Electronic Medical Records in Anesthesia: Who Will Benefit and How?

The concept of automating the anesthesia record has intrigued academicians for decades. Some of the early versions of software packages for computer-based record keeping were quite complete and impressive despite the technology's limitations during the 1980s and 1990s. A number of companies made significant investments in the various software options. There was only one problem: they simply were not selling.

While the proponents' arguments could be compelling, there was no demand. The solution was way ahead of its time, and it would take a combination of regulatory initiatives and market changes for the time to be right. Even now that market conditions are more favorable, implementation continues to be a slow and ponderous process. Understanding why this is sheds light on the fundamental challenges of implementing any meaningful change in the American healthcare system.

It is not hard to understand what motivated the early innovators. Anesthesia record keeping has remained an arcane and manual process that relies on the provider to capture and record a wide array of data trails. The first paper records were developed in the 1890s and remain little-changed. Unlike the surgeon's operative report, which provides a narrative description of the surgery, the anesthesia record is designed to document not only what was done to a patient during the surgical experience but also how the patient responded. Even a cursory review of a completed record makes one wonder if there is a meaningful correlation between what actually happened in the operating room and what got documented. Not only is much of the information filled in after the fact from memory, but as the writer of the history of the case, the anesthesia provider will tend to write the story based on how it ended and discount any adverse events or inconsistencies that may have occurred during the case.

Not only are anesthesia records an imperfect form of record keeping of a reasonably complex interaction of surgical procedure and pharmacological management, but they are also a poor tool for research and analysis. The details and data elements are captured in a format that is impossible to review except by means of a tedious manual audit. If one wanted to compare aspects of case management for a given surgical procedure one would have to identify, pull and review a sample of records prepared by different providers, which would prove incredibly time-consuming and which would probably reveal more inconsistencies in record keeping than in clinical care. Not only is most physician handwriting nearly illegible, but often what is being reviewed are carbon copies which are even harder to interpret.

As the anesthesia record has evolved, it has become a significant document on a number of levels. Initially designed as a medical record and communication tool, it has become a medical-legal document that may be used to demonstrate that standards of care were met in the care of patients who may have had adverse outcomes. As the specialty has striven to improve the quality of care provided, the details of the anesthesia record have come to be useful for a variety of quality assurance projects. Increasingly, the template is intended to ensure that critical details of intraoperative management are consistently monitored and recorded. The American Society of Anesthesiologists (ASA) has played a key role in defining the categories of information that should be documented for each case. In many ways the anesthesia record is a reflection of the evolution of the specialty.

What the early developers of anesthesia record-keeping software saw was a significant opportunity to use technology to accomplish what they believed were three notable shortcomings of standard anesthesia record keeping:

• Lack of consistency and legibility in the way the anesthetic experience was being documented;
• Inconsistency in the way in which physiological data was captured and recorded; and
• A general inability to review and research clinical trends.

It was not until the 1980s that the first automated anesthesia record systems appeared on the market. One of the first was CompuRecord, which is still available today from Philips. Computer technology provided a ready solution to what many perceived as a significant problem. A computer could do something the human brain could not: integrate a significant amount of data into a clearly legible format that could be saved in a digital format.

The creation of the actual record starts with an extract from the scheduling software that includes the patient name, date of service, scheduled procedure and preoperative diagnosis. This basic information is then supplemented with notes from the preanesthesia assessment and the anesthesia plan. Details of the case are captured in real-time via digital interface to the monitors. Buttons prompt the provider for standard services, which make the record keeping much easier and more consistent than a paper record. At the end of the case, the provider has an opportunity to review the details before closing the record. The result should be a true and accurate record. As computer technology improved, more features kept being added to make the input easier and the output more complete. The inclusion of touchscreen technology represented a huge leap forward.

One of the main features of these automated records was their ability to facilitate compliance with arcane billing regulations. The need to document that an interscalene block was performed for the purpose of postoperative pain and at the request of the surgeon could be achieved with the click of a button. Such records also held the potential to allow providers to make critical coding decisions at the time of service. Most systems included a combination of hard and soft stops to remind providers of critical pieces of information.

The concept appealed to a wide variety of practices but the reality did not live up to providers' expectations. Interfacing with hospital information systems to capture the demographic and scheduling details often proved problematic. There were similar problems capturing the digital output from the monitors. Databases were not always easy to manage and mine.

For many years the development of automated anesthesia systems was a cottage industry that attracted anesthesia providers with a particular interest in software development. Some of the systems were adopted by companies, such as Philips, that sold anesthesia machines, but many were essentially homegrown, stand-alone systems. Most anesthesia providers recognized the paper record's limitations, but there did not seem to be a viable alternative, and most anesthesia providers tended to dismiss the limited offerings out of hand. However logical and attractive an automated record might seem, the challenge of creating a system that would work in every operating room to capture all anesthetics was overwhelming.

