Electronic Health Records (EHRs), Computerized Physician Order Entry (CPOE), and the Governmental Meaningful Use Criteria (MU Stage 1, 2, 3) for EHR Implementation

Before beginning on what might be perceived as yet another Haldol-deficient Mallon rant, it is important to note a few things. In full disclosure, I am an emergency physician of that certain age where I am not considered by anyone to be computer savvy. I don’t have an iPhone. I don’t text, I don’t tweet, and I have never put a hashtag on anything. I barely type, and residents and nurses frequently make fun of me as I hunt and (violently) peck at a keyboard while working my way through the typical documentation burdens of a shift. Logging in alone might take me two or three tries. Passwords roll around in my short-term memory causing dismay and further frustration, and every EHR featured here asks me to create new ones with better security features (which also means they are impossible to remember) on a regular basis, intermittently locking me out with a new password request, as if lowly frontline clinicians were the cause of the big healthcare information security breaches (they aren’t).

Furthermore, I suffer from an EHR-related type of dysphoria not yet described in DSM-IV. This disorder has resulted from the fact that I have been clinically active during one too many “go-live” disasters. I have now been dragged kicking and screaming through Cerner, Wellsoft, Epic, and T-System all in the last few years of practice. None of the facilities that I have been involved with have Dragon (or any other dictation software), and none use scribes, contributing to my “me against the computer system” despair. Such is the substrate for the following EBM diatribe.

The AMA agrees with many of my views, and in a May 29, 2015 response to CMS, noted “As we have said many times, the Meaningful Use program is not meaningful in its current form and is not helping physicians use EHRs in a way that facilitates the best care coordination, increases efficiency or improves the quality of care for patients” (AMA president-elect Steven Stack, MD who just happens to be an Emergency Physician). In that report, the AMA was specifically requesting critical changes and more assessment time BEFORE moving on to MU Stage 3.

It might not be surprising that my favorite recent satire was the EXTORMITY Electronic Medical Record(EMR) parody. EXTORMITY EMR was a completely bogus EMR company that had a web site complete with hilarious testimonials from fake users. The fake company name “EXTORMITY” was a word salad formed by the confluence of the Latin roots for EXTORTION and CONFORMITY which just begins to explain how many physicians currently feel! The company tagline was “Expensive, Exasperating and Exhausting”! The tagline could have added a phrase about “meaningless” implementation of the “meaningful use” criteria. When the creators of this parody finally publicly outed themselves they said:

“As the healthcare community somewhat reluctantly transitions to electronic health records, many physicians, hospitals and healthcare IT experts have been frustrated with significant expenses, inflexible applications, a lack of implementation success and less than stellar ongoing support from EHR vendors. In addition, the whole industry is trying to wrap their heads around meaningful use guidelines and determine which EHR solutions will help them qualify for stimulus funds and improve the quality of patient care,” said Jeff Donnell, President of NoMoreClipboard.. “Several late-night brainstorming sessions with our EHR partner, MIE, led to the creation of an expensive, exasperating and exhaustive EHR solution (Extormity) and a “fake” certification organization (SEEDIE) — all in a good natured attempt to tap into this skepticism and shed satirical light on the industry’s situation.”

SEEDIE, the fake Meaningful Use certifier is beautifully ironic too, and stands for the “Society for Exorbitantly Expensive and Difficult to Implement EHRs”. I love it.
First, a little orientation to the real eponyms and vocabulary surrounding the topic:
CMS: Center for Medicare and Medicaid Services (what happened to the second “M”?) 
The American Recovery and Reinvestment Act: Legislation including the EHR incentive programs which resulted in unprecedented growth in EHRs. 
MU1, MU2, and MU3: The stages of implementation of EHRs and the Meaningful Use criteria to get stimulus funding. 
ONC: Office for the National Coordination of Health IT, which recently noted that a lack of interoperability, technical limitations, and excessive vendor fees are hindering physicians’ ability to meet MU2 criteria. ONC exists to help make the EHR evolution and implementation flow smoothly. MIPS: Merit-Based Incentive Payment System created by the above legislation to implement Medicare reform requiring physician quality reporting programs including MU 1–3 and their respective criteria to receive stimulus payments. 
CPOE: Computerized provider order entry, part of MU1, and supposedly, a fantastic patient safety innovation that would prevent medication errors. 
BoB (Best of Breed) EHRs: A term describing specialty-specific EHRs which are designed to meet the work flow and charting needs of a specific clinical environment (EM or Ophthalmology or Dermatology, for example). Specialists generally prefer them to “enterprise-wide” EHRs.

