"Why doesn't anyone question the ECMO machines? I'm definitely no doctor, but as a software engineer if my system failed 30-40% of the time I'd stop using it."

"Why doesn't anyone question the ECMO machines? I'm definitely no doctor, but as a software engineer if my system failed 30-40% of the time I'd stop using it." ECMO is a hail-mary treatment for people who would almost certainly die otherwise. A 40% chance of death is better than a 99% chance of death. What you're saying here actually happened with ventilators (though ventilators are nowhere nearly as invasive or complex as ECMO).