of getting and dying from colorectal cancer. Getting sick and dying from colorectal cancer—especially due to delayed screening—is real. Screening with colonoscopy saves lives.
Done!
The NordiCC Study
Why would multiple European national governments go the trouble and expense of such a massive trial if the value of colonoscopies had been conclusively proven? Hmmm. The answer is simple: colonoscopies are expensive and highly unpleasant, and there have been NO RCT's that show a causal relationship between getting them and avoiding getting and dying from colon cancer. So before taking on all the cost and trouble, those groups figured they'd better get some real evidence on the subject.
First, let's see what the expert from Mt Sinai drew from the study:
Most importantly, in the section of the study that analyzed people who actually had a colonoscopy, the risk of developing colorectal cancer decreased by 31 percent and the risk of dying from colorectal cancer decreased by 50 percent, which is huge.
Wow. Why would anyone be worried about the NordiCC study scaring people from getting a colonoscopy, when it has such a huge benefit -- and the doctor claims that in the US it is performed better, yielding even stronger preventative results!
When you read the NordiCC study itself, aided by understanding of the statistical tricks that are used to distort the results, a whole different message emerges. This is why the medical-industrial complex goes to great lengths to hide the truth.
In this case, the core trickery is a biased sub-group created from the group invited to have a colonoscopy that they call an "adjusted per-protocol analysis." The other main tricks are widely used: reporting "efficacy" (relative risk); ignoring NNT (absolute risk); and endpoint selection (dying of cancer vs. dying of any cause).
Here is a brief summary of the real results of the trial by Dr's John Mandrola and Vinay Prasad. I have verified that this summary accurately reflects data in the NEJM paper (see Table 2 of the paper):
Over 10 years of follow-up, an invitation to screening colonoscopy modestly reduced the risk of being diagnosed with colorectal cancer, but it did not significantly reduce the risk of dying from colorectal cancer. Survival from cancer was nearly identical in both groups. And all-cause mortality was the same.
The specific numbers of the primary outcome:
The chance of getting (diagnosed with) colorectal cancer in the invited group was 0.98% vs 1.2% in the usual care group. This represents an 18% reduction in relative terms, and an absolute risk reduction of 0.22% or 22 per 10,000.
The chance of dying from colorectal cancer in the invited group was 0.28% vs 0.31% in the usual care group. This 10% reduction in relative terms amounted to a difference in 3 in 10,000 and did not reach statistical significance.
In the invited group, 11.03% of patients died; in the usual care group, 11.04% of patients died.
Here is a key chart from the paper, showing the risk of dying from colon cancer. Note that the real data is a barely noticeable squiggle along the X axis at the bottom; most of the chart is a big blow-up of the bottom 1% of the Y axis.
The tiny difference (3 in 10,000) in dying of colon cancer was not statistically significant. And the most-important-of-all measure, all-cause death, was identical.
Those are the indisputable facts from the study itself.
The authors of the study did their best to fiddle with the results -- they declared themselves to be "disappointed" in the study results. They put their fiddling in the paper.
The most significant game they played involved the fact that 42% of the subjects who were invited to have a colonoscopy actually had one. They recalculated the results for just the people in the invited group who actually had the procedure, trying to get better results; in the study, they call this a "per-protocol analysis." This shows the desperation of the authors, since doing this violates all the randomness and invites a host of what statisticians call "confounding factors," exactly the thing that a RCT avoids by being random. Naturally, this improved the results. By a small amount.
The authors and all the establishment defenders of colonoscopies do the classic thing that people who want to promote a drug or procedure do to mislead patients: they focus on relative risk instead of absolute risk. Relative risk, often called "efficacy," makes you think that the procedure is terrifically effective. What do you think when you read that your risk of death is "decreased by 50 percent?" It's huge, right? See this post on vaccine efficacy for a detailed explanation. What most people care about is absolute risk, which is how likely it is that the feared thing will happen to you. This is sometimes called NNT (Number Needed to Treat), which is the number of people who have to undergo the procedure in order for just one of them to benefit.
The per-protocol analysis (the best case) showed the risk of dying from colon cancer to be 0.15% in the invited group and 0.30% in the control group. This is a 50% improvement in relative terms which sounds great! But in real-life absolute terms, it's a different of 0.15%, which is 15 for each 10,000 people And again, the difference of dying from any cause between these two groups was effectively zero. Here is a detailed discussion of per-protocol fiddling and the results of a statistically sound approach, which reduces the statistical significance of the colonoscopy benefit to zero ... as shown in a table of the supplement of the NEJM study.
And then there are the harms. What can go wrong with a colonoscopy? The USPST, the government group quoted above that recently recommended that colonoscopies start earlier than before, at age 45, discusses them deep in the supporting material. They report: "Harms from screening colonoscopy have been reported in 67 observational studies (n = 27,746,669)." You have to read carefully, Here's what they report:
14.6 major bleeding events per 10,000 colonoscopies (95% CI, 9.4-19.9; 20 studies; n = 5,172,508) and 3.1 perforations per 10,000 colonoscopies (95% CI, 2.3-4.0; 26 studies; n = 5,272,600)
That is a high confidence result of harm based on millions of patients, vs. the nearly identical low-confidence results of benefit from the NordiCC study.
Conclusion
The authors of the NordiCC study wanted to find that colonoscopies are effective -- they say so! They did their best to slant the results and obscure the real results. Experts reporting on the study in the US cherry pick and criticize the slanted results and confidently proclaim that colonoscopies are essential to health and longevity, when the clear numbers in the published study of over 80,000 subjects show no such thing. This is yet another example of self-serving advice from a medical establishment that depends on trusting, gullible patients to keep the money rolling in. For more, read these posts about flu, diet, saturated fat, cholesterol and blood pressure.
Why do insurance companies jack up premiums to shell out big bucks for drugs and procedures that don't help?
I would think they promote these tests and in turn jack up premiums because:
(a) the have created a perceived value for these tests, through all the scare mongering of disastrous outcomes if not tested, coupled with
(b) the cost of such tests for patients are kept vey high, while the negotiated rate by the payer with the physician (provider) is opaque and is ridiculously low
justifies the increased spend on health insurance.
The providers may have little to no choice since their utilization is higher and pays for the recovery of capital expenses. In the end the payers have everything to gain.
Phew!
Posted by: Muthu Krishnan | 02/15/2023 at 10:07 AM