Home  |   Subscribe  |   Resources  |   Reprints  |   Writers' Guidelines

July 19, 2010

Proposed HIT Patient Safety Program Receives Mixed Reviews
By Elizabeth S. Roop
For The Record
Vol. 22 No. 13 P. 20

Industry insiders point out potential flaws in a federal plan to develop a national program that would monitor patient safety problems in HIT systems.

The HIT Policy Committee’s call to establish a national program to monitor patient safety issues stemming from HIT system use has been met with mixed reactions from various industry sectors. While there is agreement that some sort of safety tracking system is a necessary extension of the broader adoption of HIT, there are grave concerns about the validity of the data and how they would be maintained and utilized.

Though the details have yet to be hammered out, in April the HIT Policy Committee recommended that the Office of the National Coordinator (ONC) create a national transparent database to which providers would report errors and safety issues arising from the use of EMRs, computerized physician order entry (CPOE) systems, and other forms of HIT. Reports would cover all HIT systems, encompass multiple factors, including usability, processes, and training, and would be submitted confidentially without liability.

The recommendation gave rise to a number of questions, the most prominent of which is whether such a system is even needed. The answer for many is yes, as long as it is designed to collect data on true IT errors and is used to improve rather than punish.

“I don’t think it’s a good idea for the industry in question to say ‘no, I don’t need it’ when we know that there are incidents. … Part of it is perception. To say there isn’t a problem is not realistic, so how can anyone realistically object to it?” says Bill Spooner, chief information officer of Sharp HealthCare, a San Diego-based regional healthcare delivery system that includes four acute-care hospitals, three specialty hospitals, two medical groups, and a health plan.

Noting that the proposed system could be similar to the Aviation Safety Information Analysis and Sharing (ASIAS) system that, under the auspices of the Federal Aviation Administration in collaboration with airlines, collects and analyzes safety and incident data, Spooner says success or failure is likely to depend on how the data are ultimately utilized.

 “I’d like to really understand how the airline industry uses [the ASIAS system] to engage in improvement,” he says. “To me, the danger is if [the data] is used in an overreactive manner. If what [the airline industry] is saying is accurate, they use this to improve their own practices and it isn’t displayed on the front page of The New York Times every day. It would not be a good thing if this information was used in that manner. But if it is used in a transparent manner and used constructively, not punitively, it could be good.”

A Grab Bag of Challenges
Spooner notes that the ASIAS system has successfully introduced transparency and enhanced air travel safety. However, there are enough significant differences between the two industries that lead him to question whether the ASIAS system’s success can be duplicated in healthcare. Unlike the airline industry, which encompasses a small number of carriers and manufacturers, a healthcare version of the ASIAS system would need to track data from thousands of provider organizations and types and hundreds of HIT systems.

“How do you communicate these problems in a way that they can be addressed? It’s a lot easier if you take an issue and let four or five airlines pick out problems and see what needs to be addressed,” says Spooner. “There is also always the challenge of separating the root cause between true software defects and process defects or software incompatibility. We know that change management is an issue, and we’ve had a lot of issues over the years throughout the industry about implementing software but not rethinking workflow to correspond to the software. You have to be able to decide whether it was in fact vendor X and a true defect or something I did in the way I implemented vendor X’s product.”

Shane Hade, CEO of emergency department information management systems provider EDIMS, concurs that a program to monitor and analyze patient safety issues with HIT is a good idea. While it is unknown how many adverse events are directly caused by HIT, he notes that as a general approach to mitigating risk, the ability to gather and evaluate technology-related safety issues is an important step as adoption expands. But he also agrees that how the data are gathered and utilized presents significant challenges that, left unresolved, could negatively impact compliance with such a program.

“How do you determine if an adverse event is due to the use of technology? That area is subject to interpretation, and it could be very damaging,” he notes. “I also don’t know how you would police [reporting]. It’s going to be very problematic. That one issue could torpedo this entire thing.

“Vendors will also be concerned about how to account for or necessarily determine issues that were caused by the hospital customizing away from the normal functions of a system,” Hade adds.

