Monthly Archives: July 2009

What if? What if we expand the definition of EMR?

What if I were to make a note of those occasional strange things going on in my body? You know, those annoying aches and pains that are hardly worthy of notice at the time. On the other hand, what if just a few are precursors of something going wrong? What if a few small symptoms could signal the need for attention to prevent a heart attack, stroke or some other potentially deadly disease? Or just identify something, the progress of which can be slowed by early medication? What if thousands of people made occasional notes in their PHR and these notes were examined after they had significant medical events? Most posts would probably be insignificant, but what if some proved to be useful? We have little or no access to that kind of data today, but we could.

What if a large number of yoga places were to log attendance electronically (mine does now) and then periodically post when attendees took what level and type of class to their PHR? (With their permission, of course.) Again, a large amount of data that might correlate to risks or significant benefits. There are at least two useful ways of looking at that type of data. Does yoga create a risk or benefit? And, does an unrelated change in medication or lifestyle have more or less impact on people who practice yoga than people who have no record of similar activity.

What if I could use an iPhone application to keep track of when and how long I bicycle, my top speed and average and how many feet I climb? What if I could add the results from my heart monitor? And what if I did that for several years? Risks? Benefits? Correlations with medical events?

What if I went to an acupuncturist and he included notes about what we did and why and I made notes before and after about what was happening to me?

All of this, and certainly more, is possible. Indivio, HealthVault and Google Health and their expanding network of participants such as CVS and Walgreen are moving us in that direction.

Note that all of the data collection discussed here is outside the normal scope of electronic medical records and that the value of almost all of it arises out of the ability to correlate it after a medical event with data that is within the scope EMRs.

Narrow-scoped EMRs will be able tell public health authorities how many people go to the hospital for a spreading virus like H1N1 and how long they are there. EMRs will also report how many people go to see their doctor and whether they go multiple times. PHRs could add data about how many people self-diagnose themselves with the flu. They could tell public health how many of these self-diagnosed and how many doctor-diagnosed victims stay home and for how long or go to work despite their symptoms.  Also, which family member brought it home and how did it progress through the family? There has to be a pony or two in there somewhere.

Expanding the scope doesn’t mean that the amount of effort or cost to system owners will increase. Proprietary systems will have to develop interfaces to other proprietary systems, add one more to access Indivio based systems won’t have much impact. Expanding the scope may even decrease costs a bit by moving some application development and maintenance from proprietary systems to PHRs. At a high level the cost benefit is potentially very attractive. It would be a shame to miss this opportunity because of a lack of creativity and overly narrow focus.

The Innovator’s Prescription #2

The Innovator’s Prescription (InnoRx) addresses the current state of electronic medical records: Role of Information Technology in Coordinating Care. (p 130) It deals with the complications and a possible solution. This post includes quotations, summaries from the book and comments by the author.

“As has proven to be the case in most of our other studies of innovation, there are predictable, rational reasons why good people aren’t doing what the rest of us feel they ought to be doing—and equally relational reasons explain why, for more than a decade, EHRs have always seemed to be just a few years away …

“The job that an EHR is designed to do is a systemic job, not [just] a local one. It is designed to enable different providers in different locations to see what kinds of care other doctors and institutions have given or are rendering to a patient. It would be an extraordinarily selfless act for the independent physicians’ practices that care for over 60% of America’s population to invest in and adopt the EHR systems that would make it easier for other caregivers to care more effectively for their patients. [Emphasis added] For many providers, patient records can even serve as a strategic asset, since paper records increase switching costs for patients.  [Costs in terms of dollars but also time to recreate portions of the record with a new provider and an almost certain loss of some quality in the record. But these costs are incurred today each a provider hands off a patient to a specialist.] …

“For the average doctor who gets paid on a per-transaction basis, writing paper prescriptions and keeping paper records still cost less, and are a lot more convenient, than adopting an electronic health record.” (p131-135)

The authors draw an analogy between enterprise resource planning systems (ERP) and EMR.  They both have to be designed to (a) optimize the system and then re-organize operations to fit the system or (b) the system must be optimized to fit the organization’s operations.

There are only a limited number of ERP systems in the marketplace, SAP is one of them. “Companies often pay firms like Accenture or Deloitte over $100 million to help them implement the SAP System. In the end, the company’s processes must give way and conform themselves to SAP. … It is very, very difficult!” (p 137)

ERP systems have become sufficiently standard that there are benefits in changing the way an organization does business. The two most important are: (a) the organization benefits from the prior experience of others to optimize the performance of the system and reduce the impact of periodic change and (b) the future cost of keeping the system current can be amortized over hundreds of installations instead of each installation bearing the full cost of upgrades.

