The insurance industry is 'under the gun' right now, and this allows the provider sector to avoid the fundamental issue…
Readers Write 7/1/09
Submit your article of up to 500 words in length, subject to editing for clarity and brevity (please note: I run only original articles that have not appeared on any Web site or in any publication). I’ll use a phony name for you unless you tell me otherwise. Thanks for sharing!
Hats Off to AMDIS
By Ann Farrell
Congratulations to AMDIS for saying what many of us believe and promote, but had feared was falling on deaf ears or been drowned out by politics and ego. It’s not surprising that the “Boston Docs” known MD-centric view of the world (healthcare and IT) produced a largely MD-centric, “CPOE first” meaningful use strategy. Hopefully this attitude was rejected when Version One of MU was sent back to the drawing board the day after the first draft was issued.
Chasing ARRA money already put some hospitals on a dangerous path to drop everything in hurry up mode to “install” CPOE without examining physician workflow, decision making, cultural and change management needs, and foundational applications. Some EMR companies and their advocates encouraged this — some unwittingly, others with an eye on increased or accelerated quarterly revenue recognition, the metric vendors are held to (incented by), particularly public companies.
For CPOE to be more than an automated requisition generator, MDs need to get tangible value, including the ability to make better informed decisions based on more timely data (not meaning the computer is making decisions for them). Since ancillary systems were ground zero for hospital clinical automation, lab and X-ray results are almost always online before or with CPOE.
What may not be present is assessment data entered by nurses, ideally at the point of care in near real time, e.g. allergies, height/weight, vital signs, I & O, nurse-collected lab values, and an accurate medication record. That is critical data for clinical decision support (CDS) for MDs in ordering. Not having these data available wastes MD time and steps and results in suboptimal or even unsafe ordering decisions. If data is not easily retrievable (preferably “pushed” to MDs in the ordering process at the right time), physicians are forced to look for paper charts, call for information, chase nurses down, or make ordering decisions without important or current information.
In addition to providing a clear path to CPOE, automating the eMAR/BCMA has greater potential impact on med error reduction than CPOE. Not killing or harming patients would seem a primary goal to improve quality of care. MDs and RNs make approximately same number of errors, but pharmacists or RNS catch 50% of MD errors downstream whereas 98% of RN errors reach the patient. And, nurses work for hospitals and are more easily corralled (in theory), thus making clinical and business sense to start with foundation pieces first.
Hopefully Drs. Glaser and Halamka (and Blumenthal) are listening. Some have recommending staging implementations as if it’s a pecking order — doctors first! To be effective, CPOE needs to be part of a bigger strategy –patient-centric, outcomes (not IT) focused, with staged functionality and a 21st century interdisciplinary care team approach that respects all caregivers’ roles and contributions.
For the good of all, we want CPOE to be embraced by MDs, but also for MDs and US healthcare reform to be more inclusive and patient-centric. I speak as clinical consultant, former EMR vendor exec, and RN who worked with first commercial EMR in a hospital with near 100% CPOE in early 1970s. CPOE is hardly a new phenomenon, yet some MDs and vendors act as if it started with them. We’ve known for decades how CPOE can be implemented successfully. Now’s the time to really get this right.
Ann Farrell is a principal at Farrell Associates of San Francisco, CA.
An Alternative Desktop Standard
By Mark Moffitt, MBA, BSEE
We have deployed a unique desktop configuration at our healthcare provider organization. The configuration is a Mac-mini running Windows 7 release candidate (RC) with a 17” wide-screen monitor.
The advantages of this configuration over a conventional PC are:
- Smaller footprint
- Less expensive
- Higher quality hardware
- Better cloning capabilities, i.e. ability to clone the windows partition using the OS X operating system
- Run Leopard and/or Windows 7
We skipped Vista as a desktop standard. We found W7 RC to be very stable. So, rather than install XP on newly deployed machines, we opted to deploy W7 RC. Once W7 is released, we will install it over W7 RC.
The cost of the Mac-Mini, display, and keyboard and mouse was less than the conventional PC configuration we were considering. Your mileage may vary.
