Question on probability of later problems due to coumadin

[geebee]

I hate to correct such a smiling face, but the statement "every year I have a 1-2% chance of a bleeding event" may not be true. Remember, you
are not a coin, whose outcome can only be heads or tails(Objective Probability). You are rather a data point in a distribution, where in the average of probability is 1 to 2 % depending on how a bleeding event is defined (Subjective Probability). Also, is the distribution a Poisson or Gaussian and where do your own risk factors put you on that curve. Statistics only tell you what the mythical average person?s chances are, but your chances might be much lower. I think being on this board, having access to Mr. Lodwick and home testing put you at a much lower probability.

Now that is sweet, simple and surely works for you

OK - I'll accept a lower percentage. At least I think that is what you said.:

 

[RCB]

one of my friends sent me this link, it is about congenital heart defects and a 'risk score' http://www.aristotleinstitute.org/aboutScore.asp
as i was looking at it I kept thinking about this thread so thought some of you would find it interesting
here is a little bit about how/why it was started
DEVELOPMENT
The motivation behind the Complexity Score Project was a growing frustration of pediatric cardiac surgeons over the fact that their surgical performance was being evaluated based on hospital mortality without regard for the complexity of the operations performed. A working group of Congenital Heart Surgeons from Europe and the United States decided to develop a risk-stratification method which could be adapted to our specialty.

When starting this project in 1999, two difficulties were encountered: 1) Multi-institutional databases were just starting and there was no reliable data yet available. 2) Due to the absence of risk stratification, the more prominent centers dealing with the sickest patients and potentially having a significant mortality were very reluctant to send their data. It was, therefore, necessary to base this risk-adjustment on an evaluation that was largely subjective. Following many discussions, it was concluded that a subjective probability approach based on the consensus of a panel of experts was valid, provided that the risk-adjustment score is subsequently validated based on collected outcome data.

A group of 50 internationally accepted experts has been working for more than five years on a new method to evaluate the quality of care in Congenital Heart Surgery (CHS) that is called Aristotle. Senior, experienced congenital heart surgeons considered the possible risk factors for each procedure and assigned scores based on potential for mortality, potential for morbidity, and anticipated surgical difficulty.

The Aristotle system, electronically available, has been introduced by both the European Association for Cardio-Thoracic Surgery (EACTS) and Society of Thoracic Surgeons (STS) as an original method to compare the performance of Congenital Heart Surgery (CHS) centers. Pediatric cardiologists have joined the project and are currently developing a complexity score for interventional cardiology procedures.

This subject was discussed on an AF board run by a former member of this board, Jack Drum. He had to shut it down because he couldn’t handle the workload anymore, so I can’t point you towards that discussion. It is a growing problem, because of “pre-selection” or “Cherry-picking” of pts. for procedures. The point was exactly as this article points out, you can’t always tell how good someone is by success rate because some drs. pick patients based on how simple the case is.

So read and be forewarned!

 

[Dennis S]

Emotional Motivation to comprehend statistics.

Emotional Motivation to comprehend statistics.

I read through this entire thread last night and found it fascinating. There are a lot of smart people on this board, and many different types of intelligence. Many of us have a particular interest in this subject because we are afraid we might choose, or may have already choosen, a valve that exposes us to an elevated risk of a stroke or bleeding event. Like Sue, and many others, my real nightmare is the possibility of a stroke for those taking Warfarin. I am convinced Andy's mathmatics are right. Also, I think Andy and most of us would agree that their value as a predictor of future events is diminished by the fact that they have no way of assimilating the improvements in Warfarin management. These consist of improvements in the medical field (far less in the case of some than others), and especially improvements in patient education and management. The internet in general, and this board in particular can greatly improve the odds of avoiding the dreaded stroke.

