Trees Falling in the Forest | Pam Arlotto | Healthcare Blogs Skip to content Skip to navigation

Trees Falling in the Forest

October 4, 2009
| Reprints
A focus on data collection in 2011 to performance improvement in 2015

One of the major shifts that will occur with successful Meaningful Use will be movement from a focus on data collection in 2011 to performance improvement in 2015. This, however, will require significant rethinking by the healthcare industry of how we interact with the data within our electronic systems. A clinical quality improvement consultant recent told me a story that exemplifies this challenge. In order to conduct an assessment of an organization’s quality processes, he asked to audit 50 charts. He described his excitement when he learned, “they have an electronic medical record, so this should be easy”. Summary sheets were printed from the EMR for his review, and in utter disbelief he entered a room filled with paper. The record from one patient, a 30-day ICU stay, stacked from the floor to his waist. His initial review indicated “data” from Respiratory Therapy for one day totaled 15 “summary pages”. John Halamka sums it up, “The problem today is that we have too much data. We need to find a way to distill it to what the doctor needs to know, and provide actionable information.”

Most of us are familiar with the knowledge pyramid (Ackoff 1989).
It has been used for several decades to describe the hierarchical relationship between data, information, knowledge and wisdom. Each element can be defined as follows:

§ Data – discrete, objective facts such as who, what, when, where

§ Information – linking of who, what, when and where to tell a story

§ Knowledge – information that is culturally understood, such that it explains the how and the why about something or provides insight and understanding

§ Wisdom – placing knowledge in a framework to allow it to be applied to different situations

Unfortunately, we in healthcare are still, for the most part, focused on data – making it discrete, structured and standardized. Oh and of course, interoperable. Just think back to my colleague’s experience, this was one patient’s record from one institution. What if we had “exchanged information” from multiple sites and added that to the mix. The stack for one patient would have touched the ceiling.

Peter Drucker once said, “Knowledge is like the sound of the tree that falls in the forest when no one is there: it doesn’t exist unless people interact with it”. Today we are focused on planting seeds that might one day grow into a forest. As we plant those seeds, it is critical that we consider how clinicians, healthcare business leaders and researchers will interact with, and develop knowledge and wisdom to produce the benefits we anticipate from digital healthcare.

As we work toward Meaningful Use, we must begin to explore what Thomas Davenport in his book, Information Ecology, refers to as information behavior, " individuals approach and handle information. This includes searching for it, using it, modifying it, sharing it, hoarding it, and even ignoring it.” If you consider all of the components of the EMR, basic information management strategies should begin with a series of questions:

§ What customized views needed by specific users – specialists, management level, internal/external role, etc.

§ What workflow and decision rules help collect, modify and summarize

§ What are the roles for creation, contribution, review, approval, etc

§ What actions are needed to edit, cosign, view, refer, etc.

§ Where is standardization necessary and variability appropriate

§ How is the information best packaged and distributed

§ How do we communicate the value of new information management tools and approaches





Business intelligence software, as used in other industries, is a tool that helps to restructure data into intelligent information. Clinical information systems using "Report Writer" software gathers information using discrete data elements to provide either statistics or lists that can be analyzed and transformed — into intelligence.

That said, report writing software, requires human input for proper use and we should be cognizant of the machine/human interaction that transforms this discrete data into intelligence (i.e. people, using processes, on a particular platform).

Years ago, "data warehouses" was the buzzword in business — not sure what they call them now. But, within a data warehouse is a collection of relevant data elements with a common definition used for reporting and analysis.

One of the important factors for building a data warehouse was that all "data owners" have a common, shared "data dictionary" that provided definitions for the data prior to its use. Often, this was one of the most significant barriers to its adoption since there were (and will always be) a variety of opinion and viewpoints on what a specific data element meant.

My 2 cents.

Your 2 cents is worth a great deal. BI is critical as we expand our data mining and knowledge management tools. That said, we've only beguna to scratch the surface