I had an interesting chat today with one of the senior medical administrators in my hospital.
It was the second conversation I’ve had this week which bemoaned the seemingly random variability in practice between different doctors. Both conversations were in the context of implementing information systems.
I think that we can all agree that, where good evidence exists, any variation from standard practice is not a good thing. The problem is that practice varies most where the evidence is thin, or alternatively where there is plenty of evidence that a number of alternatives are equal. And granted there are obvious advantages of for doing things a standard way, both in terms of safety and organisational efficiency.
The question is: “is it appropriate to use information system implementation as an opportunity to standardise practice?”
My opinion? (I’m assuming you’ve read this far because you’re interested in my opinion) The answer is “no.” Or at least “not straight away.”
This is for a couple of reasons.
Firstly, implementing information systems is hard enough without forcing doctors to change their opinions on what constitutes good clinical care at the same time. Let us focus on meeting their information needs first, then we can look at using the system to implement some standardisation.
Secondly, there are good reasons for much clinical variation. Doctors are not all the same, even in the same sub-specialty. We all had subtly different training, even if we are in the same age group, which we are not. Some of us have just finished our training and some are close to retirement. So even if our patients were the same we are not. We are good at different things. Five years, yet alone twenty-five years, is a very long time in clinical practice.
There is a learning curve with any procedure. The first ten or twenty times you do most things, you’re not very good at them. By the time you have done something a thousand times you are pretty slick.
So currently we all do things differently, for better or worse.
But when you become a patient and get a procedure done by a doctor do you want them to chose the procedure they have done a thousand times? Or the new fangled procedure they have done ten times? Don’t know? Neither do I. Generally I’d be letting my doctor choose to do it the way they felt would be giving me the best outcome.
Does this mean that doctors are prima donna’s? Dunno. Maybe. All I’m saying is that standardising practice sounds like a good idea, and often it is. But it’s complicated.
So when we’re implementing computer systems, can we just stick to getting the information in the right place at the right time?
Maybe having good information systems will increase organisational awareness of unnecessary and unjustifiable variation in clinical practice. Maybe it will allow some conversations to occur and allow monitoring of the outcomes of resulting changed practice. But let us not make standardisation of practice a precondition of information system implementation.
My firm belief is that if we accept and allow variation in clinical practice while we implement our systems paradoxically we may get more more standardisation in the future. We'll definitely be building more flexible and robust systems.
It was the second conversation I’ve had this week which bemoaned the seemingly random variability in practice between different doctors. Both conversations were in the context of implementing information systems.
I think that we can all agree that, where good evidence exists, any variation from standard practice is not a good thing. The problem is that practice varies most where the evidence is thin, or alternatively where there is plenty of evidence that a number of alternatives are equal. And granted there are obvious advantages of for doing things a standard way, both in terms of safety and organisational efficiency.
The question is: “is it appropriate to use information system implementation as an opportunity to standardise practice?”
My opinion? (I’m assuming you’ve read this far because you’re interested in my opinion) The answer is “no.” Or at least “not straight away.”
This is for a couple of reasons.
Firstly, implementing information systems is hard enough without forcing doctors to change their opinions on what constitutes good clinical care at the same time. Let us focus on meeting their information needs first, then we can look at using the system to implement some standardisation.
Secondly, there are good reasons for much clinical variation. Doctors are not all the same, even in the same sub-specialty. We all had subtly different training, even if we are in the same age group, which we are not. Some of us have just finished our training and some are close to retirement. So even if our patients were the same we are not. We are good at different things. Five years, yet alone twenty-five years, is a very long time in clinical practice.
There is a learning curve with any procedure. The first ten or twenty times you do most things, you’re not very good at them. By the time you have done something a thousand times you are pretty slick.
So currently we all do things differently, for better or worse.
But when you become a patient and get a procedure done by a doctor do you want them to chose the procedure they have done a thousand times? Or the new fangled procedure they have done ten times? Don’t know? Neither do I. Generally I’d be letting my doctor choose to do it the way they felt would be giving me the best outcome.
Does this mean that doctors are prima donna’s? Dunno. Maybe. All I’m saying is that standardising practice sounds like a good idea, and often it is. But it’s complicated.
So when we’re implementing computer systems, can we just stick to getting the information in the right place at the right time?
Maybe having good information systems will increase organisational awareness of unnecessary and unjustifiable variation in clinical practice. Maybe it will allow some conversations to occur and allow monitoring of the outcomes of resulting changed practice. But let us not make standardisation of practice a precondition of information system implementation.
My firm belief is that if we accept and allow variation in clinical practice while we implement our systems paradoxically we may get more more standardisation in the future. We'll definitely be building more flexible and robust systems.
No comments:
Post a Comment