Ugh ugh ugh. That’s awful for you and I wish you well in recovering from it.
I am not a medical professional or anything like one; I would just like to give further emphasis to the following things others have already said:
waveman says: base rate is 1⁄200 to 1⁄60, probability given a history of stillbirths is 1⁄40.
I remark (but maybe this is already taken into account somehow) that a couple with a history of stillbirths, just like a couple with a history of live births, is statistically going to be older than average; if, as I guess, older means more stillbirth risk, then this may mean that the difference between the “base” and “history of stillbirths” condition is really less than these figures suggest.
Elo says: the most obvious causes, including ones giving obvious reason to expect higher risk in a future pregnancy, seem improbable given what your doctors have found. It looks like you’re in that “Undetermined” category.
Most likely you will never know what the problem was. It seems like Pr(next child stillborn) is probably no worse than about 2%, and in particular not drastically worse than the probabilities anyone else faces.
I repeat that I am not offering any new analysis (aside from the remark about ages), just picking out what look to me like the key points of other people’s.
If you (very understandably) want a probability estimate from an actual obstetrician, you might try explicitly looking for an obstetrician willing to give probability estimates. That is, contact some local obstetricians and ask not “Can I consult you about this?” but “Are you willing to give me your best estimate of probabilities?”. You may of course find that they all say no, or that they have no actual understanding of probability.
If you (very understandably) want a probability estimate from an actual obstetrician, you might try explicitly looking for an obstetrician willing to give probability estimates. That is, contact some local obstetricians and ask not “Can I consult you about this?” but “Are you willing to give me your best estimate of probabilities?”. You may of course find that they all say no, or that they have no actual understanding of probability.
From my experience sitting on the other side of those conversations, I’m never going to give a number. First, producing the number is very resource intensive, likely more difficult that figuring out the correct things to do for the client.
Similarly, I’m not confident that I know (or remember) all the relevant facts about your situation that would effect my professional opinion. In particular, I’ve always found there were facts I was told and forgot or could have discovered but didn’t. Even though I can perform quality work, failure to keep all those facts in mind during this estimate is not providing an opinion to a reasonable degree of professional certainty.
Third, my clients are human, and like all humans, are bad at probability. If I tell a client they have a 60% chance of winning and we lose, the client will be mad at me. That by itself is reason to give qualitative estimates, not quantitative ones.
Yup, that sounds very plausible. Would your unwillingness to give a number be changed if your client said—as I think the OP here would—something like this? “I understand that any probability you give me may be wrong in ways it’s prohibitively hard to prevent, and I promise that I am not looking for perfection or anything like it. I understand that providing a probability may mean extra work, and I am happy to pay for that extra work. And I assure you that my own understanding of probability is extremely good and I will not do silly things like assuming that if you say something’s unlikely and it happens then you’re incompetent.”
First, I don’t believe the assertion. Second, the kind of work to generate this kind of answer is different from providing service for the client. I enjoy advocating for clients, not meta-level analysis of advocacy. Think medical care vs. MetaMed.
Third, my clients are human, and like all humans, are bad at probability. If I tell a client they have a 60% chance of winning and we lose, the client will be mad at me. That by itself is reason to give qualitative estimates, not quantitative ones.
This is a huge meta-level problem with trying to be rational as a human being, surrounded by other human beings who are not rational.
Organisations with access to quantitative information have every incentive to hide it from you because the average human is a f**king idiot who will make a total pig’s breakfast of the decision theory and probability theory, and then try to use the legal system to punish the giver-of-information.
Ugh ugh ugh. That’s awful for you and I wish you well in recovering from it.
I am not a medical professional or anything like one; I would just like to give further emphasis to the following things others have already said:
waveman says: base rate is 1⁄200 to 1⁄60, probability given a history of stillbirths is 1⁄40.
I remark (but maybe this is already taken into account somehow) that a couple with a history of stillbirths, just like a couple with a history of live births, is statistically going to be older than average; if, as I guess, older means more stillbirth risk, then this may mean that the difference between the “base” and “history of stillbirths” condition is really less than these figures suggest.
Elo says: the most obvious causes, including ones giving obvious reason to expect higher risk in a future pregnancy, seem improbable given what your doctors have found. It looks like you’re in that “Undetermined” category.
Most likely you will never know what the problem was. It seems like Pr(next child stillborn) is probably no worse than about 2%, and in particular not drastically worse than the probabilities anyone else faces.
I repeat that I am not offering any new analysis (aside from the remark about ages), just picking out what look to me like the key points of other people’s.
If you (very understandably) want a probability estimate from an actual obstetrician, you might try explicitly looking for an obstetrician willing to give probability estimates. That is, contact some local obstetricians and ask not “Can I consult you about this?” but “Are you willing to give me your best estimate of probabilities?”. You may of course find that they all say no, or that they have no actual understanding of probability.
From my experience sitting on the other side of those conversations, I’m never going to give a number. First, producing the number is very resource intensive, likely more difficult that figuring out the correct things to do for the client.
Similarly, I’m not confident that I know (or remember) all the relevant facts about your situation that would effect my professional opinion. In particular, I’ve always found there were facts I was told and forgot or could have discovered but didn’t. Even though I can perform quality work, failure to keep all those facts in mind during this estimate is not providing an opinion to a reasonable degree of professional certainty.
Third, my clients are human, and like all humans, are bad at probability. If I tell a client they have a 60% chance of winning and we lose, the client will be mad at me. That by itself is reason to give qualitative estimates, not quantitative ones.
Yup, that sounds very plausible. Would your unwillingness to give a number be changed if your client said—as I think the OP here would—something like this? “I understand that any probability you give me may be wrong in ways it’s prohibitively hard to prevent, and I promise that I am not looking for perfection or anything like it. I understand that providing a probability may mean extra work, and I am happy to pay for that extra work. And I assure you that my own understanding of probability is extremely good and I will not do silly things like assuming that if you say something’s unlikely and it happens then you’re incompetent.”
No, my answer would not change.
First, I don’t believe the assertion. Second, the kind of work to generate this kind of answer is different from providing service for the client. I enjoy advocating for clients, not meta-level analysis of advocacy. Think medical care vs. MetaMed.
Fair enough. (In so far as you’re typical, it sounds like the OP is unlikely to get any further benefit from talking to more medical professionals.)
This is a huge meta-level problem with trying to be rational as a human being, surrounded by other human beings who are not rational.
Organisations with access to quantitative information have every incentive to hide it from you because the average human is a f**king idiot who will make a total pig’s breakfast of the decision theory and probability theory, and then try to use the legal system to punish the giver-of-information.