911 Blogs

There's a blog published by the now online-only Seattle Post-Intelligencer called Seattle 911: A Police Blog. Many cities have similar sites that act as 21st century police scanner bulletins. While surfing for news this morning, I happened to find that the top three stories on the site involved patients I saw at Harborview Medical Center during my last shift. There are also pictures to help understand the injury mechanism. Here's one:

The entries each indicate transport to the hospital where I was on call. My willing compliance with HIPAA and patient confidentiality rules prevents me from saying any more about the specifics of the cases, but I will comment briefly on a facet of patient care that could use improvement. Information is often lost in the transition from witnesses to emergency response personel to emergency physicians to their hospital consultants. (I was a student on the orthopedics team at the time.) We hope that the important information is maintained, but invariably, there is something that we wish we had known at the time.

Even with excellent sign-offs between providers, patients come in to the hospital with limited histories. Patients could be 'out of it' due to shock, pain or pain medicine. There could be a language barrier. Patients are sometimes intubated. Important features may have been observed but not documented on the scene, in transit or during an initial physical exam.

One of the important questions in the patient's history for emergency docs are: How did this occur? Among providers, this question becomes: What was the mechanism? Discovering or confirming this info with the patient is one way emergency providers evaluate patient alertness and orientation while they do their injury surveys, so patients sometimes get annoyed at having to tell the same story over and over again. But that's if the patient can tell the story. Sometimes they cannot.

It turns out that the Seattle 911 blog had information that may have been helpful for providers to understand these patients' injuries. In two of the cases from Friday, the entry was made while (or soon after) the patient was in the emergency department, further underscoring the potential utility of electronic documentation of pictures. One of the patients described the accident in a way that when I saw the image, I thought, "I saw the person involved in that accident." The other image generated a, "So that's how that happened" response in me. The importance of pictures (yes, worth a thousand words) is well known in emergency care; the soon to be history Polaroids of automobile accidents are often taped to critically injured patients' charts. The photo below is more a reminder of how beautiful it was on Friday that how the accident occured.

It wouldn't have changed how we treated these patients to know the specifics documented in the blog entries; the primary determinants of treatment are derived from the physical exam and what the x-rays and CT scans reveal. But one wonders if speedy documentation of accidents and injuries in the field could ever be incorporated into the electronic medical record. iPhone medicine is already being practiced in many emergency departments. The fellow on our service used his Blackberry to photograph one of our patients' wounds. He only partially joked with the radiology tech that he needed it to plan for a surgery. The image was later used to communicate with the attending surgeon and was reshown the next morning during a sign-out conference.

Reforming and universalizing the electronic medical record is central to the Obama plan to reduce health care costs. I hope the software programmers include a mechanism for documenting accident photos. In the mean time, maybe I should keep the local injury blogs open on one of the ER's computers.

Photos are from the Seattle 911 blog and were taken by Ben Otteson and Dana Vander Houwen.

Clinical Knowledge

My USMLE Step 2 Clinical Knowledge score came back in record time. As you will recall, my score on the costly sandwich (CS) exam was PASS. Today, three weeks to the day after taking the 9 hour computer exam, I received my score report. I PASSED, and scored higher than I did on Step 1 (which I took back in 2003).

Since my score was higher than I expected, and lower than most of my peers, I've decided to share some data about how well the USMLE World practice software package prepared me and the accuracy of one practice exam score predictor, provided by medfriends.org. Those who know I'm an engineer at hear will not be suprised by the following chart.

Plotted on this graph are score estimates from sequential practice exams I took using the USMLE World software. That software grants you acces to a couple thousand practice questions and prepares exams composing of 46 questions each. I averaged the raw scores of the six most recent exams made of questions I'd never seen. That average was input into the medfriends.org tool and plotted on the date. The averages are rolling. The turnover was 4 exams (adjacent data points share from 1-3 individual scores in their average. On the scale above, the horizontal line is 184, or passing. You can see that my trend was dangerously close to that cutoff. But considering my poor showing on Step 1 back in '03, I knew I needed to buckle down.

