A survey conducted by the Institute for Health Policy between late 2007 and early 2008 found that only 4% of physicians are using a fully functional EHR.
A survey conducted by the Institute for Health Policy (IHP) between late 2007 and early 2008 found that only 4% of physicians are using a fully functional EHR, with an additional 13% using only portions of a basic EHR. Although EHRs have been available for the past 10 years, and recent incentives have been provided to incorporate EHRs into practices nationwide through HITECH (Health Information Technology for Economic and Clinical Health Act), the adoption rate has stalled. Why is this? There appear to be two major obstacles: (1) the lack of electronic patient information embedded into the EHR when a practice goes live; and (2) the amount of time it takes to enter information compared with dictation or handwriting. Let’s examine these issues in greater detail and evaluate which steps can be taken to minimize or remedy them.Data transfer from other sources
When evaluating data entry methods, we need to consider electronic data transfer from other sources. Let’s start with very basic information, such as a patient’s demographics and insurance information, which are the easiest data to enter. Because most practices already have a practice management system (PMS), physicians should require their EHR vendor to transfer all patient demographics and insurance information over to the EHR. Additionally, since the PMS application maintains visit dates and CPT and ICD-9 codes, physicians should require their EHR vendor to convert visit-level data into identifiable, discreet data showing patient visit dates tied to the appropriate diagnosis codes and any procedure codes not related to the office visit level. Using this methodology, a patient’s clinical record could be pre-populated with dates of service, problem lists, and medical history.
Another key set of patient data are laboratory values. Because laboratory organizations in most regions are required to maintain patient clinical laboratory results for 2 years, physicians should require their EHR vendor to download specific patient clinical laboratory results from local and nationwide labs. If these data are downloaded following LOINC matching criteria, 27% of the necessary clinical data could be available electronically on the first date of going live, saving the average practice more than 156 hours of data entry time per physician.
This same methodology would work for “active and prior medication history.” Using the SureScripts network, EHR vendors can obtain a patient’s medication history, assuming the medication was paid for by the patient’s local health plan. Given the number of prescriptions maintained in the SureScripts network, it is likely that more than 70% of prior medications could be electronically entered into a practice’s EHR database before going live, which could save the average practice more than 141 hours of data entry time.
The Patient Health Record
When a patient checks in at the front desk, the registration clerk usually asks the patient to fill out numerous forms that cover family, social, and medical history. Instead of asking every patient to fill out the forms when they are in the waiting room, practices could ask patients to fill out the same information via the practice’s Website or a kiosk located within the practice. Enabling a patient to electronically enter this information eliminates the costs normally associated with entering these data in practice. By using an electronic program like Instant Medical History, which interfaces with more than 40 EHR products, the practice can direct the patient to a site where discreet data can be captured and seamlessly imported into the practice’s EHR without anyone in the practice touching a keyboard. Each practice could customize their patient clinical questionnaires based on clinical protocols and physician-specific guidelines. Once the patient has filled out the practice’s questionnaires, a nurse or medical assistant can capture information on the chief complaint, allergies, medical conditions, vital signs, active medications, recent medical and social changes in the patient’s life, and other key information.
A survey conducted by the AC Group in May 2008 that included 137 practices found an 83% higher EHR implementation success rate when the practice assigned a nurse or medical assistant to entering Review of Systems (ROS) and History of Present Illness (HPI) data for patients into the EHR. These findings indicate that once the clinical data entry template has been approved by the physician, entry of ROS and HPI data could be assigned to a nurse or medical assistant, who would be trained to follow the physician’s clinical protocols and guidelines. This single change in data entry methodology could save physicians more than 100 hours of data entry time per year.
Of course, one of the best methods for discreet data collection is use of the Electronic Health Exchange (EHX). If discreet patient data have been collected by one provider, why should another provider need to re-enter these data into their EHR? We must move toward an integrated-community EHR (ICE) in which data can be collected once and seamlessly transferred to all providers who have the right to access the patient’s clinical information. Can EHR vendors accomplish this? The answer is a resounding “yes.” Following the government’s continuity of care document (CCD) data exchange standard, vendors can provide discreet data sharing between products and between practices, saving the average practice more than 200 hours of data entry work annually per provider. While all of these options will save time and money, the best method may be to combine an EHR with a new functionality that the AC Group has coined DRT (Discreet Recordable Transcription).
