Print

Creating a Clinical Database: Opening the Pandora’s Box or Mining the Treasure Trove

Aman Mahajan, MD
Chair, Department of Anesthesiology, University of California, Los Angeles, CA

Jody Locke, CPC
Vice President of Anesthesia and Pain Management Services, ABC

Background

It is often suggested that an anesthesiology department should have more complete and readily accessible data about the clinical care provided in the operating rooms and the delivery suite than any other department of the facility; but how often is this actually the case? Anesthesia providers review and document enormous amounts of clinical detail and critical events for every patient they see, but little of this information is actually captured in a way that allows for its logical indexing and retrieval. Most anesthesia groups and their billing services have been so focused on the data necessary to calculate a charge and generate a claim that they have virtually ignored what is potentially the most valuable of information of all. The implementation of electronic medical records is slowly inspiring a change in a very traditional way of thinking about the clinical details of the peri-operative continuum. Progress, however, is slow and there continues to be a eneral reluctance to invest time and resources in clinical databases. The Anesthesia Quality Institute (AQI) has made a start, but its database of 10 million cases represents only a small percentage of total U.S. surgical volume and thus far relatively few of these cases include any clinical outcomes data. Small-scale individual initiatives are being undertaken across the country but few groups or departments can truly claim a robust clinical database. This state of affairs raises some fundamental philosophical, economic and practical questions about the value of such data in the current health care environment where the purported focus of policy is on quality improvement, safety and cost containment.

Defining a Database and Its Purpose

The term “clinical database” has varying definitions depending on the community. What data elements would be most useful in the current debate? Today’s clinical documentation tends to include two types of information: specific events and ongoing data streams. If all the data streams from physiologic monitoring, drug administration and provider interactions were captured, each case would result in a massive file of questionable value, except to those seeking to answer clinical research questions. The key elements would be those potentially predictive of complications or that confirmed the outcome of care. In comparison, creating the database for billing and accounts receivable management would be considerably easier because the basic elements were defined in the form of the standard CMS 1500 claim form. Inevitably the requisite clinical elements have to be defined in reference to those already being captured for billing.

The standard anesthesia billing dataset includes patient demographics, date of service, surgical procedure, diagnosis, start and end time and the details of any incidental procedures performed. These bits of information are intended to define what services were provided and by whom. We call the current system of reimbursement fee-for-service because it pays providers for individual patient services. Although payor policies purport to identify unnecessary services, little attention is actually focused on the appropriateness of cost-effective care, especially in anesthesia. The diagnosis code is a case in point. ICD-9 is supposed to provide a reasonable justification for the services provided but today’s diagnosis codes focus only on the rationale for the surgical procedure and not the need for or mode of anesthesia. Their application to the specialty is little more than a vestigial inconvenience. There should be a code that justifies the administration of anesthetic care in terms of the pain and inconvenience, medical management and complexity of monitoring during the surgery.

If a provider indicates that a general anesthetic was administered, it is never questioned by a payor. The appropriateness of a MAC anesthetic for endoscopy or the use of a nerve block for post-operative pain management might be, but even here there is little consistency. What a curious state of affairs this is given the detailed pre-operative assessments documented for such a large percentage of surgical cases. When things go wrong this information is also usually documented, but rarely are the two ever correlated. The path is slowly being charted by a handful of visionary practices that hope to use the information to demonstrate their superior quality of care.

Groups wishing to make the investment in a relevant and useful clinical database must make some critical decisions up front. Information only becomes knowledge if it has some practical application. Data elements could logically be selected based on their utility to some or all of the following:

There is a tendency to look at projects like the development of a clinical data base as a daunting enterprise that will require expensive consultants, significant resources and the serious commitment of key members of the practice. While one should never minimize the significance of adding data elements to a database, the process need not be perceived or promoted as an insurmountable task. Consider the bits of information that are only captured haphazardly today, indicators such as ASA physical status and diagnosis. Too often these are only retained when they have an impact on charge calculation or payment. Databases should evolve logically and be driven by a set of reasonable and practical questions. The fact is that most anesthesia providers already have a good idea of the issues that merit monitoring and areas where potential improvements in care can be realized.

Building a Database: Challenges and Pitfalls

The peri-operative continuum offers us three clearly defined areas of investigation: pre-operative bservations, intra-operative events and post-operative complications. Because anesthesia risk factors have been so well studied over the years, reasonable and appropriate lists are available for each phase of care. Any of these would represent a good starting point for the formulation of a data capture strategy.