Power users in IS run both Leopard and W7 RC. They are both really good operating systems. Leopard is much better working with multimedia, while W7’s sweet spot is “corporate computing.” I run both on my MacBook Pro.
Mark Moffitt is director of information systems at Good Shepherd Medical Center of Longview, TX.
Physicians Using PCs
By Ben
I think you need additional inspiration!
Seriously, I think you’re confusing the work flow of an office based physician with the work flow of an inpatient physician (i.e., hospitalist or critical care specialists as examples). We (hospital-based physicians) spend much more time sitting down, sifting through and analyzing data (whether in electronic or paper formats) than we do with hands-on patient care. That’s NOT because the data analysis pulls us away from the bedside, but rather it is the bulk of the work: analysis, married with the patient visit and examination, tempered by experience and judgment, aided by decision support as available, leads to action.
Why do computers in patient rooms fail to attract physicians? We want to work at a desk, adjacent to our colleagues, where we can sit and work without being distracted by what’s going on in the patient’s room. Doesn’t matter whether we’re working from a computer record or a paper record.
And BTW: the “pecking away at a keyboard” has made me a vastly more efficient and informed physician than when I worked off of paper. Lawyers have the option of turning the work over to “associates”. In the absence of medical students, the patient gets the full attention of the “partner”! Score one for physicians.
I disagree with Ann Farrell’s article. She said it best herself, she was at a hospital with near 100% CPOE adoption ALMOST 30 YEARS AGO. As a nurse and a former implementer, I have to say “come on, this isn’t rocket science”, there isn’t a valid reason why CPOE adoption should be as low as it is today (<20%). There also isn’t any reason to believe an organization would not be able to achieve meaningful CPOE use in time for stimulus $$$.
Even if it’s only “an automated requisition generator”, at least they’ll get interaction checking (and hopefully be using allergy checking). At the very least, transcription errors will be minimized if not eliminated outright. Check out how many medication errors every year are attributed to transcription errors – it’s depressingly high. Ask yourself why JCAHO put such an emphasis on minimizing verbal orders and mandating read back.
Nursing resistance has never been the problem with EMR adoption. The single best indicator for EMR success is how the hospital responds to a physician refusing to use the system (Nurses are easy, adoption=continued employment). If this means a few hospitals tell their docs “No, you WILL do CPOE or privileges will be revoked”, I’ll consider it a success for the industry as a whole…
Recently there has been a trend from several EPIC customers to drink the marketing Kool-Aid and decide that a sound strategy is to unplug working, even progressive HIM (Medical Record) applications to rush into EPICs HIM and EDM (provided through partners) applications, regardless of whether they are actually superior to the practices and processes in place.
My read (and I am very experienced in this area) is that EPIC has less than great HIM applications and strategies, have only recently engaged an internal HIM resource. If not carefully strategized migrations like this could negatively impact HIM operations and even increase staffing.
HIM is a very complex, high volume envroinment that can easily be guided off the track into unproductive and more expensive practices by precipitous movements to a vendor that in reality does not have well established applciations in this area.
I’d love to hear from any sites or readers that can give guidance on their thoughts on this matter.
Thank you, Ben, for the simple and clear description of PC use by hospital-physician vs office-physician. It makes perfect sense to me.
If what you say is commonly true (and it sounds like it is), then systems at the bedside should focus almost entirely on making it easy to collect data in as much detail as the doc cares to – leaving deskside applications to focus on easy data access, presentation options, and robust analysis tools. Yes?
I’m new-ish to hospital IT… do you feel that this fundamental workflow difference is something we ITers have missed all along?
Ms. Farrell ends her article by stating “We’ve known for decades how CPOE can be implemented successfully. Now’s the time to really get this right.” Well if we have known why hasn’t it been done? Exactly how many more years do we need? I’m sure that resources have been a factor in the delay; however, we can’t discount the strong physician lobby at some hospitals that has made it all but impossible to implement.
I say this having worked in a health system that has full CPOE for over 5 years as well as and eMAR/BCMA system in place for over 3 years. Having been the lead IT analyst on the medication portion of CPOE and the eMAR/BCMA implementation I can tell you first hand the resistance we faced. One major key to success was having strong management support and a physician champion we could let loose on any physician who was “resistant” to change. Most of those physicians can’t remember life without CPOE now.