An informal poll of anesthesia providers would have identified three main concerns. The potential for artifact in records was a major concern. Suppose a surgeon were to lean on the blood pressure cuff during the case. The resulting uptick in blood pressure would be ignored in the manual record but be captured by the automated record. The resulting record might indicate a situation to which the anesthesia provider should have responded. Imagine if there were an adverse outcome. Artifact in the record might be difficult to explain.

The flipside of this argument suggested that since most manual records are not true reflections of what happened during the case, they would make automated records look problematic. Anesthesia had enjoyed the power of the pen in its record keeping for so long that there was considerable resistance to losing control. Even though most programs allowed the provider to modify or edit the data displayed on the record, this argument continued to be a popular theme of resistance to automation.

For those inclined to buy into conspiracy theories, the debate over automated anesthesia records was fertile ground. What was the point of capturing all the details of each anesthetic if not to impose clinical standards on individual providers? Anesthesiologists were quick to point out that anesthesia is a specialty that does not lend itself to "cookbook medicine."

Meanwhile, the biggest obstacle to automated records was the price tag. A typical system could cost as much as $40,000 per anesthetizing location, which was far beyond the budget of most anesthesia practices. Ultimately, then, the decision lay in the hands of the hospital, which also had little reason to make such an investment.

And so for years, the automated anesthesia record industry sputtered along. A number of companies would bring impressive offerings to the ASA annual meeting each year, but when asked, they all had to admit there were very few working installations. Automating the anesthesia record was a concept whose time had simply not come.

There is a popular saying that sometimes the way to solve a problem is to create a bigger problem. This was the case with anesthesia record keeping. The solution to the challenge of financing the implementation of a full suite of automated anesthesia records came only when hospitals had to implement electronic medical records (EMRs) for all services. As one might have expected, even though the initial offerings of the enterprise solution companies were not the ones preferred by anesthesia providers, the more systems that companies like Epic and Cerner installed, the better their systems got.

Legislation laying the groundwork for the encouragement of the use of EMRs can be traced back to President Clinton's signing of the Kennedy-Kassenbaum bill into law in August 1996, a bill that would subsequently be referred to as the Health Insurance Portability and Accountability Act (HIPAA). This legislation laid the foundation for standards for the transmission of healthcare data and created a framework for the protection of protected healthcare information (PHI). These were necessary first steps on the path to a world where details of a patient's care could be safely captured and transferred via the worldwide web.

In June 2004, President George W. Bush signed an executive order to provide federal leadership and national implementation of an interoperative EMR system by 2014. His order established the Office of the National Coordinator for Health Information (ONC) to lead efforts in health information technology (HIT). These actions clearly envisioned an evolution of what was primarily paper-based medical record keeping to an electronic format. While a variety of systems were being worked on, the most notable of which was being developed by the Veterans Administration (VA), this order clearly envisioned being able to develop national standards so that various systems could communicate with each other.

Proponents of this process argued that the creation of a national database of healthcare data would create a system that would allow patients to have better access to their own healthcare information and participate in the management of their own care, as well as:

• Improve the quality of care by reducing unnecessary and inappropriate testing and errors;
• Reduce clinical paperwork;
• Give healthcare providers better access to patient chart data in a standard format;
• Improve record keeping for more efficient and accurate billing; and
• Drive down the cost of healthcare in the United States.

Because it was clearly understood that the implementation of the necessary technology would require an investment of both money and resources, a series of incentives were envisioned to encourage the meaningful use of HIT. Over time, the use of the term "meaningful use" took on a life of its own to describe the various phases and aspects of the transition to an era of electronic medical record keeping in healthcare.

The next step in the process came in 2009 when President Barack Obama signed the Health Information Technology for Economic and Clinical Health Act (HITECH) into law. HITECH authorized the ONC to develop HIT to facilitate EMR implementation. The law also envisioned providing grants and loans to healthcare providers, hospitals and other healthcare entities to implement HIT. Most providers in anesthesia have been actively responding to the various phases of implementation related to this. They know them as PQRS, meaningful use, QCDR and MACRA, all distinct programs to encourage HIT adoption.

Navigating the changing guidelines and regulations has been no small feat. Most practices have had to expend considerable time and effort to comply. Increasingly, each new program provides financial incentives and penalties for compliance and noncompliance. It is safe to say that when one looks at the details of each year's plan, to the extent that it is actually defined in the regulations, it is easy to lose sight of the intent of the law.

One of the fundamental challenges of implementing any change in the American healthcare economy is human behavior. There is an inherent inertia in healthcare protocols. Change tends to come slowly. Much of what we are seeing in the unfolding of the meaningful use strategy is public policymakers playing chess with healthcare providers. However, meaningful use started long before "Obamacare" and will, no doubt, persist well into the next administration.

As a result of these developments, the implementation of anesthesia EMRs has been gaining momentum. Careful observers of this process, such as Anesthesia Business Consultants (ABC) (which is currently working with dozens of clients who have either just implemented an anesthesia EMR or who are in the process of doing so) note every implementation process has to overcome and address two common misconceptions about EMRs.

