Every health professional is dreaming of ways to replace the age-old, paper-based notes with new and improved electronic means. This would permit the clinician to retrieve data instantly from a patient, with minimum reading or page hunting; cut down filing and transcription costs, improve clinical skills and protect their liability, etc.
Even after coming a long way, the electronic medical record (EMR) still presents different hurdles for the newcomers.
Choosing an EMR system will depend on how you want to use it. If you have not spent some time researching what is available and then tested those options, then the odds are you will buy a poorly adapted system for your needs. I would compare the blind purchase of an EMR system to booking a foreign tour without knowing the destination or departure or arrival time. There currently are companies that do test EMR programs and then sell their services as a consultant. This series of articles comes from the solid experience of a practicing doctor with a strong academic background who also is a programmer and a software developer. This series of articles is geared to give the practicing doctor an outline of what is available. You can use the services of a consultant or go on your own. In any case, this article should prompt you to question, investigate, test and become a more educated buyer.
The questions I asked myself about EMR, the facts I uncovered, and my suggestions will have the rigor of experience and opinions. The goal is to stimulate curiosity, avoid the pitfalls and describe what the practitioner should be looking for. You will know what currently is offered and what to expect. Most users are health professionals - intelligent enough to analyze a problem as long as they have the different parameters to solve their personal equations. This article aims to provide you those parameters in an unbiased fashion.
The most adapted solution will not be perfect but will be a mixture of compromise. Perfection would imply that those systems have reached full maturity and cannot be improved; I don't know of any software that can claim that yet.
The Biggest Problem With EMR
The biggest problem is the process of inputting certain data including how to enter a patient's history quickly, some X-rays, different types of treatment and exams. The two key elements of that process are hardware and software.
The Hardware Side
Keyboard entry from a regular laptop or computer: This is the most traditional and time-consuming process, which requires fast typing skills to be efficient. It is difficult to ask a patient to use a keyboard on their own.
Tablet PC with pen entry: A tablet PC is a laptop-sized computer with a rotational screen that can lock itself flat on the keyboard. This allows the user to access the computer as a normal laptop or as a tablet. It is priced between $1,200 and $2,500. This usually requires a dedicated laptop that the doctor or therapist may conveniently carry around. The laptop also can be given to the patient to enter data with a pen. The convenience, however, should not overshadow the weight of the system. Lugging 4-7 lbs. from room to room may be cumbersome for many.
Usually, most of those systems use a resistive type of screen which requires a pen input. Pen input may be a problem for a professional who uses their hand. The need for the pen may prevent on-the-fly documentation, forcing the practitioner to first treat, then document. The entries are somewhat limited and even with text recognition from handwriting on a specific tablet, it will take a lot of time. Tablet PCs usually are associated with whole phrases picked from drop-down menus which are contextual to specific questions. This is the current leading system.
Palm-top with pen: The interest here is portability. The setback is the small screen size, which forces many practitioners to wear glasses while inputting data, with less available data on each screen from which to choose. Due to their age range, many medical professionals have less than perfect eyesight.
Palm-top with touch screen: Those systems are extremely convenient and promote quick input, usually on the fly. The setback is that the data usually are less customized, as the screen is smaller. The menu needs to be in bigger font size to accommodate the size of a finger, which is bigger than a pen. Also, the touch screen palm-top software usually requires more screens through which to sift, compared to the palm-top with a pen.
Touch screen: The average size of a normal screen is between 15 and 19 inches diagonal. The different types include capacitative, resistive and sound-wave. They permit the direct input of data without the need for a pen. Because the screen is bigger than palm devices, it allows more choices of information from which to pick. Usually, they have an onscreen keyboard that allows for limited text input. They allow less data on the screen than a pen-driven touch screen.
They are not portable and they are pricey (three times the price of a regular screen). They also require software to be specifically adapted to their use. Contrary to what most vendors tell you, if they do not use a touch screen to demo the software, the chances are that the software is NOT touch screen adapted. Touch screen software usually is more expensive to produce and requires an onscreen driven keyboard, the elimination of the right mouse click, bigger buttons, etc.
My preference is for a touch screen of at least 15 inches which makes the input seamless for the clinician. It is, by far, the fastest and most practical system for a hands-on, practicing doctor. Watch the cashier in newer department stores - they need to use their hand to move the merchandise, but they also need to type the price. At the cashier's station, the trend is toward the touch-screen. I think it will be the same trend in the hands-on therapy arena.
The Software Side
Voice recognition: The main advantage of voice recognition is that it leaves both of your hands free and, in theory, would allow you to speak to a machine while performing another activity. We speak very fast and if data are picked up as we speak, a considerable amount of information can be stored this way. The second reason for voice recognition is to not need a keyboard to still enter data at a fast speed. This system has greatly improved in the past several years and provides a reliable way to enter data. However, there still are some major restrictions.
Voice training is required with any system and voice recognition is restricted to good diction. If you find it difficult to understand some accents or regional dialects, why wouldn't your computer have the same difficulty? Many doctors have an accent. The inflection is not always done on the right part of the word and the computer interpretation becomes erratic and not always contextual. Two words may have the same sound but are spelled differently and mean different things. This remains a major reason for many errors while using the voice recognition during normal, general voice input.
It is not socially acceptable to speak to a machine in the middle of a patient interview and patients could consider it to be disruptive or rude. Voice recognition forces the doctor to document after the patient encounter. For a long history, report or exam, voice recognition is not ideal. Written documentation should use visual landmarks such as bullets, indentations or a specific formatting of the information to make it better stand out when reread by someone else. Voice recognition is better while following a specific template.
Voice recognition using a template: A template will prompt the user to insert spoken text into an already formatted form on the computer. These voice inserts are a sort of "voice field," which is pre-formatted. Those voice fields may use a numbered, drop down macros menu. Several EMR systems allow you to instruct the computer orally for a choice of a number in a drop down menu. You read the number aloud, and the computer writes the macro corresponding to this number.
In brief, voice recognition is best used while associated with a specific information field, allowing you a second choice of input other than mouse or touch screen. Voice recognition is an outstanding tool to add short phrases or notes to seldom encountered conditions. The key to voice recognition is the degree of integration and customization inside your EMR system. As an example, it can be used to add an option not already existing in a drop down menu of your system. Voice recognition is a practical, secondary way to enter data. Voice recognition cannot be the primary way to input text if the user aims at documenting the encounter while seeing the patient.
Entering the Patient History
If an EMR system forces you to read the questions to your patient and tap on a screen as you go along, chances are this system is slower than your current paper and pen format. The paper form does not need to boot up, make you aim at a drop- down menu or print every letter perfectly in order to be recognized.
A question you should ask yourself is: Does it take me the same amount of time or more time to enter the patient history with this EMR software as with a piece of paper? If the answer is positive, it reveals that the interest of this system is not for saving time, but for the "electronic storing" capacity of the EMR. It would be much better if your EMR system provided you with both saved time and electronic storing.
You also might ask yourself the following: Does this EMR system allow the patient to enter, via touch screen, their history at their own pace? Certain EMRs have delegated the major part of the history taking to the machine. While the patient is in the waiting room, a touch screen is placed in front of them and they are prompted to answer yes or no questions by touching the screen. The answers are then forwarded to the doctor's computer in the presentation of a regular form with a built-in "flag" for answers of critical importance.
There is no doubt that this system, using the patient interaction, is the only one to consider if you are looking to save time.