The Seven Myths of Healthcare Technology5. One Size Fits All

August is over, and so is, sadly, vacation. So, we came back to the office this week, to a pile of mail (I see that Canadian Tire had some nice sales in August) and a week of "start-up" appointments. One was a discussion with a partner company in Toronto. They are in marketing and were contacted by a potential client about an EMR/management system for small-office providers. As an SME, I was brought in to provide support. It sounded like a well-thought-out product serving a clear need. However, as we talked, I recognized a flaw in the initial assumptions;

Myth #5 "Solutions Are Shippable."

The potential client, a software vendor, was thinking of establishing a client base in Ontario and then expanding through Canada and the US--a reasonable and not uncommon business-development route for most products. Vendors and buyers of look on both sides of the border for new clients and better deals, respectively. This works with auto parts, fruit juice and so many other things--especially after NAFTA. And one would assume it would work with HIT as well--and be wrong.

Healthcare IT is only a device to enable a service process. It does not actually deliver healthcare. It just helps keep track of the healthcare being delivered. And that process is not the same everywhere. In Canada, the first goal of record keeping is accuracy of health data and surety of care. In the US, the first task is bookkeeping--determining payment liability and authority. In the US, there are HIPPA regulations carefully restricting access to patient information. These regulations are much more stringent and litigious than those in Canada, where a "circle of care" is more readily extended to include those brought into a patient's care process. These things change the nature of how electronic-record software should be not only managed and used, but actually structured.

In the US, software has be made to accommodate the tangled web of insurers and pre-authorizations required. Before care other than emergency stabilization can begin, financial responsibility and authorization have to be obtained from any one of a myriad of people and documented. Following that, HIPPA practices (coupled with litigation-risk management practices) require careful and full documentation of patient information and consent. Security functions (ID/password systems and access logging and monitoring) are required to assure HIPPA adherence by providers and facility staff. Then each service, each supply and each provider interaction has to be documented in a way that can later be billed successfully to whichever insurance company, HMO, employer fund or government plan may apply. This is often a complicated mix of incremental payments from a number. There often is still a significant balance to be charged the patient (or their family) which may require financing (and the related software functionality). Healthcare software must handle all of these requirements as core elements. This directs a large part of initial development to service coding, annotation of coding, coding-decision support and back-office interfaces and resources (billing office dataset displays and telecommunications with insurance company databases, for example). And these are core requirements for providers in the US.

In Canada, financial issues are far less complicated; a patient is either Canadian or not (or, in rare cases, not up to date on their Provincial premiums). This immediately clears authorization and billing for the majority of primary care services (more than 90% of all healthcare). For "non-urgent" care; optical and dental, elective services or "extras" like private rooms; billing is to one of the fewer-than-in-the-US insurance companies or to the patient. But the billing process is not a primary function in the electronic record. The "Circle of Care" allows easier involvement of providers and requires less documentation of patient information and consent. While patient privacy and security are highly valued in the Canadian practice, the limitations on provider liability in Canada reduce the urgency for liability-risk mitigation through documentation and security processes. Similarly, coding of diagnoses, treatments and prescriptions are important in the Canadian system, but for different reasons. While the US providers are working to ensure that they get paid, the Canadian providers are more concerned with helping organizations like CIHI in the development of trending data and national care standards (as examples).

Clerks in the US ask "Who's paying?" In Canada, they ask, "What's wrong?"

Software is a device to enable a process. While the goal may be the same everywhere, the process is not. And the software cannot.

Some Notes and Then Myth #4, EMRs...

In this slow summer, we are working through the Seven Myths of Healthcare Technology. Myth #2 is "Technology Can Do Healthcare." [check it out below] Here's a comment I found out there--The writer certainly seems to have a reason for his anger. But is he completely correct? What do you think?
Community College - No Way

And is this an attempt at something "Swiftian?"

Responsibility-Based Rationing of Healthcare

"Eat the Elderly" kind of stuff? Or do you think he is serious? With all of the alternate-reality based rhetoric being put out there, I am honestly unsure...

