A forum to discuss ideas, approaches, standards, and architecture to establish and support open interoperability among healthcare IT systems.

Wednesday, August 17, 2011

The Road [Not] Taken

"...I took the one less traveled by,
And that has made all the difference."
Robert Frost
I had the good fortune yesterday to having been afforded the opportunity to participate in a symposium hosted by both George Washington University and UT Health entitled The Role and Future of HIT in an Era of Health Care Transformation. The experience there catalyzed me to get off my laurels and make a few comments after what has been too long of a hiatus.

In a nutshell, the symposium (http://sphhs.gwumc.edu/abouttheschool/events/healthinformationtechnology) sought to convene a community from across the health care sector to consider the principal recommendations put forward in the PCAST report. The document released last December identifying what were identified as key challenges facing the US affecting health care, and where HIT can/should be focused to achieving national objectives in terms of improving our ability to manage care, improve outcomes, and engage citizens. (Ref: http://www.whitehouse.gov/administration/eop/ostp/pcast/docsreports ).

John Halamka has already done an admirable job summarizing what happened throughout the day, so there is no need to do so again. (http://geekdoctor.blogspot.com/2011/08/role-and-future-of-hit-in-era-of-health.html ). What I thought merited attention was the approach that was taken to come to conclusions, as this particular activity chose the road not often taken.

All too often, solutions, policies, mandates, or otherwise narrowly conceived ideas are developed in a vacuum, "matured", and then unleashed onto a much broader constituency which is expected to understand, to embrace, and ultimately to support. There are a few observations about decisions made in this way:

Visibility: We can only address those problems or challenges that we're aware of. Solution development happening among a narrow community naturally has less visibility into a problem space. As ideas are vetted and potential approaches considered, they are done in the context of the problem being solved. A significant challenge that healthcare faces, particiularly when considered in a multi-institutional or National scale, is that complexities grow and there are ever increasing considerations that were not taken into account.

Buy-In: A number of years ago, Steven Wretling from Kaiser-Permanente noted in a keynote session that "Culture eats technology for breakfast." I would assert that it isn't only technology that gets eaten. When ideas are put forward that have not had sufficient vetting (or where there hasn't been an opportunity for the community affected to participate in the vetting), there is ubiquituously pushback. Building into any maturation process the means for the affected communities to engage with the decisions, to weigh in and have their thoughts considered, and to affect the outcome is key for long-term viability and success of any initiative.

The Other Road

While it is not the easiest path (or some might argue the fastest path, but we'll come back to that), a consensus-based approach where input and pushback are actively sought as part of the vetting and validation process can be a much more effective means to an end. Let's explore that notion:

Defining Consensus. The first misconception about consensus is the perception that consensus is ubiquity -- in other words, that 100% of participants agree on an outcome. While this is clearly a happy place, when it can be achieved, it is not technically necessary to achieve a consensus. A consensus, simply put, means that an overwhelming majority can stand behind a recommendation or decision.

A simple majority may be enough to technically make a group decision, but it is generally not enough for that decision to be embraced. If 51% of a constituency agrees with something, then up to 49% does not, and without clear indicators, incentives, or penalties associated, that 49% has yet to be convinced.

Let's poke at the "simple majority" point above. When we don't have consensus, there is a natural tendency for those whom were not believers to either 'wait and see' or pushback. The result is that activities (and ultimately progress) languishes even after a "decision" is reached and deployed.

A consensus-based approach involves much more compromise (and arguably a lower target) as the elements of an approach or solution are identified that can meet the needs of a much larger audience are determined. Sometimes that means leaving something out of scope. Sometimes it means compromising on a little purity. The enemy of the perfect is good enough.

Consensus cannot be about perfect. The question is whether enough common ground can be reached so that perhaps nobody is happy, but [almost] everyone can live with the outcome. Key is to ensure that whatever compromises made do not adversly affect the intent of what was being sought -- is the result still fit-for-purpose.

The difference in this approach, however, lies in what follows a decision point. The value proposition has been thought through. The compromises and tradeoffs already made. Done well, the result is not only fit-for-purpose, but also broadly supported.

The PCAST Symposium

I applaud the organizers, hosts, and participants of the GWU/UT Symposium for having taken the other road. The PCAST report has been a "disruptive" force affecting HIT, and disruption can be either negative or positive depending upon what follows that event. This event included a broad spectrum of diverse stakeholders respresenting government, payer, provider, public health, consumer advocates, industry, academia, and others.

The event was about embracing the core message that PCAST identified: that opportunities abound about how to more effectively leverage HIT to achieve national priorities, and that we must step up and address those challenges. What had been lacking, however, was the stage to convene what are multiple diverse points-of-view and interests into a common ground: to converge the community, and to identify a plan that a diverse industry could support.

While a full consensus cannot be achieved in one day and with a limited audience, what I believe was accomplished was the first step down that path. Having brought together a thoughtful and diverse community to begin to collaborate, and capturing "nuggets" in the form of suggestions that are helpful, useful, and actionable, the seeds of what could be common ground may have been sown from which a much broader consensus can grow.

Reaching that common ground takes longer than does a 51% decision. For that reason, it is often the road less travelled. In this case, however, I believe it will make all the difference.