Electronic Health Records… what tomorrow brings?
- You are a physician at work seeing patients. Your computer detects that you are dictating a note and waits until you are done, then it presents an alert “ Mr Johnson in room 332 has a critical K of 6.5, would you like to initiate the standard adult hyperkalemia protocol or modify?”
- You are a nurse at work treating patients when your computer detects you are with a patient answering a call bell. When you are done and have stepped out of the room, your device prompts you with “Dr X has altered the heparin order for room 45, to be completed in the next 30 min”…
Both of these examples give the computer a sense of situational awareness. That means the software is aware of where you are or what you are doing and has prioritized information that needs to be delivered to you. It is aware of your workflow and when interruptions can be the most damaging, so it is programmed to alert at the least intrusive time. Could this be the EHR of tomorrow?
The electronic health record is a vast improvement over the paper record keeping systems of the past. It offers realtime access to a large compilation of medical data inside the system. It also allows for access and data entry by multiple types of providers. However, healthcare delivery is changing, and with it comes the need for our EHRs to change as well. What areas are changing and where are these gaps in the systems?
- Workflow: Increasing focus is being placed on the workflow of everyone who accesses the EHR. As hospitals and health systems seek to improve patient flow and the patient experience in the face of increasing demand, attention is shifting from the old question of “what can we enter into the computer and track?” to the new question “how can we decrease human time with the computer?” There is an appreciation for human to human interaction and it’s value in patient care. There is also a recognition that the EHR is at minimum expected to provide information, and to do so quickly and easily. The system supports the human, not vice versa. This is a growing focus as we meld lean processes with workflow evaluation at every level, reducing waste. Make no mistake, time spent with a computer is now considered waste. Counting clicks, focusing on information ease of access, and pushing results (and possibly interpretations) to clinicians is now a necessity.
- Communication: The EHR is not designed to be used for real-time communication but that is a large part of what we do in healthcare, especially in large institutions where multiple services support each other. Today, this issue is completely unaddressed by the EHR and the field is ripe for solid solutions. Often times there is reliance on third-party solutions. These tools vary from software to hardware devices, some utilizing personal smart phones, others pushing messages to pagers or proprietary devices. Some systems will store communications, others do not. Integration with the EHR is a natural focus of development. However, one of the stumbling blocks is the ever-growing amount of “metadata” being stored by systems. The legal realm in the US has been utilizing this information in medical malpractice cases and there is hesitance to have it readily accessible and tracked. As we struggle with what data is “protected” and what must be made available legally, this area has significant hurdles before integration occurs.
- Awareness: Here we are not talking about the machine becoming self-aware. This refers to awareness of the location or task a human is performing and then delivering interruptions based on a preset hierarchy. For example, the system should be able to tell when a user is entering data and what kind, since this is direct interaction with a computer. It also should be able to integrate location data and based on that determine if a clinician is in a room with a patient, at a medication dispensing unit, or off the floor. Combining that kind of awareness with alerts yields a system that can predict that Nurse X or Dr. Y are in with a patient and that an elevated lactic acid can wait 15 minutes before being alerted. It can also interrupt both providers to make them aware of an arrhythmia in the room next door. This type of hierarchy places value in human to human interaction and ranks it above most other alerts. Today, no such system exists. Clinicians face a constant barrage of interruptions which increases the chances of errors occurring. In the emergency department, studies have shown that physicians experience an interruption as often as every 4 minutes. A little awareness would go a long way in reducing these interruptions and combining notifications together at an appropriate time.
- Data processing: EHRs currently track large quantities of data. Much of it is reportable but in very basic data sets that require further processing to interpret. This is another area of third part growth that should be integrated into a complete EHR. Hospitals in the US have differences, we know. But there are more similarities than differences. In addition, struggles are similar. So why is it that a complex EHR displays data for exporting into a third-party solution for interpretation, instead of directly providing that interpretation? Why do we need yet another third-party to extract data and help us digest it?
- Patient Identification: We are still using humans to enter data into a system and create accounts or identify a patient based on a driver’s license or social security number. Why? My smart phone recognizes my voice and my fingerprint. Medication dispensing machines recognize fingerprints. Biometrics are already here and there is no reason to rely on patients to provide ID. It is actually more accurate to use biometrics and it allows for identification of the unconscious or altered patient. Fingerprint, palm print, retinal scan, facial recognition, all are here today.
There are many more areas of improvement for the EHR of tomorrow. The list will grow as we continue to examine what we do daily and why we spend so much time in front of a screen instead of a human being.