Shown in red is my actual score - a 209. This is below the national mean, but right about where I'd hoped to be. The one data point immediately before the exam - that was the free USMLE practice exam provided by the exam company. I recommend you take it. not only, will it make you feel better, my experience was that combined with the score estimator, it was a better predictor of my actual score.

Finally, you may ask: why is he advertising such a low score? Well - there's a good chance that residency programs already know my score. If you're reading this because Google brought you here, maybe you'll appreciate seeing a self-reported score that isn't in the 240 range. If you're reading this because your my friend, thanks for being my friend!

If you haven't taken the exam yet, hang in there. Take some time to focus on the material, practice with question sets, and try to take some time off to prepare. I wouldn't, for example, recommend studying at the same time as taking a required psychiatry clerkship.

UpToDate Up To Date

Earlier this year, medical students from Wyoming to Alaska were up in arms about the University of Washington's canceling a heavily-used electronic medical reference called UpToDate. I was in the minority of voices saying, "Good Riddance." Not because I didn't use or like the resource, but because the company exerts a profit-motivated monopoly on medical guidelines. UpToDate wanted to charge gazillions of dollars to provide articles that the authors write for free!

I had gotten used to the slightly less organized and less thorough eMedicine articles when my university's library portal indicated,

UpToDate - Good News!

Renewal for the coming year pending. We expect confirmation the first week of June ...

The so-called confirmation hasn't come yet, but it looks like the threat of cancellation and the uproar from UW docs and students combined to renew the subscription. Whatever. I'll still use UpToDate, but I have another arsenal of tools at my disposal now, too.

Bacertial Rasmosis

In the coming age of the electronic medical record, all of your health information will be stored as pixels and megabytes. The consequences of this are far-reaching. Hopefully, your record will be able to move seamlessly between doctors and accessible no matter which ER you find yourself in after one of your drunken tree-climbing escapades. Of course, EMRs are not yet standardized, what with 5 or 6 competing software companies hawking their services. And there is the Health Insurance Portability and Accountability Act (HIPAA) to deal with... (What we really need is a Health Information Portability and Accessibility Act!) But this is all to set the stage for a unique problem faced in health care.

How does all the information get into the computer?

Doctors from my generation (and especially my 5 year younger classmates) grew up typing. They'll be able to type their notes and orders real time. In the outpatient setting, many resident physicians carry a laptop into the office to enter patients' data. But what about the folks who can't type very fast?

Dictation.

Doctors have notoriously bad handwriting. As patient encounters became more important for the legal record and later were stored digitally, there needed a way for scribbles to be translated into typeface. Docs started using services that would transcribe speech into text. Sometimes 'in house' stenographers collect tapes at the end of the day and type them up the next day. In more urgent situations, docs can call an 800 number and connect with a service that contracts out the conversion from spoken to written word.

These days, voice recognition software is phasing out the transcriptionist. Voice recognition is good enough that docs can list the illness history, physical exam findings and treatment plans realtime between patients.

But whether it's human transcription or computer, there will be typos. Take for example my recent discovery of a patient who presented with a significant history of bacertial rasmosis. 'Bacertial' looks enough like 'bacterial' that my mind barely noticed it. But rasmosis? A new class of infection? I immediately consulted Dr. Google. About the closest I could find was the species name for the herbal called black cohosh. Cimicifuga racemosa. Black cohosh is bad enough for you - especially in women of child-bearing age. I hadn't heard of infection exacerbation. That's the funny thing about medicine - there's always something new to learn.

The other funny thing about medicine is that some of what you learn is wrong.

Enter the punchline. This woman had recurrent bacterial vaginosis. The record was an error of transcription. Welcome to the EMR.

UpToDate OutOfDate

Where do most academic medicine providers get their information about the latest understanding of disease and treatment?

A web search engine called UpToDate. This service provides review articles of the medical literature that range from pretty good to excellent.

