curative design

So what is ‘Clinical Pharmacy Integration’? At its most basic, it might involve initiatives to improve coordination around a single disease, typically asthma or diabetes. At its most sophisticated, it might encompass fully integrated hospital systems with closed staffs consisting entirely of employed physicians. Most hospitals and health systems are somewhere in-between.

When 82-year-old Jack Schlichting was admitted to the Henry Ford Hospital in Detroit five years ago with a badly broken leg, he received standard hospital treatment: initial care in the trauma center, surgery and some time in the hospital recovering. But his discharge from the hospital did not mark the end of Schlichting’s care by the Henry Ford Health System, a large, diverse health care organization that is an established leader in clinical integration.

While Schlichting recovered at his daughter’s house, under the care of the Henry Ford home health care program, he went through extensive rehabilitation at another of the system’s hospitals. All the while, his care was supervised by physicians of the Henry Ford Medical Group. He found a wheelchair through Henry Ford’s durable-medical-equipment company and ordered his medication from the system’s pharmacies.

And while this explanation focuses primarily on the ‘coordination’ of care, I believe that true integration, in order to be successful, relies on one final piece – the thoughtful application of domain experts with technology solutions in a coordinated manner.

In other words, identifying the weaknesses in our workflow and creating mechanisms to apply solutions by the folks that are considered ‘experts’ in the domain, supportive IT platforms to aid in their efficiency.

Basically, get the folks that love that particular task hardwired into the workflow for that task.

Consider our current workflow for medicine reconciliation in the ED. The ED physician does a cursory med reconciliation during his all-important initial H&P, gathering information towards a working diagnosis and management plan. Upon the decision to admit, the ED nurse is charged with the task of completing a comprehensive medicine reconciliation prior to the admitting internist or family practitioner seeing the patient.

The hard truth of the matter is that accurate medicine reconciliation is one of the most labor intensive processes that this nurse will face in his / her shift, under the guise of a simple list of medications. Aside from the accuracy of the list itself, there are myriad other considerations:

• Accuracy of the medication dosage, schedule, frequency.

• Accuracy of whether the patient is in fact taking the medication

• Accuracy of whether this list coincides with the most recent prescribed medication list from the PCP and/or the pharmacy

• The investigative work required to accurately identify medications described from memory by the patient, ie ‘That Little Purple Pill’

The simple truth is that a medicine reconciliation approaching 100% accuracy can take as much as 30 minutes of concentrated effort with calls being made to family, pharmacies and PCP offices taking up the bulk of that time.

It came to me in a vision. The proverbial AHA moment. The curling of toes, the . . . you get the idea.

It occurred to me that almost *all* of the high yield Quality initiatives in our institutions (and by default all institutions, as these tend to be national measures) are inextricably linked to Pharmacy.

SCIP / Med Rec / Transitions of Care / Re-admissions / Medication errors / CORE Measures, etc. The list goes on.

This represents a top down approach to the oncoming challenges that we face in an increasingly regulated environment. Technology is only one small piece of the puzzle. As leaders, it falls to us to effectively deploy technology and systems to meet these requirements, but also be sensitive to provider workflows.

I believe that forcing workflow changes that could possibly make the practitioner less efficient is an inelegant solution. A rework of our system, deploying domain experts to add a layer of intelligence may not only alleviate workflow bottlenecks, but also possibly invigorate it also! This to me, is the essential premise behind the term, Clinical Integration – aligning incentives so that domain experts are adequately positioned to do the most good, every step of the way.

Who is best equipped to obtain medicine reconciliation information?

Who is best equipped to make sure that the Med Rec list is the closest thing to the source of truth, using calls to Pharmacy, access to the Sure Scripts database and, sometimes, just plain common sense?

Who is best equipped to make recommendations with regards to stopping / continuing medications on the inpatient setting, based on the proposed plan of care and the admission medications?

Who is best equipped to answer questions about all of the above to all the stakeholders – the nurse, the physician, and most importantly, the patient?

Well duh – it’s the clinical pharmacist, and my aha moment was the realization that this individual holds the proverbial keys to the castle in terms of managing the transition from a paper to an electronic medicine reconciliation process. Over the course of the next few posts, I hope to share how we came to this realization.

As clinicians, we generate incredible amounts of data on a daily basis. If you think about the number of orders generated per patient interaction on admission through follow-up care all the way through discharge, we are essentially a data generating fools.