How would this additional data be captured? The good news is that computer memory continues to get cheaper and more powerful. For practices implementing automated anesthesia records, capturing additional data is easy. Others that are not so technologically advanced may want to consider including key elements from pre-operative assessment or intra-operative complication forms into their data entry process. Suffice it to say that with all the technological options available, data capture is not the practical obstacle that it used to be. There is more than ample evidence that practices that truly want to distinguish themselves in the market have found ways to consistently and cost-effectively capture the data they feel they need.

Too often the potential value of a robust clinical database is more of a theoretical proposition than a practical reality. Why is this? Standard arguments tend to fall into three broad categories: the philosophical, the economic and the practical. Considering the concerns one cannot help but wonder to what extent they are serious arguments versus veiled excuses for inaction.

Every anesthesia practice distinguishes itself based on the values, beliefs and outlook of its principals. Philosophy is an especially powerful factor in the specialty. If you want to appreciate its impact just ask anesthesiologists across the country their feelings about working with other anesthesia providers. The geographic distribution of practices corresponds quite neatly to a philosophical spectrum of views where those in the East find nurses integral to the specialty to those in the West who tend to avoid working with CRNAs at all cost.

Philosophical attitudes clearly underlie the arguments surrounding the need for more clinical data capture. A strong belief in the value of American clinical training, the appropriate use of monitoring and a broad armamentarium of powerful drugs tends to result in a belief that trending and analysis are both unnecessary and inappropriate. There are many providers who remind patients that they are at greater risk driving to the hospital than undergoing general anesthesia. How much better can clinical outcomes possibly get? And then they drive the point home with a cautionary note. Capturing risk factors and outcomes can only be used against providers who, for one reason or another, are perceived as outliers.

Further, philosophical prejudices focus proponents primarily on the evidence that supports their position than what might undermine it. The stronger the position, the more impenetrable the filter. There is no greater obstacle to change than the belief that change can only be for the worse.

The next bastion of opposition tends to cloak itself in economics. The economic realities of fee-for-service medicine have conditioned physicians to accept that if the market values something, it will pay you to do it. Conversely, if the market is not willing to pay for a service, then it is probably not a service that needs to be provided. This mentality has infused the specialty, conditioned thinking about compensation systems and ultimately proved to be one of the greatest challenges to group governance. Why should someone spend valuable time doing work for the group as a whole if they do not get compensated for it?

To a large extent data capture strategies continue to be defined by the economics of health care. Data elements such as PQRS Quality Data Codes are only added when there is a financial motivation to do so. The fact that capturing a particular indicator or piece of information would not result in a reward would be a huge disincentive. The government understands this principle all too well. “Meaningful Use” of electronic health records is a perfect example.

The reality, of course, is that there are short and long-term economic lenses. Those who believe that the Accountable Care Act is going to usher in more cost-cutting and a focus on gain-sharing argue that having the data to identify and develop strategies for improving productivity and profitability is actually the most important economic argument one can offer. They argue that the beliefs and strategies that have gotten us to where we are today will not get us to where we need to be tomorrow.<.p>

None of this is to imply that the practical challenges in capturing more data from each anesthetic are inconsequential. Every additional data element captured from clinical practice requires three distinct steps. First the information expected must be clearly defined and providers must know exactly where and how it will be captured. A process or mechanism must then be established to capture the information and include it in a database of some sort, either in the practice’s master billing database or some other database. Ideally, all required information would be captured in one large integrated database but this is not always possible. Finally, any additional data must be validated.

Here is a case in point. A practice decides that it wants to start capturing the anesthetizing location where each case is performed. A decision has been made that having this information will allow for the calculation of much more precise productivity metrics. In this case the group must first define how each location will be labeled and work with the software vendor to establish a field for the data to be captured. Providers are instructed where to note the location on the anesthesia record. A month passes and a QA process must be performed on the resulting data. 25% of the providers either did not mark the location for each case or did not report it consistently. In some cases the data entry team missed some of the locations. Ultimately, it takes three of four months for the group to achieve 95% data capture.

Capturing risk factors and outcomes that could potentially be used against the provider poses even greater challenges and issues. Self reporting requires great discipline and honesty. Inevitably the practice must ensure the confidentiality of the information and make certain that it will be used only in a blinded tatistical manner. For maximal success, the practitioners need to perceive this process as being essential, unbiased and helpful.

