Electronic Medical Records - Risky Business

This post examines the case for installing Electronic Medical Records (EMR’s) in primary care offices. The Stimulus Bill earmarks billions for EMR development, and politically, savings from installing EMR’s have been cited as a way to pay for health care reform. Many of us in the clinical world think this is all chimerical. Here is the reasoning.

Everyone can see the mountaintop there in the distance: No charts in the office. When you see a patient, little reminders pop up, indicating the need for a screening test at this time, or an immunization update, or medications that interact with each other. Lab tests, X-rays reports, and specialist consultation notes filed automatically in the chart at night while we sleep. The complete chart available at the Emergency Department when a patient arrives. Data can be collected widely so treatments can be evaluated scientifically. No paper no paper no paper. Ojala!

And it is so politically convenient. The problem with reforming health care is, everybody's ox is subject to goring. Your waste is my income. But, talk about introducing electronic health records and a beatific smile appears. Innovation without victims! Efficiency! Theorists, and Obama's early backer who built the Allscripts Company that promotes its own EMR product, say that introducing EMR's will save money that can be reinvested to cover the uninsured.

A tipoff, however, is this --non-clinicians are more enamored than clinicians, theorists over those who practice. Clinicians warn that if there are economic benefits to be had, they will be far in the future. So, beware.

In fact, EMR's have been very slow to be adopted in this country. Some specialty practices have adopted EMR's, with good reason. Their tasks are very stereotypic, so creating templates for their visits is very easy. Also, specialists make a lot of money, so they can spend money rather easily. Hospitals are also prime prospects for EMR's. They, too, have a lot of money to spend. Also, when hospitals invest in this product, they themselves realize the efficiencies, since hospitals are complex enterprises and their departments are integrated in one entity and one budget.

Primary care practices, however, have been very reluctant to invest in EMR’s. There are probably two types of practices where they can make sense. One, a small practice of nerds. Work on it together and intensively, know it intimately yourself, be able to fix it yourself, take pleasure in the geekiness. That makes sense. The other early adopter would be the large, integrated practices, where the top administrators make the decision but don’t personally bear the cost and trouble, and possibly where efficiencies can be realized internally. But aside from these two types, the bulk of primary care has not taken the EMR plunge.

Here are the reasons:

• EMR doesn’t make sense as an investment. They are expensive to buy and to maintain. With EMR, it is a challenge simply to come back up to the level of productivity you were at with paper charts, because the EMR makes the clinician a data entry clerk. The transition from paper to EMR is excruciating, taking over a year of lost productivity in many cases, and costing hours and hours and hours of work, summarizing charts for EMR entry. The savings on paper file clerks is offset by IT and other electronic jobs. So, adopting EMR is investment without financial return.
• EMR is underdeveloped. Many think the current EMR’s are not ready for prime time. They can be cumbersome to enter data and to retrieve. They are reminiscent of early Windows.
• EMR is proprietary. Get into one EMR and if you don’t like it, you’ve got problems. There is no uniformity among EMR’s, so going from one EMR to another is like going from paper to EMR, which you have already done.
• EMR doesn’t necessarily connect with other medical entities. Get one for your office, and the odds are you cannot have reports enter seamlessly into the record. All too often, one still receives paper reports from hospitals and labs and you need to scan it into the EMR, with no improvement in productivity, and in addition, it will sometimes be difficult to find.
• EMR is not set up to do medical research. One of the objectives of EMR is to be able to compare and contrast treatments. The EMR of the future may be able to do that. Not the present.
• EMR data may be useful to others against you. Perhaps paranoid, but no one trusts insurance companies, and the EMR might provide them ammunition, which you have paid for.