Further to this, consider the sheer volume of data created from the very transactional nature of medicine. Take for instance something as benign as physician referrals. Emergency room physician A refers to General practitioner B for follow-up care of the patient that he or she has seen in the emergency room. Alternatively general practitioner C refers then refers to specialist D for the same and vice versa. In many ways referrals are part of the institutional glue that holds modern healthcare together in addition to being a substantial driver of wealth. From a data analytics standpoint, referrals provide an interesting opportunity. By finding out which physicians other physicians refer to most frequently, one obtains a sense of which physicians are the most trusted amongst physician networks and groups.  This may also indirectly reflect such qualities as approachability, affability, ability, responsiveness and overall professionalism.

Sarah Mirtoff highlights this issue in a wonderful piece from Wired magazine, detailing the rise of companies that seek to explore the significance of this type of data.  The beautiful image that accompanies this post was provided by the company Healthtap, and is a snapshot of northern California doctors and the referrals to other doctors.  Each doctor is represented by a blue dot and the connecting lines represent a referral.

Furthermore, data scientists like Fred Trotter that inhabit the healthcare IT space, are busy working away at making this whole process of physician referrals more transparent and truthfully, democratizing data.  Recently, Fred took to the health IT start up crowd-funding platform MedStartr and was able to raise $15,000 towards building out this database. Sunlight is the best disinfectant and I believe that this is a positive step in the right direction. To be fair, however, I would also hope to see this kind of transparency in other industries also. For instance, why are there are no open graphs of business to business referrals in say the food industry, the fashion industry, and manufacturing industries also? Knowing how interconnected certain companies and brands are might better inform my choices as a consumer.

When I was doing my research fellowship at Hopkins a few years ago, the big buzzword then was biometrics.  Essentially using properties of the human body to generate unique identifiers that would allow us to build security protocols around them.  A few years later and now we’re inundated with various applications of this technology in the form of attempts to replace the age-old login process of username and alphanumeric passwords.

Ask any IT person.  Passwords are the bane of our existence.  Strong passwords seem to operate on a principle of the more irritating they are, the ‘stronger’ they are perceived to be, when in fact, this may not entirely be the case, as demonstrated elegantly by this brilliant XKCD comic.

Will applied biometrics in the form of single-sign on solutions save us from the pain of the ‘minimum of 7 characters, one upper case, one lower case, one symbol, one number, rotated every 2 weeks’ insanity?

We here at Cd hope so.

What is a boundary spanner?  An individual that seeks to gain understanding from the different groups of people in an organization and align disparate ideologies through dialog and progresses toward a common goal.  Sound familiar?  It is the elegant CMIO that is able to unify the various different tribes.  Furthermore, it is the elegant CMIO that is able to build upon these newly found relationships by learning what is valuable to these groups and cross pollinating with the right questions.Examples of these questions might include:

1. Who are the safest, most high quality physicians in this hospital?
2. Who are the most cost effective physicians?
3. Why?
4. Which patients cause this hospital to lose money?
5. Which diseases cause this hospital to lose money?
6. Which processes in discharge/follow up really work?

Where can I invest $50K (in which disease management process to obtain the biggest gain in terms of patient morbidity/mortality?

Analytics itself is not a panacea. Analytics is simply a series of tools and statistical methodologies that enable us to come closer to this thing I call “the truth”. I just wrote that down because it seemed like a cool thing to say. Anyhow, speaking plainly, replace the word “truth” with “answers you can believe”. So analytics is simply a means to an end. What is this truth? Well while everyone is caught up with the buzz word that is analytics, or even clinical analytics, the truth is it is simply a tool. The elegant CMIO knows that the more important skill is “Being able to ask the right questions!”.

And this comes from that series of skill sets that brought us to prominence in the first place. Relationship building, interfacing, creating shared anthologies and project management.

As you can see, once you have the right questions, then you can start upon the task of using analytics to answer them. Here is your task list to success armed with this knowledge.

• Learn the Tools.
• Make sure your answers are as Visual as Possible.
• Learn Six-Sigma and Lean to implement real tangible change in your organization.
• Present, Share and Evangelize often.
• Celebrate each successful implementation with pomp and circumstance.
• Analyze each failure with a post mortem and consider taking “holidays” from projects to allow fresh viewpoints.

We’ve spent the last few posts discussing the process of getting in and out of a system remotely and securely, and how using simple usability principles can create a smoother process. As one can appreciate, this is merely the beginning of the physician HIT usability journey, as we have yet to explore the gulf that is the clinician/EMR interaction.

Many in the industry acknowledge that a good number of the EMRs in existence today began their lives as enterprise-level software for business-to-business applications. As this technology became more end-user dependent, usability once again arose as a key factor, and some of the weaknesses in systems that were not designed from the ground up with the clinician in mind started to become evident. Human factors engineering is the term given to the study of how humans, both physically and cognitively, engage with computer systems.

