Suppose there is a slow-developing disease D that can be fatal if not detected early enough. However, there are various screening methods intended to find D early enough to be stopped before it becomes fatal. But the various screening methods have different costs, risks, and tradeoffs.
Option 1: Use X as the primary screening method occasionally. X is considered the most effective screening method at preventing deaths from D. However, it is expensive, somewhat invasive, inconvenient for the patient, and has some risks of harm (due to X, not D) to the patient.
Option 2: Use Y as the primary screening method frequently, with X as the secondary screening method if Y indicates a potential problem. Y is inexpensive, non-invasive, not inconvenient for the patient, and has no risks of harm from screening in the common case where no potential problem is detected and X is not done. The expected number of times a patient needs X is fewer than half as many as in option 1, so the expected cost is half or lower, and the risk of harm due to X is lower as well. Option 2 is considered about 90% as effective at preventing deaths from D as option 1. On average, for every two deaths from D that option 2 fails to prevent relative to option 1, five patients will be spared harms (usually non-fatal, but can be serious) due to having to do fewer X.
Which option would you choose if:
A. You are a patient who falls under the screening guidelines whose costs are covered by third party insurance.
B. You are a patient who falls under the screening guidelines who self-pays the costs.
C. You are a physician who is making a screening recommendation for a patient who falls under the screening guidelines.
Sounds like colonoscopy vs one of the stool-sample methods to me.
Unfortunately, if you don’t have enough money for basic survival, money has to enter into it. If you have your basic needs and maybe a little bit more taken care of, you might spring for the better test.
However, neither option is uniformly better (even without considering monetary cost). The one that is better at detection also has a greater risk of harms.
You left out a crucial number. What is the incidence of X in the screening population? What is the number needed to treat in order to prevent one death?
Suppose D is a rare disease; I have a 1 in 1000 chance of getting it, or 10 in 10,000, and if the disease is detected by X, it can be cured early enough. Then if I get use X, I will not die of D, and if I use Y, I have a 1 in 10,000 chance of dying of D. Then I pick Y.
Suppose D is common; I have a 1 in 2 chance of getting it. Again assume if I use X, my D will always be detected and I will be cured, whereas if I use Y, I only have a 9 in 10 chance of being cured. Then if I use Y instead of X, I have a 1 in 20 chance of dying of Y. Too high! I choose X.
Compared to a population of unscreened persons, option 1 would save 2.4% of persons from dying of D, while option 2 would save 2.2% of persons from dying of D.
So I have a 1 in 500 chance of dying of D (in my lifetime, I assume) by choosing Y instead of X. But I also have a 1 in 200 chance of having a complication from X (again, I assume this is in my lifetime). How serious are the complications? What is the death rate from X, over my lifetime?
X does not look very good to me. I’m very unlikely to die of D, and X is somewhat dangerous.
How much money do doctors and hospitals have invested in infrastructure for X and how many screenings do they need per month to make a lot of money off X?
I think most people will be challenged to pick either option A or B (especially method X) based on vague hypotheticals that they can’t relate to. Add in facts like where is this disease (eg brain, skin (face v upper back), breast, prostate, etc) may yield different answers. How expensive? How inconvenienced (eg the morning, a day, a week, etc). What are the risks of harm (eg loss of full/partial cognitive function, two inch scar on my face v my upper back), loss of limb, inability to have sex, loss of appearance (esthetics/personal identity (eg mastectomy), etc. Should an outcome of invasive procedure result in long term harm, am I and/or family prepared for financial disability or death. I’m just saying the above hypos are difficult to wrap one’s mind about when important facts may need to be considered.
As to option C, I think many patients faced with either option would say what do you think MD, and more often than not, go with MD’s opinion as to course of treatment expecting their MD to act in patient’s best interest.
I’m also not a big fan of value to patient of data such as some option having X% of success, or incidence of X in screening population. Air travel is very, very safe, but if I’m on the plane going down, how do such numbers help me.
The problem with scenarios like this is that they rarely happen in real life. There are very few health problems where the invasive procedure isn’t proceeded by significantly less invasive procedures. There are some, sure, but most have some kind of blood or tissue test for.
