Why not cheat when improving healthcare quality?

Dr. James Holly

In 1993, John Patrick set IBM on another course and changed the company’s future. Reading his story made me wonder if it’s possible for SETMA to set medicine on another course and to change the future. Patrick did not want people to work “collaterally,” side by side, maybe going in the same direction, maybe even having the same goal but working independently and at best in a cooperative manner; he wanted people to work “collaboratively,” synergistically, leveraging the generative power of a team in creating a new future that they partially envision but which even they could not control.

What can we do today in healthcare that would mirror the change that IBM experienced? How can we change “collaterallists” into “collaborativists”? How can we use the power of electronics, analytics, and informatics principles to energize radical change to create a new future in healthcare?

Why not cheat?

Testing and measurement is a science. In most industries, quality was determined by testing performance. But, in healthcare we are involved in a new kind of “testing.” The tests used to measure the performance of healthcare providers are unique. Therefore, if you are going to measure the quality of care given by healthcare providers with a test whose questions they have beforehand, and if the test is open-book, and there is no time limit, why not “cheat”?

Look up the answers before the test so you can know your performance before you get the test results. Don’t wait until an insurer, an ACO, or an agency measures your HEDIS performance. Know your performance beforehand by measuring it yourself. In fact, know your performance at the time you see a patient. The ultimate “game changer” in healthcare is when the provider knows how he/she is doing in the care of an individual patient or a panel of population of patients and then when the provider turns around and shares this information with patients and with the public at large. The game is changed because the motivation to improve is maximized.

Of course, ethically there is no “cheating” in this context. Unlike traditional medical-education tests, this test is not measuring what you know; it is measuring what you have access to and to what you pay attention. It is measuring how efficiently and excellently you are applying what you know. The test is not measuring what you remember; it is measuring what you are reminded of. If you have Clinical Decision Support (CDS) that reminds you of what needs to be done and if you have CDS tools that allow you to measure your own performance at the point of care, you can consistently improve your performance.

Go around barriers

After a 3.5-hour presentation of SETMA’s system to eight Medicare Advantage executives, they asked how they could get other providers to perform as well as SETMA. I told them that they need to develop leaders who will help improve the processes and outcomes of care, but that they must recognize also that some times physician leaders use their positions to resist, or to obstruct change rather than to facilitate it.

Determined never to be invaded by Germany, in the 1930s the French constructed a fixed defensive fortification between France and Germany called the Maginot Line. The French did not know what Gen. George Patton intuitively knew. In an era of mechanized warfare, fixed fortifications could be and were easily ignored. The Germans went around the Maginot Line. Similarly, when the barrier to healthcare improvement is created by the refusal of healthcare providers to accept new realities and new standards of care, health systems will simply go around them. The obstructing providers become irrelevant to the process. The reality is if healthcare providers become fixed fortifications against the future, the process and the system will go around them.

There is also the fact that excellent healthcare makes much greater demands of healthcare providers than ever before. But many of those demands can be met without active participation by the provider. For instance, one of the most complicated things healthcare providers are asked to do is to report to the local and state Departments of Health the diagnoses of certain illnesses. In Texas there are 78 such of these. Just remembering all 78 conditions is daunting; add to that the need to stop in the middle of a long day to fulfill the reporting requirement is another and perhaps even bigger problem.

Value: Quality divided by cost

The lessons of the industrial revolution give us guidance here. Rather than handmade tools and machines made by artisans who were creative geniuses, machines were now made by other machines and they were reproduced in mass. The cost went down and the quality went up so the value escalated geometrically. If we apply the lessons of standardization, automation and reproducibility to healthcare, we can get to our goals much faster. Henry Ford made a new machine on an assembly line that was nothing more than a standardized, automated method for producing a product that also required human input.

If we look at every process and outcome in healthcare as a sum of that which can be automated and standardized, and of that which still requires human input, we can improve quality predictably, we can reduce cost consistently, and we can improve the satisfaction of the provider and the patient. Some things in healthcare cannot yet be standardized and automated, but the satisfaction of receiving the care that can be will be increased by determining what can be automated and standardized and then by doing so.

Complexity demands systemic solutions

The Texas State Health Department’s Reportable Conditions illustrates the standardization and the automation of parts of healthcare processes. Remember, “The more complex a problem is, the more systemic the solution must be.” Now, SETMA providers make a diagnosis and when that diagnoses is one of the 78 reportable conditions, automatically, the following takes place without them doing anything else:

1. The Reporting Conditions Template is populated automatically.

2. A note is sent to Care Coordination automatically.

3. Care Coordination reports the condition automatically.

4. The provider is automatically notified that the report has been made.

If an important task is not being done either because the provider is resistant to doing it or because the provider has “too much” to do, automate it.

How many tasks can you get a provider to do? That question was asked at a conference in Boston in May 2012. There were many answers, but the best answer was based on the need to ask three other questions:

1. How important is the task?

2. How much time does it take?

3. How much energy does it take?

If you were to create a formulae to represent this process, there would be a direct correlation between how many tasks a provider can or will do and how important the tasks are; the more important the tasks, the more tasks a provider will do. There would be an inverse relationship between how much time it takes and how many tasks will be done; the more time it takes, the fewer tasks will be done. There would also be an inverse relationship between how much energy it takes and how many tasks will be done; the more energy it takes, the fewer tasks will be done. The key to getting more done is to determine what is important and only to do that, and then to make the completion of the tasks require less energy and less time.

How can we change the future?

Make it easier to do it right than not do it at all! There are many aspects of patient care which can be automated. Classically, SETMA has used clinical decision support as reminders to providers, but now we are realizing that many of the tasks that were the object of CDS actually could and should be automated, requiring no input from the provider. For instance, the value of the flu immunization is not enhanced by it being ordered by a healthcare provider, or by it being given by a registered nurse. And the process of a flu immunization can be automated.

1. When a patient is given an appointment and the system determines that the patient has not had a current flu immunization and the appointment time is in the appropriate time frame to receive the vaccine, the system should order the flu immunization, and send the order to the nurse, to the chart and to charge posting. The provider is not involved, which increases the probability that it will be done.

2. Additionally, the system should be programmed so that every patient who has not made an appointment in the time frame for a flu immunization should be notified electronically at the beginning of the flu-immunization season that they need to have a flu shot and toward the end of the immunization season, the system should check again to see who has not had the shot.

This principle can be expanded to all chronic conditions for which the patient is being treated and/or for all screening and preventive care the patient requires. In the future, all healthcare process will be evaluated for:

1. That which can and should be automated, all based on evidence-based medicine

2. That which requires human input based on patient-centered care

This will give the healthcare provider more time to focus on the patient – patient-centric – while fulfilling the processes (care), which we believe will improve the health (outcomes) and which will decrease the cost of excellent care.