CDS: It’s About Process, Not Just Tools

Clinical decision support (CDS) is more than just tools to execute evidence-based care.

For many, CDS is synonymous with alerts, but it’s really about “the processes that you are using to support information flow and workflow related to targets, and how to do that well. The interventions are the means to an end,” says Jerry Osheroff, MD, principal and founder of TMIT Consulting.

This “vast misconception” about CDS was driven by Meaningful Use (MU) Stage 1 requirements, he says, for which providers interpreted the CDS “rule” objective to mean they needed to deploy an “interruptive alert.” Considering CDS as alerts “only emphasized negative connotations that they interrupt people,” he says.

MU Stage 2 requirements from the Office of the National Coordinator for Health IT (ONC) and the Centers for Medicare & Medicaid Services (CMS) tried to broaden providers’ decision support approach by requiring “CDS interventions” rather than rules. “Stage 2 is a step in the right direction, but the focus is still on the interventions and not the process.”

What is CDS?

If CDS is more than alerts and reminders, clinical guidelines, order sets, documentation templates, data reports and summaries and clinical guidelines, then what is it?

According to CMS, CDS is “health IT functionality that builds upon the foundation of an EHR to provide persons involved in care processes with general and person-specific information intelligently filtered and organized, at appropriate times, to enhance health and healthcare.”

The CDS 5 Rights framework, first published by Osheroff in a 2009 guidebook on improving outcomes with CDS, helps providers understand and improve the what, who, how, where, when information flow dimensions, which are essential to implementing effective CDS and quality improvement approaches.

To improve care processes and outcomes, the interventions must provide:

• The right information
• To the right people
• in the right intervention formats
• Through the right channels
• At the right points in workflow

ONC contracted with Osheroff to develop resources to help providers, regional extension centers and vendors improve care utilizing this CDS framework. The yearlong project concluded in September 2013, and the resulting CDS/QI resources are available at the CDS for Performance Improvement Collaborative website as well as HealthIT.gov.

CDS/QI worksheets are “the heart and soul” of these resources, says Osheroff. They help guide providers through all the opportunities to improve decisions and actions that determine performance on the targeted measure. This includes interactions with individual patients, both inside and outside any specific encounter, and opportunities for population management—such as through registries.

Osheroff, through his consulting firm, is working with a number of organizations on fine-tuning their CDS and QI processes and improve outcomes. “We help organizations apply our guidebook’s clinical decision support definition. This includes reinforcing the broad range of CDS interventions and helping them use the CDS/QI worksheets to document, analyze, share and improve their target-focused information flows and workflows,” he says. “And everywhere, we hear, ‘Oh my God, no wonder we have crappy performance. What we’re doing doesn’t make sense.’”

Redwood Collaborative and Others

California-based Redwood Community Health Coalition (RCHC) is harnessing these CDS tools with the goal of increasing its percentage of patients with their blood pressure under control.

Clinical and QI leaders from RCHC, eight community health centers and their EHR vendors are using the worksheets to map and enhance blood pressure control strategies. Project goals include the following:

• Improve blood pressure control and QI capabilities in participating community health centers
• Use CDS/QI worksheets to achieve goals
• Produce a blood pressure module to augment tools
• Prepare to widely scale project results to other targets, settings and stakeholders

The health center leads weekly calls with CMIOs, chief medical officers and quality improvement leaders, as well as offers webinars and other instructional material on how to benefit from worksheets. “It’s all about quality improvement, improving care processes and outcomes and getting people to the right channels at the right time,” says Osheroff.

His firm also is working with N.J.-based Trenton Health Team (THT) to improve targets on diabetes and blood pressure control. Similarly, he is working with clinical leaders to build shared excitement about implementing enhanced care processes and tools.

Preliminary results for RCHC are beginning to demonstrate improvement in the targeted measures. Participants report they have gotten value out of the worksheets, says Osheroff. “This process is a helpful process. It has spawned quality improvement initiatives, and spawned development and deployment of new sources.”

