Feedback Loop

flying cadeuciiRight now there are two patients in every room. One is made with flesh, bones, and blood. One is made with a monitor, a mouse, and a keyboard.

Both demand my time.

Both demand my concentration.

A little over two weeks ago I wrote the short story Please Choose One. I posted it online. The response it generated exceeded anything I could have ever imagined. It struck a nerve. People contacted me from all over the world, from all walks of life, about the story. Everyone, it seems, can relate to the challenge of having to choose between a person and a screen.

People sent me all kinds of suggestions and ideas. A few sent words of encouragement. Yet, what struck me the most about the people who contacted me was what they did not say. Not a single IT person argued the computer was more important than the patient. Not a single healthcare provider stated they wanted more time with the screen and less time with the patient. And finally, most importantly, not a single patient wrote me and said they wished their doctor or nurse spent more time typing and less time listening.

Medicine is the art of the subtle- the resentful glance from the mother of the newborn presenting with the suspicious bruise, the solitary bead of sweat running down the temple of the fifty three year old truck driver complaining of reflux, the slight flush on the face of the teenage girl when asked if she is having thoughts of hurting herself. These things matter. And these same things are missed when our eyes are on the screen instead of the patient.

I get it. We need to collect the data on patients. In the modern world, medicine is also a business- a business of collecting, sorting, and collating data for billing purposes. I am not naïve enough to believe or argue otherwise. But maybe right now we need to step back and ask ourselves the one question no one seems to want to ask:

Has the data we store about the patients somehow become more important than the actual flesh and blood patients themselves?

One of the most difficult things to do in the practice of medicine is to recognize when a previously established diagnosis is incorrect. It requires having an open mind that maybe, just maybe, the prior five doctors have been wrong. I wonder if we are at a similar point. Maybe we do not need another screen in the room, another page of data, another flag popping up on the screen warning us to address some incomplete part of the patient’s record. Maybe instead, we just need to spend those thirty seconds interacting with our patients.

Computers, EMRs, and patient databases are ultimately a good thing. We need them. I have no doubt that we will reach the point when they can collect all the data they need without inserting themselves between the doctor and the patient. But we are not there yet.

To the IT people out there who were offended by the story, my message to you is clear. You are the very ones who can help save us. Keep working, keep innovating, keep looking for ways to build a better, more invisible system that still does what it needs to do. After hearing from so many concerned people in the IT industry, I have nothing but faith we will find our way together. Ultimately I am reminded that we all want the same thing: to do what is best for the patient.

I am looking forward to the day when I step into a room and there are two providers. One made with flesh, bones, and blood. One made with a monitor, a mouse, and a keyboard.

Both advocates for the patient.

Philip Allen Green is an ER physician based in Walla, Walla Washington. He is writing a book based on his experiences as an emergency room physician. More information can be found at PhilipAllenGreen.com

Categories: Uncategorized

Tagged as: , , ,

14 replies »

  1. Great article. As a nurse, sometimes I feel as though I spend too much time on the computer and less time on the patient. However, I know how important documentation is and how EMR, patient databases and computers are actually a blessing to us. I would prefer computer documentation over paper charting any day.

  2. Phil, there is an EMR based on Google glass that a company is doing some work on here in Seattle. We just saw a presentation from them at University of Washington last week…looks promising, from what I could tell:


  3. I am curious, is anyone aware of an EMR system that is currently in use or in development that is a icon based, touchscreen type system for physicians? Something along the lines of an ipad or iphone where use is intuitive and simple? If yes could you post a link here? I would love to see them.

  4. The hospital system of EHRs crashed this week. Aside from the fact that no agency wants to record the problem or its aftermath of adverse events from the delays in transitioning back to paper, once the paper system was running, care was much better.

    Nurses and doctors actually spent quality time with their patients.

  5. Yes, I got the JFK allusion.

    I like the surety bond idea. Fraudulent claims are a risk, but that is true in many areas and it is known how to deal with it.

    Re the politicians…..I surmise there are some very good concierge medical practices around DC where the only records are paper ones kept in the office.

  6. @Slobodisn
    I was just making a tangential remark about how JFK kept his disease out of the public’s eye.

    Yes, I dont know how VIPs will silence sensitive data in the future That is why I believe patients will avoid revealing certain medical history and deny giving permission to do certain tests and images.

    But shucks, isn’t a partial answer a privacy bond or a surety bond issued by the government or a privacy default swap type instrument issued by a firm like AGE, or the HIT vendor like EPIC, etc.?

    This would not be like a personal injury tort because one wants immediate recompense if data is stolen or hacked or lost. You do not want to have to wait to discover actual damages. You want money to be paid to the patient immediately on discovery and proof that data was revealed to unauthorized parties. This would do a lot of good things: every stakeholder handling the data would be very energetic in securing it. It would quickly maximize software security and concentrate minds. It would assuage patient fears and enhance participation in insurance.

    But it might also stimulate collusion and fraud by patients joining with others to reveal spurious data and collect the bond money. I guess administrative judges would have to decide these issues.

