EMR ROI depends on workflow improvements

Some of the toughest obstacles to EMR are the physician doubters.

These people say, “My charts are going to be on a computer.  So what?  All I know for sure is that it will take longer for me to finish my charts every day and we will have another component in fixed overhead.  Why this is a good idea?”

The concerns are valid.  When my practice chose to get EMR 6 years ago we made a decision of faith and vision, not from an ROI analysis.  But for most practices, faith and vision are not good enough.  We need a return on investment (ROI) rationale that justifies EMR adoption to the Doubters.  The IT experts talk in vague terms about workflow and re-designing your practice to take advantage of EMR, but these arguments are not concrete or specific enough.  Yet after 5 years of EMR no one in our group has ever suggested that our EMR investment was unwise.  I am convinced the ROI argument exists.  My next few posts will attempt to make the case.

Let’s start with an unusual example.  Your car needs new tires.  You live in a beautiful rural area but there is only one car shop, staffed by a single mechanic.  He is glad to put on new tires but the job will take all day.

Why so long?  How many steps does it take to put on new tires?Any interruptions such as other cars needing work, a phone call, emergency, etc. will make the job take longer because these events interrupt the work on your car.

1.     Remove the first wheel from the car
2.     Take the old tire off the wheel
3.     Put the new tire on
4.     Balance the wheel
5.     Put the wheel back on
6.     Repeat the above with other 3 wheels, one at a time.

Our solo auto mechanic must operate by sequential processing – defined as one operation at a time.

Now consider the other extreme.  You are an Indy racecar driver going 180 miles per hour around the track.  You need new tires fast.  You pull into the pits and the pit crew changes all 4 tires at the same time.  You also get mechanical adjustments, a full tank of gas, and the windshield cleaned.  A pit stop that takes more than 8 seconds is considered a failure.   This is parallel processing – defined as multiple operations taking place simultaneously.  Thanks to parallel processing the Indy pit crew can do in 6 1/2 seconds what takes the solo mechanic all day.

Now go to the doctor’s office.  The physician sees a patient with a suspicious nodule in his thyroid gland that needs surgery.  How many steps does it take to get that patient to the operating room?

1. Create a chart note that supports the need for surgery
2. Schedule the operation with the surgical facility
3. Preoperative labs, imaging, EKG
4. Specialist clearance (i.e., cardiology)
5. Pre-certification with insurance
6. Generate and complete documents
a. Surgical consent
b. History and Physical
c. Preop and Postop orders
7. Communication with the referring physician
8. Handle the unexpected – patient calls with questions, abnormal lab values, scheduling conflicts, etc.

How does the paper chart office handle these tasks?  In all but the smallest practices these tasks are each handled by different individuals.  Every step requires access to the paper chart, which can only be in one place at a time.  The chart won’t be available to anyone for at least 24 hours until the transcription comes back and is filed.  The paper chart office must therefore accept the slowness and inefficiency of sequential processing.  Workflow is defined by stacks of paper charts – stacks waiting for transcription, stacks waiting for labs, waiting for scheduling, etc.  And if the patient scheduled for surgery calls with a question…what stack is the chart in?  Will the chart find its way back to the right stack after the phone call is handled?  Everyone competes with each other for access to the chart.  Not only is the process slow and inefficient, it carries a high risk of workflow failure.

How is the same process handled in a doctor’s office that has EMR?  With the power of parallel processing:

1. The chart note, including the diagnosis codes, is immediately available to support preoperative workflow.
2. The chart note is paperless faxed to the referring physician the same day, sometimes before the patient leaves the office.
3. The staff is immediately notified of the new workflow via the EMR system
4. Consent, history and physical, and orders are all generated with a single button click
5. All workflows are performed simultaneously, greatly improving speed and efficiency and reducing the risk of a workflow failure.

With parallel processing there are no stacks of charts and no competition among staff for access to the chart.  Copying and faxing charts within the practice is eliminated.  The chart is everywhere, all at once.  Any phone call regarding a patient is easily handled without having to search for a paper chart and without the risk of killing a workflow because the chart was not put back in the right stack.

So where is the ROI?  The same work gets done with fewer people, fewer resources and less space.  These initial benefits happen without having to “re-engineer the practice” or change anything else about how things get done.  After electronic documenting becomes second nature it will be time to employ the concepts of remote access, computerized provider order entry, workflow design/automation and  “e-patient” functions like secure e-mail and patient portals to really get things cooking.

Mike Koriwchak is an otolaryngologist who blogs at The Wired Practice.

Submit a guest post and be heard.

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

  • http://wholelottarob.com Dr Robert Schertzer

    Great points re EMR. Without looking at the longterm improvements in workflow, it is hard to rationalize the ROI. However, these benefits more than pay off.

  • pheski

    I’m curious to see where you go with this series of posts.

    My own experience is that it is hard to get people to re-evaluate and re-engineer their work flows to get the improved quality and efficiency potential of any new tool – EHR included.

    The larger the entity is, the more complex the issue becomes, the greater the number of people whose life must change, and the less likely true re-engineering will occur. At one extreme, the EHR is a large and very expensive word processor and filing cabinet, with a large bureaucracy committed to asking the new tool to do the old job in the old way. At the other extreme, the group reinvents the process based on the desired outcome and the tool(s) available.