Finally, installing an EMR is risky. Here is a real life account of one practice’s trying to install an Allscripts EMR – that’s right, the EMR from Obama’s friend’s company – and associated hospital EMR experiences:

We have a large 2-site pediatrics practice (12 providers, 9FTE, about 45,000 patient visits last year). Workflow changes were the biggest challenge we faced when we converted to EMR 2 years ago, and we are still working through some major changes. We are using Allscripts Healthmatics (with Ntierprise practice management system). Although the EMR conversion team addressed basic workflow issues during the planning period prior to going live, there were many issues that were never addressed in sufficient detail with the appropriate stakeholders. This was espcially true for our billers, advice nurses and our laboratory. We feel that this process was rushed, mostly to save money, but partially because there doesn't appear to be a surplus of EMR employees with clinical experience out there who truly understand workflow issues.

It was frightening to convert to EMR....We found that we had to be MUCH more vigilant during the year following conversion to make sure that things were not falling through cracks with the inadvertent click of the wrong button. This involved every department and employee in the the office....phone personnel, referrals manager, care coordinators, advice nurses, medical records, billing, floor nurses and MDs. There are large burdens on nurses for procedure and immunization entry, and the overall management of immunizations turned out to be extremely complex and frustrating. Because of multiple errors that occurred during the rushed coversion, we found many immunization entry errors and We ended up having to HIRE a full time experienced RN to review every single patient coming in for a physical and reconcile differences in immunizations between paper charts, NC immunization registry and EMR records. This process has been going on for over a year. We have recently interfaced with our lab vendor (LabCorp), and once again, we have had to expend unbelievable amounts of time and effort to figure this process out, and have had to hire additional lab help so that our lab director can put full-time effort into this ongoing conversion.

THere is so much cost competition going on between EMR vendors, and the salesmen are not being forthright about what you are paying for in terms of services. The educational and training process for MDs in our office was totally inadequate. You get what you pay for, and we found that out the hard way. I was "On Call" the day we went live, and the EMR conversion team was only able to help me hands-on with ONE patient. This is all the hands-on training I had. The biggest problem was that, of the 4- member training team sent by Allscripts, only ONE knew the product well enough to train us. About two months after the conversion, when everyone was drowning, we had to pay to fly our favorite company consultant down, then pay $180/hour for her to work with each of the MDs to help them customize. We have had to do this 4 or 5 times in the past 18 months.

The time requirements for documentation for MDs has significantly increased with EMR, and has been hugely stressful. We had one salaried physician who resigned shortly after EMR conversion, with the complexities of EMR cited as a reason. Documentation efficiency for mds has begun getting a little better....however, charting times remain unacceptably high....in large part because the successful use of EMR with automatic billing REQUIRES MD INPUT in many more areas than paper charting did. There were many tasks that could be delegated to other personnel with paper charting and billing, then reviewed and signed off by MDs, and this is simply not the case with EMR. I do believe it is necesssary for practices to convert to EMR, and I have found it very useful to be able to chart at home, and to send electronic prescriptions. I expect that we will continue to discover new ways to make our EMR work for us.......but collaboration with other users will be key to doing this efficiently.

In addition to our practice EMR system, we have also experienced conversion to electronic ordering and charting at two hospitals. One hospital did a decent job with CPOE conversion, despite the complexities of trying to train university and community physicians. Their EMR system is basic and relatively easy to use, even without formal training. The other hospital did a decent job with CPOE, but the conversion to EMR in the hospital nursery has been TERRIBLE. The choice of products was poor, and the implementation even worse. Now, When we round on one newborn in the hospital, a task that used to take about 15-20 mintues, we must access several different non-interfaced systems, and the process for ONE newborn takes at least an hour.. After examining the infant and talking with the parents, we have to acceess the paper chart, where many records are still kept, then access the CPOE system for labs and orders, then access the university hospital browser, which then provides access to the EMR systems for mother and baby, where we find charting of vitals, hand-entered lab results and progress notes. There have been numerous examples of incomplete charting, and operator error in entering lab results. It is frightening, and very dangerous for the patients, and we are considering pulling out the hospital.


On reading this account, our Medical Director at Bayside, Rich Ash, who joined us this year after experiencing an EMR installation in his former practice associated with the University of Pittsburgh, said this:

very interesting, and rings very true. ...brought back memories :)

So, EMR? Probably not ready for prime time.

Budd Shenkin