EHRs Are Inevitable, Yet Studies Still Pose Serious Questions

1 Comment

Bukata RMAs studies emerge regarding the implementation of electronic medical records in emergency departments, we’re seeing a dramatic increase in testing and a shocking amount of physician time being spent on data entry. 

This article is one of two concerning the measurement of EHR effectiveness. Click here to read the article by Nicholas Genes, MD, PhD.


ADVERTISEMENT

 

As studies emerge regarding the implementation of electronic medical records in emergency departments, we’re seeing a dramatic increase in testing and a shocking amount of physician time being spent on data entry. 

Bukata W
 
I get it. EMRs are here to stay. And I think the data generated by EMRs are absolutely terrific. Finally we have some detailed information to help analyze what we do so that we can do it better. I get it.


ADVERTISEMENT

But there are lots of issues with EMRs – some are obvious and some not so much. Nobody disagrees that the transition from paper records to EMRs can be painful. But suck it up for a few months and we should be back to baseline – at least that’s the story. 

But let’s look a little deeper. What do the patients think when we sit in their room with our backs to them doing data entry. Do we look over our shoulder periodically to try and face them a bit? Do we convey the same sense of concern sitting at a computer terminal that we would if we were beside the patient with our hand on their arm as they tell us of their concerns?

I went to the UCLA ED a couple of years ago for a tour. It was striking – there were easily 10 computer terminals in the nursing station and every single one had a healthcare professional at it. At what cost was this data entry being done? Was the patient really benefiting? 

The literature is hard pressed indeed to show that outcomes are better – especially given that few studies have been done in the ED setting. And do we see more patients when we use an EMR? Not likely. So we can’t show if EMRs do either of two essential functions – improve the quality of care or help us go faster. 


ADVERTISEMENT

So I think the patients really get no substantive benefit, nor the providers (especially in the ED setting. Who gets the benefit? The billing office, those doing charge capture, the coders and the data analysts. 

I realize that my moaning about having the most expensive staff in the building doing data capture is a broken record. It is not only their salary, but the potential for them to see additional patients (the opportunity cost) that makes me crazy. 

But there must be some other benefits. Since the government is paying a big chunk of the initial capitalization of hospital computer systems you would think that they are expecting some return on this investment.

Seems there is some data suggesting that they are getting an ROI alright – but not anything that was anticipated. Consider this. The hospital levels of service are going up disproportionately since the advent of EMRs (and guess who gets to pay these higher charges – CMS, insurance companies and patients) and physician charges are going up as well.

A story in the September 21st. 2012 edition of the NY Times entitled “Medicare Bills Rise as Records Turn Electronic” demonstrated national trends reflecting substantially increased Medicare payments in hospitals that have installed EMRs compared to those who have not. The payment increases are both on the side of the hospital as well as physicians. 

More specifically, the number of patients being billed out at the CMS level 4’s and 5’s has gone up dramatically. “Over all, hospitals that received government incentives to adopt electronic records showed a 47 percent rise in Medicare payments at higher levels from 2006 to 2010, the latest year for which data are available, compared with a 32 percent rise in hospitals that have not received any government incentives, according to the analysis by The Times.”

Many specific examples of gross increases in billings after the installation of EMRs were provided in the story. The ease of creating a record larger than life was detailed and the use of macros and cut and paste charting were noted by the Feds as causes of potential concern regarding fraudulent documentation.

Bottom line – CMS and other payers are paying more for care since the introduction of EMRs than before. Excuses by administrators that this is because patients are sicker now and charge capture is better sound pretty weak.

But now for another surprise. One that I intuited was occurring all along but which was unequivocally demonstrated in the paper abstracted below. It is a before-and-after study of a single academic, suburban, community hospital. In an analysis performed 24 weeks after the installation of an EMR with CPOE, it was found that medication administration increased 58% (125 per 100 patients to 216), lab testing increased 60% (from 225 per 100 patients to 375) and ECGs increased by 66% (from 23.7 per 100 patients to 35.7).

With regard to imaging, there were very modest increases in MRIs and ultrasounds, but overall, there was no significant change. Why would this occur when there was such a dramatic increase in drugs, labs and ECGs? It’s just speculation, but it may have been impossible to order any more imaging studies than was being done at baseline!

The culprit for the increases in testing has to be due to order sets. I can just see how order sets are created. A bunch of doctors get around a table and start throwing out tests for the “chest pain” order set. Before you know it you have all sorts of tests that one doctor or another likes – we have the INR, the magnesium, the lipid panel, the two ECGs 90 minutes apart – you name it. Everybody gets to add their favorites. The next thing you have is a 60% increase in testing.

And why are order sets done – supposedly to save time. They can be ordered at triage and by the time the patient is in a bed they are back – all $1,000 worth. And who pays for this “time saving”? You got it: CMS (at least 60% of all healthcare is paid for by some government entity), private insurance companies and worst of all, uninsured patients.

Interestingly, this study found that throughput was actually improved in association with the installation of an EMR – a very big surprise to me. I was positive length of stay would be longer.

