Is Cerner Millennium killing the kiddies?
For those, especially in Fujitsu's LSP cluster in Southern England, who are now looking forward to receiving Cerner's Millennium EPR as part of the NPfIT nCRS, here's an interesting blog. It's a comment by a US hospital IT director on the recent article in Pediatrics on CPOE (that's computerised physician order entry): Does Cerner Millennium kill children? I don't think so. (HIStalk)
The Pediatrics paper looked at outcomes at the Children's Hospital of Pittsburgh after the implementation of a Cerner Millennium system. They noticed that deaths in the PICU more than doubled after the Cerner installation.
HISTalk takes a long, cool look at the paper blaming Cerner for the increased death rate.
But one or two salient things emerge that are just a little bit worrying - and could have had a lot to do with the increased paediatric death rate.
Cerner Millennium locks the patient record completely while a pharmacist is processing an order. So if there's a critical episode in PICU, while a pharmacy assistant is footling around processing an order, or just goes out to lunch leaving the record open (couldn't happen, could it?), well, that's just tough. No access. Bye bye baby. Raises the interesting question - if Cerner locks up the patient record, just how old a system is it? We've already had the entertainment of IDX's steam-powered green screen COBOL system imposed on the NHS - only to be unceremoniously booted out of the the Southern cluster because it doesn't work. One consequence of CfH choosing IDX is that the whole of the National Programme will now be running on Tandem mainframes, using the Guardian operating system, from the early 1980s. And HP was thinking of phasing these dinosaurs out after they acquired them in Compaq's goody bag (other sweeties included Digital's exciting Alpha range: good for running VMS - what was HP up to? Trying to set up a computer museum?).
Millennium also worsened medication turnaround times, because it imposes a rigid, highly bureaucratic work process. Anyone familiar with American healthcare IT systems, with their cost counting, budget-conscious and dogmatic work process systems, will recognise this scenario. Partly it's because systems like McKesson's, Cerner's and IDX have been built to help hospitals lodge claims with US health insurance companies, while keeping a tight control on costs. These are businesses, after all. And partly it's thanks to that general American assumption that anyone in front of a computer screen is an uneducated moron who can't be trusted to tie their own shoelaces, never mind make a life or death decision.
Millennium proved even worse than this at Pittsburgh. The order entry system actually stopped nurses from administering meds to patients until a pharmacist had reviewed and approved the medication.
It gets worse and worse. The system was obviously designed by the hospital's finance director. Read the piece, it's unbelievable.
Pittsburgh deserves to join the hallowed ranks of London Ambulance, Charing Cross PACS, etc etc, as great healthcare IT disasters. It reads like a blueprint for just how not to implement an IT system in a hospital.
The Pediatrics paper looked at outcomes at the Children's Hospital of Pittsburgh after the implementation of a Cerner Millennium system. They noticed that deaths in the PICU more than doubled after the Cerner installation.
HISTalk takes a long, cool look at the paper blaming Cerner for the increased death rate.
But one or two salient things emerge that are just a little bit worrying - and could have had a lot to do with the increased paediatric death rate.
Cerner Millennium locks the patient record completely while a pharmacist is processing an order. So if there's a critical episode in PICU, while a pharmacy assistant is footling around processing an order, or just goes out to lunch leaving the record open (couldn't happen, could it?), well, that's just tough. No access. Bye bye baby. Raises the interesting question - if Cerner locks up the patient record, just how old a system is it? We've already had the entertainment of IDX's steam-powered green screen COBOL system imposed on the NHS - only to be unceremoniously booted out of the the Southern cluster because it doesn't work. One consequence of CfH choosing IDX is that the whole of the National Programme will now be running on Tandem mainframes, using the Guardian operating system, from the early 1980s. And HP was thinking of phasing these dinosaurs out after they acquired them in Compaq's goody bag (other sweeties included Digital's exciting Alpha range: good for running VMS - what was HP up to? Trying to set up a computer museum?).
Millennium also worsened medication turnaround times, because it imposes a rigid, highly bureaucratic work process. Anyone familiar with American healthcare IT systems, with their cost counting, budget-conscious and dogmatic work process systems, will recognise this scenario. Partly it's because systems like McKesson's, Cerner's and IDX have been built to help hospitals lodge claims with US health insurance companies, while keeping a tight control on costs. These are businesses, after all. And partly it's thanks to that general American assumption that anyone in front of a computer screen is an uneducated moron who can't be trusted to tie their own shoelaces, never mind make a life or death decision.
Millennium proved even worse than this at Pittsburgh. The order entry system actually stopped nurses from administering meds to patients until a pharmacist had reviewed and approved the medication.
It gets worse and worse. The system was obviously designed by the hospital's finance director. Read the piece, it's unbelievable.
Pittsburgh deserves to join the hallowed ranks of London Ambulance, Charing Cross PACS, etc etc, as great healthcare IT disasters. It reads like a blueprint for just how not to implement an IT system in a hospital.
posted by William Payne at
3:51 pm
0 Comments:
Post a Comment
Subscribe to Post Comments [Atom]
<< Home