A computer worm that has alarmed security experts around the world has crawled into hundreds of medical devices at dozens of hospitals in the United States and other countries, according to technologists. The worm, known as "Conficker," has not harmed any patients, they say, but it poses a potential threat to hospital operations.
In a country where just 1.5 % of US hospitals have fully computerized records, West Virginia has created a paperless records system for its state-run hospitals and nursing homes serving the indigent elderly and mentally ill. The state did it on the cheap by using an electronic medical records system built by the Veterans Administration with taxpayer dollars, saving millions in software licensing fees charged by commercial software vendors. The VA software, known as VistA, is open-source software and it includes features such as a bar-coding system to track drug dispensations.
Electronic health records need a nudge from the government if the technology is to become widespread, the nation's new health information technology czar said. "It is clear that this field has not advanced (enough) ... when left exclusively to the private sector so there is a public role," said David Blumenthal, MD, head of the Department of Health and Human Services' Office of the National Coordinator for Health Information Technology.
In the push to digitize America's hospitals, Midland Memorial Hospital in Texas faced a need for information technology to replace archaic paper records, but a shortage of funds to pay for it. But Midland Chief Executive Russell Meyers found an unexpected freebie of sorts: the software used to power the electronic medical-record system of the Veterans Health Administration. Created with several billion dollars in taxpayer funds over two decades and used in more than 1,400 VA facilities, the source code is in the public domain and software developers around the world can freely build features into it. Add the cost of hardware and the services of a company that has adapted the VA software for commercial use, and Midland paid less than $7 million for a full electronic medical-record system.
Researchers at Boston Medical Center (BMC) are investigating whether a virtual patient advocate can improve real-life health literacy.
Developed by Timothy Bickmore, PhD, assistant professor of computer and information science at Northeastern University in Boston, the virtual patient advocate is on clinical trial at BMC to increase patient understanding of post-discharge self-care regimens. Bickmore and his team of researchers hope the system can decrease patient readmissions within 30 days of hospital discharge.
"Nationally it's been shown that about 20% of patients get readmitted within 30 days," says Bickmore, who adds that a third of that percentage are preventable. "There is a lot of information patients need to know before they go home. The typical discharge in the U.S. lasts about eight minutes and it's like 'here are your prescriptions and a pat on the back.'"
According to the U.S. Department of Education's 2003 National Assessment of Adult Literacy, 36% of the U.S. adult population has only basic or below basic health literacy levels. And while the virtual patient advocate doesn't have a medical or nursing degree, or even a pulse, she can provide patients with the information they need to know—via a computer. She can also devote all of her time to doing so.
The system operates by having a clinical trial nurse enter a predischarge patient's information into a computer database. This generates an "After-Hospital Care Plan" personalized booklet for the patient. After that, the nurse wheels the computer to the patient's bedside on a kiosk and gives the patient his or her pamphlet. The patient then converses with an animated virtual patient advocate character using a touch-screen display to go over their care instructions.
The system enables the patient to review his or her medication list, follow-up appointments, primary diagnosis, pending lab tests, medical durable equipment needed at home, and diet and exercise recommendations. It also creates a list of questions the patient can review with a nurse at the end of the interaction.
In addition, the pamphlet features a patient activation page that encourages patients to write down questions and issues they may want to discuss with their primary care physician prior to the actual visit.
To date, 220 patients have participated in the three-year clinical trial of the virtual patient advocate that began at BMC in fall 2008, which will conclude after enrolling 750 patients. These patients are getting 40 minutes on average and sometimes more than an hour, depending on the number of medications they are taking, to go over their care instructions, says Bickmore.
Researchers have already compiled data showing that low health literacy patients find the system user-friendly, and sometimes even preferable to receiving the information from a physician or nurse.
"What I find most helpful is the ability of the virtual patient advocate to drill down to patients' concerns," says Lynn Schipelliti, RN, who is participating in the study at BMC. "I see it as an asset to healthcare. Patients are going home sicker and have multiple medical problems to manage at home. Talking about discharge when patients are admitted allows them time to think beyond today and what it will be like when they go home."
Ultimately, it is the human-like nature of the virtual patient advocate that seems to increase the effectiveness and understanding of patients' care plans.
"The virtual patient advocate is patient and kind, shows empathy and humor, has medical knowledge, and shows confidence. All of these things make her believable and trustworthy to patients," Schipelliti says. "They appreciate the private time they are allowed to listen and ask questions—and that they decide when they have had enough."
When the uninitiated think of electronic health record implementations, they focus on build and rollout. Most likely, the implementation is considered an "IT project," and the communication machine starts rolling just before staff members are affected. However, the initiated know that EHR implementations—successful ones, that is—are process, workflow, and operational in nature. They are considered operational improvement projects with a healthy dose of change management, and communication begins when the decision to move to an EHR is made.
With the American Recovery and Reinvestment Act's HITECH incentives, healthcare organizations are being urged to roll out EHRs and use them in a "meaningful" way. The following are three areas that often get the short shrift during an EHR implementation, but they are as critical to success as the functionality itself.
Communication. One of the first steps in an EHR implementation is to carefully create a communication plan that focuses on all classes of end users. The message should address the benefits of the new system's functionality, as well as, the changes that will occur post-implementation to people's everyday workflow. From implementation experience at academic medical centers, ambulatory facilities, and community hospitals, my colleagues and I have identified the need to better prepare end users for the effects on their daily processes.
