Converting from paper chart to an EMR: Keep daily use in mind

Maybe you’ve been assuming that when it comes time to go live on your EMR, you’ll simply scan each patient’s old paper chart into the electronic system.

Maybe you haven’t given it much thought because you’re focusing on the change in your work flow when you start to use the EMR to document patient encounters.

Well, it’s time to pay attention to the transfer, because the conversion of the paper chart can significantly ease your transition to using the EMR at the point of care if you plan it with daily use in mind.

And simply scanning the paper chart isn’t the answer.

Why not? It may seem that if you just scan the entire chart into the EMR, you’ll have it at your fingertips whenever you need it.

But, scanning even 50 pages without some indexing — separating documents into predetermined folders similar to the tabs in a chart — makes it difficult and frustrating to search those pages to find that consult report or radiology report or lab test that you want right now!

So, what about scanning into predetermined electronic folders — such as imaging, procedures, labs, correspondence, etc.?

There are two things to consider in that decision: the actual act of scanning, which will require prepping the documents and inserting separators so that each item scans to the appropriate folder, and the end result.

And, although you’ll get some presorting for the documents, you’ll still have no way to find the specific radiology report within the imaging folder. Imagine having to click on page after page to open each one until you find the specific document you need.

Even if you decide you’re going to scan in such a way as to locate and identify pages quickly, you’ll require a filing nomenclature and standard. For example, you’d want to name each document in the imaging folder with the type of image study (e.g. mammogram, chest x-ray, CT-abdomen, etc.) and the date of the study.

If each specific piece of paper that is scanned is not identified and named, each one will default to a bland name and the date of the scanning — not very helpful when you want to see the cath report from last November!

On top of it all, scanning and naming the entire chart this way will take too much time to be a practical solution.

But if scanning is out, discrete data is in! That means you transfer only key information from the paper chart into the EMR.

That way, you won’t have to “chart from scratch” when you see an established patient for the first time after your transition.

One key piece of information to get into the EMR for each patient is his or her existing problem/diagnosis list. That will mean you don’t have to search the entire ICD-9 database.

There are other key pieces of data from the chart that should be in the EMR on “go live” day. The following can be reasonably captured and entered by the clinical support staff:

  • Medications
  • Allergies
  • Diagnoses/problems (ongoing, not acute/inactive)
  • Immunizations
  • Procedures/surgeries (what procedure and date) and health maintenance screenings and dates (just the item, not the result/report)
  • Maybe the most recent specific lab value based on specialty/patient diagnosis (e.g., PSAs in a urology practice or HbA1cs for all patients with diabetes), but remember that the data entry can take time, so don’t look for a lot of historical lab data to be converted.

Abstract these key components from the paper chart, use the mouse/keyboard (do not scan) to get them into the EMR, and you’ll successfully transition into go-live.

Rosemarie Nelson is a principal with the MGMA Health Care Consulting Group.

Originally published in MedPage Today. Visit MedPageToday.com for more practice management news.

Comments are moderated before they are published. Please read the comment policy.

  • http://drpullen.com Edward

    Even in 1997 when we converted from paper to an EMR there were a few key things we did.
    1. Effective about 6 months prior to conversion we had our transcriptionist add a header to each transcribed note that allowed the progress note to be flowed into the new EMR chart so we had 6 months of progress to begin.
    2. We made no real attempt to proactively scan a bunch of stuff into charts, rather saw patients for the first year with a paper chart and the EMR. At the first patient visit we selected documents to be scanned in.
    3. We see kids, so we focused on immunization info, now with the state database this may not be needed, but we hired someone to enter as discrete data the immunization record for each child under 5 years old. Think about what info for your patients it key.
    4. Don’t start live on an EMR until you can get your labs to flow into the chart. We got this done retroactively for 6 months from our lab after we had an interface made.

    To go to the average physicians 2500-3000 patient charts, and spend say 30 minutes each entering problem lists, allergy lists, med lists, etc is just too expensive and overwhelming to consider ahead of time. I say you can staff up and do it on the fly. Insist patients bring their pills in a bag to their first EMR visit, and get this info then.

  • http://www.heartlandclinic.org/plattecity David Voran

    Can’t stress the importance of discrete data. Blobs (Binary Large OBjectS) are the bain of our ability to have the EMR help us manage our patients and our patient pools. This extends to dictated daily notes since dictation is marginally better than scanned in documents.
    This became so apparent when we created our summary pages that aggregate data and give us stop lights as to whether we’re to goal on various disease states. The patients who’s insurance plan forces us away from our existing lab interfaces to scanned in documents show up as deficiencies because the computer can’t see the HgA1c or whatever other discrete value is being measured.
    The same is true for dictated notes in that the computer right now isn’t able to pick out the foot exam on diabetics whereas when structured templated documentation is used the last time the foot was examined is easily documented.
    We’re a long way from the point where we can leverage all of the computing power but if you don’t continually try to get data in discretely you’ll be much farther away from having the computer help you manage your patients.