Why Workflow Engines Must Work Right

Physicians work differently. What they do and how they do it depends on their specialty. Radiologists and cardiologists, general practitioners and orthopedists — like brothers, sisters and cousins; dads, moms and grandparents — all are in the same family.

People routinely take pictures that show things in different ways. Yet they can all be posted on Instagram and Facebook and sent by e-mail, cell phone, even Twitter. Now enterprise imaging wants to do for physicians what social media has done for the public.

Not surprisingly, there are challenges. One of them is really big. It has to do with the different ways that physicians work.

People in different walks of life can post pictures easily, because regardless of what they photograph, their pictures are all taken the same way. Radiologists take a multitude of pictures — magnetic resonance imaging (MRI), computed tomography (CT) and single photon emission computed tomography (SPECT), to name a few — but the work that underlies taking these pictures is always the same. This led years ago to DICOM, which is as close to being a hero as a communications standard can get.

DICOM saved radiology from information overload and markedly increased productivity. In the process, it made possible the most successful healthcare IT networks ever developed, ones that — quite frankly — proved that medical records could be digitalized.

Re-engineering PACS

Now it’s time for the next phase — the ingestion and management of many types of medical images. Given the extraordinary success of picture archiving and communication systems (PACS), and human propensity for making the future from the past, it will be tempting to make enterprise imaging in the likeness of PACS, retrofitting its components and reorienting them to accommodate new management, workflow and visualization in nonproprietary ways (i.e., vendor neutral). That idea has merit, so long as the abilities of individual physicians are not impaired.

Critically important is developing workflow engines that allow individual physicians to do what they do best. Ideally an enterprise imaging network would rely on a single workflow engine, one that harmonizes how different physicians care for their patients; allows the identification and exchange of different types of images; weaves these images into the fabric of data storage; and supports their management by vendor neutral archives (VNA) and integrated imaging platforms. The trick will be doing so without causing any significant adverse effects.

I know from personal experience that bad things can come from good intentions.

In the late 1950s, the St. Lawrence Seaway opened, clearing the way for ocean-going ships to travel from the Atlantic through the Great Lakes. The Seaway opened the door not only to commerce but sea lampreys, which devastated the trout population of the Great Lakes. These revolting creatures, called “vampires of the sea” because they latch onto and suck the lifeblood from their prey, caused an explosion in the population of alewives, which had been held in check by the trout.

Of Lampreys and Alewives

Speaking as someone who, as a child, pulled a lamprey from his leg in waist deep water, the alewives were worse. These saltwater fish, which followed the same path as the lampreys, peaked in the sweltering summers of the mid-1960s, died in droves, washed ashore — and rotted. Bulldozers pushed their carcasses into piles and dump trucks took them somewhere out of town to be buried. But the stench was inescapable.

Life didn’t return to normal in my hometown until ecologists found a way to destroy the lamprey larvae and introduced coho salmon to eat the alewives. Today the lampreys are in check and ports along the Lake Michigan shore host coho fishing derbies.

Just as the locks of the Great Lakes produced an economic boom for the Midwest, developing workflow engines that harmonize the various practices of medicine could improve and reduce the cost of patient care. Mistakes in healthcare flow through knowledge gaps that can only be plugged with teamwork. Better communications and increased collaboration are the stuff from which these plugs are made. But before putting in place the needed workflow engines, it makes sense to examine whether or how these engines might negatively affect the way physicians work.

Enterprise imaging needs an environment stocked with tools that allow free-thinking experts from different specialties to develop a consensus on how best to manage the patient. The challenge is to create a common image management structure that accommodates the diversity of medical practice and supports best-of-breed tools.

Like the locks and canals that connect the Great Lakes, workflow engines could inadvertently do harm, if they take away the uniqueness that make specialized physicians special.

It would be nice to see medicine go straight to the derbies — and skip the lamprey and alewife phases.

Editor's note: This is the third blog in a four-part series on enterprise imaging. The first blog, “The Impracticality of a Truly Universal Viewer for Enterprise Imaging,” can be found here. The second blog, “Will Enterprise Imaging Save Hippocratic Medicine?” can be found here.