That child could have easily died running involuntarily from bullies or dogs. Or from having his heart race due to anything else: being startled, nervous, or excited on a first date or whatever.

He didn't die while undergoing a medical procedure; he died of a bad ticker.

The medical establishment didn't cause his heart ailment, it just neglected to inform him. That is unfortunate, but it is somehow not the in the same category as, say, being treated for a broken leg, and given a fatal overdose of some painkiller.

Honest accounting for deaths due to medical error must only count situations when someone dies of something irrelevant, not directly connected to the condition for which they are treated. (Certainly not of the condition for which they are treated, especially if that kind of condition itself carries a reasonable probability of causing death!) That cause of death which occurs must not be a recognized risk factor in that kind of treatment (when that treatment is correct).

The key reasoning is that without any medical intervention at all, the child would also have likely died, and of the same thing.

If we count as a "death due to medical error" a situation in which medicine merely failed to prevent a death, that has to be very well justified. This means that it was very probable that if the mistake had not been made, the patient would have made a full recovery: their condition would have cleared away to the point of that condition no longer being a death risk, and that this positive outcome is virtually guaranteed in such cases when the correct treatment is applied.

The advice not to ever strain yourself physically for the rest of your life or get excited is not a treatment which results in the condition being cured. The condition remains in place, along with the risk of dying.