In today's era of health care, documentation has become almost as central to patient care as the actual care itself. As far as many credentialing and reimbursement entities are concerned, if it is not documented, it was not done. Unfortunately, these expanding demands for daily medical documentation have led to excessive busywork that frustrates providers in their attempts to achieve the increasingly unrealistic goal of "100% compliance." More importantly, these demands compete with providers' time spent in the actual care of their patients. We absolutely must design novel and sensible ways of automating the documentation process.

Documentation Demands

As an example of a poorly met documentation challenge in the intensive care unit (ICU), consider the series of interventions targeted at reducing ventilator-associated pneumonia (VAP). Few experts would argue over the importance of complying with these preventive measures, yet even in high-performing medical centers with incentivized protocols, documented compliance with these best practice measures often falls short of targets. For example, a key element of the VAP prevention bundle, keeping the head-of-bed incline greater than 30[degrees], routinely improves with education and engagement of staff, but can still fall as low as 60% to 70%,1 even when a device is used to continuously record head-of-bed incline and alert providers when this incline drops to less than 28[degrees].2 To make matters worse, even when documented compliance with one or more separate elements of the VAP bundle is maximized through intensive training, comprehensive compliance with the entire bundle falls short of the target. In one study, compliance with head-of-bed elevation climbed to more than 95% following the institution of nurse and physician champions, educational sessions and materials, bedside cues, engagement posters, and regular performance feedback. Yet compliance with all elements of the entire VAP bundle was still only 70%.3 Many possible reasons for this shortcoming include the more appropriately prioritized and time-critical tasks of providing care, which often trump tedious charting and documentation, even though the latter is vital for auditing performance.

Given the current workload imposed on health care providers today, how realistic is it to expect them to achieve 100% compliance on the ever-growing list of items they must document, particularly when even the most dedicated and conscientious of our professionals do not always believe that it is a judicious use of their time? Solving this problem requires more than simply saying, "We'll try harder," and it is certainly not going to be solved by thinking we are all going to someday magically "get on board" to improve our compliance to 100%. Even the best-intentioned and hardest workers are flawed simply because they are human. Being human, however, is not a problem; it is a statement of fact and one that we must acknowledge and move beyond. The real problem at hand is our failure to properly use existing technology.

Lack of Device Interoperability

In today's era of advanced technology, with digital wireless communication and intuitive user interfaces, it is inexcusable that most important devices used in our ICUs, often designed by several different proprietors, have no standard means of communicating with one another, with our electronic medical records (EMRs), or even in an efficient manner with health care providers themselves.4 This lack of interoperability not only creates unnecessary tedium for personnel, it becomes a threat to patient safety.5 Imagine going to the grocery store today to find that the cashier, instead of swiping each item across a bar code scanner, takes out a pencil and ledger, writes down the price of each item, and then uses longhand addition to calculate the bill. Such a scenario sounds unthinkable to us today because it would waste so much time, as well as introduce an unacceptable allowance for human error. Yet this is what we are asking our physicians and nurses to do every day in the ICU. We manually determine each patient's weight and height, use a calculator (or possibly a smartphone app) to estimate ideal body weight (IBW), and then calculate the respiratory tidal volume on the basis of 6 mL/kg IBW before plugging it into the ventilator. How many times have these steps led to human error, resulting in deviation from best practice in a patient with sepsis or acute respiratory distress syndrome? Imagine if the patient's bed could report weight and height directly to the ventilator so that the ventilator could then automatically determine and deliver the 6 mL/kg IBW tidal volume. Imagine this same bed providing a regular update to the EMR on the incline of the patient's head and upper torso. Imagine software in the pharmacy dispensary automatically telling the EMR every time chlorhexidine kits are checked out and used for oral hygiene. Imagine intravenous infusion pumps and ventilators, respectively, telling the EMR when sedative agents are turned off and when the patient is placed on a spontaneous breathing trial as a "wake and wean" event. The travesty in all of this conjecture is that many providers have already dreamed of these advances, and the technology has long existed to make them possible. Unfortunately, the very people who recognize the need and the market for these advances, we the providers, are not driving the technological design of our own tools and instruments.

