Technophobia
CNet reports: Electronic records not helping outpatient medicine
Electronic records hold the potential to improve medical care by flagging problems such as drugs that shouldn’t be combined, but a study by Stanford and Harvard medical school researchers has concluded that so far they haven’t improved the quality of outpatient health care.
The researchers studied a database of 1.8 billion doctor visits in 2003 and 2004 and examined performance on 17 indicators of quality. The results were mediocre, according to Stanford.
“In essence, we found little difference in the quality of care being provided by physicians with electronic health record systems, compared to those without these systems,” Dr. Randall Stafford, a Stanford associate professor of medicine and senior author of the research, said in a statement. The research is scheduled for publication Monday in the Archives of Internal Medicine.
I’ve done some work for doctors, and an amazing number of them have a “real thing” about computers. You point out how much of the equipment they use and tests they run involve computers, and they still won’t budge.
Typing your notes in a computer is not computerized medical record keeping. Garbage in – garbage out. The computerized drug database at the local KMart has stopped three prescriptions that my Mother was given by doctors. The drugs would either react badly with something else she was taking, or she was allergic to them based on an earlier incident. This is what electronic systems are designed to do, cut down on the problems.
When people talk about how electronic record keeping is going to save money, they need a reality check – they can’t do a thing until they are used as designed and that requires changing the attitudes of health professionals.
12 comments
My primary caregiver and his staff carry laptops everywhere they go in the facility, and it works great. They can transfer their notes and records directly to any of the local hospital, fax prescriptions to any pharmacy, etc.
When I took my mother to the emergency room, the triage nurse also had a laptop that she entered everything into. But when Mom was admitted – I had to go through the entire rigamarole about medications and history all over again. WTF?! The hospital can’t communicate with their own emergency room?
Furthermore, after a week she was transferred upstairs to the Restorative Care unit. It was the same thing….everything had to be reentered.
I’m certainly glad the hospital is taking baby steps, but you’d think they’d link their various units. I have a feeling a software company somewhere is making a pile of money, someone on the hospital administration is pretty stupid, and no doubt the patients are paying for shoddy work.
Brought her home today, incidentally, without having to pay a cent to do so. Thank God for Medicare and Tricare.
The emergency room is a separate “profit center” and may, in fact, have an entirely separate record keeping system from the hospital. I have copies of all that crap to hand to the people at the hospital when Mother has to go in. The hours wasted by everyone concerned on this duplication of effort is absurd and gives lie to the fact that “private enterprise is more efficient.”
The emergency room would have given your Mother a plastic ID bracelet which is barcoded these days. The hospital should have been able to scan the barcode to get all of the necessary information. I suspect that they get paid for admission, like any other procedure.
I’m glad to hear that your Mother is out, so she can get well, something that hospitals are not good at.
“When people talk about how electronic record keeping is going to save money, they need a reality check – they can’t do a thing until they are used as designed and that requires changing the attitudes of health professionals.” [emphasis mine]
One thing I’ve learned designing and using databases (limited though my experience is) is that record-keeping systems should be designed around the way people are going to use them — not the other way around. It really cuts down on the learning curve if the machine is the tool of the user, and not some inexplicable video game.
I’ve been in too many places where a system was plopped down front of an operator who was told “You will use this” only to find that half their data was not accounted for and their previously functional procedures totally ignored. Why would anyone ask the peons how the system would/should be used? The human part of the system breaks down — and that why medical professionals resist.
Maybe it’s the attitude of the software designers that needs changing.
i’m agreeing with anya on this one. what the end user actually needs and how the end user actually thinks aren’t always top priority.
“Maybe it’s the attitude of the software designers that needs changing.” – Anya
Maybe. Or maybe the bad decisions about what is and is not included in those systems are made long before IT professionals get their hands on them.
Most such software is specified by committee, with the result one expects from committee work. In my experience, the software design team has at most one member on that committee, and when I’ve been that member, I’ve found I have very little influence over what the “suits” ultimately choose to do. They have their agendas, and they often don’t much care about your convenience.
I do not mean to excuse bad interface design or implementation, which is an IT professional’s problem… only to say that the higher-ups can be very insistent about functionality that is or is not included, and they often do not consult people in my shoes any more than people in yours, Anya.
Regarding doctors, it’s been probably 15 years since I’ve worked for an MD as such, and I can’t say I miss the work. Many of them are quite certain that an MD confers godlike knowledge on them. And one MD newly employed by an institution I worked for actually went into my desk space, rearranged everything and threw away my coffee cup (the china kind, not the Styrofoam kind). That’s not the way to get good work out of your programmer!
that’s just cold, throwing away someone’s real coffee cup.
The system I’m most familiar with was designed by a doctor who is an emeritus professor at a medical school and taught himself programming because he didn’t like what was available.
I’m going to take a wild guess and assume that most of these systems were designed with billing in mind, not medical records.
Over the last 30 years I have learned that the last two groups of people consulted on any major IT project are those who are tasked with creating it, and those who will be tasked with using it. I have never seen a real, authentic systems design document, normally you are provided with a wish list that doesn’t even specify input and output.
My default procedure is to get copies of a company’s forms, and, wherever possible, design the screens to look like the existing forms.
The problem with doctors doing input is that they generally use a form that has three blocks: patient’s name, date, and text. There may be other boxes on the form, but those are the only ones that get filled in, and the name and date are filled in by someone else. It is disconcerting when halfway through an office visit you determine that the doctor thinks you are someone else.
With the minor alteration of putting medication names in an appropriate box, they could be checked against a database. And if the diagnosis is put in an appropriate box, more information could be provided from other databases. When everything is dumped in a text box, searches are made more difficult and the system is less useful.
“a real, authentic systems design document”
the design document i got for this project was a walk through the office:
–get those [points to wall full of filing cabinets] off of paper and into the computer.
–put those [points to boxes on top of filing cabinets] into the system if they’ll go.
–those [points to opposite wall full of filing cabinets] will be put into xyz system, will your system talk to their system?
–make it all as user-friendly as possible, but you can ignore suggestions from the users in department x. if this makes them all take early retirement, we’ll give you a bonus.
well, ok, that last one is a bit of hyperbole, but not by much.
I always loved the rather common admonition – don’t bother the clerical personnel, their supervisors will train them.
As I have mentioned, I’ve had hardware thrown at me.
i lucked out. sort of. the laptop was already snugly cabled into place before i got into the truck. no flying misguided missiles to worry about, unless you count the vehicle itself.
During the design phase of my sole database project, I was fortunate enough to be both designer and end user. Consequently, I got to do things my way — but mostly because the higher-ups were clueless, and because my immediate boss hired me because a) he knew he was clueless and b) he was desperate for the reports my system was to generate..
Of course, I was not, ultimately, the sole user of my system, so I designed with that in mind. The company for which I created my database continued to use it for several years after I left — until the company finally folded (due to bad business decisions).
My proudest achievement was that I made the data entry forms “ergonomic” — I could enter that damned data all day long without getting a cramp.
I avoided parking lots where I knew there was opposition, Hipparchia. Wlking is good for you.
The annoying part about not being able to talk to users is that even though you can design from forms, it’s nice to watch someone fill out the form so that you can put things in the proper order. Most people jump around forms, and if you can determine a pattern or fill out part of the form for them, people are much happier.
Then you have the mouse versus keyboard considerations [I’m a keyboard partisan], which are never easy.
Designing for yourself is the easiest path, Аня, but the “ergonomics” are more important if people are going to be happy.