If you go and take a look at the ACCORD study
reports and
here, you'll find that what it says is:
Suggesting that the key risks for intensive therapy and CV events were either a previous CV event and/or a higher starting Hba1C that was lowered rapidly. Or in other words, those with neither of the above given intensive therapy had fewer CV events than the standard therapy cohort.
Even the authors of the various papers that have been written about ACCORD recognise that there are many questions that remain to be answered and need further research.
Worth having a read through if you are concerned. I'm not sure the results were fiddled, but as it's a set of statistical data, you can report this in many ways, and that seems to be his biggest complaint in the video. That the headlines that were made from the study reflected an incorrect view of what the study really found.