Today’s pace of technological innovation means that almost any large computer system is obsolescent before it is fully installed and will be seriously handicapped in communicating with newer versions in a matter of just a few years. At this stage in the implementation and development process, there are no comparable levels of standardization or concurrent benefits with regard to EMR. As more and more systems are tailored to the needs of specific organizations it will become even more difficult and costly to standardize on a limited number of types of systems. As some systems are allowed to fall further and further behind, the interfaces for all will become more complex.

The pace of change is illustrated by a blog’s note that “SAP’s shift from 6 to 8 months for each enhancement package to 1 year may need to be reexamined in light of the quarterly pace of innovation found with the SaaS vendors.” Another simple illustration is that documents written and saved using Word 2007 cannot be read by people using Word 2003 without the aid of conversion software; relatively simple systems and just four years of differentiation.

The authors proceed by suggesting that the most likely course is “virtualization—essentially a technology for translating ‘foreign languages’ into a common one that allows previously incompatible formats to work seamlessly tougher.” They discuss VM Ware but not Microsoft Amalga.

“We see something like this beginning to emerge from … the Indivo system.” (see prior post: Indivo – part of the InnoRx disruption) “The Indivo system resolves [the problem] by inserting a layer of virtualization … It makes the data open, modular, and conformable, to that the applications using the data can be optimized. … [in the future] profits in the industry will be made by firms that build applications that use the data.”  (p 138 and 142)

See notes in the prior post that draw an analogy with the application flexibility of the iPhone. Other articles even suggest that gym workouts and other health related/non-medical information could be included to the patient’s PHR if the patient chose or perhaps their doctor’s urging. Who knows when or how this might be useful as part of the process of assessing outcomes: What role, if any does physical condition or exercise play in the results of a particular medication?

“We cannot overstate how important PHRs are to the efficient functioning of a low-cost, high –quality health –care system. … We thing that the Indivo system, or something like it, is a good place to start.” (p 143)

Indivo – part of the InnoRx disruption

The Indivo system is cited in The Innovator’s Prescription and showing up in other material I have been reading. It is part of the Children’s Hospital Informatics Program of Children’s Hospital Boston

I started this post to facilitate reference to Indivo. Found much more than I expected. See the link below that includes: Ten Principles for Fostering Development of an “iPhone­-like” Platform for Healthcare Information Technology.”

The Indivo system is essentially an inversion of the current approach to medical records, in that the record resides with the patients and the patients grant permissions to institutions, clinicians, researchers, and other users of medical information. Indivo is a distributed, web-based, personally controlled electronic medical record system that is ubiquitously accessible to the nomadic user, built to public standards, and available under an open-source license. …

Indivo is built to be extended and customized: users can connect their record to third-party applications that enhance the management and analysis of their health information. Indivo accelerates the development of these third-party applications by providing a core set of common features:

  • secure storage, categorization and aggregation of health data.
  • single sign-on and standards-based data-access delegation.
  • a simple, open, web-based Application Programming Interface (API).
  • unified user notification. …

Google, Microsoft and Dossia emerged as giant PCHR platform providers all sharing the basic PCHR platform approach. Microsoft Healthvault launched with Indivo open source code in the founding companies. Google later announced the Google Health model, building a platform around the Anvita (formerly Safemed) knowledge base and analytics. [links added]

Am article titled: An “iPhone-like” Platform for Healthcare Information Teechnology
June 18, 2009 argues “for the development of a platform model – similar in nature to the approach of the Apple iPhone – that would support an ecosystem of “substitutable” health care applications.”

The post includes: Ten Principles for Fostering Development of an “iPhone­-like” Platform for Healthcare Information Technology — truly disruptive.

Code Red

Code Red: How software companies could screw up Obama’s health care reform.
By Phillip Longman

This is a well written discussion of open source systems, specifically “the largest and most successful example of digital medicine is an open-source program called VistA.” It notes that, “For nearly thirty years, the VA software’s code has been continuously improved by a large and ever-growing community of collaborating, computer-minded health care professionals, at first within the VA and later at medical institutions around the world. Because the program is open source, many minds over the years have had the chance to spot bugs and make improvements. By the time Midland installed it, the core software had been road-tested at hundreds of different hospitals, clinics, and nursing homes by hundreds of thousands of health care professionals.”

