I will let Horst tell the story for himself – which appeared a little while ago today on the GPCG_TALK mailing list.
Horst has given me specific permission to republish this saying:
On Tue, Mar 9, 2010 at 1:26 PM, David More
“Hi Horst,
Happy to post your account on the blog. That might get some attention. There are a lot of influential readers I understand from the feedback I get.”
Thanks.
Anything that might contribute to positive change is welcome – the status quo is not acceptable
Horst
----- End Message
For those who don’t know Horst is a resident polymath on the General Practice Computing Group e-mail list being a GP in Dorrigo as well as being a very experienced system designer and implementer among other things.
Take it away Horst:
[GPCG_TALK] Cerner software - the saga continues –
From: Horst Herb
To: General Practice Computing Group Talk
I was offered AUD 400 to sit for an hour in a teleconference and see the improvements in Cerner's Firstnet software for rural hospitals, and -silly me- I accepted.
So I sat and watched.
And I left very, very angry.
In essence, they tried to convince somebody who has been using a Microwave oven for the past decade that the best way forward is to buy their (most expensive) two flint stones, because then you can bang them together, light some fire, and eventually cook on it. Actually, the analogy is wrong, because flint stones can be very useful in the right circumstances.
In detail:
at the time of entering a progress note, the doctor cannot
- browse, add or modify allergies
- browse, add or modify past history
- browse past consultations
- browse or request test results
- browse or manage medications
(in fact, the concept of medication is still alien to that software it seems) etc without leaving the data entry screen (and potentially losing all data entered so far if he leaves the screen the wrong way), navigating a complex menu that makes Homer's odyssey look like an easy travel guide for nursing home patients, and then eventually find the way back to the data entry screen.
The number of mouse clicks required to perform even the simplest tasks remains legion, and the screens remain cluttered with (to the clinician) useless administrative details while utterly neglecting the needs of the clinician.
The presentation of data (e.g. vital signs) remains in a form where it is very difficult to find what is actually relevant (even to those who are familiar with the format of the information presented in) .
This software is not just extremely poorly designed (in fact I cannot recall any clinical software I have seen in 20 years in multiple countries that was anywhere near as bad) and time wasting - it is dangerous to patients because it makes information access needlessly difficult and confusing.
It saddened me to witness that they failed to improve anything in a meaningful and significant way after so much time spent after the initial assessment, and it maddens me that they refuse to take advice on board from those who KNOW what the requirements are because they actually do the work AND have experience with other (much better) implementations.
We VMOs at Dorrigo Hospital refuse to work with that software and we fail to see any realistic way forward given the time they wasted between the last presentation and today without producing any MEANINGFUL changes. I learned that the doctor from Bellingen Hospital who was meant to watch left disgusted even earlier than I did, so I suppose the verdict there remains the same too.
Horst.
----- End Message.
I pass this on without comment other than to say it seems this software is just not suited for small hospitals in small towns. There really needs to be an alternative to the one size fits all approach!
David.
Some free advice for NSW Health - do a Google search on "Horst Herb" + "Open Source" and see how many hits you get. Somehow I doubt you will ever win this guy over...
ReplyDeleteSounds like advice from someone with a product to sell.
ReplyDeleteIt also sounds to me like the tool provided does not suit the workflow of the user. I am also hearing that the tool is inflexible and does not appear to take advantage of advances in user interface design made over the last 20 years.
This may be barely acceptable as a cost saving measure if the workflow implemented suits the job at hand, but it sounds to me like it does not. From the descriptions and previous writings from Jon Patrick on this topic, I would be surprised if the workflow supported was suitable for a majority of potential users.
Why was this not identified as a problem prior to the decision to purchase and deploy?
Sounds like someone who may know what they are talking about - i.e a clinical user who has had a go at producing his own clinical application.
ReplyDeleteJust to clarify - while I have written clinical software in the past in various countries and indeed participated in open source projects too, I have no intention to be involved in or benefit from any type of software that could be seen as a competitor for the Cerner software I criticise.
ReplyDeleteMy criticism stems solely from being a end user of it (as VMO at Dorrigo hospital), realizing how such appalling software not only destroys my productivity but also endangers my patients. That, and my frustration as a major tax payer over the fact that money is wasted on such product that does not do what it is supposed to do.
I am not a luddite - I desperately want a paperless record system at my hospital. But I am not desperate enough to forfeit productivity and patient safety in the process.
Hmmm, not to be too controversial but in my 20+ years of Health IT experience, the doctor who is also an "IT Expert" is generally the LAST person you want anywhere near your project if it to be accepted by the majority of clinical users.
ReplyDeleteRegardless of whether the Cerner product is good or bad, has anyone pointed out an alternative product available for installation now? Comparing Cerner's product to a vapourware solution - open source or otherwise - doesn't seem entirely fair.
ReplyDeleteTo Wednesday, March 10, 2010 10:56:00 PM
ReplyDeleteSo the author of the dominant GP software in Australia should not be involved in Health IT?
To Wednesday, March 10, 2010 11:13:00 PM
There are lots of competing products and no reason why in smaller hospitals one of these should not be used.
David.
Just one more point. Many of these products are highly configurable, and so it may not be the actual software so much, as the way it was configured and implemented. Ideally, clinicians need to be involved up front and their workflows mapped (which may be different in smaller health sites). Then systems can be configured to suit. I don't work for Cerner, but am sympathetic - when things go wrong it is too easy to blame the vendor's software. Also, we need open minds when it comes to the possibility of improving the way we do things. But one thing is for sure - get the clinicians involved early! Even if you fluke it and get it right, they may not like it because you didn't ask them!
ReplyDeleteI think those respondents who dismiss the concerns of practitioners such as Horst Herb should read this paper from Ash et al, 2004, which very clearly enunciates why products such as Cerner, which follow a Talyorist philosophy, will never work in the multi threaded, multi layered emergent clinical decision making processes of the typical busy clinical inpatient unit, or for that matter a comprehensive rural health service.
ReplyDeletehttp://www.ncbi.nlm.nih.gov/pmc/articles/PMC353015/?tool=pubmed
"There are lots of competing products and no reason why in smaller hospitals one of these should not be used.
ReplyDeleteDavid."
Like what?
Obvious examples are the later versions of the EDIS Cerner replaced and products from people like iSoft to name two.
ReplyDeleteDavid.
And Ascribe's Symphony and Intersystem's Trakcare
ReplyDelete