Though the HIT Policy Committee intentionally focused its recommendations on outcomes rather than design or implementation, it did endorse the idea of integrating a “feedback button” into user interfaces. The belief was that such a button would streamline the process by enabling the immediate submission of error reports and eliminating the need for providers to identify the appropriate agencies to which the report should be submitted. It is an idea that is unlikely to win much support on the provider or vendor sides of the equation. Spooner notes that such a button would be “about as effective as when I’m in heavy traffic and hit my horn.”

He adds, “I think it would really be a shame if a system were created which encouraged people to rampantly report potential safety issues because you’d have a lot of false alarms, and people would be spending a lot of time investigating those false alarms instead of investigating process improvements.”

Hade also sees serious compliance challenges with such a reporting system. First, integrating the reporting mechanism directly into the care process heightens the risk that technology will be blamed for human error, especially if providers are not completely convinced they will be protected from liability by participating. Second, stopping to report an error moves the clinician’s focus off the patient, disrupting the clinical workflow.

“Part of this will be bound by who has access to this information. … If information is very controlled in terms of use, you’ll have a higher level of compliance as opposed to accusatory [reporting],” says Hade. “It is also a consideration toward the accuracy of the information because if it’s not reported at the point of incident, you’re relying on someone’s memory. But in the moment, people are more concerned with dealing with the issue rather than reporting on it.”

This dovetails with a third challenge facing the successful establishment of a patient safety program: Who should be responsible for oversight and ensuring that the data collected are both valid and meaningful in the larger context of improving overall safety?

Keith J. Figlioli, senior vice president of healthcare informatics for Premier, Inc, a performance-improvement alliance of more than 2,300 U.S. hospitals and 67,000 other healthcare sites, notes that patient safety laws were enacted long before the concept of meaningful HIT use came into being. So it is important that any HIT safety-monitoring program contribute to, rather than cloud, the bigger picture.

“It’s a classic stance where we get very married to going down a path and someone doesn’t look at the other laws. Patient safety organizations are still trying to figure this out,” he says. “Technology is one widget. Patient safety is a much bigger umbrella. How do you make heads or tails if you’re only assessing CPOE or components of an EMR when you might have things from a process perspective that aren’t even being taken into consideration?”

To that end, Figlioli recommends that oversight be handled by an existing patient safety organization that is in a position to draw parallels between errors attributed to IT and errors attributed to other aspects of the care process.

“I believe in this concept because I think there are going to be issues. The problem is that you could end up reporting on and making assumptions about things that the outcomes [of which] might not be linear,” he says. “The bigger question for all of us around patient safety is how it can be managed in the macro sense.”

A Thoughtful Approach
For its part, the ONC agrees that whatever form the final program takes, it should advance the goals of improving access to and the delivery of care by leveraging safe and efficient technology use.

In the months since the HIT Policy Committee sent its recommendations to the ONC, the agency has been working with other relevant Health and Human Services agencies to evaluate the recommendations and develop implementation strategies. Much of the early focus is on information gathering, including evaluating whether the Agency for Healthcare Research and Quality, the FDA, or another existing safety organization is best suited to manage it.

“We’re open-minded, but we’re definitely looking at where we can leverage existing approaches,” says Jodi Daniel, JD, MPH, director of the ONC’s Office of Policy and Planning.

For now, there isn’t a set timetable for when the agency will issue its preliminary strategies, although the hope is that it will happen within a few months. Meanwhile, “We’re looking carefully and taking time to go through and come up with an approach,” Daniel says.

“We are taking the issue seriously and putting a priority on safety issues in the context of all the other work we’re doing in health IT,” she adds. “Generally, we believe, and there are studies to suggest, that HIT can improve quality of care and safety. What we want to do is make sure that if there are any issues that are raised by technology that we identify them quickly and are responsible for making sure that we have electronic health record technology that provides the best care for patients and provides a safe and trusted way to assist in the delivery of care.”

— Elizabeth S. Roop is a Tampa, Fla.-based freelance writer specializing in healthcare and HIT.