This is an op-ed written by Jerry Avorn. His bio from WaPo:

Jerry Avorn is a professor of medicine at Harvard Medical School and co-director of the Program on Regulation, Therapeutics and Law (PORTAL) at Brigham and Women’s Hospital in Boston.

The author argues that the accelerated approval process within the FDA has led to drugs reaching the market that fail to show efficacy when administered more widely. They write:

For these medicines, the FDA has granted “accelerated approval” based not on patients’ clinical improvement but merely on small changes in lab tests. After accelerated approval, drugmakers are asked to do follow-up studies to prove that their drugs actually help patients. But the government may allow many years to complete such studies, and they are often started late or done poorly, my research group has found.

As an example of the accelerated approval process gone wrong the author points to Makena:

In 2011, a medication meant to prevent premature birth, Makena, was approved on an accelerated basis, but then was never clearly shown to work. Getting it off the market took until this past April.

The author proposes a middle-ground between the pass/fail status that the FDA gives medicines currently:

The FDA’s simple all-or-none decision framework is a poor fit for this evolving scientific reality. The agency should instead create a new status for some drugs that is neither full approval nor outright rejection — a conditional approval that would allow a new drug with ambiguous credentials to be used only in a limited way until better evidence is collected.