Open Thread

This is the 1.5 year clinical trial for the GLP-1 drug Ozempic (Semaglutide) made by Novo Nordisk.
Like all clinical trials it was funded, designed, performed and selectively released by the pharma corporation selling the drug.
The authors report being paid and being employees of Novo Nordisk and a dozen other pharmaceutical companies.

The trial excluded those with a history of mental health problems, pancreatitis, endocrine and thyroid problems, renal impairment, history of cancers, and various other medical conditions.
It is standard practice for clinical trials to exclude people who will have higher rates of adverse effects from the drug. This biases the results in favor of the drug compared to the real world.

The drug group lost an extra -12.7 kg (27 pounds) compared to the placebo group. Weight loss in the drug group ended by week 52. By week 60 the drug group began regaining weight. From weeks 60-68 those in the drug group regained around 5 pounds (figure 1).
89.7% and 86.4% of people in the drug and placebo group respectively experienced adverse events.
9.8% and 6.4% of people in the drug and placebo group respectively experience any life threatening event.
https://www.nejm.org/doi/full/10.1056/NEJMoa2032183

If you read the adverse event list there is a curiosity. The trial reports similar adverse events rates in both groups but in table 3 the drug group has around 20% more participants reporting adverse events in many of the sub categories.
The reason for this discrepancy is relatively easy to find in table 3. The study didn’t report the NUMBER of adverse events they instead reported the number of PEOPLE experiencing ANY adverse event.
People can experience many adverse events. Someone with 10 adverse events will be reported the same as someone with 1 adverse event. This statistical slight of hand is used to make the drug appear safer than it actually is.

In table 3 you can see the number of adverse events and find out how much this statistic falsehood does.
With the trick you have the drug causing 50% more people to have “any” life threatening event. Without it you get the drug causing 75% more life threatening events.
Likewise, the drug caused 50% more adverse events if you don’t want to state it in a way to deceive people.

Another trial was pretty much the same and used the same lying with statistics the previous trial used.
https://pmc.ncbi.nlm.nih.gov/articles/PMC7905697/#H1-3-JOI210014

We didn’t perform these studies and we don’t have access to the full study data. If the pharma study and authors do things this borderline fraudulent in the open with no consequences what do you think they’ll do in secret? What do you think occurs in an environment where fraudulent practices are this common?