Some of our members have done a great job so far on educating the rest of us. I think I have 1 major question that needs to be addressed to make this thread as helpful as possible. If I understand correctly, the discussion so far concerns the use of statistics to predict the possibility of a future event, e.g. a stroke. But the reason many of us have invested time and energy following this topic is we want to know how we effect or might effect this possibility with our valve choice . I would guess that any type of valve problem that requires a replacement (tissue or mech.) increases the risk of stroke. But if the valve is bad, replacement (of some type) is a no brainer. So far, no difficult decision to make. Now (if you are lucky) you get to decide what type of valve you want-a very difficult decision. Now, the statistics become very important to many of us. So here is, for me the big question: I now have an idea (see through a glass darkly) what statistics can tell me about strokes for valvers on Warfarin. What do the statistics tell us about strokes for valvers that are not on Warfarin? The reason I want to know is that I made a choice (and others are considering a choice) that will put them in 1 camp or the other. My bad valve left me no choice but to go to "camp". I would like to know f the choice I made expose me to a greater risk of stroke-and if it does, what can statistics tell us about the extent of that greater risk? I would love it if you research/statistic folks have the energy to go this last mile. And thank you for what you have contributed so far!

 

[Andyrdj]

This one's for Dennis

This one's for Dennis

What do the statistics tell us about strokes for valvers that are not on Warfarin?

I think the young people's study in another post mentioned both similar stroke rates and similar overall survival rates in both tissue and mechanical patients for aortic, though not so for mitral - this for patients who either had a mech when young, or a tissue when young and multiple re-ops.

http://www.valvereplacement.com/forums/showthread.php?t=16301

to quote The ten-year cumulative incidence of embolic stroke was 6.3+/-2.4% for mechanical AVR patients, 6.4+/-2.9% for bioprosthetic AVR patients, 12.7+/-3.9% for mechanical MVR patients, and 3.1+/-3.1% for bioprosthetic MVR patients.

now, this gives us 93.7% chance of NO stroke in 10 years for the mech AVR (very similar figures for bio)

So (yearly_chance_of_free)^10=0.937

therefore, taking the tenth root of each side, we get

average yearly_chance_of_free=0.9935=99.35% on average free, or 0.65% per patient year

now, using the upper and lower bounds of the +/-, we get 3.9% lowest and 8.7% highest probs of event, therefore 96.1% freedom at best, and 91.3% at worst

best_yearly =(0.961)^(1/10) ( something to the power of a tenth is equivalent to the tenth root of that number)
best_yearly =0.9960 = 99.6%, therefore 0.4 % prob of event per patient year

worst_yearly =(0.913)^(1/10)
worst_yearly=0.9909= 99.09%, therfore 0.81% per patient year.

so a low of 3.9%, and a max of 0.81%.

I hope these figures are pleasing to people when compared to the 1-2%
I've re-check the error on the estimates using another method and the range of values checks out

(method described here for masochists
http://people.hofstra.edu/Faculty/Stefan_Waner/RealWorld/calctopic1/linearapprox.html - see "Estimating the error of y = f(x) ")

So anyway, this trial seems pretty good for younger AVR patients in terms of stroke risk

 

[Andyrdj]

And this bit's in English!

And this bit's in English!

The good news is that relatively recent valves (15 year old or more) have low stroke risks.

The survival rate overall? Well, we haven't seen the causes of that, but there are clearly a bundle of causes other than stroke risk affecting it - since the stroke incidence cannot account for the death rate.

Some will be just because of related heart damage - from the ops, and years of being worked through an inefficient valve. Some will be from non-heart related causes.

Best news is that modern valves seem to be addressing the other causes that limit the effectiveness.

Mech - on-x and others look to be moving towards a lower INR (I think internal bleeding is the main problem, other than stroke). Any others?

Bio - More effective anti calcification treatments, big improvements in orifice size, haemodynamics, and guarding against patient/prosthesis mismatch.

other reasons to be cheerful

Technology to build new heart tissue is on its way - your damaged heart tissue could be patched up!

Minimally invasive surgery is making ops (and presumably re-ops) less traumatic.

Percutanous valve replacement has already replaced both native valves and an ailing bioprosthetic valve.

It's a good time to be alive!

 

[sue943]

What do you do for a living Andyrdj? So with my addled brain, what is my risk factor with mechanical mitral and aortic valves and taking warfarin?