Ms. Farrell is right to say we need to “examining physician workflow, decision making, cultural and change management needs, and foundational applications.” However, there are solutions to all these issues. She continues to say with key data, “physicians are forced to look for paper charts, call for information, chase nurses down, or make ordering decisions without important or current information.” Without a proper information system that is precisely what is physician is doing now. There will always be more data requested in a CPOE system, there will always be requests to change the way that data displays – but to hold off because of those reasons is self defeating. I know from experience: someone will always have a reason it cannot work.
Experience tells me that the journey to CPOE is a series of small steps and a few big ones. But every journey has to start with a single step. I’m not fond of the tack of the federal government forcing it to happen, but it does need to happen.
If properly implemented a CPOE system can not only bring the benefits of medication safety, but also safety in other clinical areas. It can also save money by avoiding duplicate testing. For example, not reordering a CT with contrast that a patient had a few days ago can save the kidneys and the cost of the test. It could also prevent lawsuits.
A dentist once told a friend of mine “only floss the teeth you want to keep.” It is similar, albeit an oversimplification, for CPOE: only use it on the patients you don’t want to harm.
Ann and Ben are both right. I have seen all things CPOE from a glorified typewriter that prints orders on a dot matrix printer in the receiving department (as far back as the 80s yet still I have seen this recently) to the most ungodly complex unlearnable systems that would require 3 years minimum to build, configure, and train users decently, not just minimally usable which is the standard. Not to mentions dozens of IT analysts full time configuring, troubleshooting, and teaching. I won’t name any company. Insert usual big names. There needs to be some sort of standard for institutional software. Something scalable from simple to relatively complex, yet not so labor intensive that it will take manyfold much more time to create and change patient orders and maintain the system. I don’t think such a thing exists yet. Somebody that can create such a thing stands to make some good money.
Happy to debate CPOE phasing and mandate strategy but want my positions to be clear.
CPOE fails for many financial, IT and deep-seated cultural reasons – 4 primary:
1) Lack of “financial incentive” – Most MDs paid on volume, claim CPOE slows them down / lowers income – need more volume to offset lower reimbursement rates- worst fear (private focus groups) is “big brother” and loss of autonomy.
(never reconciling not supporting patient safety benefits, e.g. drug interaction, or other advantages for MDs, and hospitals, teams and patients)
2) Poorly -designed and -integrated EMRs – some “whining” a smokescreen, some legit. Original EMR user-friendly, sub second response time and integrated. Today, more GUI and “bells and whistles” but hard to describe some EMRs convoluted logic and “design”.
Unless hospitals demand more usable systems from vendors, it won’t happen. Design takes time and costs money, but can pay off. IT (vendors and hospitals) focus on “build” and are measured/incented by schedule, budget, not adoption or benefits realization.
3) Poor implementation strategies – lack of prerequisites
4) Lack of CEO or CMO “success strategy” – fear-based (some realistic) decisions about MD revolts or exodous. More hospitals now see higher risk in not mandating use.
I’ve been alone in many forums for years in recommending CPOE be in MD bylaws (“mandated”). We need carrots, but in the end, also a big stick. Voluntary “for the good of all” and small financial incentives haven’t worked.
Agree drug interaction key CDS but med orders complex and require allergy data (often collected by nurses first) to fire alerts and MDs need other clinical data, e.g. most recent vital signs, to know what to order.
I’m PRO-CPOE-, in timely, logical, incremenal steps with usable systems. Considering workflow doesn’t imply there aren’t solutions, I offered one that suport business and clinical goals. I’m not advocating either “whatever the docs want” or “slam in the software” approach.
MD resistance will die out as older MDs retire, with younger ones expecting automation. In the interim, IMHO we need to move forward, not let old guard hold us all back till they retire – rather lets bring them along with us, if possible.
As a longtime CPOE/EMR evangelist, I’m especially dismayed by lack of MD support. After courting MDs, for die hards suggest “we’ve done all we can and are sorry to lose you but to work here you have to use our systems.” More than ever it in best interest of MDs, hospital, care team and patients.
Thanks Tim for forum for lively discussion!
Ann