Why are hospitals spending huge amounts of start-up dollars for EHR systems that have not been compellingly demonstrated to improve safety or quality of care?
I think that I have made my view of electronic health records very clear to anyone who’s cared to listen. I think they are great for monitoring the movement of patients in larger EDs (electronic tracking boards). I have concerns that CPOE (computerized provider order entry) through the use of order sets has the potential to result in over-ordering (and will result in higher bills, unnecessary tests and less thought by providers). And I think making providers do computerized histories and physicals is a huge waste of precious provider time (and is likely to result in records that are one repetitive macro after another, creating a chart that really does not reflect the care provided). But, I know, the train has left the station.
As the result of being involved with our Emergency Medical Abstracts courses, I have had the opportunity to interact with thousands of physicians and discuss EHR innovations. The vast majority indicate that they like CPOE but don’t like charting histories, physicals and – the most important part of the chart – medical decision making, via computers. Why? Because generally, these records take significantly more time to generate than what we did in the past. Plus, the quality of the records is not as good, particularly for those of us who had the option of dictating.
What I see happening is an explosion in the use of scribes and mid-level providers as a “work around” for the degradation in efficiency that has resulted from physicians being required to generate computerized charts.
Fundamentally, I believe physician and nurse work should be facilitated, not encumbered. Doctors and nurses are the engines of the department. They are the ones who move the patients through and generate the revenue. Slowing them down by mandating that they be computer data entry personnel is just economically nutty. And I’m not even talking about physician salaries, though they may be considerable. I’m talking about the lost revenue opportunities that result in going from 2.7 patients an hour to 2.2 patients an hour. With hospital collections of about $500 or more for discharged patients (which is routine for most community hospitals) can we truly afford to slow down our doctors and nurses?
And is the quality and safety of care any better as the result of using EHRs? The research just hasn’t proven it yet. On the contrary, take the study below from the Archives of Internal Medicine. While not specific to emergency medicine, it makes some compelling points. The quality with and without EHRs (whether they had clinical decision support systems or not) was not improved to a degree worth talking about – only 1 out of 20 quality measures improved.
ELECTRONIC HEALTH RECORDS AND CLINICAL DECISION SUPPORT SYSTEMS: IMPACT ON NATIONAL AMBULATORY CARE QUALITY
Romano, M.J., et al, Arch Intern Med 171(10):897, May 23, 2011
BACKGROUND: Outpatient care in the US often fails to comply with existing guidelines and is, thus, considered by some to be of suboptimal quality. Widespread implementation of electronic health record (EHR) systems has been called for by the Institute of Medicine and is a component of the American Recovery and Reinvestment Act of 2009, based on the belief that this will improve quality of care, particularly if clinical decision support (CDS) is included. However, the available evidence does not consistently support a relationship between quality of care and the use of EHR systems with or without CDS.
METHODS: The authors, from Stanford University, examined the association between the use of EHRs and CDS and compliance with 20 quality indicators relating to five broad categories in 243,478 outpatient visits not resulting in hospital admission that were included in the National Ambulatory Medical Care Survey and the National Hospital Ambulatory Medical Care Survey during the period 2005-2007.
RESULTS: EHRs were utilized in 29.6% of the visits, and CDS was utilized in 56.3% of these visits (16.7% of all visits studied). Performance of most of the quality indicators was considered suboptimal (exceeding 90% for only one of the 20). When compared with paper records, the use of EHRs was associated with statistically significant improvement for only one indicator (provision of dietary advice for high-risk adults, 28.2% vs. 19.7%, p=0.01). Similarly, in the EHR visit group, the use of CDS was associated with statistically significant improvement for only one indicator (no routine ECG in low-risk patients, 97.7% vs. 93.0%, p=0.001)..
- no reprints) (21263077 [PMID])
Copyright 2011 by Emergency Medical Abstracts - All Rights Reserved, 7/11 - #9
Despite the research, hospitals are being driven to computerize. CPOE will come first since it is the component least likely to piss off the medical staff. Our hospital has about 700 doctors on staff. We have no residents (who will do anything they are told). Each of our medical staff doctors thinks they are doing the hospital a favor by admitting their patients to our hospital. Each is truly a customer of the hospital. Can you imagine convincing 700 of your “customers” that it is better to move to CPOE rather than to quickly write their orders and hand them to a clerk as currently done. Good luck.
That is why most of these systems seem to start out in the ED. Emergency physicians are not customers of the hospitals at all. They work in the ED at the pleasure of the hospital, thus they are dispensable. Another group will come in and take the contract in a heartbeat – and if the physicians are employees they still need to listen to the mandates of administration. And, not only is the ED probably the easiest place to get CPOE started, given the huge number of orders generated in the ED, use there will go disproportionately far in helping hospitals demonstrate their adoption of CPOE to CMS.
The Obama administration has promised billions to facilitate the adoption of EHRs – although the amount of money offered will ultimately be a pittance compared to the actual costs of these systems – and many have drunk the Kool-Aid. What about the fact that one hospital’s system will not talk to another’s or to the systems in the offices of the 750,000 physicians in America? Details, details. This is supposed to be handled by some big bucks being spent on “black boxes” that will allow disparate systems to share data. Good luck.
But, I acknowledge I am baying at the moon. How can emergency physicians deal with these challenges given that they have virtually no options? What I’ve heard from physicians who have been involved with the installation of these systems is always the same – emergency physicians must actively get involved from the beginning. They need to review and give feedback to how the systems work. We must develop the order sets carefully, be smart about the macros we develop for generating the charts, and basically immerse ourselves in the process. I’ve heard this time and time again.
Fee-for-service emergency physician groups are particularly vulnerable since most have all of their income generated from patient billings. Can they afford to have a decrease in their productivity forced upon them? No way. So more and more of these groups are bringing on scribes and more mid-level providers in an attempt to not have to increase physician staffing.
Unfortunately, the hospitals seem to view this drop in productivity as a problem solely to be borne by their contracting emergency physicians. Somehow the hospitals don’t seem to realize that a drop in productivity will impact them more than it does the emergency physician group – the hospitals likely collects $4 to $5 for every one dollar collected by the contracting emergency physicians. But have the hospitals shared in providing for the extra staffing? I know of no contracting group where the hospital has chipped in to pay for the scribes.
So here are the sad facts. Hospitals are spending huge amounts of start-up dollars for EHR systems that have not been compellingly demonstrated to improve safety or quality of care. They will incur very substantial ongoing costs for maintenance, upgrading, IT staff, etc... in the coming years. Their medical staffs are likely to substantially resist the changes due to a widespread belief that they will significantly decrease productivity. And all this will happen in the face of ever-increasing ED volumes and wait times that remain an industry embarrassment.
W. Richard Bukata, MD is the Editor of Emergency Medical Abstracts