but had marginal scientific relevance in terms of statistical significance (as is the case with many low-number case studies)
As a professional engineer [and professional pilot], I would deduce that zero evidence of error in over 15000 flight hours of 3 pilots would provide ample 'scientific relevance in terms of statistical significance' to support the premise that reliable colour discrimination is not required in the task of piloting an aircraft.
Further, those with protanopia, like these 3 pilots, are supposed to have more severe colour deficiency than the vast majority of colour deficient pilots.
I think some aviation doctors and regulators alike are too scared to listen to logic and historical evidence that colour deficient pilots pose no increased risk to aviation safety.
Road transport authorities world-wide gave up banning colour deficient drivers long ago. Why? - because the evidence shows they are able to perform the task of driving as safely as a colour normal.