Diabetes QI

The North Carolina Regional Extension Center also is engaging in quality improvement activities to improve diabetes management among its citizens.

“Initially we wanted to pick the measures that essentially had the biggest bang to improve outcomes over time,” says Sam Cykert, MD, the center’s clinical director who spoke during an ONC webinar on quality improvement and CDS.

Aside from quality and quantity of life, diabetes costs an estimated $245 billion including $176 billion in direct medical costs, $69 billion in reduced productivity and $13,700 per year per patient.

The initial pilot in 2007—which entailed a partnership with Community Care of North Carolina, the state’s Medicaid program, medical societies and primary care practices—began with 18 practices.

“We trained quality improvement coaches to go into practices, work with them on being able to measure diabetes care and then work on their practice workflows and the roles of practice personnel so quality improvement could really get done,” Cykert says. Eventually they worked with 180 primary care practices to help them report monthly on diabetic quality measures.

Then, once EHRs took off, “our job [at N.C. Regional Extension Center] became getting practices on EHRs to be able to achieve Meaningful Use, which we would segue into our quality improvement work.”

As practices gained EHR functionality and progressed in MU, they saw greater improvement in their outcomes. Practices that fully attested to MU, used automated quality reports and actively facilitated QI activities performed the best in terms of controlling blood pressure, cholesterol and blood sugar levels.

Cykert cited one town in eastern North Carolina that managed to control 90 percent of patients with hemoglobin AlC levels of less than 7. “This is a town that has lost manufacturing jobs, has many uninsured patients and is economically struggling, and yet it is essentially producing benchmark numbers.”

Process played a role in this success. Consistent data, the ability to drill down into populations and using point-of-care reminders and templates only go so far. Even when generating a list of high-risk patients, you still have to know how to intervene at all points of care, he says. “I have to know what to do with patients when I bring them in and how to teach them about medicine, make sure they have self-management skills and make sure I have protocols where I can intensify care between visits.”

Sepsis Improvement

People, processes and products are important in CDS--in that order of importance, says Osheroff.

“If you don’t get the people part right, you are dead in the water. When people are engaged, they can collaborate effectively on understanding and improving care processes,” he says. Only at that point should intervention tools be explored, whether order sets, documentation templates, registry reports and other tools to achieve shared goals.

Johnston Memorial Hospital (JMH), a 116-bed hospital in Abingdon, Va., looked to its people in its successful quality initiative that decreased sepsis mortality and patient costs through multidisciplinary teamwork.

The hospital implemented a screening tool in the emergency room that flagged nurses if patients were meeting criteria for sepsis. They then developed an order set in-house to handle these cases.

“It’s had a pretty amazing impact on sepsis,” says Hughes Melton, MD, MBA, vice president and chief medical officer for the Virginia operations of the Mountain States Health Alliance, who presented findings last October.

For the tools to succeed, it required “changing the behavior of the whole hospitalist and emergency teams,” he says.

The hospital’s quality staff recruited physician leadership to assemble a team of physicians, nurses, pharmacy and organizational team members.

The team followed a people, processes and technology approach to guide this intervention. It first mapped the current process of sepsis identification and treatment, and identified the following gaps: no sepsis nurse screening at triage, no sepsis automatic alerts, no sepsis order sets available and no capability for management reports.

The team worked with JMH’s emergency department to bridge these gaps by building screenings, alerts and order sets into the EMR system. Mandatory education was rolled out for the entire care team.

The initiative yielded benefits in the areas of days, dollars and deaths, according to Melton, saving 26 lives, 175 hospital days and $1.6 million over six months.

Physician comprehension of the underlying processes was essential.

“You’ve got to have physicians understand those kinds of dynamics and be willing to work with clinicians to better utilize resources.”

These initiatives show that while CDS tools are important, it’s the processes in which the tools are used that drive QI success. With CDS an even greater priority for Meaningful Use Stage 3 and other quality initiatives, the stakes are high for getting it right.