Terry Loftus MD Weblog

There was a recent post on the Medscape discussion forum which referred to an article on the use of mnemonics in psychiatry.  The title of the article is: Mnemonics in a nutshell: 32 aids to psychiatric diagnosis.  Many of these memory aids are simple acronyms such as the CAGE assessment which is used to quickly determine a person’s risk for alcohol abuse.  Of note, among the 32 aids listed is the number of discrete units used for each mnemonic.  CAGE is one of the simplest with 4 discrete units and under delirium diagnosis is “Deliriogenic Medications” which has 15.  For those fans of Miller’s Magical Number Seven this may seem excessive.  In 1956 George Miller, a cognitive psychologist from Princeton, published a paper in which he demonstrated the capacity of our working memory was about seven, plus or minus, two elements (for English speaking people).  It is interesting to note that the number 7 is frequently referenced in our language to describe a list of various elements such as: 7 days in the week, 7 numbers in a phone number, 7 deadly sins,  7 hills of Rome and , of course the 7 wonders of the world. As it turns out the average number of discrete units for all 32 aids is 7.5 and the most frequently occurring value (mode) is, no surprise, 7.  The range is 4-15 which appears to deviate from Miller’s claim of 7 plus or minus 2.  Since we are referring to a memory aid, the aids with a low number of elements are not critical.  Of the 32 aids only four of them appear to violate the upper part of the range.  Fortunately  a simple mnemonic trick is used to facilitate these memory aids.  Each of these is broken up into smaller discrete units much the way we break up phone numbers.  The 15 element “Deliriogenic Medications” is given the mnemonic “ACUTE-CHANGE-IN-MS”.  Even “WWHHHHIMPS”, which lists the life threatening causes of delirium can be seen as “WW-HHHH-IMPS”. With the massive growth in information in healthcare it seems like we are rapidly approaching the point where mnemonics will no longer be adequate to support human memory.  It has served as a useful bridge but the information age is becoming more and more dependent on information technology as the primary memory aid.  Miller’s Magical Number Seven will not go away though, it will become more important in how information technology presents information to us.  This is the realm of information management.  For healthcare to progress we are going to need to become much better at how we manage information.  It is going to take more than mnemonics to overcome our memory capacity limitations, not to mention the other limiting features of bounded rationality.  The information age will allow us to move beyond satisfactory solutions and endeavor toward optimal solutions.  This assumes, of course, we learn how to manage this vast amount of information in a way that compliments our natural capacity to understand and use it.

Hospital administrators view things differently. One of the primary reasons any firm in any industry (health care included) brings an individual who provides a service or produces a product in as an employee is to reduce transaction costs. Firms exist because transactions that occur within them are more efficient than if those same transactions occur outside of the firm.  Once the marginal cost of a physician’s services consistently exceeds the marginal benefit the physician provides to a hospital it is in the hospital’s best economic interest to employ the physician in order to control the costs or improve the benefit. For example, as physician reimbursements decrease and their costs increase they must seek alternative revenue streams or practice efficiencies in order to maintain the same income. With the paltry 1.1% raise coming from Medicare and the July 07-08 inflation rate at 5.6% it is not difficult to understand why this is happening. As physicians struggle to keep up it seems it is inevitable they will lean on hospitals to either boost revenue (stipend for call, reimbursement for uninsured patient care) or cut costs (EMR support, medical office space). These are transaction costs administrators are willing to support in order to keep the hospital doors open. When it is less costly to employ a physician than pay these costs hospitals must as a practical matter move toward employing physicians. It’s an expensive choice but when the alternative is more costly it becomes the best alternative to a negotiated agreement (BATNA).  In order for physicians to remain independent they must be able to generate revenue and decrease costs in a manner that does not make them dependent on the hospital. I don’t claim to know where this will lead to in the coming years but I think as long as physicians squeeze more of their revenue stream out of the hospitals they will ultimately increase the probability they will become employees of the hospital.

It’s been a busy couple of months between returning to school and starting a new job so I thought I would share a recent assignment I wrote for school.  It’s based on my understanding of a Harvard Business Review article (HBR 9-808-072) titled, Transforming Arizona’s Health Care System: Developing and Implementing the Health-e Connection Roadmap.

In his recent bestseller, The Age of Turbulence, Alan Greenspan recounted an article written by Paul David, a professor of economic history at Stanford, who explained why it took nearly four decades for the electrical motor to displace the steam engine as the primary source of power in America.  When Thomas Edison first demonstrated the superiority of electrical power in 1882 nearly everyone could appreciate its merits.  It was also assumed that it would be just a few years until electrically powered motors would be everywhere and still it would be decades before it happened.  As Professor David explains the reason it took so long is because the process was capital intensive and required a restructuring of the existing cultural, social, architectural and technological infrastructure supporting the status quo, the steam engine.  The same could be said of the use of computers in the evolution of electronic data exchange in our healthcare system. 

            Everyone can see that computers should be able to improve the delivery and efficiency of healthcare.  Just like electrical power it is obvious that computers will displace the status quo, the paper chart, but it has not happened as yet on the grand scale and many are getting restless.  Governor Napolitano’s Executive Order 2005-25 was an attempt to jump-start the process by bringing the state of Arizona’s healthcare stakeholders together to create a roadmap to “achieve statewide electronic health data exchange between various entities in the health care delivery system”.  The driving force behind this is the increasing awareness of the problems that exist in our healthcare system and the recognition of the role information technology can play in the solutions to those problems.

