What are medical data?
This section looked like it would be not very helpful, but it actually got me thinking about the different aspects of data, and what I usually automatically do with it. For example, one of the questions after the chapter asks you what it would mean if your pulse were 100. It can mean many different things, depending on the situation. If I had just ran to catch up with my favorite ice cream truck, it would be a normal response to increased exertion. If I were a pediatric patient, it could just be normal physiology. If I were losing blood, it could be a sign on intravascular volume loss. This type of thinking is helpful on wards. When I analyze the vital signs, I should do more than just check if they are in the normal range. Sometimes if they are in the "normal range" they aren't necessarily normal. And I can think about the patient's condition beforehand, and then look for certain changes in the vital signs that could be a signal of pathology.
Another part of the chapter talks about how medical terminology isn't standardized, and how it would be helpful if it were. If the computer knows what the information actually means, it can perform tasks to learn more about the patient, or to survey patient charts, being very useful in research. However, like I said in the previous paragraph, medical data are more than just a number. They change depending on the situation and the patient. This might be one of the reasons medical terminology is not standardized, yet. It is complex and conditional. If a program searches for all the patients with "hypertension" it would need many conditional statements to get the information that the researcher needs.
Another part of the chapter has an interesting statement: "it is crucial that they do not trust their memory when caring for patients." I've actually received conflicting training on this. One of my medicine residents said that we should have our patients in memory, and that we should be able to present them without any notes. Later, an attending taught me that it is foolish to do this, and that things can be forgotten so easily, which could result in errors. I think there are things that are useful from knowing your patients from memory, though, even if you don't solely rely on our memory for decision making. I should know my patients well, and not just rely on the chart to tell me everything. When I do this, I can think about the patient and the disease process even when I am away from the chart.
I was especially in accordance with the section on "Weaknesses of the Traditional Medical Record System". As I posted before, with the capabilities we have today, computerized records provide so much more to the patients. It's not that the doctor is lazy and doesn't want to flip through a few pages. The charts can get huge, making it very difficult to find what your looking for. Additionally, probably every time I've used a paper chart, I've had parts that I couldn't read. That is not good medicine, and not good for the patient. If it is not legible, it's like it was not written.
All in all, this chapter got me thinking differently about the data I use and interpret. It also gave me some ammo for convincing people that paper records are insane.
You've made some great comments and observations about the contextual nature of medical data.
ReplyDeleteI also like your comment about the conflicting directives you've received on wards - one doctor says do it from memory and another says that's foolish. Obviously the reality is somewhere in the middle.
The doc who asked you to do it from memory is putting you to the test to make sure you've synthesized the data. A large fear of an EMR and the large amounts of data it can readily provide is that a doctor will simply rely on the presentation and amalgamation of the data to provide care.
Care comes in the form of synthesis of the data - some of that synthesis will come from an EMR but most of it will come from you.
Good stuff.
Again, very insightful. I think you'll agree that the current EHR is really nothing more than organized electronic paper. Don't get me wrong, infiitely better than a paper chart where no one can find or read anything.
ReplyDeleteWe have yet to unleash the power of the EHR which really comes from clinical decision support and physician order entry (CPOE). However, this is where you will find the greatest pushback from providers.
Another thought: the EHR needs to work more like Facebook:
1. No training needed (even my 10-yr old can do it)
2. When something is updated, it goes out to everyone (the age-old push-pull question).
3. More collaborative like Facebook, AND the patient is actually a contributor.
Thoughts??
Kevin Pho has a great Blog that discusses this and other great EHR/social media questions. Highly recommended.
I was thinking about how data can be used...longitudinally, I guess. What we talked about yesterday helped me refine some of my ideas. As far as the social media style of lots of comments being added, I think especially for ambulatory, it is still meaningful to have a note in the old format. It kind of gives a snapshot of the hpi, pe, labs and plan for the current visit.
ReplyDeleteHowever, there are ways we could make that record more than just an electronic representation of the old way, even if it still looks like it. I envision each section(hpi, pe, plan) divided into sections for each problem. Many do this already. They could be numbered or bulleted, whatever. The power of the computer could come in making it possible to track the individual problem in history. It'such like you were talking about yesterday. Having it organized by problem, but being able to still fit it into the traditonal note style when needed.