How reports of a promising COVID-19 treatment was found to be fraudulent

FiveThirtyEight looks into how a paper from a major medical journal about hydroxychloroquine could be published using fabricated data.
8:27 | 06/26/20

Coming up in the next {{countdown}} {{countdownlbl}}

Coming up next:



Skip to this video now

Now Playing:


Related Extras
Related Videos
Video Transcript
Transcript for How reports of a promising COVID-19 treatment was found to be fraudulent
And the okay. You've probably heard about Hancock's core. A malaria can happen as a treatment Africa that nineteen Haitians. It. And I feel it is just a few you know support. Construction iteration Yemen has become a political issue. Last month. Research and had dropped to Clark and started a different kind of protectors. Scandal in fact it is shaken. Community. In any change helmet clintons are published from here on out. And it all came to light on Twitter. Don't started on May 22. In the medical journal the lancet published painter alleging that Jackson clerk and actually harmed included. It's clear if they were so many. Thousands of individuals and hunt. In response to World Health Organization. And other agencies extended to large ongoing trials of the drug as treatment for it. It's directly to say OK I'll take these since Tuesday. Pulse Biggio. There's nothing country. Coming into paper used to. From 96000. Patients 671. Hospitals. And six continents. This data seventies teen had cut out papers hijacks a clerk went steady and another on the possible effect coded and red blood cells. There was also a third paper a prepayment on the use it can't take care safe medication costs I connected to treat the current actress. It claimed I connecting lower at her patients ask me I just treat percent. He was like yeah. The other countries as well behind me hey it was getting so much traction that we've decided to force and it's. Deepest you'd pass. Talking hunters ulcers thousands of homes. Are by principal issue. Restriction on selling the drug. It's much Denton some horror stories. Tara Conner crew. Our can't. Use two product. This is set was having at peak impacts on policy decisions around the world. But soon after the lancet paper was published scientists untreated pinky and questioning whether the paper's findings were real. People we're treating this blog posts and and one of the TV bulls in the lancet Ebert didn't make any sense to the number of deaths and Australian. In salute. And that's sort of meeting wondered. How sloppy was a steady. And has sleuths counted on should floppy ears like this for instance. Doctors claimed that they had he hailed health data from 4400. People hospitalized in Africa. That's about 25%. Of all the cases reported in African state. That that number isn't reasonably high in South Africa. Only 5% of coated patients were hospitalized. And it seemed very unlikely that the percentage would be much higher throughout the rest of the continent. And it's doubtful surgeons here in the capital that data anyway. This company names. Hospital working room Davis. Actually records leeches or Tokyo their eight or DOD team. Two days company. Seven if it wasn't. Working the African choose to workers are scarce to say something. Com. It sounds. And super high quality. Coverage. This. Another red flag in the papers according to the database around the same number of very sick and mildly sick patients had received the drug that's who we years. The FDA only approve the drug for the sickest patients EE. Perfect. And sleep well and it's. And then perhaps the most glaring problem when journalists and Twitter excellence with buddhists or disappeared south of the candidate only had a few employees. Most of whom were not even experts in the field. The scientists on Twitter came to an unsettling conclusion. Rod. How could one of the world's top medical journals be tricked this way. Let's look at how the purity process works. After a painter is submitted to a journal the journal sends it to round three experts who provide their feedback and recommend whether the paper should be accepted. The journal editors and decide whether to publish it based on these (%expletive) That's. Accurate view is that it's not veto asked us comes through. And this. About you herbs that are eight those answers. That's not something that most of us. In our. Looking forward into you because it is so rare. Once these concerns came to light the lancet acts are just here to share their data. But the company refused. So the lives formally with convinced that after treatment for authors at that they had no faith in concerts near Davis. The fourth Loughner was found your searches here sometime desk. Never was there been a situation. Where and one group. One person published two papers within a few sure beats in the humans premier medical journals that were fabricated. We reached out to the paper's authors are common stock and death again it's time. But the paper's lead author Monday Mara who is not surgeons carefully wrote. It is now clear to me that in my hope to contribute this release are hearing kinda great did not do enough to ensure that data source with appropriate. Or that and for all the distractions. Both directly and indirectly I am truly sorry. Think about what good. Process. I think that this post publication accurate figure is one so you. But I think that there ear actors. Out there in goofy happy. I'm focus its dealer to ease hits you. Dismissed. Yes two. Dismiss evidence based medicine and Z it's this and those are some things. Reluctant to say and this won't happen again. At least for now every journal in the world is gonna be their first question is going to be what does this database that you're using. And there's ways of preventing this that are pretty easy but we don't currently do it. A lot of key it is going to be commitment for open science and sharing things more so. A lot of journals now require you to statewide can't share your data. Where human subjects please there's always because you know people could be identified and you can't share the database so that's always going to be the case. But there's no reason why. The authors should be required to share their statistical. This whole episode but tonight a number of problems with the scientific process. But perhaps it's also taught us some lessons about how scientific publishing can improve in the future. Particularly in the age of hope it. Because of the art dean. Have a critical crisis moment. It's not a reason to lower standards actually recent increase standards that they're so much commission meets this news. We want to make sure it ourselves and nation are being cautious. In terms of how the police and news that's believed that it. And it need circuit equal or higher standards brought in thinking. Granite Cray suspension lower standards and policies that.

This transcript has been automatically generated and may not be 100% accurate.

{"duration":"8:27","description":"FiveThirtyEight looks into how a paper from a major medical journal about hydroxychloroquine could be published using fabricated data.","mediaType":"default","section":"ABCNews/fivethirtyeight","id":"71392340","title":"How reports of a promising COVID-19 treatment was found to be fraudulent","url":"/fivethirtyeight/video/reports-promising-covid-19-treatment-found-fraudulent-71392340"}