While these are all legitimate and practical concerns, they should not be used as reasons not to forge on. Rather, they should be part of a serious conversation about an appropriate approach. The fact is that many practices have already embarked on this path. Those that find the process too daunting are likely to be left behind by those who rolled up their sleeves and worked it out.

Process changes such as those presented here should ideally be implemented with an eye to addressing issues of specific relevance to the practice or to anticipating changes believed to be taking place in the market. So what are the big clouds on today’s anesthesia landscape? Most observers would agree that virtually all anesthesia practices must deal with three general practice management concerns. These could be defined as (1) the revenue challenge, (2) the security challenge and (3) the strategic challenge. In order to remain viable, every anesthesia practice must secure a revenue stream sufficient to recruit and retain appropriate numbers of qualified providers to meet the expectations and service requirements of their facilities. Given the reality of expanding coverage requirements and limited revenue opportunities, most practices must find ways to justify hospital support.

Asking for financial support from a hospital comes with its own set of risks. When administrators pay for services they typically want assurances that they are getting value for the money they spend. Today’s hospital contracts are more extensive and complicated than ever. They often include metrics and standards of care that must be met. Anesthesiology practices find themselves caught between accepting the hospital’s data and metrics and building their own databases.

This new reality has given rise to a significant concern about the security of a practice’s contract. The increasing use of Requests for Proposal (RFPs) and the growth of national management companies only heighten the anxiety of the providers. Clearly, many practices are seriously at risk of losing their franchises. The problem is that many do not appreciate just how much they are at risk of being displaced. The anecdotal evidence from around the country can be quite disconcerting.

Practices now have to consider options and formulate fall back strategies. The fundamental strategic question being discussed by large numbers of practices around the country is whether it is worth trying to remain independent versus joining a larger entity. In other words, every group must consider whether it can sufficiently distinguish itself and provide consistently superior care or let some other entity define its destiny.

Ironically, for all the observations and experience of its providers, anesthesia has let others define the quality of care provided rather than defining it themselves. The result has been a fairly myopic focus on negative or unexpected anesthesia outcomes. The ASA supports considerable research in anesthesia safety and modalities, but most of the papers, their conclusions and recommendations are intended only for the specialists themselves. The result, of course, has been dramatic increases in quality and significant reductions in anesthesia morbidity and mortality. As clinicians, anesthesia providers have tended to live in the shadows in their respective medical communities. This is why an article about the chairman of anesthesia at North Shore University Hospital in Manhasset, New York was entitled, “Who is This Masked Man?” While a sense of humility has served the specialty quite well thus far, it is unclear it will serve the specialty well in the competitive medical marketplace of the future.

These general themes are playing themselves out in three specific arenas. There is an increasing focus on the cost of anesthesia care. Customer service has become the issue of the day. All practices are scrambling to find new services and venues so that they can remain viable. Each of these has significant practice management and governance implications. Growing numbers of physicians are simply throwing in the towel rather than radically redefine what they do and how they do it.

The cost question is particularly challenging to many practices. How do you reduce the cost of a service where the only significant factor is the compensation of the providers? Physician-only practices have to consider care team options. Care team practices are looking at using more CRNAs. There are no really good options, however, when the real driver of their costs is the unrealistic coverage requirements of administration. Few practices have many good tools or strategies to reset administration expectations. This is where having reliable data and a compelling way to use it becomes so critical.

Customer service is another particularly frustrating issue for practitioners who have always seen theirs as the quintessential customer service specialty. The fact is, though, that in the customer’s eyes good clinical outcomes are a given and not the definition of good customer service. They want all three of the traditional anesthesia As: ability, affability and availability.

Today’s hospital administrators have high expectations. They want accountability, collaboration and innovation. They want business partners who are willing to share ideas and risk. They want anesthesia to take ownership of what happens in the O.R., not simply to profit from it. Committee involvement and the sharing of data and ideas are the new reality of today’s medical staff.