When a hospital or anesthesia practice selects a system from a qualified vendor, such as Epic or Cerner, two of the most common vendors, there is a tendency to think that such well-established firms have developed a comprehensive solution that can easily accommodate the average anesthesia practice. This is simply not the case. Despite the systems' features and functionality, few installations do not involve considerable modification. Nowhere is this more evident than in the implementation of anesthesia EMRs. Knowing this need for customization, ABC developed a dedicated team of EMR specialists to guide clients through the implementation process. The team's mantra: "If you have seen one Epic/Cerner implementation, you have seen one Epic/Cerner implementation."

Anesthesia practices have a preconceived notion that the EMR will simplify record keeping and streamline the submission of charge data to the billing company. Vendors often tout the potential for a direct electronic interface, but the reality is otherwise. Virtually no anesthesia practice with an EMR is able to rely on an electronic interface to transmit all cases from clinical venue to billing system. Most coding for practices using EMRs is done based on a manual review of a digital pdf of the anesthesia record. The notion that an EMR will expedite charge submission is a fallacy. The implementation of most Epic anesthesia systems, for example, requires a three- or four-day hold on charges to ensure completeness of batches.

The selection process complicates these basic challenges. Few anesthesia practices actually choose the system they will be forced to accept. In most situations, the facility picks the system, and the facility's goals and objectives rarely align with those of the anesthesia practice. What most facilities want is an enterprise-wide solution that can accommodate all medical specialties and services; the anesthesia record option is often little more than an afterthought.

This is why the implementation process for anesthesia is so challenging—and so critical for success. Any anesthesia practice contemplating the implementation of an EMR should ask the following questions:

• To what extent will the proposed solution accommodate all aspects of the practice's services?
• How will providers get trained and what kind of support will they receive?
• How will the interface to the hospital and its scheduling program work?
• Will the provider have the ability to readily edit the details of the surgical procedure and the postoperative diagnosis?
• How will rules for hard and soft stops in record keeping be defined?
• What is the process to confirm that all cases are captured and transmitted to the billing office?
• Who will be responsible for assigning billing codes, the provider at the site of service or the coder in the billing office?
• Who owns the anesthesia data and how easy will it be for the department to mine its clinical database?
• How will requests for software modifications be handled?

This last question can be a major issue. When a hospital picks a system and enters into a contract with the vendor, it is usually based on a limited budget. The anesthesia department's functional requirements must be agreed to up front. Anesthesia must be involved early and interact regularly with the developer and vendor. To facilitate this process, the ABC EMR team has developed an extensive list of features and functionalities essential for most practices.

Clearly, the time has come for EMRs in anesthesia. Before long, any practice that has not implemented an EMR will be considered a dinosaur. As the percentage of anesthesia practices with automated records increases, the state of the industry improves. Each version of each system offers more user-friendly features and functionality. There is no doubt about it: just as banks did away with tellers for most transactions, so too most practices are slowly eliminating the paper anesthesia record. An anesthesia resident graduating now may never complete a paper record.

A review of the initial objectives of meaningful use yields a rather disappointing scorecard. It is not clear that the implementation of EMRs in anesthesia has had any meaningful impact on the quality of care provided, which has been impressive for years. Maybe some facilities have made improvements in the consistency of care. Probably, though, the real impact of EMRs in anesthesia will not be measured in improvements in the quality of care, but rather, in the establishment of quality benchmarks. How much can we reduce the cost without impacting the quality?

EMRs were supposed to reduce clinical paperwork. By this measure, EMRs have been a success, because they capture considerably more data than was ever captured in a paper environment. But meaningful use has also forced providers to worry about and document many more data elements than they ever had to before.

There is no question that EMRs provide greater access to details of a patient's medical history. This is clearly one of the biggest selling points. Details of previous surgeries, evaluations and test data are now readily available to anesthesia providers online. It is in the areas of record keeping and billing accuracy and efficiency that EMRs have failed to meet expectations. To calculate billing charges, anesthesia coders now routinely review 20-page pdfs rather than one-page paper anesthesia records. Integrating arcane billing rules into the EMR has been almost unresolvable. Additional days typically have to be added in the charge capture process to ensure that nothing gets missed.

All of which raises the question of whether EMRS have had any impact on healthcare costs. So far, there is little evidence that they have. In fact, the evidence suggests they have simply added a layer of expense to an already expensive system.

And so we are left wondering what has been accomplished. So many practices now have state-of-the-art record keeping but what has it gotten them? It is like having more than 100 cable channels but nothing worth watching. These systems are capturing a lot of data but there does not appear to be any more knowledge. All the expense and time that has been invested to get the specialty to this level of automation is probably just the end of the implementation phase. Now that the specialty has its new tools, it has to figure out how to use them effectively to achieve the original goals of the initiative.

There is a business saying that you cannot manage what you do not measure. However imperfect they may be, EMRs have provided much better and more complete management tools. They provide a mechanism to capture more data more efficiently than ever before. The questions now are what to do with this data, how to validate it and how to analyze and apply it. Much work remains to be done to make meaningful use truly meaningful.