Now that I'm done with my HIMSS07 wrap up piece for MX magazine, I'm going to be posting on items that didn't make it into the story. (FYI - here's last year's HIMSS report.) For obvious reasons magazines don't want to publish content that's already been published here. This first post reports on a presentation by Julian Goldman at HIMSS 2007. Much of his talk was about the OR of the future program at CIMIT, but what really grabbed me was what he had to say about the dearth of medical device interoperability and the resulting impact on patient safety.
In his presentation, Goldman succinctly described many of the challenges resulting from the confluence of HIT and medical devices at the point of care – and the potential impact on patient safety. In his “Views From the Top” presentation (pdf), Goldman used research done at the Massachusetts General Hospital OR of the Future and elsewhere to explore interoperability, “error-resistant” systems, and patient safety.
The frustration for providers is the fact that caregivers represent the only point in which pertinent patient information is accumulated, processed and correlated to create the contextual awareness necessary to provide nursing vigilance and deliver safe and effective therapy. The creation of this contextual awareness is a manual process done in a frequently chaotic and interrupt driven environment. Dependence on these manual processes combined with over crowding, high nurse to patient ratios and increasing patient acuity in general care areas, represents a systemic failure in health care. It is no wonder that failure to rescue remains the most common patient safety incident at 134 per 1,000 at risk admissions.
According to Goldman,
“The acute care endgame is to automate the workflow that provides contextual awareness through improved communications and by linking the various medical devices connected to the patient to clinical information like orders and diagnostic results. In an interoperable environment, this data can then provide safety interlocks to generate smart alarms and to ensure or interrupt harmful therapies.”
Goldman provided 5 examples of simple safety interlocks between devices or between devices and information systems that would save patient lives but don’t exist. Goldman’s example that results in perhaps the greatest number of adverse events is the patient-controlled analgesia (PCA) pump. When PCA pumps are used in place of nurse administered analgesics, the chance for patient harm is increased more than 3.5 times. Interoperability between a continuous patient monitor and the PCA pump could detect a decline in the patient’s condition, lock-out additional pain medication infusion, and generate an alarm. The adverse events that result from standalone PCA pumps represent system failures rather than a lack of alarms or inadequate vigilance on the part of clinicians.
Despite broad interoperability in other vertical markets like manufacturing, telecommunications, IT, and consumer electronics, health care has yet to move past proprietary end-to-end solutions. Creating “one-off” solutions is not practical for vendors, systems integrators or hospitals. “Patient and clinician interaction with medical devices has not received the same attention from vendors as medical record-related data communications,” according to Goldman.
The patient safety market requirement is to automatically create context for decision support using data from devices and information systems. Not surprisingly, HIT vendors, start-ups and smaller device companies view patient safety and the point of care as important product differentiators in an attractive and untapped market. Established device vendors must develop business strategies and products that meet these long standing patient safety market requirements, or they will face declining market share.
Goldman dropped another very interesting nugget of information during his presentation. Using current technology, simple connectivity to integrate medical device data into EMRs is prohibitively expensive, representing 40% of the total cost of ownership of medical devices. This is the first time I've heard this cost quantified, and it represents a huge hidden cost to providers. Medical device connectivity, systems integration, management and support is something device vendors don't want to provide; third party vendors are way expensive, and is a substantial burden for internal resources. Large providers are starting to include a "connectivity clause" in their standard purchase agreements putting the device vendor squarely on the hook for delivering fully implemented connectivity. Of course the lack of any standards in this area makes things more costly and complex for everyone. It has been estimated that an interoperability standard would save a health care delivery organization like Kaiser Permanente an estimated $40 million annually for 10 years - that's some serious money.
Pictured right is a view of the OR of the Future at MGH.