Grading Drugs Within a Class: Level 1-2 of Evidence

The paper¹ categorizes levels of evidence for comparison of drugs within a class from 1-4, each with unique threats to validity. Levels 3 and 4 will be discussed later.

Superiority trial

Is Drug A better than Drug B or placebo?

Equivalence trial

Are Drug A and Drug B close enough in effect that we can treat them as clinically equivalent?

Noninferiority trial

Is Drug A not unacceptably worse than Drug B? Though the two may appear to be roughly similar, this is different from the equivalence trial since the study works under the acknowledgement that Drug A is worse. How acceptable that is, upon a multitude of factors, is another consideration.

Primary Prevention Trial

whether a treatment prevents a disease or event before it has happened for the first time.

Secondary Prevention Trial

whether a treatment prevents recurrence/complications after the person already had the disease/event.

Level 1

This includes RCTs that compare drugs head-to-head via clinically important outcomes. Trials that conduct such studies are referred to as Level 1 in the literature.

Because head-to-head studies (drug A vs drug B) only compare drugs against eachother and do not necessarily include the placebo (drug A vs drug B vs placebo), one of these drugs, A or B, must be anchored to some previous trial that confirms its benefit. That is to say, we must verify that we are indeed comparing beneficial drugs rather than ineffective drugs.

As level 1 trials are the strongest basis for claims on the superior drug, researchers, and hence, acute readers must make sure to consider the appropriate dosage for the drug. This is especially consequential in level 1 trials as one drug at one dosage may confer more/less benefit than another drug even though they are equally efficacious. Of course, this is not unique to level 1 trials.

Trial size and methods also must be considered before concluding equivalence, as equivalence trials require larger sample sizes compared to other trials. Why?: proving clinical equivalence requires presision. When comparing two drugs, researchers set an acceptable range in the difference of efficacy; and oftentimes, these differences are marginal.

Level 2

Same thing as level one, but rather than using clinically significant outcomes, they use surrogates; so firstly, one must make sure to verify the validity of the surrogate.

Level 2 evidence can be useful. But, this does not allow one to compare across different level 2 trials of the same drug class. Samples may contain individuals with different risks, different methods of study, different geographical regions, etc. As such, one cannot compare treatment arms across trials—usually. It is theoretically possible but verification of various aforementioned factors are necessary.