Funnily enough, accuracy is a somewhat inaccurate term when it comes to medical testing. What we are interested in are sensitivity, specificity [0] and prevalence of the disease in the population being tested. With those three we can get positive prediction value (PPV) and negative prediction value (NPV) [1]. But PPV and NPV always depend on prevalence! That's why we don't want to test healthy people for all kinds of exotic diseases.
I do agree with your skepticity. A better study design could use midazolam as placebo. And I think a good idea would be to ask the participants amd their physicians, which one do they think they received. That would give some idea about how well the double blind design did. Also, many commenters seem to disagree with me, which I also have to be take into consideration. My n is very low as my hospital has just started ketamine treatments.
Thanks, I appreciate your comment and experience. I agree that if one was to get those effects, one would be quite unlikely to suspect placebo. I hope you got the help you needed from ketamine.
Advice not to operate an automobile is for minimizing risks. It doesn't mean that most would be unable to handle a car, although it could mean that their reaction times would be slower. The standard way to administer ketamine is to give 0,5 mg/kg under 40 minutes. It would in no way be immediately obvious to everyone in, if someone began to receive ketamine or not. At least I couldn't differentiate them at that point.
I agree about you second point. I was too quick to make my comment and I was conflating stuff I've read about nasal esketamine, which didn't have as robust evidence behind it as i.v. ketamine. I should not write comments while holding a baby that is trying to sleep and failing at it.
About the first point, I still disagree with the parent comment. The standard 0,5mg/kg ketamine is given under 40 minutes. I've seen two patients receive it. The room was dimly lit, silent and peaceful, as to not induce or exacerbate any possible dissociative or hallucinative side effects. The patients didn't report any side effects at first and later reported a little sedation. I wouldn't be a able to tell with certainty if they had received a placebo or ketamine, and certainly not within minutes.
Source: psychiatrist in training
Edit: I still think we need hard data about anti-suicide effect and not just surrogate markers. Variability in suicidal ideation is suggested to be a predictor of suicide attempts [1]. If ketamine first gets someone better and later on they relapse, they could, in theory, be at increased risk for suicide. More long term data is also needed about it's long term effects. Note that I do think it's a good thing that ketamine is becoming more available! I will continue to suggest it for e.g. patients with treatment resistant and severe depression.
I’m sorry but you are strongly over exaggerating the (side) effects of i.v. ketamine. It’s not a dose that sends you in a K hole. It’s way less than that. Some get sedation and some do get dissociative symptoms which can unmask the study arm of the patient. Approximately one out of ten got at least one of those side effects. Sedation can be mimicked with midazolam, so if you use that as placebo, you get 90 % of the patients that are not obvious, which medicine they got. I do agree that a better ”placebo” would be good to use.
The problem with ketamine and suicidality studies is that suicides - thankfully don’t happen very often - so we have no direct evidence that ketamine prevents suicides. Of course if it would only alleviate suffering and wouldn’t increase the rate of suicides, that would be a win too.
[0] https://en.wikipedia.org/wiki/Sensitivity_and_specificity [1] https://en.wikipedia.org/wiki/Positive_and_negative_predicti...