Thanks for the note. I do have e-mail notification turned on for replies to my comments, at a much lower threshold than +3.:)
My browsing at +3 is not an indication of my "faith in the moderation system" so much as it is an indication of my limited time. I can only afford to read these comments in so much depth. When I get really interested in a thread, I turn down the threshold to take a closer look.
I actually skimmed these comments at a threshold of 0. I did miss your posted corrections from John (sorry!), but that's only because there were >400 comments and I was moving pretty fast.
There is an eternal tradeoff between efficiency and fidelity. In medicine, we refer to the tradeoff between sensitivity (finding something important) and specificity (not finding something unimportant). It's kind of the same here, and I have chosen specificity over sensitivity.
I said: I read slashdot because it's so hard to find anything else intelligent to read.
People replied: And Slashdot intelligent? Haw.... It's worse out there on the Internet than we thought.... Keep searching...
Wow, that was a robust response. No kidding, guys&gals. Without the time or patience to surf, I used to have to tolerate the big-media news feeds, who seem to think that Winona Ryder Busted for Shoplifting is big news.
Admittedly, to refine the/. content, I read at a +3 threshold with -2 for Funny and +1 for Insightful. I only lower the threshold if a thread interests me enough to consider replying (so that I won't be redundant). Try those settings for yourself; suddenly,/. seems pretty damn smart.
P.S. Sorry if this is offtopic; mod me down if you must. I know that those replies were just friendly jesting, but the topic touched a nerve. I hate stupidity.
OK, I'm a physician, not a techie (I read slashdot because it's so hard to find anything else intelligent to read). So perhaps I can provide the "Joe Schmoe" perspective.
I wanted to confirm what people are saying about the average Joe Schmoe not needing computer power. I'm still running an AMD K6 200 MHz processor from 1997. I have a DSL connection and Win 98SE (shudder). I surf the web with IE6, run Yahoo! Messenger, and check email with Eudora, typically all at once, often while also playing bridge on M$N Zone (sorry, M$-haters, it's the best free bridge I've found!).
I couldn't sell this box for anything, let alone $200, yet it does everything I want to do. Sure, it slows down a bit when I use everything at once, but not enough to go out and spend money. If you're a Joe Schmoe like me.
So you're right. The common man doesn't need a monster CPU, etc. That being said, don't forget that usability is key. Most people will be completely unable to surf the web and send email if they have to do too much more than plug it in, turn it on, and follow some very user-friendly instructions. If the Lindows box can't do this, it's not going to do very well.
So, can it? I hear people saying that it doesn't even come with a monitor?
I saw Bill on Letterman last year... he behaved exactly the same way. Letterman tried to get him to talk about his reputation as a bad actor, and basically he joked about it and didn't seem to care.
Personally, I believe that sees this whole thing as the best joke ever. Who cares whether he's a good actor; he has a permanent career as the ex-captain Kirk, a lot of great stories, and a heck of a lot of fun. He can ham it up whenever he wants and get tons of attention from a huge community of fans. I think he knew exactly what he was doing when he recorded "The Transformed Man," when he did the SNL skit, and whatever else.
He was very funny on Letterman. He's got a great sense of humor, and love it or hate it, he has a history of answering probing questions by joking around. This is not special behavior reserved for slashdot. He's just having a great time, and the whining, criticism and complaining from his "fans" are all part of the joke.
I don't know... people weren't too happy about the virtually meaningless math inaccuracy in the original Pentium.
Also, you desperately need a spell-checker.
OK, this is my first post, which I know demotes my relevance in this forum. What's more, I don't know much about the technical details of networks. But I am a physician, I was there, and I think I might be able to provide some perspective. I tried to read the large part of the posts made before writing this, and I hope that my post is not too irritating to the locals. So, here are some things to think about.
As at least one commenter suggested, the reason that the network wasn't better prepared is because the IT budget is woefully unfunded. You might be unaware of the ridiculously poor fiscal situation that most academic medical centers are in. Caught between increasing costs for diagnostic tests and pharmaceuticals, and decreasing reimbursements from the government (N.B. this post is in no way intended to take ANY position on government funding of health care), all hospitals are in an increasingly difficult position. When you throw in that academic centers will not turn away indigent / uninsured patients, and that there are ~40 million uninsured Americans, it is nigh impossible to even break even.