And it looks like telehealth is happening without complex, expensive proprietary software systems...

The doctor is in and logged on

And we continue with…

The Seven Myths of Healthcare Technology; #4 EMRs are for the Doctors

As hospitals and medical facilities move to meet the technological mandates and requirements coming their way (from managers, regulatory agencies and insurance companies), they are confronting the issue of stakeholders--who are they and who is most important. Historically, the final word in healthcare has been that of The Doctor. What the MD says goes. As software is implemented in more and more systems, vendors and internal IT staff are having to customize and personalize software for each facility and clinic.

Healthcare software is not “off the shelf,” as some vendors might present it. It is more “out of the crates, custom-tailored, re-measured and stuck with unusual plug-ins and attachments to serve the vicissitudes of legacy systems.” In the process of making these applications work in the particular digital and business environment, sensitivity to end-user experience is considered essential. And doctors are the obvious end-user stakeholders.

Doctors are characterized as (just a little) demanding. Nurses everywhere tell tales (or at least roll eyes). This is no different, and perhaps worse, with software. Doctors are a technically saavy group and not shy about speaking up. But they are the wrong users on which to model EMR use.

Doctors use EMRs to mostly see data. They look at one or two views of data; to review intake information, view images; check results, or, watch for drug allergy alerts--then they look at the patient. They use systems to place orders or prescriptions--sometimes. Often, these orders are completed by PAs or nurses. Doctors do not move through an EMR very much--they tend not to do patient intake or to record new data very much. They are not the primary users of the digital tools---their skills lie elsewhere. There are certainly providers who are more “hands-on” with electronic systems. These are usually providers in record-intensive practices such as psychiatry or in the unique practice structures of the military.

Clerks, Nurses and PAs are the big users. They are the people using the record systems most--from patient intake through scheduling to checking for labs and rads. Very often it is they who bring up the patient record for a doctor and file orders. And it is the desk staff who are using the EMRs to communicate with insurance companies and other referral facilities. It is their workflow that should be modeled--and theirs are very different workflows.

People will say “Nurses really run the hospital.” I won’t get into that debate, but I will say “If they don’t control the hospital, they certainly control the information in it,” and that is the EMR…..

The Seven Myths of Healthcare Technology 3. Technology Can Do It All

In a letter sent Tuesday to Sens. Edward M. Kennedy (D-Mass.) and Max Baucus (D-Mont.), President Barack Obama reiterated his commitment to promoting the use of information technology as a means of reducing healthcare costs.
healthcareITnews, June 4, 2009
A month later, HealthcareIT news reported
"President Obama called for fixing the broken healthcare system by building upon investments made in electronic medical records in a town hall meeting held Wednesday." [July 1, 2009]