Imagine the horror medical students, residents and providers across the University of Washington system experienced today when they read this email:

"As a result of extraordinary price increases to provide UpToDate access for our distributed community of UW Health Sciences students, faculty and staff across the Pacific Northwest Region I have had to make the painful decision to cancel this heavily used resource. In spite of extensive negotiations with the publisher over the past two years we simply were unable to negotiate a price that our Health Sciences Libraries budget could afford. The price for UW is much higher than for other institutions, in large part, because we are a regional health sciences center with regular faculty and students across Washington State as well as in other states throughout the WWAMI region. This means that simple onsite access as provided by some Health Sciences Libraries (e.g. walk into the Health Sciences Library physically and access the resource) will not work for many of our students and faculty who would need to get into a car to come here or worse, fly for several hours!"

For more details about the specifics of this situation at the University of Washington, visit the UW UpToDate alert page. As a medical student spending the majority of his third year outside of Seattle, I am particularly aware of the crux of this issue. What I want to know is how they can charge so much money for what is basically a bunch of review articles. Just how much are their writers being paid? Do the authors get paid per hit? I kindof doubt it. The last time I checked, basic science professors don't get paid jack for writing review articles. (Actually, I just (this evening) reviewed the preprint proofs for a review article I was the first author on. Far from being paid, we had to pay the journal because we include color plates...)

Johns Hopkins refused to subscribe last year, and now the UW. As of July 1, 2008, UW students, residents and faculty will no longer have access to the most used resource for medical decision-making. In a way it is too bad, because the articles on UpToDate were the best annotated and easiest to read of any online medical resource. Because I am a (small potatoes) employee of WebMD, I've done a little bit of research in to the reference services provided by that company. So far, I haven't been able to use it for comprehensive reference information as seamlessly as UpToDate, but I think that will have to change. And, the WebMD portfolio includes a large amount of other cool information that is more media-friendly than UpToDate. Other services available to providers in the UW system include:

MDConsult (WebMD's healthcare provider page)
Cline-guide
DynaMed
AccessMedicine

If you are logged into your browser with your UW password, these links will send you directly to all of the subscription services.

Despite the inconvenience it will be for me not to have an UpToDate subscription, I am glad that UW just said no to big publishing. This info should be open access, anyway!

Anyone out there have tips for an UpToDate-free world?

As I discover tips, I'll try to post them under the tag, OutOfDate. But that reminds me, I've got a series about presidential health care politics I was going to write, too...

Asus Eee

My laptop is on the fritz. Five years of service and it doesn't want to start up sometimes and blue screens once week. Usually it just needs a timeout. Anyway, since I just will be using it for word processing and email, I am thinking of replacing my 6 pound monstrosity with an Eee PC, weighing in at just under 2 pounds. The beauty of this one is that it's all flash drive, fits in your palm and is going to cost around $250... if it ever comes out!!! It was supposed to be on sale last month, but everything seems to be hush hush. Anyway, I am on the lookout for it. I'll probably want a test drive to see if my gigantic fingers are too clumsy for the thing... What does Eee stand for?

A Clipboard for the Wards

I recently stumbled across a tool that could be of use for medical students on the wards. I haven't even started my first clerkship in earnest and it's come in handy already! Called the White Coat Clipboard, it's a sturdy aluminum clipboard that folds in the middle so that you can tuck it into one of your white coat's pockets. The part that is different about this one is that there are about a hundred important quick reference items printed onto the waterproof sticker, and the hinge allows for the board to be folded without creasing papers. It costs $29.95 on the website, but if you get a group together of of fellow medical students, you can get the price lower. (I heard about $20.) Here's a picture from the website where you can order the product:
Update 9/25/07: I emailed the folks at whitecoatclipboard.com to ask what their bulk pricing is, and this was their reply:

A bulk discount is available for purchases of 6 clipboards or more. Some students have found that there is enough interest among their classmates to make bulk purchases possible. If so, for any order of 6-10 clipboards the bulk price is $19.95 each. For orders of 11- 20 clipboards the price is $18.95 each. For any order of 21 or more clipboards the price is $17.95 each.

So there you have it.

UW Email on your Palm

For the one or two readers out there that happen to own Wi-Fi enabled Palms, wish to collect email from the University of Washington server, and have not set that up yet, the UW Computing and Communications Department has a step-by-step guide on their website just for you! Other Wi-Fi enabled devices can be set up from this page.

This is the first in what may be a series of posts related to my adapting to the technology used by medical students.