The initiating event for this essay was this month’s EMA paper on medication errors in the era of CPOE. While there are limitations to this paper and its methodology, it represents the most detailed look at medication errors from within a CPOE setting, and found that the computer system was blamed for 6 percent of medication errors. The paper is: “Computerised physician order entry-related medication errors: analysis of reported errors and vulnerability testing of current systems,” by G.D. Schiff et al in BMJ Qual Saf 24(4):264, April 2015.

The authors reviewed medication errors within the US Pharmacopeia MEDMARX system in detail and developed a taxonomy based on what went wrong and why. Of the 1.04 million medication errors between 2003 and 2010, 63,040 were recorded as CPOE-related! After analysis, the authors then checked to see if 375 “test case” errors could be entered using 13 CPOE systems at 16 sites. 79.5% of the erroneous errors were able to be entered, including 28.0% being “easily” placed, and 28.3% being placed with minor workarounds and no warnings.

It seems that CPOE systems fix some errors while introducing new ones, and along the way frustrating clinicians. The “test cases” prove that these CPOE-related medication error problems are endemic, repeatable and easily made across several CPOE systems in multiple institutions. Apparently the promise of safety by CPOE systems is premature indeed.

What price have clinicians paid for this CPOE wasteland? Alarm fatigue, difficulties finding the drug or solution you want (D5 1/2NS with 40 of KCl is about 45th on the Cerner IV fluids menu), and patient weight demands all gang up to make clinician life miserable and disrupt EM work flow. Verbal orders which in reality are frequently a needed feature of EM have become very difficult. Doing something atypical like nebulizing fentanyl or tranexamic acid is nearly impossible. The ordering process becomes so difficult you might decide to skip a treatment even though initially you felt it was the best treatment for the patient. This is the CONFORMITY part of EXTORMITY coming home to roost! Whatever you do, forget being innovative with your orders, and never ever try something new (evidence-based or not), because the CPOE system will torture that instinct out of you. CPOE will never keep up with the rapid evolution of treatments and potential “off-label” usages. The literature evolves much too fast for CPOE programmers to keep up.

In another paper titled “Unintended effects of a CPOE nearly hard-stop alert to prevent a drug interaction” by B.L. Strom et al. in the Arch Intern Med 170(17):1578, September 2010, the study was terminated early as a result of four CPOE-related unintended consequences that were identified where a hard-stop alert resulted in a delay in appropriate treatment with TMP-SMX or inadvertent discontinuation of warfarin. In this study, the hard-stop alerts prompted the “desired result” in only 57.2% of cases which, while better than the controls, can hardly be seen as a home run.

How about resource stewardship? Does CPOE deliver on the promise of decreased resource utilization and operational cost savings? You already know the answer — NO!

Another related paper on this topic is: “Is CPOE use associated with a decrease in hospital resource utilization in hospitals that care for children?” by R.J.Teufel et al in J Med Syst 36(4):2411, August 2012. Here the authors reviewed 4,896,004 discharges of children from 3,438 hospitals in 2003 and they failed to find a significant difference between CPOE hospitals (which actually spent $70 more per case) and non-CPOE hospitals. No LOS (length of stay) differences were found either. Mind you, this study was done in pediatric cases and pediatric facilities where medication errors and weight-related problems are described as an important problem and confirmatory literature exists (see “Prescription errors before andafter introduction of electronic medication alert system in a pediatric emergency department” by Sethuraman U. et al in Acad Emerg Med 2015; 22:714–719).

The entry of fake patient weights (in kilos) for adult patients is the newest industry of systematic “lying” imposed by CPOE, because no ED actually measures all their patients’ weights on a scale! This is just the beginning of the pragmatic disconnect between CPOE and the lowly emergency physician. In a paper by D. Schwartberg et al. in J Surg Res 2015 Apr 30 titled “We thought we would be perfect: medication errors before and after implementation of CPOE,” the authors found CPOE to significantly increase ordering errors. The 26-week pre-CPOE period at this university teaching hospital had 1741 order errors and the 26-week post-CPOE period had 2226 (p<0.01). So you might assume it will get better over time, but that was not the bill of goods we were sold. The computers were supposed to fix it all and stop errors immediately. And lastly, I offer one more meta-analysis that looked at 22 studies regarding CPOE. Both nurses and physicians spent more time at the computer after CPOE. Overall prescribing errors went down, and when decision support systems were included, there was improved dosing for targeted renal disease medications. The authors still concluded that the evidence in favor of CPOE was neither consistent nor comprehensive. (“The effect of CPOE systems on clinical care and work processes in emergency departments: a systematic review of the quantitative literature”. Georgiou A. et al. Ann Emerg Med 2013 Jun;61(6):644–653)

Mark Anderson, a health IT consultant in Montgomery, TX, told InformationWeek Healthcare that “specialists like their EHRs less than primary care doctors do because most systems were developed for primary care and lack many of the features and templates that the specialists’ work requires.”