Once the user experience piece has been optimized, then it is possible to focus on the “fun stuff” – for instance real-time, evidence-based, clinical decision support. When the physician no longer feels beholden to the technology solution, a different relationship emerges; a relationship in which the clinician is provided relevant, helpful information at the point of care; a relationship in which the system anticipates the needs of the clinician, offering reminders that are contextually sound and situationally appropriate – a true safety net.

In the coming months, we will look at some of these technologies and the potential that they have to be transformative in the field of hospital medicine.

(Image courtesy of Orange Aura)

So how do you round? Any good assessment of rounding workflows requires serious self-reflection regarding your current practice. Do you document each encounter individually? Do you see three or four patients first and then document three or four progress notes in a sitting? Is there a “right” way to round on multiple patients?

Once you’ve looked at these specific considerations, it is reasonable to build technology solutions around this process. In our facility, we engaged a low-level pilot study using tablet PCs. Tablet-based rounding was deemed suboptimal for a number of reasons:

1. Our current EMR architecture was not optimized for touch-based input, making data entry and documentation laborious and time-consuming.
2. The Citrix-based workaround would not suspend appropriately and required a repeat login every time the tablet went to sleep.
3. A dictation/transcription solution could not be utilized via the tablet modality.

We learned a great deal from the small and relatively inexpensive pilot study. From a hardware standpoint, the tablet is, I believe, still the ideal solution. Substantial pluses are battery life that is able to withstand a twelve-hour shift, the ability to turn on instantly, multitasking capabilities, and excellent screen resolution. However, for the reasons stated above, in addition to the absence of having a dedicated, streamlined way to enter data efficiently (via dictation or keyboard entry), we were prompted to look at other options. As a result, we are looking at Ultrabooks as an interim solution for our rounding hospitalists while we allow tablet-based workflows to mature to fit our needs.

As a rounding clinician, the technology solution has to work seamlessly for me before I force it upon my colleagues. If, between patient encounters, I can open my Ultrabook and have it use a biometric system to seamlessly log me in to our EMR system, allow me to quickly enter my orders, quickly and accurately dictate my progress notes, and be intelligent enough to know that closing the laptop lid suspends the system, allowing me to move on to my next patient without incident, then I will be happy. And because the EMR that I am using exists virtually on the hospital servers, I am comforted that should my laptop be stolen, no sensitive data exists on the laptop itself.

If I were able to use such a system daily in the care of my patients, day in and day out, then I might consider this system robust enough for daily hospital use.

(Image courtesy of OHSU)

The BYOD approach is often most successful when used in tandem with a virtualization architecture. Simply put, a virtualization architecture means that a physician is not actually interacting with software on the physical machine in front of him or her, but a representation of it served up by a server in a secure location, often on the hospital premises. Think of it as using binoculars to look at the screen of a computer located in a building across the street, only that you are able to enter information as if that computer were directly in front of you.

From an information security standpoint, this is more secure than carrying potentially sensitive information around with you on your day-to-day device. This information could be compromised if ever the device were stolen or misplaced. Such devices are, in effect, used as thin clients to access a suite of applications virtually. If this can be implemented with a single sign-on solution that is robust and non-intrusive, then we have the makings of a truly evolved user experience.

Think of it… being able to sit down at a PC and log in instantaneously by way of a biometric scan or a “tap-in” system in which an FRID-enabled physician’s ID badge initiates the login on your behalf. The screen then instantaneously opens a window into your hospital system’s virtual desktop, instantaneously logging you in to your EMR, upon which you can commence writing orders and/or documenting a patient encounter. The session is then closed and/or suspended with a tap-out or biometric logout, allowing you to continue with your day.

This episodic, instanced, virtual desktop access is slowly becoming the norm, and represents a step towards true decoupling from the patient’s chart. Much has been written about the promise of electronic medical records freeing up physicians’ time to allow for increased interaction/education with the patient. In my experience, we have yet to reach this panacea. However, by integrating systems (biometrics and single sign-on technologies) and being respectful of medical staff workflows, we are slowly, surely, making gains.

(Image courtesy of TechBabu)

The tool that one uses to access this information also directly affects user experience. Am I unable to access this information remotely? Is this information secure? Do I require a desktop or can I use my cell phone, laptop, or tablet? Can I see the relevant information correctly?

There is a tremendous pull from the medical staff side to integrate popular consumer electronic devices (smart phones and tablet PCs) into the clinical work environment. The challenges are the need for rigid information security protocols and maintaining a high level of satisfaction among users. BYOD stands for Bring Your Own Device, which is seen as a way to attain a compromise with an increasingly sophisticated medical staff from a technology standpoint. BYOD is increasingly being seen as a way to meet this compromise and build towards "device agnostic" systems.

(Image courtesy of Digital Marketing Blg)