Many would argue that the fundamental problem is our own failure, as professionals, to put sufficient pressure on the medical equipment industry to provide greater value by having better usability and interoperability.4,5User interface design refers to a process that focuses on simplifying a product's user interface to most efficiently achieve the user's goals, a process that is desperately needed in the ICU. Interoperability refers to "the ability for medical devices to exchange information with each other and with patient data repositories such as electronic health records."6^(p2) As things currently stand, each ICU device appears to have been designed in isolation and later thrown into the hospital setting and expected to do its job efficiently in collaboration with other devices-devices with which it cannot communicate. This is not the path to operational efficiency; it is the path to chaos. If health care providers had given sufficient input into the design of today's instruments, we would almost certainly already have a wireless inclinometer on every bed that continuously checks head-of-bed incline and records it in the EMR, thereby providing a reliable index for auditing performance, relieving frustration for nurses, and most importantly allowing more time for nurses to be where they belong-with their patients.

Examples of Patient Safety Initiatives

Fortunately, institutions such as the Armstrong Institute for Patient Safety and Quality at Johns Hopkins University have begun to tackle this problem. Their initiative, Project Emerge, seeks to improve patient safety and care through innovative ICU redesign.6 The goal is to improve safety and efficiency by focusing on systems integration and collaborative design. The Institute hosts collaborative "innovation days," in which health care professionals, engineers, patients, and family members come together to brainstorm more efficient means of achieving high-quality care in the ICU (and hospital). Collaboration between health care providers and engineers supplements the process of device design with firsthand understanding of our goals and the obstacles that exist in our daily workflow. In fact, researchers at the nonprofit West Health Institute have estimated $30 billion in waste within the US health care system that could be mitigated with device interoperability alone, $2 billion of which would be attributable to avoidable adverse events, and $12.3 billion of which would be attributable to the skilled labor time currently wasted in manual documentation in the EMR.5

Call to Action

So what can we collectively do about the problem? One might argue that we should engage our local vendor sales representatives with questions such as, "Why doesn't this intravenous pump talk to our ventilator and EMR?" One would hope that the more often they hear these questions, the more likely they might be to pass them on to their superiors, or to speak up at their company's next strategic planning meeting. Those of us serving on program committees for our professional societies can lobby for "Innovation Workshops," such as those recently offered at the Annual Congress for the Society of Critical Care Medicine. We could invite vendors to attend these sessions so they might witness our frustration with the current lack of systems integration, and our oftentimes brilliant solutions to the problems. One would hope that forward-thinking product development specialists would be willing to listen and make the investment in product integration to get a head start on the future marketplace, eventually putting pressure on other developers to catch up in the systems integration "arms race."

Unfortunately, the reality is that hospitals and providers stand to reap an estimated 93% of the projected $30 billion in health care savings from moving toward greater device interoperability.5 Industry, on the contrary, has little to gain, so there could be a disincentive toward universal interoperability when trying to market a number of their own "interconnect-able" devices. Furthermore, most hospitals are equipped with devices from several different manufacturers, in various stages of their natural life span, and getting them all to speak to one another and the EMR through interoperability is no simple task. In fact, although many devices on the market are already equipped with similar standard serial interface ports that in theory can enhance connectivity, simply sending data from one device to the other is not enough. Each device must be able to understand and incorporate those data in a meaningful way. In other words, they need to either "speak the same language" or communicate through a converter or intelligent medical device hub.7,8 Fortunately, experts recognize this problem and are working on a single communications standard for medical devices, but this solution is still far beyond the immediate horizon.7 We should also consider ourselves fortunate that pioneering philanthropic organizations, such as the West Health Institute and the Patient Safety Movement, are currently lobbying on our behalf (and the behalf of our patients) for universal device connectivity platforms and interoperability.

At the same time, while doing what we can to collaborate with industry, we can also engage the leaders within our own respective health information and supply chain services to raise awareness of the need for device integration and petition them to make interoperability part of the standard terms and conditions for purchasing. This process will ultimately require executive leadership buy-in to make device interoperability a strategic goal for their organizations. In doing so, experts can collectively address all the device interfacing needed by the nursing staff, establish whether it is possible to interface each device with the EMR system, and determine the projected time and resources required to establish a comprehensive integration with the EMR. This will, of course, involve a determination whether the cost-benefit balance supports what is certain to be a time- and labor-intensive project, and in that process, we need to be there to remind them of the costs to the organization of wasted labor time, failed documentation audits, and avoidable medical errors. Whether it be through political pressure on industry for improved product development or engaging executive leadership for homegrown device-integration solutions, it is ultimately up to us, the users of this technology, to define the future market's demand, and to promote the adoption of systems integration. Whether tomorrow or 10 years from now, this is where the health care industry is moving, so wouldn't it be to our hospitals' benefit to get there early? Just think of the improvements not only in our own daily workflow, but more importantly in patient safety.

REFERENCES