Talk to me about the new Remdesivir data!

Treatments

A (short): “Remdesivir isn’t a home run but looks active and can be part of a toolbox of drugs and diagnostics that substantially lower our risk heading into the fall,” says former FDA commissioner Scott Gottlieb.

A (long): The stock market and news media are brimming with optimism around the release of clinical trial data indicating that the anti-viral Remdesivir has demonstrated efficacy and safety in an important NIH-funded trial.

How should we think about these results? When I teach students how to diligence the results of clinical trials, I talk about the 3 C’s Framework: control; chance; context. Let’s think about each in turn.

CONTROL: How credible is the comparison between those who get the treatment and those who don’t? Best protection: randomization. The NIH study is indeed randomized, perhaps the most important marker for we data peeps tracking the various studies.

CHANCE: How likely were the results to have arisen from the play of chance? Best protection: large sample size. “Randomness doesn’t look random” is a common maxim among researchers, and the small samples are particularly vulnerable to data flukes that represent statistical noise as opposed to meaningful signal. The NIH study has a relatively large sample size, around 1,000. Dr. Fauci indicated that this number of observations is sufficient to make statistically precise statements.

CONTEXT: How well do the results translate beyond the research setting? Best protection: Replication across multiple geographies, time periods, and health care settings. Here’s one big reason that scientists are urging caution in interpreting these result: It is unwise to put too much faith in one individual study, even one that is well-controlled with adequate sample sizes. Replication is the very best protection we have against overgeneralizing – of note, it is also an important protector against the vagaries of chance.

OTHER CONSIDERATIONS: Anti-virals aren’t typically “miracle drugs” — and the efficacy impacts, while meaningful, aren’t a cure. Moreover, the data indicate that efficacy may depend upon administration during a specific critical period during the disease course. It’s an intravenous drug, which also adds to administration difficulties.

As Al Tompkins of the Poynter Institute wisely concludes “Stay home, stay tuned, and wash your hands.”

Link to original FB post