It then goes on to contrast the failures of proprietary systems developed commercially.

What purports to be a defense of open source software does a better job of explaining how complex and difficult it is to develop and install an EMR system. Almost 30 years ago, enterprise resources planning systems, ERP, were the knights in shining armor that would save American industry. They were difficult to develop and brutal to install. Even today, they are difficult to install. Because of the life and death consequences of EMR systems and the sheer size of a nationwide set of linked data bases, an EMR system is more difficult to develop and deploy than ERP. We may need to rethink our objective of implementing fully functional EMR systems in the next few years.

I am optimistic. Our technology and our skills are vastly better than they were 30 years ago when ERP was being developed and VistA was starting. We can do a lot more a lot faster now. But there are limits on what we can accomplish in a limited period of time. A single example of an open source system does not justify relying on just one technology model, i.e., open source. On the other hand, I agree with Philip Longman that success with the system does justify doing whatever is necessary to keep it as an option.

Would we be better off at the national level to concentrate on a limited set of applications, e.g., fully integrated management of lab reports and pharmaceutical orders and results? Can we link larger applications to a backbone built on HealthVault or Google Health? Are there modular approaches we can take? Are personal health records a better place to start? We need to keep working on solutions but we also need to keep working on the questions that determine what solutions we can expect and when.

Comments on the same article by Bill Crounse, MD Senior Director, Worldwide Health Microsoft:

The Innovator’s Prescription #1

The Innovator’s Prescription: A Disruptive Solution for Health Care by Clayton M. Christensen, Jerome H. Grossman, and Jason Hwang (New York: McGraw-Hill, 2009), 441 pages, $32.95

I am reading The Innovator’s Prescription (InnoRx) and tracking it on Twitter and the blogosphere.

As described in InnoRx, disruptive change is never led by the established members of an industry. Change led by members of an industry is simply the normal evolutionary process of business improvement. That kind of change is almost always gradual; disruptive change often has a gradual phase but eventually becomes revolutionary.

The value of studying disruptive change is not to find a prescription about what to do. InnoRX does not tell practitioners what to do. Instead, it provides models to use to examine the practice or business of medicine. The models provide both suggestions for improving the business—more revenue, less cost, better service—and radar warnings of threats. Forewarned is forearmed. Failure to pay attention to threats on the horizon leaves any business vulnerable to marketplace obsolescence.

Until recently patients’ options were pretty much limited to hospitals, doctors’ offices and variations on those two themes. Change in that environment is almost always evolutionary. Problems with America’s health care—particularly cost—have reached a point where evolutionary change is not adequate to meet current and future needs. The current state of health care is increasingly being referred to as a state of “crisis.”

Alternative practices are being developed. Most patients are skeptical about new and “unproven” alternatives so the rate of adoption of changes in clinical and business practices begins slowly. Because of this slow rate of change, these changes pose little or no apparent threat to current members of the industry. Using the models in InnoRx, identifying the potentially disruptive changes on the horizon, looking at vulnerabilities in current practices, and listening to the public/ corporate/ governmental expressions of concern about the current state of the industry provides new ways of assessing the threats and developing appropriate responses.

We can never be certain about the future. However, a carefully considered plan that is periodically updated provides a much higher probability of long term business success that relying on a late-term knee-jerk reaction to whatever the future may bring. InnoRx provides models and examples of threats useful in developing and maintaining carefully considered plans. It is a prescription for planning; not necessarily a prescription for action. It provides illustrative examples of disruptive threats based on experience in other industries and extensive analysis of healthcare. It provides illustrative responses to these types of threats and examples of the consequences of ignoring them. It strongly recommends using the examples to plan and, as part of that plan, to set triggers for more detailed planning and/or action.

Disruptive changes are initiated at the edges of an industry or outside of it. InnoRx uses MinuteClinc by CVS and other “retail clinics” (RCs) as an example of an external threat and provides models to assess that threat to current business practices. The existence of RCs does not require any action on the part of physicians. RCs will do whatever they do to expand their business and earn an acceptable rate of return on their investment.

On the other hand, physicians will almost certainly see some of their practice disappear. For many patients, but not all, RCs are cheaper (on the order of 30% cheaper), faster even without an appointment, and more convenient (up to 24/7). Physicians who see the threat and take appropriate action will minimize the loss and will develop alternative services to replace the lost revenue.