    With only anecdotal evidence, my impression is that this is not a bell-shaped curve It is heavily skewed to the wrong side.


  • http://onhealthtech.blogspot.com/ Margalit Gur-Arie

    I support EHR adoption, but parallel processing is not exactly what happens here. You can’t perform the 8 tasks in the manual list simultaneously, like race cars crews switch 4 tires at the same time. You cannot create a chart, send orders out, receive results and handle patient calls, all the same time since they all have dependencies on each other.

    You can perform each task quicker and more efficiently, except the portion where that pesky visit note is created, and that portion is the main focus of your interaction with the system, and that part will become slower for you and remain slower for a very long time.
    I think docs should get EHRs, and I think there are efficiencies to be gained, but nobody should go in to this with expectations of Indy 500 performances. And one of the reasons for EHR failure is wrong expectations. I would suggest that a good dose of faith and vision is still necessary.

    • r watkins

      Exactly. In a well-run paper-based office, all the tasks listed are accomplished much more efficiently than the author implies. A poorly-run practice is not transformed into a well-run one just by adopting EMRs.

      The crucial event is the physician’s interaction with the chart, and there is no evidence that this is improved by using EMRs.

  • http://drpullen.com Ed Pullen

    One issue with automating all these processes is the immense up-front work required to accomplish the automation, I agree they often pay off, but maybe only for expensive, time consuming processes.

  • http://drpauldorio.com Paul Dorio

    Margalit has a very important take on the EMR/EHR. I have watched the implementation of the EMR at my hospital and it is quite impressive. But one thing that stands out is its ever-increasing complexity. You need continuous tutelage to be able to effectively navigate and understand and find everything you need. How this makes for improved efficiency is beyond me. I hope that the docs using it (I’m a radiologist, so I consult it infrequently) will find better information flows as a result. That’s really the most important potential benefit. Time will tell whether outcomes are improved and whether CMS/Medicare bonuses/penalties are at all justified.

  • http://curbside.posterous.com Nuclear Fire

    The theory and ideals behind EMRs are wonderful but the actual products are junk.

    In many ways it reminds me of the early generation of PDAs. Always on the cutting edge of technology, I had the early Palms, Visors (Handsprings), Clies, Newton, etc. The idea was to be more efficient, save trees, be more organized, be able to find the data more quickly. All the wonderful things we want from EMRs. But hardly anyone else had them because they really didn’t make life easier. Data entry was too slow (sound familiar?), different systems didn’t communicate, the products were too expensive, etc. They were fun toys and as a hobby something to waste money on, but not a good business or productivity decision. It wasn’t until the technology became good enough that people started to adapt them, on their own, en masse. Now it is the user that bugs IT if the hospital platforms don’t support the latest Android or iPhone OS whereas 5 years ago, my colleagues didn’t know what Exchange or Active Sync were.

    The problem with current generation EMRs are they are not good enough. When they are, the providers will adapt them on their own. Forcing people to adapt them will only encourage stagnation rather than encourage innovation. Force people to buy now, they’ll be stuck buying a crap product whose maker will then get a strangle hold on the market rather than keeping the market open for good innovation. If your product is good, you don’t need to beg for faith and vision or legislate adoption. Those with faith and vision who are right will do so, gain market share and the slower to adopt will follow along when they see the success. Those who just in too early will lose a lot of money (or waste a lot of money on a hobby).

    I’ve been installing EMRs on my laptop since I was in college. I still haven’t found one I’d trust my business to or that I like half as much as the products I use in the other aspects of my life.

    Until those that have drunk the Koolaide gain some sanity and stop demonizing those of us who have legitimate reasons not to move forward, real innovation and change in medicine will not happen.

    • r watkins

      Very well said. Thank you.

  • Marc Gorayeb, MD

    Where is the data? It’s not difficult to design an unbiased, well-controlled time-management study of a number of office practices both before and after EHR adoption. Produce the hard data, and maybe then we can have an intelligent discussion.

  • stargirl65

    This is ideal. Reality is:
    Cal the surgeon, office closed. Notes faxed before patient leaves anyway. Call back surgeon the next day. They are busy for 3 weeks. Patient on wait list. They get patient in. They call and ask for notes. Advised notes sent. They cannot find. Resend notes. They call for referral and can you back date to yesterday because they forgot. Also need referrals for the ultrasound they want and the ECG for preop and the surgery center and the surgeon. Generate all 5 eferrals by EMR and fax without printing to other sites. (They all call back later because they lose them. Easy to resend since in EMR) Get patient in for preop since it has been 6 weeks since things started. Notes sent before they leave office to surgeon, anesthesia, and given to patient because always lost by OR before surgery.

    EMR allows one person in my office to perform all these functions. Unfortunately system is still too disorganized to manage things yet. Either that or other offices are very poorly managed.

  • jsmith

    Adopting an EHR based on faith and vision. Wow, that’s a pretty amazing statement. I sure hope you wouldn’t treat a patient based on faith and vision.
    The ROI for our EHR has been well less than zero. Sometimes visions are hallucinations.

  • http://www.theblackribbonproject.org Beth Haynes

    EMR may be a good idea–but what isn’t a good idea is having the government decide whether of not it is a good idea for me personally and to have the power of rewarding me (with other people’s money) if I go along or punish me (with my own money) if I don’t.

Most Popular