In the US, many of the players--drug makers, provider organizations and insurance companies--have been calling for increased IT investment as the key solution to the problems of healthcare in the US. The advantages of increased IT efficiency are clear. Yet the healthcare business process lags far behind other businesses in the use of electronic automation. Most providers and facilities are just beginning to look at electronic record management, and many are still using only paper records. The most advanced use of IT for healthcare is by the Federal government. The Department of Defense is well along in development of a "lifetime electronic record" and the VA continues to set a high standard for management efficiency. And Medicare/Medicaid manages one of the largest patient bases and payment processes electronically.
But does electronic automation provide ALL the answers?
In Canada, aggressive support of HIT has been a policy of both the Federal and Provincial governments since 2000. A number of hospitals in the greater Toronto area became very active in digitizing patient records and integrating digital patient records into the care process. Record access has been greatly increased. There are years of data recording increased record accuracy and improved efficiency in data entry. Many of the hoped for results have been realized.
Yet, this past year, the Province of Ontario presented a grant pool of millions of dollars to support possible solutions to wait-time reduction. It seems that even after all of the HIT implementations of the last eight years, patients are sometimes experiencing emergency-room waits in excess of five or even 10 hours (these grew even longer in the 2007 flu season). The back-ups waiting for hospital admissions are staggering and seemingly without a solution--even from "fully integrated" HIT systems. According to a 2006 survey by Ipsos Reid, 42% of Canadians surveyed felt that '"a patient wait time guarantee that would reduce wait times for key health services' was the most important to them personally."
So, why has IT not been the solution to the Number One issue for Canadians (lower taxes got only 19%)? Because the solution is not about the technology. According to providers at a number of well-digitized hospitals in Ontario, patient records and patient interaction are great in the ED--patients' records can be created, their histories can be accessed, and intake moves pretty quickly. In-patient care has also been greatly improved; medication conflicts are avoided, patients are correctly identified and prescribed, etc. These providers are quite clear that the breakdown is the connection between in-patient and out.
Because of the differences in workflow and practice, HIT systems are different in the ED than in the rest of a facility. And connecting the data from the one to the information from the other to create usable knowledge that would enable efficiencies is not as easy as just installing a data-mapping agent. It takes people and, more importantly, it takes changes in the ways those people work.
Emergency Department staff do not access bed-management resources. Why would they? Emergency care is just that. And as we have discussed earlier (Myth #1), healthcare providers focus on the immediate task at hand--the here and now. They only look for a clinic/bed assignment when they are done with a patient--and then it can become a rushed, time consuming task for staff both in the ED and on the wards. Perhaps, Upon initial diagnosis of a patient, the search for a clinic/bed assignment were begun (to run in concert with the continued efforts in the ED) in anticipation of an eventual need for admission. Then patients might move more easily and quickly into the inpatient population, and that would reduce the long queues of people waiting in the ED.

A non-technical solution to a problem of technology--Sometimes, Healthcare IT cannot do it all--and should not......

The Seven Myths of Healthcare Technology2. Technology Can Do Healthcare

"All involved in the healthcare industry will have to understand the entire healthcare system much more deeply than at present."
Dr. Steven F. Collier, CEO White River Rural Health, Arkansas, US

At the HIMSS GHIT conference this week, Doctor Steven F. Collier presented an overview of his work at White River Rural Health in Arkansas. As CEO, he has overseen the development and implementation of an EMR solution over the past 7 years. With a population spread across 10 counties living in towns (usually) of which the largest is less than 50,000, White River clinics are often the sole providers of healthcare to one of the poorest and most poorly served communities in the US. White River has been lauded as a model of best practices in implementation of EMRs to serve diverse, rural, under-resourced populations. The staff of White River has more than 6 years' experience in driving their technologists to develop and deliver solutions that would serve White River's understanding of their own business; so, when the doctor speaks of "understanding the entire healthcare system," he speaks from great experience and not a little pain.

As discussed in "Myth #1: Healthcare Does Technology," the very dialectics of healthcare are the opposite of those of technology. Healthcare is urgent, responsive and independent where technology development is planned, scalar, and collegial. Healthcare personnel neither understand nor welcome technologists in their environments--and the requirements of technologists do not track to the common expectations of healthcare organizations. Doctor Collier made clear that he is a healthcare provider. He spoke of results in clinical terms, of motivating problems as patient needs, and of success metrics in terms of patient wellness.

Technologists do not share this worldview. They speak of results in processing terms (operations/sec), of motivating problems as "needs for new calculus," and of success metrics in terms of technical performance. And while internal IT personnel may want to deliver solutions that serve the organizations' needs and timelines, external technologists are even less likely to succeed at healthcare. In addition to the cultural gap between technology and healthcare, they have the additional gap of being outsiders to the organization. And they are not just technologists--they are vendors.

Vendors have their own priorities, many of which are counter to both the organizational and the cultural interests of their healthcare clients. "Close-out" delivery of a completed solution is a goal of healthcare. Vendors in HIT have a number of self-serving goals that require more time and less closure. Often, technology solutions involve hastily or imperfectly written pieces of software which require more customization or "shake-down" time than a vendor might care to admit. As the HIT industry is growing so rapidly and as many of the vendor companies are so new, solutions are sold before they are fully developed. This leads vendors to use clients as "beta sites" to extend their development time while garnering revenue. Healthcare organizations are so eager for solutions and are often so averse to technology that they will buy "total solutions" with "full support and maintenance." This puts the IT infrastructure of an organization in the control of the vendor being paid to deliver it--where is the motivation for close-out in that? And vendors are businesses. They sell things; software, services, support, training. While interfacing with clients to deliver one product, they are motivated, even driven to upsell more products. which just pushes out the horizon to "solving the problem," when all healthcare wants is to solve the problem.