 

[Dennis S]

Hi Sue

Hi Sue

I too will be waiting for Andy's response. I must admit that, to some extent I posed my question with you in mind, particularly what seemed like remorse for a valve decision made without as much information as you would have liked. If I am understanding things correctly, you should not feel that you made a choice that exposes you to a substantially greater likelihood of stroke. (I think). I think this is more true for your Aortic valve than for the Mitral.

When talking to Andy I sometimes I wonder if we are talking to the next Bill Gates!

 

[RCB]

Funny thing about lawyers

Funny thing about lawyers

They are trained to ask questions in court, that they are certain
of the answers. It doesn't work that way unfortunately in biostatistics.
Your specific probability is a combination of risk factors such as age, type of valve, family history, etc. Then you can compare that the "normal" (someone who is dead center in a Gaussian distribution [I have added this so
Andy doesn't start a lecture on The Central Limit Theorem]) person probability who has your same risk factor, absent the heart valve and ACT.
I'm not sure those statistics exist.

A more general question could be what is probability of someone who is on ACT and has a mech. valve and what is that same probability for the "general"
population of the USA, which together would yield a comparison. The problem with that statistic is that it is so general as to not be very useful.
The young have very low rates, the old, higher rates. Remember too, there are three basic causes of strokes, e.g., thrombosis, hemorrhage, embolism which can affect the probability of stroke for anyone person. Precise answers, like precise questions are hard to come by.

But wait- There is more!

Having compiled those statistics, to get a good feel as to how to evaluate your decision, you must recalculate the same risk of stroke from a tissue valve ( yes folks, I?m sorry but have a tissue valve does not mean no chance of stroke) and make that comparison.

Bottom line is questions about ACT statistics should be asked in the ACT forum where we a have a national known resident expert- Al Lodwick, who I think you know.

 

[Dennis S]

A retired court room lawyer.

A retired court room lawyer.

RCB-you are right about the training of trial lawyers (which I was until very recently). But, as far as this forum is concerned I have never posed a question where I was sure of the answer. In fact, how could anybody be sure of anything based on this complicated (to me) thread.

What I want to see is a statistical comparison of the likelihood of a stroke for heart valvers on Warfarin as opposed to heart valvers who are not on Warfarin. I can assure you that I do not know the answer to this question, and am quite interested in finding the truth, even though I have already made my choice. I did have the feeling that a discussion concentrated on the likelihood of stroke for those on Warfarin-with no discussion of the fact that you would encounter some risk of stroke even if you choose a tissue valve & no Warfarin-doesn't allow valvers to compare apples to apples.

Finally, to me I would see Al as a person who can help tilt the odds (that otherwise apply to the general population) a little bit in our favor. But what I am looking for here is, I believe, a mathmatical answer. Al is a resource that might save your life, or quality of life. To me this thread is not about my health, but a question of fact that I find very interesting.

 

[Andyrdj]

Thanks for the flattery Dennis and Sue

Thanks for the flattery Dennis and Sue

Sue
Well, I'm a computer programmer - just moving to a new job on Monday to do web work with a health insurance company.

Will look at doing a similar analysis to the one already done for Mitral valves - although I should say that this particular study is possibly outside your own age group.

"The next Bill Gates"? Very flattering Dennis, thanks.

But It's actually only really A Level Maths probability that I'm using (for the americans, it's equivalent to High School age).

My background is Physics and natural sciences, which I studied at Cambridge University, England. Hence these stats are familiar to me, although I haven't reaally been stretching my maths skills in my IT job that much - so this online debate is a keen opportunity for me to refresh rusty skills.

RCB raises some interesting points about comparative statistics - will get on Wikipedia and look up the Central Limit theorem - I'm sure I've used it before, but in probability you have to be extremely careful what questions you ask - some averages look very much the same in english but can in principle be very different.

I did somewhere find the average for all people (which isn't the same as someone just like you minus the valve and ACT, but it's a start)

Would happily see this post in the ACT section if anyone wishes to move it - just tell me, i'd miss the pontificating bit.

 

[RCB]

RCB-you are right about the training of trial lawyers (which I was until very recently). But, as far as this forum is concerned I have never posed a question where I was sure of the answer. In fact, how could anybody be sure of anything based on this complicated (to me) thread.