TRANSIENT AND SUSTAINED CHANGES IN OPERATIONAL PERFORMANCE, PATIENT EVALUATION, AND MEDICATION ADMINISTRATION DURING ELECTRONIC HEALTH RECORD IMPLEMENTATION IN THE EMERGENCY DEPARTMENT Ward, M.J., et al, Ann Emerg Med 63(3):320, March 2014
BACKGROUND: Several studies have questioned the presumed operational benefits associated with implementation of electronic health records (EHRs). 
METHODS: These authors, from Vanderbilt University and the University of Cincinnati, performed a longitudinal analysis at a single, suburban, academic, 34,000 visit ED to evaluate the effects of EHR implementation (EPIC ASAP, Epic Systems Corp., Verona, WI). Primary outcomes were time intervals representing patient throughput. Secondary outcomes included patient satisfaction, medication administration and radiology utilization. Data were collected for four weeks prior to implementation of an EHR and for 24 weeks after implementation.
RESULTS: The median length of stay was 185 minutes pre-implementation, increased to 198 minutes at eight weeks post-implementation, and then decreased to 163-179 minutes from 8 to 24 weeks. The median interval from clinician to disposition increased from 95 minutes pre-implementation to 115-134 minutes after implementation. Median patient volume was higher during the pre-implementation period (2,588 vs. 2,180-2,258 per four-week block). Post-implementation there were increases per 100 patients in medications administered (from 125 at baseline to 216-265), laboratory testing with results available (from 225.4 pre-implementation to 374.5 per 100 patients) and EKGs performed (from 23.7 to 35.7 per 100). Imaging remained relatively stable. Post-implementation there was a planned 33% decrease in unit clerk hours, largely reflecting increased administrative work in the EHR by clinicians. CONCLUSIONS: Implementation of an EHR was associated with both transient and sustained increases in ED performance metrics.
41 references. mward04@gmail.com (PMID: 24041783). Copyright 2014 by Emergency Medical Abstracts – All Rights Reserved8/14 – #10

A similar before-and-after study, with the same (Ann Emerg Med, June, 2014), looked at eight performance measure at 23 EDs staffed by the Schumacher Group and found no meaningful difference in four length-of-stay measures and four operational characteristics. No analysis of testing was included.

As an editorial note I must add that the Schumacher Group and other sophisticated practice management groups would likely do everything in their power to minimize any negative effects that EMRs would have on throughput and patient satisfaction. Such tactics could include the use of scribes (the scribe industry has exploded in response to the widespread adoption of these systems) and the rapidly expanding use of advanced practice clinicians (PAs and NPs) and dictation, even if by voice recognition software. So a study of the Schumacher group EDs is unlikely to be able to be extrapolated to groups that are not able to bring to bear similar tactics to minimize any potential throughput issues. Anecdotally I’ve heard of many EDs for whom throughput has never returned to baseline after adoption of an EMR.

Finally, here’s a paper reflecting just how much of a clinician’s time an EMR can consume. This paper did a time study tracking 16 attending physicians, residents and advanced practice clinicians for 30 hours in a busy ED. Bottom line, 44% of provider time was spent on documentation while only 28% was spent on direct patient care. During a busy 10-hour shift, mouse clicks approached 4,000. If this doesn’t represent a backwards system, nothing does!

4000 CLICKS: A PRODUCTIVITY ANALYSIS OF ELECTRONIC MEDICAL RECORDS IN A COMMUNITY HOSPITAL ED Hill, R.G., et al, Am J Emerg Med 31(11):1591, November 2013
BACKGROUND: The US federal government has mandated “meaningful use” of electronic medical records (EMR). EMR is meant to improve communication and patient safety, and to reduce error, but these benefits are unproven, and also may come with significant efficiency and operational costs. Industry has suggested that use of EMR could reduce yearly costs by up to $100 billion, but the cost of software for an EMR system can range from $150 to $200 million, and actual studies of the effect of EMR on cost have been inconclusive. Furthermore, the impact of EMR on provider efficiency, which would greatly affect overall cost, is uncertain. 

METHODS: The authors, from St. Luke’s University Health Emergency Department in Allentown, PA, tracked 16 attending physicians, residents, and midlevel providers for 30 hours and recorded time spent in data entry, patient care, staff interaction and reviewing test results. Time spent on individual activity types and total mouse clicks were tabulated.
RESULTS: Participants spent 44% of their time on data entry, 28% on direct patient care, 13% in consultation with staff and consultants, and 12% in review of test results. Extrapolated mouse clicks per case averaged 160 across selected cases, and approached 4000 during a busy ten-hour shift. 
CONCLUSIONS: This observational study is consistent with others showing that 30-40% of a physician workday was used solely for EMR data entry.
25 references (lynnmsears@ptd.net – no reprints) (PMID: 24060331). Copyright 2014 by Emergency Medical Abstracts – All Rights Reserved8/14 – #9

In conclusion, we still need to ask the fundamental questions of EMR. Faster throughput? Higher quality? Are EHRs helping or hindering? The studies by Ward and Hill suggest that we need to look dispassionately and carefully at what happens when an EMR is introduced to an ED. The evidence simply isn’t compelling that they make patient care better, more cost-effective or more efficient. Other than that they are great.

 
Richard Bukata, MD is the Editor of Emergency Medical Abstracts at www.ccme.com

This article is one of two concerning the measurement of EHR effectiveness. Click here to read the article by Nicholas Genes, MD, PhD. 

 

1 Comment

  1. Rick has been the voice crying in the wilderness for several years, to no avail. The suits have decided, on no basis in fact, that computerization of the medical record will be a good thing. The bureaucrats are licking their chops at the thought of near-instant denial of payment based on something omitted, as well as number crunching for … what?

    We the docs, and our patients, are paying the absurb price for this foolishness.

    Go figure!

Leave A Reply