Example: Crohn’s. Definitive test is by colonoscopy but before that, you’ll have blood and fecal calprotectin tests.
In this case, option 1 is commonly recommended by physicians, which is the use of the invasive procedure as the primary screening method, rather than as a secondary screening method after a non-invasive primary screening method as in option 2.
Note that some risk factors are lifestyle factors – so in addition to screening regimen, a patient can potentially reduce their risk with lifestyle choices.
So at least in theory, a person who is a physically active, thin vegan and goes with screenng options Y might be doing more to protect themselves from the risk of D than the overweight, sedentary smoker who opts for option X – and perhaps that is a factor that ought to be considered by physicians in making recommendations to their patients.
I think that the reason the OP left out the information about tests is to avoid the emotional bias that comes with the belief that the colonoscopy is “gold standard” for screening – even though the stats suggest a more complex picture, and its routine use for screening seems to be a largely US-thing. (In Canada, where everyone gets health care for free, the recommendation is for either the FIT or FOBT test every other year, or flexible sigmoidoscopy every 10 years – see https://canadiantaskforce.ca/guidelines/published-guidelines/colorectal-cancer/ & https://www.healthnewsreview.org/2016/02/americas-closed-information-border-news-of-canadian-colon-cancer-screening-guidelines-denied-entry/) But of course in Canada, there is a different cost/benefit anallysis going on than in the US, where colonoscopy is readily available and the fee-for-service model may incentivize doctors and hospitals to favor more costly tests.
Your situation is different because you are getting a colonoscopy for some other medical reason - not routine screening – as you are no where near the age (50) when routine screening is recommended. And when you do reach age 50, there will likely be more medical advances and data so the equation may be very different then.
I’ve learned in my lifetime that a lot of things labeled “gold standard” really aren’t the best approaches – but it’s great marketing!
I personally think that for a low-risk individual the FIT test combined with healthy lifestyle is probably the best approach, because that lifestyle stuff carries over into other cancers and life risks too.
But the downside of the FIT test is simply compliance – it’s easier to do than colonoscopy, and certainly less expensive – but if the choice is something that is done once every 10 years vs. once a year … there’s going to be some people who just blow off the once-a-year test and end up getting no screening at all. So I can see why some doctors might favor the test that is done less frequently.
True that once in 10 years versus every year may make a difference. But giving 10 pieces of stool to the lab seems like a lot less effort than dealing with colonoscopy, including the colon cleanse needed beforehand. Note that FIT kits come with the special postage-paid container to mail the piece of stool to the lab to make it more convenient for the patient if a mailbox is more accessible than the lab.
In the US, it appears that most (over 60%) people age 50-75 who get colon cancer screening get colonoscopy as the primary screening method (option 1), with relatively few (under 10%) using FIT or gFOBT as the primary with colonoscopy only if FIT or gFOBT detects blood in the stool (option 2). California seems to have a somewhat higher rate of option 2 than other places, possibly because Kaiser has taken to automatically mailing FIT kits to every member due for screening ( https://lookinside.kaiserpermanente.org/kaiser-permanente-successfully-screens-for-colorectal-cancer/ ).
I’m getting the colonoscopy for reasons having nothing to do with colorectal cancer; my leukemia chemo can cause GI bleeding, which is exacerbated if there are polyps. I’m on the three-year colonoscopy cycle, since I’ve had two bleeds in the past three years, including one significant enough to require hospitalization.
@ucbalumnus — my comment about frequency was essentially addressed to the nag factor and the doctor’s point ov view – that is, just because FIT is easier doesn’t mean that patients will do it. It sounds like Kaiser has resolved that problem with it’s policy of mailing out the kit to to all appopriate patients – and I think that would be the way to go. But even so you can see from the report you linked to, only about half of the recipients actually did the screening.
I do think that for low risk individuals, FIT is the way to go. Cheap & easy and I think that it when it is done regularly the data shows that it is highly effective. The problem is that there aren’t any stats about what happens when a patient relieson FIT alone at irregular intervals – for example, a person gets the test at age 52 and then doesn’t bother again until age 57.
