The 2% solution
The 2% solution
There has been discussion about the 2% (per year) risk of complications associated with mechanical heart valves and Coumadin. I have thought about it quite a bit. The 2% figure is not actually that straightforward. Suppose you follow 1,000 patients and you identify a 2% occurrence of complications. It's possible that each and every patient has the identical risk. But if half of them have a 1% risk and the other half have a 3% risk you will get pretty much the same overall results. So it would be important to know what factors affect the risk - valve type, age, control of Coumadin level, other health problems etc. The same way there are multiple risk factors for heart attacks, there may be multiple risk factors related to valve replacement that have yet to be to fully quantified.
In addition, how did they grade the severity of the complications? Are they only including serious, life-threatening complications, or are they (more likely) including any complication, as long as it was reported to a doctor?
If you find the math tricky to follow, try this example. A 2% risk is 1 chance in 50. So take a deck of cards and remove the ace of spades and the ace of clubs. Now pick one card from the remaining 50, look at it, then replace it. If you picked the ace of hearts, let that represent the risk of a complication.
The odds are cumulative in the sense that statistically the chance of you picking the ace of hearts at least once in 20 tries is much greater than the chance of picking it on the first try.
However, this assumes that the odds remain constant over time. This is probably better than your chances with cigarettes and cancer where your odds probably get worse over time as your lungs are progressively damaged. e.g. your chance of getting lung cancer the very first year you smoke is probably a lot smaller than the chance after 20 years of smoking.
Alan