    Perhaps money would not be the choice recompense. This could be free hospital care or physician care or clinic care, whatever. This might reduce fraudulant temptations.

    Anyway, some kind of privacy default bond as above might help to reduce this security problem to its minimum.

  7. Here’s an analogy:

    If you’ve ever used an old program to create a web page (and come to think of it, many of the newer systems have the same problem) you may have noticed that code that starts as a relatively clean page of simple commands and tags can become pretty crazy as you work with it.

    Add a couple of lines, change a couple of fonts, make a correction or two and your beautiful code starts to become a little messy. Add a couple of authors, a couple of major revisions, drop in graphic or two and a couple of files and your code becomes a living, breathing monstrosity.

    It’s almost impossible to make heads or tails of what’s happening.

    Whatever logic guided the creation of the document has been lost. It’s virtually impossible to tell what goes where. Good information has been corrupted by bad. Sorting through it all can take hours. An experienced coder knows the easiest solution.

    Delete everything and start over.

    The same thing happens as information is added to an electronic medical record. What started out as a relatively easy to understand document becomes increasingly complex, as the automated system and doctor after doctor dumps information on it.

    What we need are tools that help us go back in and clean up.

    What we also need to be doing is asking ourselves if we’re asking too much of this technology too soon.

    A tool that has was built for administrative purposes and clerical purposes is being pushed out to the front line. We need to give doctors lightweight new tools that don’t get in their way, tool that give them the magical access to data that we keep talking about.

    Is anybody listening?

  8. Dr. Palmer:
    you said “Eg there is no politician who is going to allow the fact that he has Addison’s disease be publically known.”…..or have an addiction or a sexually transmitted disease. I assume they are not letting these things enter their health records. How are they keeping these things out of the EHR?

  9. I think we might end up with a hydrid system, the billing demographics and vitals and some lab and image data would be in digital format and the sensitive personal data in writing. This latter could be scanned by character recognition software with the patient’s permission and used anywhere he/she allows. Fairly good algorithms might be written to filter, or alert, possible sensitive personal data. Eg 1. Any data that might interfere with getting or keeping a job. 2. Any data that might shock or interfere with personal relationships like STDs or pregnancies or abortions or mental disease or drug issues.. 3. Any data that might interfere with success in elected office attainment or celebrity status. 4. Any data that might disparage ancestry relationships ( eg blood types showing this man is not your father) . 5. Data that might interfere with obtaining insurance policies or entering LTC facilities or receiving loans such as mortgages. Probably several more categories are in here.

    I think it would be refreshingly healthy for users of health IT data to have to go back to the patient to get these permissions to scan the written entries. It centers the patient as the key player in this economy and reminds the stakeholders what 18% of the GDP should be swirling about.

    But, alas, it may be too costly and inefficient to run a hybrid system. It seems we may be heading into an impossible brick wall where the easiest way out is going to be refusal by the patient to give all the truth and all the permissions. Thus we are pushing health care into a two tier and boutique or incomplete system. The 45 year old patient who is looking for a job and who is running high sugars is not going to enter a care environment that is connected to any computer system.

    We all love computers. It is sad that we were recently (sort of ) pushed to apply it to health care from the top down. What we had done with digital data in health care before was very useful but was bottom up: lab information systems; imaging and pathology transcription systems; discharge summary sheets; billing entry systems.

    Now we find ourselves having been pushed through a trap door. We must use EHR systems, yet we can’t use it completely. Eg there is no politician who is going to allow the fact that he has Addison’s disease be publically known.

  10. I don’t advocate getting rid of EHRs….I do advocate getting rid of the coercive inducements/subsidies. Getting rid of them will allow the health systems most committed to continue on the process of getting the software to work and at some point in the future enhance patient care….not interfere with it. It also will mean Epic and the others will focus more on making the software better so that the purchase will be compelling because of the merits….rather than because of govt. penalties.

  11. I agree with you. I am not advocating we get rid of the EHR and go back to paper charts. At some point in the future EMRs are going to be an amazing resource without ever interrupting patient care. We just aren’t there yet. I know the designs will evolve over time, I just wish they would evolve a little faster.

    Thanks for reading.

  12. Thanks for this great piece. In my opinion, the question isn’t EHR yes or no. The useful question: when we need the best user interfaces available to minimize the distractions you describe, why do we accept such poor quality in this regard? Much, if not most of our ‘screen distraction’ is a reflection of poor design.

  13. The EHR has become the patient to the doctor forced to use these ill conceived devices and the EHR has become the doctor to the patients forced top suffer the neglect associated with the use of these devices to run medical care.

    The EHR is an emerging disease rife with multiple unexpected and unpredictable complications.

  14. I believe that though helpful, these technologies may create an additional barrier between a patient and his doctor. This unfortunately forces the doctor to focus on one thing and ignore other factors that may contribute to the disease (i.e. what’s the patient’s perspective on his illness? how does he feel about it? how are his relationships with his family or friends?) when you create this barrier, it lessens your chances to sympathize and reassure, which may be intangible but might go a long way.

    Thanks for this great article! I enjoyed reading it.