Tuesday, June 23, 2009

A Debate on Gradual versus Radical Change

Following is a discussion I'm having on a LinkedIn forum about adopting a very gradual approach to HIT evolution versus implementing a strategy promoting evolutionary leaps coming from disruptive innovation and a focus on radical change.

Someone wrote: I have proposed in the past and believe you might agree that a better approach would be to create a model of data sets that could be sent and received by providers in order to "qualify" their systems. This does still require a data standard but it allows the vendors to create their own tools and for innovation to thrive. Without a well defined data structure, we…will never be in a position to leverage those innovative tools such as cognitive support.

And I replied: While I agree that a model data set would be useful for qualification purposes, I've learned from experience that they tend to be "minimum data sets" with minimal usefulness, instead of being comprehensive/complete multidisciplinary data sets that approach "maximum usefulness." The problem with such minimal data standards is that it sets the bar way too low because vendors typically make that minimum their ultimate goal, rather than developing tools that can handle the maximum. I contend, therefore, that we ought to evaluate HIT tools in terms of (a) the depth and breadth of information they can manage cost-effectively and (b) whether that information enables patients and clinicians to make ever-better decisions by providing patient-centered cognitive support (PCCS). That is how innovation is sparked and meaningful change is encouraged; anything less promotes a glut of me-too commodities with minimal usefulness.

He also wrote: So, first we need to exchange information before we can take action on the data. In order to exchange the data we need a universally agreed upon data format. There are RHIOs and HIEs [i.e., organizations designed to enable data exchange in closed networks of providers] in place now that are functional but are not appropriate for a broader exchange of information. For the most part, those early implementations are proprietary systems that are scaled out to serve a region. While they have provided a valuable service, they need to adapt to a common data exchange that can be used to share data among other HIEs thus, a Nationwide Healthcare Information Network.

And I replied: Information exchange is certainly crucial. And I agree that intra-RHIO/HIE data exchange alone is not enough; there must also be inter-organizational information exchange across RHIOs/HIEs. While data and technology standards offer a way to handle information exchange challenges, they come with issues posing serious problems in terms of cost, effort, time, hassle, complexity, inefficiency, usability, reliability, information loss, political influence, etc. One innovative low cost solution we've been proposing is a publisher-subscriber node-to-node architecture with universal translation. I discussed these issue two years ago in a series of posts at this link.

He also wrote: Other than HIMSS desire to see CCHIT as the certification body, I believe their proposal is sound. That only brings us to the first step. I'm sure we would all agree that this will be an iterative process. ONC [The Health and Human Services' Office of the National Coordinator for Health Information Technology] needs to place a stake in the ground so we can move forward. Give us version 1 of a data object and we can start building Exchanges to get data to the EHR. Now we can start making use (even meaningful use) of the data and work toward improving healthcare. In time give us version 2 of the data standard and we will ratchet up functionality.

And I replied: Under the last administration—during the ONC NHIN initiative—$18.6 million was awarded to four large organizations, after deciding to renege on a promised 2/5 small business set-aside despite receiving many applications from small businesses around the country (see this link). Those millions, in the end, did not bring us any closer to an NHIN and left a very bad taste in the mouths of many innovative small businesses. Thus, it is important to guide ONC in guiding us by assuring that innovative ideas from "weak voices" (i.e., individuals and small businesses) be heard loudly and clearly, and be influential!

He also wrote: We have the data, we just need to improve the delivery and presentation of the data otherwise we'll be buried under information overload). We have the knowledge; we just need direction from the powers above. Give us clearly defined objectives and we will build an Information System that will be forward thinking and will enable us to use the patient's records as part of the treatment process. Keep us in the dark and we will go a hundred different directions, building dozens of perfectly functional systems that are each islands of information. One of the big failures of our healthcare system is not that we don't have the technology or the data; it's that we don't have the data where we need it when we need it.

And I replied: Yes we have data, but we do NOT have the data needed to create the necessary information and emerge the knowledge required to realize substantially improvements in care effectiveness and efficiency. And this goes well beyond the data distribution challenge. But even if we did have the necessary data, information, and exchange mechanisms, mainstream HIT tools would still NOT provide adequate decision making assistance because they fail to implement the patient-centered cognitive support process. Furthermore, I previously commented, there are serious problems with being overly reliant on restrictive standards on the "powers above."

So, we ought not to be squabbling about minimal data set and the use of conventional technology standards, but instead focusing our collective efforts on widespread collaboration to define the requirements for PCCS-enabled HIT that are able to manage comprehensive/complete multidisciplinary data sets with the potential to be maximally usefulness in both clinicians and patients/consumer in their decision making. I much rather see us going in "hundred different directions" in search of truly useful, disruptive HIT innovations, than to have the powers that be encourage adoption of the same old commodities that simply cannot do the job.

Having said that, there is a place for today's EHR and CPOE commodities, i.e., to define the strengths and weaknesses of the "AS IS" HIT model, so we can start building the "TO BE" model by adopting the most advanced and promising HIT tools that can overcome the AS IS model's deficits. These next-generation tools could be add-ons to today's commodity products, or they may be completely new types of tools that will replace what's currently available.

He also wrote: I'm an optimist and believe that we will pull it all together. The timeline that we've been given however is not realistic. How many practices are going to be in a position to satisfy the requirement that have not yet been issued before January 2011? The bonus payment that takes the 2011 reimbursement to $18,000 is based on meaningful use before January 2011.

And I replied: I contend that meaningful use of next-generation HIT can very well happen by 2011 and without great expense! Success will depend on the promotion and adoption of radical (disruptive, discontinuous) innovation. [Full disclosure note: My company is offering a novel cost-efficient way to exchange and evolve computational models in loosely coupled professional and social networks, which is an essential function for developing and deploying PCCS-enabled HIT.]

No comments: