3 ms·

Wouldn't it be a lot clearer if it were something like "issues at three sites doing Pfizer vaccine trials" instead of "issues in Pfizer’s vaccine trial"? The s

Wouldn't it be a lot clearer if it were something like "issues at three sites doing Pfizer vaccine trials" instead of "issues in Pfizer’s vaccine trial"? The singular "trial" in the headline implies that the flaws are in the trial as a whole rather than in a particular distinct subset of the trials.

hcrean 5y ago
Well, there are: The provable lack of regulatory oversight and demonstration of the existence of poor practices at the company. The argument that just because one of a company's trials are unsound doesn't mean that the rest aren't is not a good one, when the whole objective is to provide assurance in proving the safety of a product.
- quadrangle 5y ago
  I get that it's hard to find the best balance in a headline let alone in a summary article. Yes, evidence of a problem in a portion of a trial is indeed evidence of problems in the trial. I can see room for misunderstanding no matter what. Emphasize the portion of sites where the problem was observed and it wrongly can be inferred that the other sites are fine even if they have no evidence either way. Emphasize the problem as a generic feature of the trials overall and it can be wrongly inferred that the whistle is being blown on some aspect of the trial that doesn't have any site-specific aspect. The story that I'm hearing in all this is that the whistleblower brought out issues that were specific to certain sites in a way that has no direct bearing on other sites besides being a reason to scrutinize the rest of the sites as well. Maybe an analogy would be helpful. If a whistleblower observed a disregard for safety protocols at a food processing facility, that's not a randomized sampling that indicates a high likelihood of the problem being throughout all similar facilities, but it doesn't indicate otherwise either. It opens the question whether this is a typical problem in such facilities or if that facility is an exception. If the facility was supplying brand X, there's at least some sense to criticizing a headline like "whisteblower reveals brand X unsafe food handling" instead of "whistleblower reveals unsafe food handling at facility run by brand X". The former implies wider issues, the latter errs toward not making that presumption.
  - srcreigh 5y ago
    Your argument is basically that this article may be largely speculation, assuming that no information was withheld (even though the article contradicts itself in its own communication about FDA review of the trial)... and that this speculation is the same as minimal conclusions based on available evidence. I disagree. One bad study doesn't mean all scientific conclusions are bad. One bad study doesn't mean Pfizer's trials are bad. Ignorance doesn't prove anything...
    - quadrangle 5y ago
      I don't know if you imagine me to be partisan here. I was arguing simply that headlines are hard to balance, and that people can interpret articles with a slant. I wasn't saying that one bad study means the overall trial was bad. The issue is that SOME people might think that. (Also others might think that if it's ONE study that's bad, the overall trial WAS fine, when that's not clear either). The risk is whether real people take away a too-strong message and run with it. In other words, journalists have SOME responsibility to not mislead people even when the main way people get off-track is their own erroneous jumping-to-conclusions as readers. We KNOW that readers do such things, so we have to be careful to minimize it.
  - NoPie 5y ago
    Yes, the problems in one site, especially if not dealt with properly, might indicate larger problems. We all know that the trials were rushed and for a very good reason, there was a pandemic. Still, if issues are discovered they should be resolved. Pfizer is a private company, they are not obliged to be very transparent with everything they do. FDA is the main organisation that oversees such trial in the US, so this is their responsibility to investigate.
NoPie 5y ago
You apparently use the word trial differently than it is used in pharma. Pfizer had one large trial that involved many sites. Three of those sites run by Ventavia had problems. That's why the headline says “in trial”. But even if you didn't know that it is explained in BMJ article “... Ventavia had enrolled more than 1000 participants at three sites. The full trial (registered under NCT04368728) enrolled around 44 000 participants across 153 sites that included numerous commercial companies and academic centres.”

hcrean 5y ago

Well, there are: The provable lack of regulatory oversight and demonstration of the existence of poor practices at the company. The argument that just because one of a company's trials are unsound doesn't mean that the rest aren't is not a good one, when the whole objective is to provide assurance in proving the safety of a product.

quadrangle 5y ago
I get that it's hard to find the best balance in a headline let alone in a summary article. Yes, evidence of a problem in a portion of a trial is indeed evidence of problems in the trial. I can see room for misunderstanding no matter what. Emphasize the portion of sites where the problem was observed and it wrongly can be inferred that the other sites are fine even if they have no evidence either way. Emphasize the problem as a generic feature of the trials overall and it can be wrongly inferred that the whistle is being blown on some aspect of the trial that doesn't have any site-specific aspect. The story that I'm hearing in all this is that the whistleblower brought out issues that were specific to certain sites in a way that has no direct bearing on other sites besides being a reason to scrutinize the rest of the sites as well. Maybe an analogy would be helpful. If a whistleblower observed a disregard for safety protocols at a food processing facility, that's not a randomized sampling that indicates a high likelihood of the problem being throughout all similar facilities, but it doesn't indicate otherwise either. It opens the question whether this is a typical problem in such facilities or if that facility is an exception. If the facility was supplying brand X, there's at least some sense to criticizing a headline like "whisteblower reveals brand X unsafe food handling" instead of "whistleblower reveals unsafe food handling at facility run by brand X". The former implies wider issues, the latter errs toward not making that presumption.
- srcreigh 5y ago
  Your argument is basically that this article may be largely speculation, assuming that no information was withheld (even though the article contradicts itself in its own communication about FDA review of the trial)... and that this speculation is the same as minimal conclusions based on available evidence. I disagree. One bad study doesn't mean all scientific conclusions are bad. One bad study doesn't mean Pfizer's trials are bad. Ignorance doesn't prove anything...
  - quadrangle 5y ago
    I don't know if you imagine me to be partisan here. I was arguing simply that headlines are hard to balance, and that people can interpret articles with a slant. I wasn't saying that one bad study means the overall trial was bad. The issue is that SOME people might think that. (Also others might think that if it's ONE study that's bad, the overall trial WAS fine, when that's not clear either). The risk is whether real people take away a too-strong message and run with it. In other words, journalists have SOME responsibility to not mislead people even when the main way people get off-track is their own erroneous jumping-to-conclusions as readers. We KNOW that readers do such things, so we have to be careful to minimize it.
- NoPie 5y ago
  Yes, the problems in one site, especially if not dealt with properly, might indicate larger problems. We all know that the trials were rushed and for a very good reason, there was a pandemic. Still, if issues are discovered they should be resolved. Pfizer is a private company, they are not obliged to be very transparent with everything they do. FDA is the main organisation that oversees such trial in the US, so this is their responsibility to investigate.

NoPie 5y ago

You apparently use the word trial differently than it is used in pharma. Pfizer had one large trial that involved many sites. Three of those sites run by Ventavia had problems. That's why the headline says “in trial”. But even if you didn't know that it is explained in BMJ article “... Ventavia had enrolled more than 1000 participants at three sites. The full trial (registered under NCT04368728) enrolled around 44 000 participants across 153 sites that included numerous commercial companies and academic centres.”