The concept of a DRT-enabled EHR is simple. Physicians would be allowed to continue to dictate their findings and clinical assessments in their own words, but the transcribed output would be entered directly into the EHR as discreet recordable data. Using this methodology, physicians would not have to change the way they practice medicine or the manner in which they interact with patients. Instead of reviewing a paper chart, the physician reviews electronic clinical data that have been created via the aforementioned data entry methods (eg, data conversions, data interfaces, data entered by the patient or nurse, and ICE) before entering the room with the patient and then follow their usual workflow. The physician talks with the patient, performs the required physical examination, and then discusses their clinical interpretation and treatment plan as per their usual protocol. After the physician dictates their findings, they plant it directly into the EHR as an electronic wave file, which would be transmitted to a local or remote transcriptionist for electronic transcription. Since 63% of the typical transcription is already gathered electronically via the various data collection methodologies, the cost for the creation of the final note is cut by more than 50%—an average of $6,000 annually per physician who has elected dictation over handwritten notes. More importantly, the transcription comes back into the EHR as discreet clinical findings, thereby improving clinical documentation, coding, and outcomes. A DRT-enabled EHR allows physicians who have elected to dictate in the past to continue this practice while cutting their transcription costs in half and generating a clinical note via the EHR.
Two companies that are pioneering DRT-enabled EHRs are CureMD and McKesson Practice Partner. Both of these companies have embraced the concept of extracting discreet clinical data from dictations. Using this novel method of data entry, physicians can use an EHR without having to perform any data entry. Although the concept of a DRT-enabled EHR is quite new, numerous physician focus groups have shown that a high percentage of physicians (87%) are extremely interested in the DRT-enabled EHR concept; thus, it may be an answer to boosting adoption rates. Regardless, if the EHR marketplace is not revolutionized by 2012, we might still be looking at EHR adoption rates of less than 20% across the nation.
Mark R. Anderson is the CEO and founder of AC Group, a market research and consultative company specializing in the Healthcare IT marketplace.
What to look for in an EMR
• The record must be totally collaborative to allow anyone in the office to open and chart without regard to others having the chart open.
• There must be security (an audit trail).
• It must be very customizable for the practice. No EMR company in the world has all the medical expertise to have the latest and greatest
templates for every specialty. Customizing must be simple and intuitive, but a template that is customized now cannot in any way harm the documentation done previously with that template. Just think of the new procedures, treatments, lab tests, medications and more that arrive daily.
• It must be capable of collecting that information, slicing and dicing it with great discrimination, and conveying that information to other health systems software.
• It must be capable of running off a client-server or be self-contained on a laptop.
• It must be affordable; $25,000 or more per user is ridiculous. A government-funded EMR should be affordable out of cash flow—that is, no upfront purchase of the software, but rather, turnstile pricing.
• It must be intuitive.
• It must be “graded” in its operational capacity. A new user can use obvious features, but as they mature in their EMR awareness, more features can be accessed. There is nothing like “need” to inspire to user to learn another step. That is, filling out a lab form is too slow and so if the user wishes to switch to a bidirectional lab interface, it should be available. If they don’t care, then at least give them the option of the software filling out the lab request form.
• It must be easy to assist users who experience difficulty. This is one of the most important items by far. The ease of assisting a user will make or break many EMR installations. My preference is to have the ability of the user, with a single click, to show their desktop to technical support, whether that technical support is in their large facility or in another prearranged site.
• It must be easy to update, including all the SQL schema changes, executable versioning, new clinical content, and so much more without the use of IT staff. If for every update, someone in the practice must go from computer to computer to update it, updates will never get dispersed. Already the bar is too high.
• It must be relatively simple to install, not requiring a dedicated IT professional.
• It must be capable of allowing the practice to be paperless. To design it short of that would ignore a significant percentage of the market. That means document management in the many forms of documents tif, jpg, doc, txt, pdf, Outlook e-mails, html, and even CCR. Additionally, it should be capable of outputting all those scripts, excuses, referrals, letters, and more. It should handle telephone triage (as it’s called in pediatrics) without generating a sticky note for the chart. It must have a forms feature.
• It should have alerts, messaging, and reminders for those who wish to use them.
• It must be fast. You want no one complaining of speed.