I am in a position to know personally just how underfunded the BIDMC IT department is. Without going into details, let me just say that BIDMC can barely replace 5-year-old desktop platforms. It is casually miraculous that BIDMC has been able to computerize laboratory reporting, medical records, and physician ordering, not to mention supply, billing, bed management, and dozens of other things we don't even think of when we think about running the hospital.
The IT department, led by John Halamka, has been turning straw into gold for years. Every year, they cut the IT budget, and every year, the computer system gets better. (Of course, I say this from an end-user perspective.) Don't let his MD and his emergency med training fool you, as it did one poster, into thinking that John is a duffer. If his description of the reason for the network crash, which I didn't understand, didn't convince you that he knows his stuff, let me add my bias. I know John personally. He works ridiculously long hours to keep his ship running. He is constantly on the lookout for ways to improve patient care with the computer network, and constantly soliciting the advice of parties in all specialties. To my knowledge, he holds the distinction of being the only CIO in history that the Hunter Group (a well-known health care consulting group) has not recommended firing.
So, if the CIO's so good, why did the network crash? I think that an agglomeration of your posters have already figured this one out. BIDMC already knew that the risk of network failure was increasing. For several months before the disaster, the IT department was upgrading network hardware and software as fast as their budget could allow. They were trying to prevent this, and their luck ran out before their ship came in.
Redundancy is expensive. New equipment is expensive. New software is expensive. Personnel are expensive. Look at the financials of BIDMC sometime. The hospital lost $26 million dollars this year, and that was considered a victory, because at the beginning of the year, BIDMC was projected to lose $40 million. The hospital hopes to be breaking even by the end of 2004, without compromising quality of patient care. All the prophylaxis that you've suggested "should have been there" needs to be taken into context with the larger financial picture. Should BIDMC have fired nurses to pay for routers? Cut back lab services to buy newer software?
So it happened. Next question: were patients endangered? Obviously I'm bound by all sorts of privacy concerns, but it's fair to say, probably, a little, but:
First of all, think about how the network impacts patient care. They mostly DO impact data retreival, in the following ways: * Getting diagnostic test results quickly, rather than having to call, or go to the lab * Getting old patient data off of the online medical record, rather than waiting for the patient's old charts (ALL data is duplicated in the paper record) * Entering patient care orders without handwriting them and making sure that a nurse sees them
But, they DON'T impact in the most important ways. The following things work without a network: * The computers that monitor vital signs of sick patients and patients in operations. * The computers inside emergency medical equipment such as defibrillators and respirators
Nobody drops dead instantly because of a network outage.
The network didn't crash all at once: it was up and down intermittently for about 24 hours. After several attempts to get the network running without shutting it down, they finally decided that they needed to shut the whole thing down and start it up again, piece by piece.
I'm not sure who got the idea that a "scramble" to restart paper ordering implies that BIDMC didn't have a plan in place. The hospital has paper backup systems prepared for everything. But you try to orchestrate a quick return to paper on dozens of inpatient wards, with a thousand patients, in short order. Good luck; that's a system involving hundreds of health care providers and separate physical locations. Suggesting that BIDMC ought to be able to throw the railroad switch and just do it easily is rather unrealistic.
That all being said, once BIDMC gave up on keeping the network up while fixing it, we had the whole switched to paper in a matter of hours. This made the system slower and more error-prone, which is why we switched to computers in the first place!
In theory, such a situation could endanger patient care. Slower data retrieval, and the possibility of missing relevant data, could both cause medical errors and patient injury. But, lest you did not realize, medical errors and patient injury are part and parcel of daily healthcare. So many decisions are made on so many patients in a day, that errors happen all the time. Due to the multisystem nature of health care and the multiple levels of safeguards and error-checking, no patient injury happens for one reason alone.
Don't fool yourself into thinking that when the network's up, nothing ever goes wrong, and once the network's down, scores of patients are unjustly injured. Any difference would be incremental. In any theoretical particular case, it would be virtually impossible to prove that the network outage was the crucial component that caused the error.
Is BIDMC at fault? Well, if there were a snowstorm, would BIDMC be at fault if they didn't have enough snowplows on hand? If someone slipped on a banana peel, would BIDMC be at fault for not hiring enough janitors? If there were a fire, would BIDMC be at fault for not having appropriate fire safety?
