Comment by boplicity | Hacker News Reader

Comment by boplicity | Hacker News Reader

Parent Full thread

This is exactly the reasoning that my gastroenterologist used. There's no placebo controlled, double blind studies that support "diet" as a treatment for Crohn's disease.

Well, you know what: The same is true for open heart surgeries. (One of the most incredibly invasive procedures possible.)

Actually, there are studies that support diet as an effective means of treatment.

No, there aren't enough of the "right" studies that are as "high quality" as the ones supporting the efficacy of various drugs. However, "evidence" doesn't change reality, instead, it simply reflects it.

You asked about success: For me, success means no active inflammation, at all. Which was achieved through a combination of drugs and diet.

For others, success means no longer going to the bathroom twenty times a day. Having "regular" bowel movements. No obvious symptoms.

Your comment brushed off "symptoms" as something unimportant. But, for people who are really debilitated by the disease, a change in diet that eliminates the symptoms is life changing. No need for "evidence" on that.

You're right that there's no "one" diet that works for everyone, as I mentioned in my original comment. It's a learning process. However, it's also my observation that everyone with Crohn's disease gets very aware, and very specific about what they eat. (Or, they eventually face the consequences.)

I'm not sure I follow the point on open heart surgeries. What would the study be? Is there some question of fact that needs to be resolved? There are many kinds of open heart surgery too. Which one do you purport does not have any randomized trials associated with it? All of them? That would actually surprise me a little.

Placebo-controlled RCTs are only feasible when it's possible to ethically apply the placebo or the treatment. Generally speaking, the types of things we do open heart surgery for, like valve replacement, coronary bypass, or pacemakers, it is impossible to ethically apply a placebo.

You only need RCTs when there is some question of fact that needs to be resolved. There is no need for a random controlled trial about whether stopping all water intake will kill someone. For some types of heart surgery, this doubt does exist, but for any of those types which are not being studied with RCTs, I would guess there is not much doubt. If your heart valve is busted and it is causing you to be unable to pump blood, you can look at the pumping efficiency before and after replacement. No RCT needed.

You'll note that these two factors are very different from diet. It is possible to ethically apply a control and a treatment where it comes to diet, and there is much doubt as to their efficiency for some purposes.

Also if it works it works. I think people worry too much about a solution being a placebo.

Yes, it may be that your solution only worked because you believed it would work. But take the win!

Certainly solutions backed by double blind controlled studied are way better. But those often aren't an option, so use what is.

This is only true insofar as the "solution" doesn't cause other problems itself.

People bankrupting themselves on snake oil, or having other health issues due to weird diets, or avoiding actually proven medicine are all extremely harmful.

For a disease with highly temporally variable symptoms, you don’t even need a placebo effect. You could try some treatment and randomly get better. (Go read about almost any Crohn’s RCT for which the participants had active flares. The control group generally improves a decent amount over the course of the study.)

I agree w/ you, but I'll also add that the premise that the evidence on diet is equivocal is pretty offbase. The evidence for EEN is pretty robust and held up over a couple dozen studies - see my other comment https://news.ycombinator.com/item?id=29453808#29456208

Reply on news.ycombinator.com