            Since the Institute of Medicine’s 1999 report, To err is human: Building a safer health system, there has been a growing urgency to fix the long standing problems in healthcare.  It is now estimated that there are thousands of injuries and deaths due to avoidable medical errors.  In addition, millions of dollars are wasted every year as a result of redundant tests, missing information and lost productivity.  It has been shown that the use of computers and information technology can not only reduce medical errors but also lead to a significant cost savings over time.  The Governor’s roadmap (Arizona Health-e Connection) was intended to provide a path to improving the delivery of healthcare through the collaborative use of information technology and act to build momentum toward achieving these goals.  So if the merits of this roadmap are so obvious what is holding back the universal acceptance and rapid rollout of this project?

            It may be, as Professor David explained regarding the electrical motor, that we are undertaking a restructuring of the legacy systems that have maintained the status quo and that replacing them involves disrupting the established cultural, social, architectural and technological infrastructure we currently depend on to deliver our healthcare.  It’s not as easy as unplugging the old system and replacing it with a new system.  The new system must be built upon, co-exist, and ultimately replace the old system.  Fortunately the Arizona Health-e Connection Committee anticipated some problems with implementation and attempted to avert them by building a coalition based on broad representation of healthcare stakeholders.  Despite broad representation and the obvious merits of the project roadblocks remain and some may simply take more than a roadmap to overcome.

            Arizona has some unique challenges to overcome.  One issue that must be addressed is that at least a quarter of the state’s population is rural and spread out over a wide geographical area.  At the time of the committee meetings only 13.5% of physician practices used electronic health records (EHR).  With many of these practices classified as small or medium in size and located in areas that exist outside of the main information technology networks it is difficult to bring everyone on board and deliver the same standard of care to all state and Native American populations.  Electronic health record technology also has significant start-up costs which are considered prohibitive in an environment of declining reimbursement and rising costs.  Physicians are skeptical about the immediate return on investment of high cost technology and like everyone else who has seen the price of computers, plasma screen TV’s and cell phones plummet are willing to take a “wait and see” approach.  While waiting may lead to continued errors and cost issues for the system as a whole it may also lead to lower IT costs for an individual practice and the likelihood of a markedly superior EHR when it becomes more affordable.  This is not to say physician practices are the only stumbling block.

            The massive build-up of information technology hardware and software is the result of numerous innovators and firms with a vested interest in the proprietary rights of their intellectual property.  This has resulted in a wide range of vendors, products and services as well as a healthy dose of competition to facilitate the innovative process.  Each hospital and healthcare system has, to varying degrees, invested in these various products and services.  In an effort to seek a competitive advantage for both the vendors and healthcare systems many of these innovations were not necessarily designed to enhance interoperability with one’s competition.  For example, if you are in the business of selling electricity and lamps it makes sense in the early phase of a product or service life cycle to sell only lamps that can plug into your power grid.  As the electrical companies focused on delivering power and got out of the lamp business the new focus became interoperability, which means getting a common standard for electrical outlets so everyone can plug into the power grid.  This may be analogous to what we are experiencing in our healthcare system.  Without standardization within the electronic data exchange networks it will be impossible for everyone to focus on innovating more improved and less costly plug-in products and services. 

            The Arizona Health-e Connection should focus on network standardization and interoperability.  The early days of electrical power dragged on while standards were being established but once they were the realization of the true merits of Edison’s vision materialized.  Once standardization and interoperability are established it will also be critical for stakeholders to agree on what information can be shared and with whom it can be shared.  Any risk of creating a network which provides asymmetric information to privileged entities will sabotage efforts at collaboration.  In addition, once standardization and interoperability are established this will articulate a clear message to the market.  The health information technology industry will be able to focus more capital on innovations that specifically are designed to reduce errors, redundancy, misinformation and costs in healthcare.  Instead of each healthcare system differentiating them-selves by owning a unique EHR that lacks network interoperability they can now focus their attention on systems that utilize network information for improving care.  The real goal after all is to improve patient care and decrease the cost of delivering that care through enhanced integration of the statewide health information system.  Standardization and interoperability eventually worked for delivering electrical power.  Let’s hope it doesn’t take as long to do the same for delivering better healthcare. 

It was one of the big healthcare stories in today’s New York Times.  It started out by saying:

      ”A growing chorus of discontent suggests that the once-revered doctor-patient relationship is on the rocks.  The relationship is the cornerstone of the medical system — nobody can be helped if doctors and patients aren’t getting along. But increasingly, research and anecdotal reports suggest that many patients don’t trust doctors.”

Right there in the opening sentence, written between the lines, is the crux of the problem.  The author like so many of us put the “doctor” first in the “doctor-patient relationship”.  It’s a subtle but important distinction.  It is essential that the public and those of us in the medical profession learn to appreciate this distinction.  If there were no patients there would be no need for physicians.  Ergo it is the patient that comes first in this relationship.  The patient’s healthcare needs are what created the physician’s healthcare solutions.  Like I said it is a subtle distinction to reverse the order but it says so much about how we view the world of healthcare services.

While I am on this topic you may have noticed a couple of other related observations.  One is that I specifically did not refer to the “patient-doctor relationship”.  In the United States the conventional use of “doctor‘  refers to a medical doctor.  This is not always the case and “physician” or “medical doctor” is the more appropriate term.  Another point is that recently the “patient-physician relationship” is quickly being replaced by the “patient-provider relationship”.  I’ve discussed some of the reasons for this in a previous blog.  Suffice it to say I believe it is because the “cornerstone of the medical system” is changing.  I suspect this is really the reason the trust is breaking down.  The real question is if patients don’t trust doctors who will they come to trust?