Was BIDMC at fault? No more than for any other disaster; you can't be 100% prepared for everything, ever. Will they be sued? Probably. Will the suits be just? Probably not. Will they win? I hope not. Hopefully you agree with me.
And, by the way, the computerization of our hospital is multifaceted, and has taken place slowly over 6 years. It's not like we've had our current network in place for 6 years with no changes. Rather, it has grown geometrically with added functionalities as time goes by.
OK, let's end with some responses to comments that I think are informative, but which qualify as "personal agenda," so if you're not interested, you can stop reading here with my compliments.
First issue. > Also, it is very common for doctors to reject > any spending on IT because it will bring their > 8 figure salaries down to 7 figures and that is > totally unacceptable!!!
If you're going to pillory doctors, perhaps you should actually know what you're talking about. The average physician makes $180,000 a year. The most well-paid MD in Rochester, NY (a city of.5M) makes just under seven figures.
Academic centers pay less than average; many grown-up MDs at my hospital don't even make 6 figures. Nobody who's in it for the money works at an academic center like BIDMC. These hospitals lose money, and every expense, yes, including doctor's salaries, suffers from it. Those who stay perceive intangible benefits beyond the monetary compensation.
Believe me, doctors are not cutting the IT budget to line their pockets. I can't speak for the administrators, some of whom are MDs and some of whom aren't, but BIDMC is a not-for-profit institution, and nobody is walking away with fat profits.
Next issue. Whoever suggested that we were unable to play Quake for 4 days: Probably you were just trying to be clever, but it's worth noting that we can't install software on any of the hospital computers.
And finally, whoever made fun of senior managment for "running around like errand boys": Good for them! This was truly a crisis, and all hands pitched in to try to prevent any patients from being hurt. Laugh at them if you like; they could have stayed in their offices, but like the rest of us, they did whatever they could.
Hopefully you have found this informative. A disclaimer should not be necessary, but since it is, let me say that my opinions are in no way intended to reflect those of BIDMC, its administration or employees, the federal government, John Halamka, you, your dog, or anyone else other than me. Have a nice day.
Slashdot Mirror
Thanks for the note. I do have e-mail notification turned on for replies to my comments, at a much lower threshold than +3. :)
My browsing at +3 is not an indication of my "faith in the moderation system" so much as it is an indication of my limited time. I can only afford to read these comments in so much depth. When I get really interested in a thread, I turn down the threshold to take a closer look.
I actually skimmed these comments at a threshold of 0. I did miss your posted corrections from John (sorry!), but that's only because there were >400 comments and I was moving pretty fast.
There is an eternal tradeoff between efficiency and fidelity. In medicine, we refer to the tradeoff between sensitivity (finding something important) and specificity (not finding something unimportant). It's kind of the same here, and I have chosen specificity over sensitivity.
I'll contact John.
Like yours? :D
Sorry, that was mean. I hear what you're saying.
I said: I read slashdot because it's so hard to find anything else intelligent to read.
People replied: And Slashdot intelligent? Haw.... It's worse out there on the Internet than we thought.... Keep searching...
Wow, that was a robust response. No kidding, guys&gals. Without the time or patience to surf, I used to have to tolerate the big-media news feeds, who seem to think that Winona Ryder Busted for Shoplifting is big news.
Admittedly, to refine the /. content, I read at a +3 threshold with -2 for Funny and +1 for Insightful. I only lower the threshold if a thread interests me enough to consider replying (so that I won't be redundant). Try those settings for yourself; suddenly, /. seems pretty damn smart.
P.S. Sorry if this is offtopic; mod me down if you must. I know that those replies were just friendly jesting, but the topic touched a nerve. I hate stupidity.
OK, I'm a physician, not a techie (I read slashdot because it's so hard to find anything else intelligent to read). So perhaps I can provide the "Joe Schmoe" perspective.
I wanted to confirm what people are saying about the average Joe Schmoe not needing computer power. I'm still running an AMD K6 200 MHz processor from 1997. I have a DSL connection and Win 98SE (shudder). I surf the web with IE6, run Yahoo! Messenger, and check email with Eudora, typically all at once, often while also playing bridge on M$N Zone (sorry, M$-haters, it's the best free bridge I've found!).
