There are two ways to scan data in my EMR:
- Through the ADM program, which is separate from the EMR, and automates much of the process
- Directly to the EMR, through an upload and attach process in the application
However, I heard from several colleagues that the new ADM program was significantly better; I went to see it, and it was better. Another physician in my group came to my office to have a look at our scanning process (she was using ADM). We had a look at her scans during lunch, and figured out that some of the quality problems were due to the fact that she was not using the right software to open the files. Her scanned files were saved in tiff format; we switched to MS document imaging--it looked better and we could use OCR (Optical Character Recognition) very easily. I had found out how to associate file extensions with different programs because of a home computer problem, so I was able to do the switch for her.
The ADM software does not work out of the box. My IT guy had to install it and to tweak some other files to get it to work properly. However, once that was done, it worked. He showed us how to use it; we started and had to make a couple of changes, but now it works well.
My scanning tech can now scan papers in batches: she puts a whole stack of papers in the scanner, and scans everything into a single file. The program automatically saves it to the right area without the need for naming the file. Once it is in, the ADM software shows her what has been scanned, and she uses it to attach different pages to different patients and different areas of the chart. The software then uploads the file to the right area of the EMR.
I then see it as a tiff file attached to a patient chart in my EMR inbox (separated into lab, DI, correspondence). I can click on the file's link to open it up. Once I see it, I run the OCR process (using the little "eye" icon in MS doc imaging), which is very fast, and then I can highlight what I want and copy and paste it into the "comments" section.
OCR in tiff seems to work better than in pdf--it is easier to highlight the section you want to copy. I find that I am copying more of the letter to the EMR.
This is still problematic. OCR is not perfect, and there are always errors. You have to proofread and correct the text, which takes time. I have a saved copy of the original, so what I do is look for bad errors (numbers being wrong etc), and leave minor problems alone--example: MRI OP BRAIN.
Anything which is OCR'd and copy/pasted is now saved as part of the EMR record (not a scan), and is searchable. If you are parsimonious with what you put in, you end up with a nice summary which is easy to look at (CT chest: granuloma RUL. Echo: Normal). If you put in lots of stuff, it becomes harder to wade through the information or you have to do a text search. If you put everything in via OCR, you don't have to individually load each scanned document when printing a referral or a transfer, but you can end up with a lot of misspelled garbage, and there is no formatting--it is hard to look at.
I'm kind of in between the two. I'll copy the relevant paragraph to the EMR (diagnosis, management suggestions), and leave the rest as a scanned document. When we transfer a chart because a patient is moving to a new family physician, my secretary copies the EMR chart to a CD, but not the scanned documents. I figure the EMR really contains the relevant summaries of everything that is needed. I don't know if I should start including only the EMR summaries instead of the scanned documents when sending referrals; I guess it depends on what the referral is for.
You can see what it took for me to change my process:
- Better software from the EMR company
- Seeing for myself that the quality of the images had improved
- Figuring out the file attachment problem (over lunch with a colleague)
- Having a good IT person who could both do the installation for me, and troubleshoot it afterwards
- Training on the new processes and revising how the secretary scans at the front and how the doctor looks at the scan
I don't mean to imply that I like scans now; I still think that it is a waste of everyone's time to have to re-digitize documents that were originally produced in digital form. However, the hardware and software is getting better at handling this necessary evil.