The learning and change process begins with these early communications. In addition to the "training" concept inherent in it, early adoption questions can surface that may alter the build and the training program. In addition to end users, leadership and the project team require early and frequent knowledge. You can use e-demos and training materials based on actual scenarios to help assimilate everyone involved to the new environment.
"IT" project vs. "operational improvement" project. It is a common mistake to label these types of implementations as IT projects. Regardless of whether it's an EHR, PACs, laboratory, or other system, the purpose of the implementation is to improve operations. Even though the IT department provides technical direction and support, the owners are the users. Early communication and adoption of this concept results in a more engaged user base, and the implementation feels more like a cross-functional team effort. This heightened participation leads to input that makes the system configuration and optimization more relevant to users. It should also eventually improve patient satisfaction, as end users across the organization represent various aspects of the patient experience.
EHR training approach. All too often, the topic of training is an afterthought. Yet if people aren't adequately trained, an EHR implementation can fail. By failure, I mean that delays mount along with frustration, and inversely, patient satisfaction plummets. Although it's understood that training is required, early focus is on acquisition, build, configuration, technology, and implementation. Therefore, the training team is usually not identified until later in the process and thus not involved in the build and configuration processes. Early involvement with the project team allows trainers to:
Have adequate time to create a good curriculum based on the build and configuration requirements.
Develop an understanding of operations and workflows that should be integrated into training scenarios.
Develop relationships with application and user teams to enhance communication and response time regarding issues, questions, and resolutions.
The following are some suggestions for how to approach the training aspects of an EHR implementation.
Create a training plan that clearly communicates the vision, mission, and approach. The plan should identify the organization's approach to staffing, curriculum, and process based on an assessment of the technology, environments, and audiences involved. This document will be used to obtain support during early communication sessions and with the organization's leadership.
Obtain leadership support for your training strategy. The training mission, vision, and delivery strategy should be reviewed and approved by administrative and physician leadership. It is important to communicate the critical training requirements to physicians, clinical care providers and general staff. For example, training may be optional for some and mandatory for others. Or, access to the new system will be granted only to those passing a competency assessment. My clients have found that a best practice is to tie system access to passing a competency assessment. Executive buy-in is absolutely critical for this to be successful, however. It should also be communicated by executive leadership versus the project team.
Staff up. Identify potential trainers early in the process. Look for people with good communication skills at all levels, experience in training environments—preferably with a system implementation, and other characteristics like patience. Don't wait. Identify trainers on staff or hire outside trainers during the initial phases of the project to begin involving them in all activities, including design, build, validation, and any re-engineering processes sessions. Training in and of itself is a significant endeavor.
Determine training methodology. There are many questions to be answered, such as: Will there be a dedicated training team? Will a train-the-trainer approach be used? How will the role of super user be defined? Will training be instructor-led or computer-based? How much can a user learn about the process and the software application in a short formal classroom session? Based on my experience, the approach most often used is a blend of instructor-led and computer-based training. Here are some suggestions:
Don't rely entirely on computer-based training. Given the number of staff who may not be very computer literate, it will not be the most effective way for them to learn.
Combine computer-based training, instructor-led training, hands-on exercises, and practice time. This blended approach should meet most of the diverse end-user characteristics.
For end users who need additional training, computer-based training is a good option.
Create curricula based on the design, build, and validation processes that focus on the specific goals or requirements of the role and the associated workflow that individuals need to know.
Cover computer basics. Early in the implementation process, evaluate end users' computer knowledge. The quickest way to frustrate staff is to give them tools that they cannot use or understand. Provide basic skills to those who need them to shorten class time during training. Our experience with clients has shown this strategy should lead to better adoption rates, a clearer understanding of application flow and workflow, and increased use of the system.
Communicate changes. In the training communication plan, include a process for distributing updated training materials to staff. This includes specific updates for managers and super users.
Set up post-training environment. You will most likely have users with a variety of unique training needs. Therefore, it's important that the communication plan clearly identifies the post-training requirements for all users. For example, a "playground" or test environment is a great way to enable users to practice without worrying about making mistakes.
Transfer knowledge. Training is a collaborative process. Trainers will need to work closely with physicians, nurses, technologists, financial staff, and other members of the healthcare organization to successfully develop and deliver training. The objective of training is to transfer knowledge to end users so that they can enhance the quality of patient care. The EHR is one tool to do this. The other tools that are just as important are processes and workflows that complement the technology. Although the initial training sessions will focus on application functions, the actual "learning" and optimization of the applications will come after the training, through use. That is why post-implementation support is so important. Whether it's the training "playground" or additional classes, users will have more questions once they start using the system and living in the new workflows. Processes may need to be fine-tuned; therefore, trainers and technical staff should be readily available post go-live.
These three areas: communication, classifying implementation projects, and end-user training can create a foundation for the technology to succeed. Some mistake them as "fluff," but that is a costly mistake.
Rob Drewniak is a consultant with Hayes Management Consulting in Newton Center, MA, and the former senior vice president of clinical resources at Glendale Memorial Hospital. He can be contacted atrdrewniak@hayesmanagement.comor visit www.hayesmanagement.comfor more information.