What I want to see is a statistical comparison of the likelihood of a stroke for heart valvers on Warfarin as opposed to heart valvers who are not on Warfarin. I can assure you that I do not know the answer to this question, and am quite interested in finding the truth, even though I have already made my choice. I did have the feeling that a discussion concentrated on the likelihood of stroke for those on Warfarin-with no discussion of the fact that you would encounter some risk of stroke even if you choose a tissue valve & no Warfarin-doesn't allow valvers to compare apples to apples.

Finally, to me I would see Al as a person who can help tilt the odds (that otherwise apply to the general population) a little bit in our favor. But what I am looking for here is, I believe, a mathmatical answer. Al is a resource that might save your life, or quality of life. To me this thread is not about my health, but a question of fact that I find very interesting.

Dennis, the point of my post is that the answer to your question is not as simple as the math (statistics). It is in the understanding of how all data is relevant to you as a person and not all "those on Warfarin", which is a very large and diverse group (distribution). It has never been an exclusive "mathematical answer", it is more about interpreting the data and it relevance to you. That is why all mathematicians are not statisticians, but all statisticians are mathematicians and also why all statisticians are not biostatisticians, yet all biostatisticians are statisticians.

That is why you would call Al as an expert witness in a trial about warfarin and not Andy no matter how much he may seem like software giant.

 

[RCB]

You see

You see

Sue

But It's actually only really A Level Maths probability that I'm using (for the americans, it's equivalent to High School age).

Now you have done it- you have just overated the American High School
Education System, Andy!

 

[Karlynn]

a little levity

a little levity

For some reason, every time I read this thread, this old joke comes to mind:

There once was a business owner who was interviewing people for a division manager position. He decided to select the individual that could answer the question "how much is 2+2?"

The engineer pulled out his slide rule and shuffled it back and forth, and finally announced, "It lies between 3.98 and 4.02".
The mathematician said, "In two hours I can demonstrate it equals 4 with the following short proof."
The physicist declared, "It's in the magnitude of 1x101."
The logician paused for a long while and then said, "This problem is solvable."
The social worker said, "I don't know the answer, but I a glad that we discussed this important question.
The attorney stated, "In the case of Svenson vs. the State, 2+2 was declared to be 4."
The trader asked, "Are you buying or selling?"
The accountant looked at the business owner, then got out of his chair, went to see if anyone was listening at the door and pulled the drapes. Then he returned to the business owner, leaned across the desk and said in a low voice, "What would you like it to be?"

 

[Lynlw]

I was wonderring if anyone looked at the link to the rsik scale i sent. I have a question while all the brains are working so hard today

the surgeries are given a number value based on a 0-15 scale, so I was have any thoughts on what thenumber would be for 2 procedures at the same time,for example a CABG has a rating of 7.5 and a Aortic root replacement, Bioprosthetic has a 9.5

i know itwouldn't be as simple as just adding the 2 numbers, but wonder if it
would be close to using the bigger number and adding like 1/2 or 1/4
of the smaller number to find the difficulty and risks,
Lyn

 

[geebee]

Karlynn,
Being an accountant, I totally agree with this. Even tho we can't actually change numbers, we can present them in so many different ways and please just about anyone.

 

[Mary]

I was wonderring if anyone looked at the link to the rsik scale i sent. I have a question while all the brains are working so hard today

the surgeries are given a number value based on a 0-15 scale, so I was have any thoughts on what thenumber would be for 2 procedures at the same time,for example a CABG has a rating of 7.5 and a Aortic root replacement, Bioprosthetic has a 9.5

i know itwouldn't be as simple as just adding the 2 numbers, but wonder if it
would be close to using the bigger number and adding like 1/2 or 1/4
of the smaller number to find the difficulty and risks,
Lyn

I think you would go with the higher number Lyn. That's not mathematical, just logical in my opinion.

 

[Lynlw]

I think you would go with the higher number Lyn. That's not mathematical, just logical in my opinion.

but wouldn't it be more risky doing both so I would think the number would be higher