I couldn't sell this box for anything, let alone $200, yet it does everything I want to do. Sure, it slows down a bit when I use everything at once, but not enough to go out and spend money. If you're a Joe Schmoe like me.
So you're right. The common man doesn't need a monster CPU, etc. That being said, don't forget that usability is key. Most people will be completely unable to surf the web and send email if they have to do too much more than plug it in, turn it on, and follow some very user-friendly instructions. If the Lindows box can't do this, it's not going to do very well.
So, can it? I hear people saying that it doesn't even come with a monitor?
I saw Bill on Letterman last year... he behaved exactly the same way. Letterman tried to get him to talk about his reputation as a bad actor, and basically he joked about it and didn't seem to care. Personally, I believe that sees this whole thing as the best joke ever. Who cares whether he's a good actor; he has a permanent career as the ex-captain Kirk, a lot of great stories, and a heck of a lot of fun. He can ham it up whenever he wants and get tons of attention from a huge community of fans. I think he knew exactly what he was doing when he recorded "The Transformed Man," when he did the SNL skit, and whatever else. He was very funny on Letterman. He's got a great sense of humor, and love it or hate it, he has a history of answering probing questions by joking around. This is not special behavior reserved for slashdot. He's just having a great time, and the whining, criticism and complaining from his "fans" are all part of the joke.
I don't know... people weren't too happy about the virtually meaningless math inaccuracy in the original Pentium. Also, you desperately need a spell-checker.
OK, this is my first post, which I know demotes my relevance in this forum. What's more, I don't know much about the technical details of networks. But I am a physician, I was there, and I think I might be able to provide some perspective. I tried to read the large part of the posts made before writing this, and I hope that my post is not too irritating to the locals. So, here are some things to think about.
.5M) makes just under seven figures.
As at least one commenter suggested, the reason that the network wasn't better prepared is because the IT budget is woefully unfunded. You might be unaware of the ridiculously poor fiscal situation that most academic medical centers are in. Caught between increasing costs for diagnostic tests and pharmaceuticals, and decreasing reimbursements from the government (N.B. this post is in no way intended to take ANY position on government funding of health care), all hospitals are in an increasingly difficult position. When you throw in that academic centers will not turn away indigent / uninsured patients, and that there are ~40 million uninsured Americans, it is nigh impossible to even break even.
I am in a position to know personally just how underfunded the BIDMC IT department is. Without going into details, let me just say that BIDMC can barely replace 5-year-old desktop platforms. It is casually miraculous that BIDMC has been able to computerize laboratory reporting, medical records, and physician ordering, not to mention supply, billing, bed management, and dozens of other things we don't even think of when we think about running the hospital.
The IT department, led by John Halamka, has been turning straw into gold for years. Every year, they cut the IT budget, and every year, the computer system gets better. (Of course, I say this from an end-user perspective.) Don't let his MD and his emergency med training fool you, as it did one poster, into thinking that John is a duffer. If his description of the reason for the network crash, which I didn't understand, didn't convince you that he knows his stuff, let me add my bias. I know John personally. He works ridiculously long hours to keep his ship running. He is constantly on the lookout for ways to improve patient care with the computer network, and constantly soliciting the advice of parties in all specialties. To my knowledge, he holds the distinction of being the only CIO in history that the Hunter Group (a well-known health care consulting group) has not recommended firing.
So, if the CIO's so good, why did the network crash? I think that an agglomeration of your posters have already figured this one out. BIDMC already knew that the risk of network failure was increasing. For several months before the disaster, the IT department was upgrading network hardware and software as fast as their budget could allow. They were trying to prevent this, and their luck ran out before their ship came in.
Redundancy is expensive. New equipment is expensive. New software is expensive. Personnel are expensive. Look at the financials of BIDMC sometime. The hospital lost $26 million dollars this year, and that was considered a victory, because at the beginning of the year, BIDMC was projected to lose $40 million. The hospital hopes to be breaking even by the end of 2004, without compromising quality of patient care. All the prophylaxis that you've suggested "should have been there" needs to be taken into context with the larger financial picture. Should BIDMC have fired nurses to pay for routers? Cut back lab services to buy newer software?