So how does more data make a practice more secure? In and of itself, it doesn’t. Today’s large anesthesia practices and staffing companies are taking a very different approach to data management. Perhaps inspired by the Googles and the Facebooks of the world, they are not just looking to validate what they know, but to identify issues that no one else has thought about. Differentiation and innovation are the driving factors in today’s marketplace. The new anesthesia mega-group strategy is to use the power of the anesthesia database as leverage to gain recognition and acceptance. These anesthesiologists are no longer willing to accept a role of subservience. There is no security in being useful to the administration, the surgeons and the rest of the medical community. Power comes from control and influence. Their goal is not simply to profit from the existing paradigm but to change it. Vital to that goal is having valuable data that can be shared with hospital administrators to identify rate-limiting steps in existing processes and to propose solutions that uniquely reflect the value anesthesia brings to the facility. It is only when this is the focus and intent of the data that it becomes truly useful.

Ironically, this is what most hospital administrators have been waiting for. The following is a short summary of a typical hospital administrator’s wishlist for its anesthesia department.

Why would they have such expectations? In large part they have such expectations because this is what the biggest and most aggressive of the nation’s practices are offering and providing. Once the bar gets raised every group must compete at a new level.

The landscape of anesthesia has clearly changed. Is it too late for most practices? Have they already fallen so far behind as to make it impossible to catch up or compete? Absolutely not. For one thing, not every administration wants to deal with a mega-group or national provider organization. For another, being competitive is less about the “what” than the “how.” While commitment and enthusiasm will never overcome substandard service or inconsistent care, they still count for a lot.

Expanding a practice’s database to include more than just the requisite details necessary to bill a patient and his or her insurance is a fundamental exercise in change management. Like many other initiatives that an anesthesia practice might consider, this one must be clearly framed and sold to the membership. Selling it is about overcoming concerns and objections. Effective change management inevitably requires three things: a champion, a vision and a plan. The importance of leadership in managing change in an anesthesia practice cannot be overstated. Given the independent nature of so many anesthesiologists, it is essential that there be a unifying force and focus to the initiative. This is not something that will ever happen spontaneously. Independent thinkers need to have all their issues addressed and their objections overcome. The leader must be able to address each category identified earlier in this discussion: the philosophical, economic and practical.

What is the vision that inspires physicians to report more details of each clinical encounter that could potentially be used to identify them as outliers, or worse yet, as providers of inferior care? It is not an easy question to answer, especially for doctors who believe that the quality of care they provide is already very good. It must be a vision of something more profound than clinical outcomes. It must remind providers that their very success has consistently diminished the perceived value of the services provided. It must inspire the specialists to think beyond their own individual value and compensation. It must remind them of the fundamental nature of the specialty, as the quintessential service specialty. It must speak to the heart and core of customer service, which always seeks to provide a safe, comfortable and compassionate surgical experience. It must remind each and every member of a practice that quality is defined by the least effective clinician in the practice. It must offer a compelling argument for doing things differently, and for being willing to innovate and take risks. Where these kinds of thoughts are persuasively communicated there will be a more enthusiastic endorsement.

Too often, however, the vision is neither clear not compelling and that makes the challenge ever more difficult.

There is a saying in sales that when the customer is confused, he will not buy. Being able to sell the concept of a robust clinical database is important, but it is only the beginning of the process. Anesthesia providers tend to think through issues very systematically and to solve problems based on their well-ingrained sense of decision-tree models. This is why the planning process is so critical. Most people prefer off-the-shelf solutions. The leader must not only sell a vision, he or she must clearly outline the roadmap to implementation.

This may all sound daunting. Effecting a change that affects provider behavior and requires the commitment and involvement of IT resources that might not yield the desired results is a risky proposition, but is there really an option? That is the question that every practice must address in today’s rapidly changing economic landscape.


Aman Mahajan, M.D., Ph.D., FAHA, is Chair at the Department of Anesthesiology, as well as Professor of Anesthesiology and Bioengineering and he holds the Ronald L. Katz Endowed Chair in Anesthesiology at the David Geffen School of Medicine at UCLA. Dr. Mahajan is a leader in the field of cardiac anesthesiology and cardiac electrophysiology & biophysics. A holder of numerous patents, Dr. Mahajan serves on various medical and scientific committees including the National Scientific Research Board. He can be reached at (310) 267-8680 or amahajan@mednet.ucla.edu.

Jody Locke, CPC, serves as Vice President of Pain and Anesthesia Management for ABC. Mr. Locke is responsible for the scope and focus of services provided to ABC’s largest clients. He is also responsible for oversight and management of the company’s pain management billing team. He will be a key executive contact for the group should it enter into a contract for services with ABC. He can be reached at Jody.Locke@AnesthesiaLLC.com.