Emergency department physicians who use “best of breed” (BoB) electronic health record systems rated their experience 59% higher than did ED physicians who have EHRs that are part of enterprise systems, a new KLAS Research Report reveals. But many healthcare organizations still impose enterprise EHRs on their EDs because it’s simpler to have a single integrated system.

Upon review of the MU2 criteria above, it is the third component of the table that is punishing the BoB Emergency Medicine-specific EHRs. Specifically, it is the requirement for the transmission of patient care summaries across multiple care provision sites. The enterprise-wide EHRs are specifically blocking the BoB EHRs from access and interoperability. This makes perfect business sense but it has been terrible for clinicians in specialty settings that really need a specific EHR to meet the environmental needs of their practice.

And so it was at the Los Angeles County / University of Southern California (LAC+USC) ED where our Best of Breed (BoB) EHR (Wellsoft) was replaced by Cerner….Ouch! I had just mastered Wellsoft, and was able to create charts where resident supervision was integrated with radiology and lab results in a way that accurately reflected care in a teaching environment. The intuitive relatively “click-free” Wellsoft product is generally agreed to be near the top in the EM BoB world. Even I could do it! Unfortunately, the consultants and inpatient services could not access these ED records. They then dropped our BoB EHR for an enterprise-wide EHR: Cerner. This was done because the BoB program could not meet the MU2 requirements that were necessary for ongoing federal stimulus funding. See the MU2 table again, which is torturing emergency physicians and their patients by killing off the BoB EHRs all across the country.

Clinically, this means decreased productivity, decreased revenues and, in some institutions, horrific increases in LBTC, LWBS and AMA rates. Decreased patient access is the clear result. For the LAC+USC Medical Center, which is a safety-net hospital, this represents a divorce from the mission statement of the facility. After the Cerner “go live,” the LBTC+LWBS+AMA (Left Before Treatment Completed + Left Without Being Seen + Against Medical Advice) numbers all increased substantially. Clearly this enterprise-wide EHR was not ready for implementation in a large teaching environment with a census approaching 200K visits annually. Will these numbers improve? We are now over three months post implementation and therefore some improvement is likely, but conversations at other centers years post-implementation reveal persistent decreases in productivity. Where did the productivity go? Data entry! See the following abstract for illustration of my point.

Hill RG Jr, Sears LM, Melanson SW. “4000 clicks: a productivity analysis of electronic medical records in a community hospital.Am J Emerg Med. 2013 Nov;31(11):1591–4

OBJECTIVE: We evaluate physician productivity using electronic medical records in a community hospital emergency department. METHODS: Physician time usage per hour was observed and tabulated in the categories of direct patient contact, data and order entry, interaction with colleagues, and review of test results and old records. RESULTS: The mean percentage of time spent on data entry was 43% (95% confidence interval, 39%-47%). The mean percentage of time spent in direct contact with patients was 28%. The pooled weighted average time allocations were 44% on data entry, 28% in direct patient care, 12% reviewing test results and records, 13% in discussion with colleagues, and 3% on other activities. Tabulation was made of the number of mouse clicks necessary for several common emergency department charting functions and for selected patient encounters. Total mouse clicks approach 4000 during a busy 10-hour shift. CONCLUSION: Emergency department physicians spend significantly more time entering data into electronic medical records than on any other activity, including direct patient care. Improved efficiency in data entry would allow emergency physicians to devote more time to patient care, thus increasing hospital revenue.

The promise of EHRs to enhance care and eliminate errors has largely been unmet. The MU2 criteria are also causing EM unrest and dissatisfaction by driving out the “Best of Breed” EHRs that actually showed some promise for emergency medicine. The AMA says we are not ready for MU3 and I agree. We must demand interoperability regulations to allow the best EHRs to succeed rather than promulgate the worst. Finally, I have to ask the most painful question of all: Did I really go to medical school, and then residency, all while dreaming of humanism, only to become a data entry computer jockey?

William K. Mallon, MD
Professor of Clinical Emergency Medicine
Keck School of Medicine at USC
International EM Fellowship Director
Los Angeles County / USC Medical Center

Jan Shoenberger, MD
Associate Professor
Clinical Emergency Medicine
Keck School of Medicine at USC
Resident Program Director
Department of Emergency Medicine
Los Angeles County / USC Med Ctr