Question

7. A randomised control trial has been conducted that demonstrated a difference in the positive response to a medication in the intervention group compared to the control group. Describe the concept used to ensure that the detected difference between the two study groups did not occur by chance alone.

Answer 1

A study design that randomly assigns participants into an experimental group or a control group is called a Randomised control trial.

Advantages:

1. It reduces the bias .
2. It uses the probability theory to express the outcome.

Concept used to ensure the study is as follow:

Schematic linear causal model is used to clear that difference detected in the group is not by chance alone.

where, Yi is the result for unit I (which might be an individual, a town, a clinic ward), Ti is a dichotomous (1,0) treatment sham showing whether I is dealt with, and βi is the individual treatment impact of the treatment on I: it speaks to (or controls) how much a worth t of T adds to the result Y for singular I. The x's are watched or surreptitiously other straight reasons for the result, and we guess that (1) catches an insignificant arrangement of reasons for Yi adequate to fix its worth. J might be enormous. The unlimited heterogeneity of the individual treatment impacts, βi, permits the likelihood that the treatment collaborates with the x's or different factors, so the impacts of T can rely upon (be adjusted by) some other factors. Note that we needn't bother with I addendums on the γ's that control the impacts of different causes; if their belongings vary across people, we incorporate the communications of individual attributes with the first x's as new x's. Given that the x's can be undetectable, this isn't prohibitive. Use here varies across fields; we will regularly allude to factors other than T spoke to on the right-hand side of (1) by the term covariates, while taking note of that these incorporate both what are now and again marked the 'freely working causes' (spoke to by the x's) just as 'impact modifiers' the point at which they cooperate with the β′s, a case we will re-visitation of beneath. They may likewise catch the likelihood that there are various baselines for various perceptions.

To show, assume that T is dichotomous. For each unit i there will be two potential results, commonly labelled Yi0 and Yi1, the previous happening if there is no treatment at the time being referred to, the last mentioned if the unit is dealt with. By review of , the contrasts between the two outcomes, Yi1 − Yi0, are the individual treatment effects, βi, which are ordinarily unique for various units. No unit can be both treated and untreated simultaneously, so just one or other of the results happens, however not both—the other is counterfactual with the goal that singular treatment impacts are on a fundamental level imperceptible.

It states that the average treatment effect is the difference between the average outcome in the treatment group minus the average outcome in the control group so that, while we cannot observe the individual treatment effects, we can observe their mean.