0
$\begingroup$

I am experimenting with few treatments to see which one is effective. I do not have luxury of control group to compare against. In such scenario, is there any method to see if the treatment has been effective?

Note: Can you please suggest answers which does not involve longitudinal study.

Improve this question
$\endgroup$
8
  • $\begingroup$ I assume the treatment is to be considered 'effective' if it makes a change in some value from the current value without the treatment. What do you know about the current value? $\endgroup$
    – BruceET
    Aug 21, 2019 at 6:55
  • $\begingroup$ @BruceET agreed, I do know the current value. $\endgroup$
    – Artiga
    Aug 21, 2019 at 6:58
  • 1
    $\begingroup$ Well then, I'd say that a control group is a necessity, not a 'luxury'. How do you assess change if you don't know change from what? $\endgroup$
    – BruceET
    Aug 21, 2019 at 7:02
  • $\begingroup$ Unfortunately, everyone involved have been sent the treatment already. $\endgroup$
    – Artiga
    Aug 21, 2019 at 7:03
  • 2
    $\begingroup$ You might try to compare those who accepted with those who didn't. But that would be risky. If there is a difference is it due to the treatment itself or due to being sufficiently flexible, gullible, or forward-looking to accept it. (My new box of dishwasher detergent claims to be 'new and improved' and I bought it without proof that it is any better.) $\endgroup$
    – BruceET
    Aug 21, 2019 at 7:13

2 Answers 2

Reset to default
3
$\begingroup$

The simple answer to your question is, unfortunately "no".

Without a control group and without longitudinal data you have no way of figuring out the effect of the treatment.

You can compare treatments to each other - since you appear to have data on several treatments - but what those comparisons will mean will depend, in part, on how treatments were assigned.

Improve this answer
$\endgroup$
1
  • 1
    $\begingroup$ +1 My impression from the question is that it will take a lot of work (matching, propensity, etc) to even have a shot at comparing between treatments. $\endgroup$
    – Wayne
    Aug 22, 2019 at 14:03
1
$\begingroup$

You mention several treatments so you can compare them and tell which if any is better than the others but you have no direct comparison with no treatment so that comparison is ruled out.

If there is historical data and the effect of your treatment is very strong then the historical control might be enough. If your treatment for Ebola virus disease leads to 100% survival at one year then most people would not ask for a control group. You might be interested in looking at an article entitled "When are randomised trials unnecessary? Picking signal from noise" available here which refers to some examples.

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.