So it happened. Next question: were patients endangered? Obviously I'm bound by all sorts of privacy concerns, but it's fair to say, probably, a little, but:
First of all, think about how the network impacts patient care. They mostly DO impact data retreival, in the following ways:
* Getting diagnostic test results quickly, rather than having to call, or go to the lab
* Getting old patient data off of the online medical record, rather than waiting for the patient's old charts (ALL data is duplicated in the paper record)
* Entering patient care orders without handwriting them and making sure that a nurse sees them
But, they DON'T impact in the most important ways. The following things work without a network:
* The computers that monitor vital signs of sick patients and patients in operations.
* The computers inside emergency medical equipment such as defibrillators and respirators
Nobody drops dead instantly because of a network outage.
The network didn't crash all at once: it was up and down intermittently for about 24 hours. After several attempts to get the network running without shutting it down, they finally decided that they needed to shut the whole thing down and start it up again, piece by piece.
I'm not sure who got the idea that a "scramble" to restart paper ordering implies that BIDMC didn't have a plan in place. The hospital has paper backup systems prepared for everything. But you try to orchestrate a quick return to paper on dozens of inpatient wards, with a thousand patients, in short order. Good luck; that's a system involving hundreds of health care providers and separate physical locations. Suggesting that BIDMC ought to be able to throw the railroad switch and just do it easily is rather unrealistic.
That all being said, once BIDMC gave up on keeping the network up while fixing it, we had the whole switched to paper in a matter of hours. This made the system slower and more error-prone, which is why we switched to computers in the first place!
In theory, such a situation could endanger patient care. Slower data retrieval, and the possibility of missing relevant data, could both cause medical errors and patient injury. But, lest you did not realize, medical errors and patient injury are part and parcel of daily healthcare. So many decisions are made on so many patients in a day, that errors happen all the time. Due to the multisystem nature of health care and the multiple levels of safeguards and error-checking, no patient injury happens for one reason alone.
Don't fool yourself into thinking that when the network's up, nothing ever goes wrong, and once the network's down, scores of patients are unjustly injured. Any difference would be incremental. In any theoretical particular case, it would be virtually impossible to prove that the network outage was the crucial component that caused the error.
Is BIDMC at fault? Well, if there were a snowstorm, would BIDMC be at fault if they didn't have enough snowplows on hand? If someone slipped on a banana peel, would BIDMC be at fault for not hiring enough janitors? If there were a fire, would BIDMC be at fault for not having appropriate fire safety?
Was BIDMC at fault? No more than for any other disaster; you can't be 100% prepared for everything, ever. Will they be sued? Probably. Will the suits be just? Probably not. Will they win? I hope not. Hopefully you agree with me.
And, by the way, the computerization of our hospital is multifaceted, and has taken place slowly over 6 years. It's not like we've had our current network in place for 6 years with no changes. Rather, it has grown geometrically with added functionalities as time goes by.
OK, let's end with some responses to comments that I think are informative, but which qualify as "personal agenda," so if you're not interested, you can stop reading here with my compliments.
First issue.
> Also, it is very common for doctors to reject
> any spending on IT because it will bring their
> 8 figure salaries down to 7 figures and that is
> totally unacceptable!!!
If you're going to pillory doctors, perhaps you should actually know what you're talking about. The average physician makes $180,000 a year. The most well-paid MD in Rochester, NY (a city of
Academic centers pay less than average; many grown-up MDs at my hospital don't even make 6 figures. Nobody who's in it for the money works at an academic center like BIDMC. These hospitals lose money, and every expense, yes, including doctor's salaries, suffers from it. Those who stay perceive intangible benefits beyond the monetary compensation.
Believe me, doctors are not cutting the IT budget to line their pockets. I can't speak for the administrators, some of whom are MDs and some of whom aren't, but BIDMC is a not-for-profit institution, and nobody is walking away with fat profits.
Next issue. Whoever suggested that we were unable to play Quake for 4 days: Probably you were just trying to be clever, but it's worth noting that we can't install software on any of the hospital computers.
And finally, whoever made fun of senior managment for "running around like errand boys": Good for them! This was truly a crisis, and all hands pitched in to try to prevent any patients from being hurt. Laugh at them if you like; they could have stayed in their offices, but like the rest of us, they did whatever they could.
Hopefully you have found this informative. A disclaimer should not be necessary, but since it is, let me say that my opinions are in no way intended to reflect those of BIDMC, its administration or employees, the federal government, John Halamka, you, your dog, or anyone else other than me. Have a nice day.