That statement still reads wrong. Assuming the risk is 2%, by year 30 the probability is still 2% in that coming year.
The probability that something might have happened over the prior 30 years is 45%. But it’s not 45% in year 30 if you happened to have no issues in the prior years. And it’s not 0% either if you’ve already had one. It’s still 2%.
Thank you for the clarification but did you not read my previous posts?
If so, I'm not sure why you think I was/am arguing that the risk specifically for year 30 or any other specific year was 45%.
I believe I clearly spelled out the 45% was the risk over the 30 year time period and the 2% was for each individual year. I even exactly stated that individual years risk does not change.
I'll have to assume you haven't read the posts or didn't understand what I was saying (or you were too emotional/upset?) so I'll copy and paste them below so you can see that I had made it clear multiple times that the risk for any particular year was 2% but the risk over the 30 year time period was 45%.
I even provided the formula and calculated the 45% over the 30 year period. Seriously, read my first post. That's what it was all about.
Where did you think the 45% over the 30 year period came from?
Do you realize that the .98 number I used in the formula came from subtracting the 2% yearly
risk from 100%?
This was also the very reason why I said that you were incorrect when you told the original poster that his odds of encountering a bleeding event was 2% over the 30 year period he was asking about when they were actually 45% over this period.
What I am glad to see is that you seem to have now fully grasped this concept that the probability over a time period differs from the probability over an individual slice of time within that time period.
It's a good thing and should help you avoid giving out bad information in the future. Win for everyone here.
Now to the posts... with helpful highlighting in bold to help you find the relevant information... and a note on the probability formula...
This was my first post to the original poster who was asking for the odds of having a bleeding event over a 30 year period
"The answer is:
1 minus the probability of the event NOT happening
(1-(.98^30))100= 45% over 30 years at 2% per year.
(Helpful highlight: Superman, the above is the formula for probability of an event happening over multiple years, not for a specific year, and is where the 45% number comes from. Please note the "over 30 years" and the "2% per year" mentioned as inputs for the formula.)
So a little less than half of people receiving a mechanical valve should be expected to encounter a “stroke or bleed” event at some point over a 30 year period
If the risk of stroke or bleed is lowered to 1% the answer would be roughly 26%, much better and why it’s important to keep your INR in range.
Note: Your probability of having an “event” is not 50-50 on the specific year 35, your probability of having an “event” is 50-50 by the time you reach year 35
The longer the time period, the higher the probability that the event will happen within that time period."
This was my second post responding to you specifically:
"Yes, your chance this year is still 2% but your chance of having had a bleeding event over the last 32 years was around 48%.
As far as stats, they aren’t cumulative like that. I’ve been event free for 32 years on warfarin. There is still a 2% chance this year of an event just like there was the first year I received my valve. But that’s a 2% chance among all patients, not me individually.
Essentially a coin flip and you did well.
Over the next 32 years, your chance of having a bleeding event is also 48%. I wish you luck once again on that coin flip.
Note: 1-(probability of not having an event ^32) is 48%
No, Dick's chance of encountering a bleeding event over 55 years was around 67%, not 110%.
If they were cumulative, our friend @dick0236 would be having adverse events annually right now (55 years on warfarin) at a 110% chance! He’s at the same 2% risk.
Note: The percentage will never go over 100% (and it's not linear). Look up Limit Theory in any first year Calculus book for why this is.
Think if it this way. X% of people die in car accidents annually. Now that you’ve been driving 30 plus years, do you feel like any time now you’re going to die when you get in a car?
Again, no. See above.
Superman, you're a reasonable guy, can you not see that this is a perfect example of why it's not a good idea for you guys to give out medical advice on this forum?
What you believed and offered as fact actually isn't fact.
What other "facts" are you guys giving out that aren't facts either? And how would you know?
Don't you think telling this guy that his chance of a bleeding event (significant or not) over a 30 year period is 2% when in fact the probability is actually 45%
could make a difference in his decision making?"
My third response to you:
You’re reading me wrong. I was saying if this misconception (odds increasing every year) were true, then….
But it isn’t true. This year, like any other year, it remains roughly a 2% chance (assuming that statistic is accurate). Of course I don’t think anyone has a 110% chance of a negative event.
I don’t disagree with the statistics as you say them and I don’t see how they disagree with any point that I’ve made. I also don’t see where I gave any medical advice beyond what you did (aiming to clear up a statistical misunderstanding) so get off your high horse please. Your incessant beating of that drum is more than obnoxious at this point.
Click to expand...
"You may be reading the original poster wrong.
The original poster was asking what his risk was over a 30 year period based on a risk of 2% per year, not simply for year 1 or any specific individual year within the 30 year period.
Your answer of 2% was incorrect and bad information if the poster was to rely on this information to make his valve decision.
(Again, the odds for the individual year doesn't increase but as the number of years increases, the probability of an event occurring in the total number of years does increase.
That's how the probability of 2% the first year increases to 45% by year 30 and grows higher with each year after
Have a good day."
End of posts.
To wrap this up, I will take responsibility for not adding any wording specifying "30 year time period" at the end of the sentence italicized above that you seem to be referencing. My bad here.
The 30 year time period is alluded to in the sentence above the italicized sentence you have issue with and specifically stated in the sentence 3 sentences above the italicized sentence... and mentioned multiple times and the entire point of all three of my posts on this subject but I still should not have expected you to make this connection.
So yes, I agree with the 2% per year risk and the 45% risk is 30 over the year time period, especially since I explained this and provided the formula and calculated this number out in my original post for the original poster.
P.S. Superman, you're better than this and I expected a better effort, this was little more than grasping at straws here.