Around the time my son Jason was heading off to begin college, he sent me a brief text message. The apparent purpose was to summarize some useful insights he'd gleaned in the run-up to this milestone. The message was, "the most important thing is to find the most important thing."
This kernel of wisdom can be very useful in translating the current flurry of activity surrounding health information technology adoption generally, and CDS more specifically, into productive action.
So what is the most important thing for our CDS efforts?
An attractive option is, "to ensure that it helps drive maximal, measurable improvements in care processes and outcomes for as many stakeholders as possible." This high-level goal can inform local implementation efforts, as well as broader research, policy, product development and other related dimensions.
The relatively brief (few-decades) history of CDS has been an exploration into the possible, but in these challenging times we must now focus much more sharply on the imperative. We have learned that pioneering organizations can harness CDS to make good things happen, but even thoughtful CDS deployments don't guarantee the results that the implementers expect and need.
Taken together, even the most widely praised and cited CDS success stories highlight the need for much tighter linkages between our CDS efforts and priorities for measurable healthcare quality improvement. There has been some reassurance recently that there is a there there. Also encouraging are reports from 'real-world hospitals' that their CDS efforts are yielding significant performance gains; e.g. a 15.6 percent mortality rate improvement, and 35 percent increase in patients receiving congestive heart failure core measure interventions.
We appear to be at an inflection point between the possible and the essential in our approach to CDS. Last year, the top recommendation from an AHIC ad hoc CDS workgroup to HHS and AHIC was to prioritize and align federal CDS-related efforts with national performance improvement priorities, monitor the effects of the federal CDS portfolio on priority outcomes, and widely disseminate the federal CDS priorities and effects so as to help entrain other CDS efforts in the goal-oriented motion.
In parallel with these policy developments, care delivery organizations are beginning to feel the imperative to improve care processes and outcomes, and leverage CDS effectively toward this end. This increasingly critical business need is driven in response to pressure from payers, regulators, accreditation bodies, and others. These latter stakeholders aren't being shy about turning up the heat to address some of the most important things they think should always or never happen during patient care.
These forces on providers led me to assert during a lively session at the AMDIS Physician Computer Connection last summer that before too long, CMIOs' jobs would depend on driving measurable improvements in outcomes key to their organizations. Many noted that this was not yet a top organizational priority, and that their current focus remained supporting successful HIT/CDS adoption, without a tight link to specific outcomes as an essential ingredient. This narrow focus on adoption is an important step in getting to where we want/need to be, but should be taken with the end goal in mind. Otherwise this 'successful implementation' step may not represent efficient progress toward ultimate outcome goals.
For those who do believe that the most important thing for CDS is to improve top priority outcomes, how do you do that? ONC and AHRQ asked that question from a policy perspective in 2005, and an answer came in the form of a Roadmap for National Action on CDS. This roadmap has informed actions at the federal level, such as the AHIC ad hoc CDS Workgroup, and its recommendations mentioned above.
For care delivery organizations, where the rubber meets the road, there have been a variety of efforts to synthesize and disseminate best practices for improving priority outcomes with CDS. HIMSS has taken a leadership role in key parts of this, including publishing (just before this meeting) a guidebook on improving medication use and outcomes with CDS. If this 'outcomes-focused CDS business' isn't the most important thing, it has at least touched a big nerve; nearly 100 contributors, several leading informatics and medication societies, AHRQ, and several CIS vendors participated in the initiative. Follow-on collaboratives led by HIMSS and others are building on this work.
Is measurably improving priority outcomes the most important thing for CDS? Even if it's not, in some ways, your job, your life and the wellbeing of your loved ones may well depend on it. Let's all give this the attention it deserves, and get it right.
Jerome A. Osheroff, MD is the chief clinical informatics officer for Thomson Reuters, and serves on the faculty and clinical staff of the University of Pennsylvania. He's the lead author on the HIMSS CDS guidebook series and national CDS roadmap mentioned in this article, and served on the AHIC ad hoc CDS Workgroup.
