It’s no secret that physicians and nurses have a tough job— physician burnout is common, and medicine is generally considered to be a stressful field. Every TV show set in a hospital constantly talks about staff overload and features nurses going on strike or doctors having to consistently work overtime in order to meet demand. And in the last few years, the burden has only gotten worse— all thanks to electronic health records (EHRs).
What are EHRs?
EHRs are the bread and butter of clinical documentation. Every member of the hospital staff — nurses, doctors, or others who impact patient care— must provide detailed reports about a patient’s medical history and stay in the hospital. While EHRs were originally designed to bring medical records up to date with the internet, making them easier to share, the flip side of this is that clinicians have to document every interaction with a patient in electronic software suites that can be hard to use. Insurance companies and hospitals have demanded that the data entry is precise and specific so that there are fewer debates about payments and billing. Malpractice claims are a lot harder to win when the physician has a complete record of what information he/she collected from the patient and what their thought process looked like in coming to a diagnosis. Billing is also tied to compensation and money. All in all, physicians feel compelled to over-document everything, which takes a huge amount of time. Some doctors have spoken out about this burden; for example, Stephen Bergman, who is a professor of medicine at NYU, recently co-wrote this popular article in The New York Times, which discusses physician burnout as a result of EHRs.
What are the components of an EHR?
There are many different building blocks that make up an EHR, but the easiest way to explain how to use the system is to walk through the typical flow of a patient’s experience. I’ll use the Emergency Department (ED) as an example because it’s easy to conceptualize what a patient’s stay in the ED looks like.
First, the patient goes to the front desk of the ED, which is usually called the triage section. A designated triage nurse will have the patient fill out some boilerplate information in an administrative/billing data section, which will include the patient’s name, age, sex (what are called “demographics”), insurance (if applicable), and other core information. Sometimes, this may have to be skipped if a patient comes in with an urgent issue, usually with an EMT or paramedic on the scene. The triage nurse will also begin to document the problem and why the patient is coming into the ED in a triage assessment, and will likely record triage-time vitals including blood pressure, pain, oxygen, etc. In some hospitals, the triage note is summarized in a one- or two-word phrase “chief complaint” which will try to encapsulate the problem in a pithy way. For example, if a patient comes in with a cardiac complaint, the chief complaint might be “chest pain”. If the patient has been in the hospital system before, the demographics and insurance information will automatically sync up with other databases to provide a complete picture of the patient, including past notes and discharge summaries, what medications they have been prescribed, any former lab or test results. Some of this information is completely unstructured — like the free-text notes that doctors have written about the patients —, while others contain quantitative information (labs, medication dosages). However, because there are so many places to save information, it can be overwhelming to look through a patient’s complete set of EHR data because of the barrage of information thrown at you.
Once the patient is checked into the ED and is done with triage, a nurse will likely come to start the patient’s diagnostic plan by recording more vitals and potentially administer a few mild drugs or start an IV before the doctor sees the patient. The nurse might perform a physical exam of the patient, and his/her observations will go in a “Nurse Comments” field. The nurse may also start to get a sense of the patient as a whole, including asking about past medical history, family history, and social history. This will help the doctor have an initial sense of the patient before seeing them in-person, which might help narrow down the diagnosis or determine whether the patient is in a critical condition such that they need to be admitted into the hospital.
The doctor will then visit the patient and take notes on all relevant information. He or she might order labs, tests, or imaging through the EHR software, which will then link any radiology or blood test results to the patient so the doctor can electronically view results as soon as they are ready. This also adds to information overload, though— it’s hard to synthesize the huge amount of data that is being collected about the patient, especially when looking for the correct report isn’t just a question of picking up a file from your desk. You have to navigate through the intricacies of the EHR to find what you are looking for.
After the patient is diagnosed, the doctor will write a report about why they made the diagnosis they did (called the “Medical Decision-Making” section), which is usually to prevent malpractice suits, and then convert their notes into a full-fledged summary of the patient stay, which is called the “Discharge Summary”. This is the document that will be filed in the patient’s records so future doctors can understand what occurred when the patient visited the ED. Finally, designated staff in the hospital will take this report and determine the ICD diagnosis codes for the visit, which will then go into a formula to determine how much the patient needs to pay.
Why are EHRs sources of complaints for doctors?
It’s totally true that EHRs provide a clean way of aggregating information without ever losing it. It also makes it a lot easier to share data between hospitals or give patients access to their medical records. At the same time, EHRs take away from the art and science of medicine by adding an extra documentation burden on physicians. EHRs are not always formatted in a way that makes the most sense for doctors. A common complaint is that it requires constant context switching from documentation to exploration — there isn’t a singular page to synthesize information. There are dozens of pathways and websites and databases, with no intuitive way of synthesizing it all. Epic, which is the leading EHR software and the one in use by all 20 of U.S. News’s “top-rated hospitals”, is often a sore subject among doctors. In a recent opinion piece, one doctor noted that physicians are constantly confused by Epic’s “overwhelming complexity” with “a dashboard… that feels like the cockpit of an airplane.” Medical notes are filled with “superfluous, robot-generated text,” largely because Epic is intrusive and will assign patients diagnosis in the middle of a doctor’s documentation process.
We’re finally in an age where design and user interaction matters a lot, though. There are entire fields of computer science and art devoted to studying how people interact with digital systems. We’re also living in an AI-crazed world— shouldn’t we able to integrate machine learning and artificial intelligence into the systems that determine the quality of healthcare that Americans are receiving? Hopefully, doctors will be able to use their voice and their power to demand better systems to work with. Maybe you’ll be hearing about an Epic 2.0 in the near future.