In a recent KLAS survey, of all organizations who reported a vendor promising tremendous ROI on an investment, not one was able to report actually realizing it within the original purchase. Not one.

So, if technologist can't do healthcare and if healthcare organizations can't do technology for themselves, what is the solution? Is there a third option?

Myths Will Wait--What's a PHR?

Today's column was supposed to be "The Seven Myths of Healthcare Technology; 2. Technology Can Do Healthcare." However, this evening was the opening reception for the Government Health Information Technology (GHIT) Conference in Washington, DC (June 4, 5 at the Ronald Reagan Center). It was hosted at the Canadian Embassy and there were senior representatives of some of the leading organizations in GHIT. It was a good time; There was talk--and rumour. And news.
The word was that there is an urgent push on the Hill to modify the HIT legislation to include Personal Health Record (PHR) with the Electronic Health Record. This led to an impromptu roundtable of core people from HIMSS and the GHIT conference to consider the question;

"Why a PHR? What Would It Be? How Would It Work?

There were different ideas of "Why." A regulatory thinker felt a PHR to be nothing more than a comfort to paranoid citizens who want to "control" their personal information, but that it could never be used as a trustworthy resource by a provider. A sociological view saw the PHR as an opportunity to gather massive amounts of statistical data to drive cultural, epidemiological and other study data--all identity-scrubbed, of course. Others saw it as a traveler's safety, providing accurate information for use in a critical-care situation.
The group came to see it as having value in each of these ways, but requiring two types of data; Mandatory information written and managed by authorized providers (drug interaction/allergies, for example) and optional data which the patient could modify or erase as they wish (treatment for a wart).
This dichotomy would have specific requirements. Among the many discussed, were;
Data requirements to qualify as a PHA (what information MUST be in the record to qualify as a PHA)
Standards of encoding
Clearly defined, documented and controlled workflows for access, addition, modification and deletion of data.
Security and privacy considerations
Most importantly, a standard storage, transportation and access method
This storage method must have access control; For instance, a method to "sign" for "consent to view" by the patient.
A physically secure method to "sign" for "access to write" by providers.

One method posited for serving these requirements was a smartcard. Specifically, a smart-Social Security-card. As a number is issued at birth (or immigration), data collection can begin shorty after the card is issued, ensuring that even children have a PHA to carry into adulthood.This would, however, have to be a very smart card. It would require two physical sections; one secured against patient access and the other enabled for patient access. These securities would have to be "countersigned" biometrically, perhaps in combination with passcode. There were some high-level workflows discussed as well.
A patient enters a new facility and meets a new provider. They present their smartSScard and swipe a fingerprint reader to authorize read-only access to their data.
Following on treatment, the provider wants to add data to the record. Some may be defined as "mandatory." For example, a new drug allergy is discovered. The provider must also identify biometrically, with crosscheck at a centralized, government record. This data would then be written to the "secured" side of the smartchip.
Some of the data may not be mandatory; Perhaps, notes of elective plastic surgery. These would be written to the "optional" side of the smartchip. All of the data would still be entered into and maintained in the regular medical record.
This is only one workflow. There are still questions; How would a PHR be used when traveling? Would an international standard be developed? How could providers from outside the system add data? If the data is pertinent to critical care, how to access the data when the patient is unresponsive? How could data be added without the specialized equipment needed to access the card, contact external databases and read biometric data? What, if any liabilities could arise from use of PHR data? How would PHR data be validated with the EHR?

Lots of questions. Lots of ideas. The PHR/EHR discussions are picking up, and at GHIT there should be quite a few as they come into play with the panel discussion of remote and rural communities......

Let us know what you think....