RCs offer links to personal health records (PHRs) such as HealthVault (Microsoft) and Google Health. Patients can now use an RC, have the results added to their PHR and upon their next visit to their primary care physician he/she can have—between the PHR and the physicians’ records—a complete medical history. A physician who participates with the patient’s PHR can update their records so the next time they go to an RC, the RC will have a complete medical record to work with. Most patients probably won’t change primary care physician just because their physician does not participate in a PHR, but if they do need to find a new physician, that participation will be a factor in their selection—particularly if they ask the RC for a recommendation.

A physician with a plan will see the threats coming and will take small, deliberate, less expensive and less disruptive steps to counter the threat, e.g., consider participating with one or more PHRs. A physician without a plan will probably not see a slow moving threat until it is too late to protect some of his/her established patient base. Again, the value of InnoRx is not in a set of prescribed changes for business practices as found in the book but in a plan to identify and think about responses to those threats.

Stay tuned.

Shakespeare & Christensen – “disruptive”

“Disruption is about the disrupting of incumbent companies. It isn’t about disrupting technology, disrupting customers, or any of that. It is about removing the floor from the other companies.

“There is a Shakespearean quote from Hamlet that keeps coming up in my head when talk about disruption is afoot:

For ’tis the sport to have the engineer
Hoist with his own petard: and ‘t shall go hard
But I will delve one yard below their mines,
And blow them at the moon: O, ’tis most sweet,
When in one line two crafts directly meet.
This man shall set me packing:
I’ll lug the guts into the neighbour room.

“Disruption is about the disruptor, innocently seeming, delving one yard below their mines and blowing the incumbent companies at the moon. The incumbent doesn’t see it coming until it is too late. The disruptor is just a ‘fad’. It is a crappy product with crappy customers… until it is too late.”

Source: Malstrom’s Articles News: “The game has changed, … and the way the game is played has to be changed.” -Iwata

Petard: a small bomb used to blow up gates and walls when breaching fortifications. More at:

EMR 2.0 on the horizon?

Electronic Medical Records 1.0 (EMR 1.0) is still in its early development and implementation stages. Out near the edges, EMR 2.0 may already be evolving. First a little context:

System development 1.0 — the process leading to EMR 1.0 — follows a methodology something like this:

  • Define a user group
  • Determine their requirements
  • Develop a system that satisfies those requirements
  • Deliver sequential improvement, i.e., versions

System development 2.0 uses a different methodology:

  • Define a capability
  • Develop a system that exploits that capability
  • Deliver it to the market place as “beta” version
  • Work in the marketplace to learn what is really wanted and needed
  • Stay in beta and make evolutionary enhancements as fast as you can

For this discussion, let’s use Microsoft Word as an example of a 1.0 system. The team in Redmond Washington developed Word 1.0. They then made minor changes, e.g., 1.1., 1.2. and a series of major changes that they stopped numbering several years ago. They listened to customers but the emphasis appears to have been to incorporate advances in the state of PC technology and the results of their own usability labs. In its era that worked and MS Word is the dominant system solution for word processing worldwide.

Electronic medical records (EMRs) – the real topic of this post – started as version 0.1. Basically a large number of standalone systems, most of which were developed for a single client or small group of clients. The primary focus was data capture and delivery within a single office of complex. There is still a very large number of those systems in use and for sale.

As the data was being captured it became apparent that sharing the data among a patient’s doctors would be valuable to the doctors and the patient. Efforts to add networking and sharing of information led to the dawn of EMR 1.0. The 1.0 model says, in effect: “We are developing an integrated system you can use when it is ready.” That version represents most of what is happening with EMRs today including the Administration’s inclusion of EMRs as part of the economic stimulus package and health care reform. A recent Twitter post defined the current status as: “an endless barrage of information and discussions and debates regarding various combinations of information.”

Google and Microsoft are, individually, taking 2.0 approaches. From here on I will use Google as my example because as a user of Google Health, I know more about and have easier access to what they are doing.

Google has taken some basic capabilities of the Internet and developed a rather simple model of a Person Health Record (PHR.) This 2.0 approach says and then asks “Here are some applications that could be part of a larger system. How can we work with you – patients and service providers – to make this and future enhancements useful to you? What role should we play? What role will you play?”

The market place is responding with uses that are almost all outside the current definition of the 1.0 version. Patients now have the option of linking 47 services to their Google Health profile. Whether or not these uses are valuable will be determined by the patients and the providers of services, not in some executive office or committee. Patients and providers will vote with their money, yes or no.

Google Health links to additional services

Google Health links to additional services