It’s Time for CDS

For many years, quality and safety improvements through health IT (HIT), and the clinical decision support (CDS) systems that help facilitate them, were regarded as healthcare curiosities that were clearly capable of making a difference to patients and providers, and successful to an impressive degree at some institutions, but not considered a necessity for all. Given a lack of financial incentives related to quality and safety, and the quirks and problems that sometimes accompanied an implementation, many institutions were understandably reluctant to make the leap to CDS-enabled electronic health record (EHR) systems.

In 2008 and 2009, all of that has changed significantly. Several waves of policy statements, national and state programs, and financial incentives from the Centers for Medicare and Medicaid Services (CMS) and private payers have brought quality and safety improvement through HIT ever more directly into the “must-consider” list for nearly all healthcare providers. It is clear that the necessity, the financial justification and the probability of success for CDS implementations have expanded significantly, and will continue to expand as requirements are further clarified and resources continue to develop.

Research and Policy Statements

In the realm of research, a study published by Rubin Amarasingham, M.D., and colleagues in the Jan. 26, 2009, issue of the Archives of Internal Medicine determined that “hospitals with automated notes and records, order entry and clinical decision support had fewer complications, lower mortality rates and lower costs.” This was one of the largest studies of actual outcomes and showed that, when physicians actually use the clinical information system to its fullest, significant improvements result.

CDS also gained recognition when an extensive report from the National Research Council criticized HIT systems for failing to provide more cognitive support for clinical decision making and problem solving. Fulfilling the Institute of Medicine’s vision for 21st century medicine will require increased cognitive support through the seamless integration of evidence-based practice guidelines, research results and patient-specific data into daily practice, according to the report.

Such candor and directness are bound to continue in 2009 as the government follows President Barack Obama’s intention to “wield technology’s wonders to raise healthcare’s quality and lower its cost.” The president mentioned HIT and quality improvement in both his inaugural address and his speech to Congress in February, and has made increased funding and national support a visible priority.

A Major Stimulus

Financial carrots and sticks supporting HIT and CDS have never been more prominent. First came the National Quality Forum’s list of “never events” — adverse events that should just about always be preventable. Then CMS, the nation’s largest payer, announced that Medicare would no longer pay for certain inpatient conditions and events that it deemed preventable. Although no hospital worker ever wants adverse events to occur, at times such events actually resulted in increased revenue as the hospital used additional resources to deal with such events; however, now there is a greater financial onus, as well as a healthcare responsibility, for hospitals to minimize such events. The federal government is now committed to providing both incentives and disincentives for high and low quality care delivery. Providers, in turn, are increasingly likely to create internal quality dashboards, measurement strategies and proactive interventions to avoid such events. Later in the year, CMS announced a 2 percent reimbursement bonus for providers who effectively use e-prescribing systems, adding direct bonuses for technology use to the quality conditions above.

Most significantly, of course, the American Recovery and Reinvestment Act (ARRA) contains unprecedented levels of funding, policy requirement and support for HIT and CDS that foster improved quality and interoperability. The ARRA includes a major $20 billion investment in HIT infrastructure and incentives. These incentives will encourage physicians and hospitals to provide “meaningful use” of HIT to facilitate quality enhancement, quality reporting and the exchange of health information. In turn, it is estimated that Medicare could save $12 billion over 10 years through enhancements in quality of care and care coordination, and fewer medical errors and redundant treatments.

Coordinated National Action

The ARRA legislation also re-affirms and codifies the Office of the National Coordinator for Health Information Technology (ONC) within the Department of Health and Human Services (HHS). ONC will have wider ability to provide leadership through: defining the terms of “meaningful use;” through guiding certification of HIT systems; establishment of interoperability standards; and, implementation expertise through the proposed HIT research center and regional centers. ONC, along with the Agency for Healthcare Research and Quality (AHRQ) and the HHS Personalized Healthcare Initiative, also provides leadership through the Federal CDS Collaboratory, a unique collective focused on ensuring broad enhancement, adoption and coordination of CDS across government agencies.

National activity for advancement of CDS extends past the government, in a number of industry and public/private initiatives. As the American Health Information Community (AHIC) transitioned into the new National eHealth Collaborative last year, it approved and released recommendations from the Quality Workgroup and a cross-disciplinary CDS planning workgroup. An AHIC Quality Workgroup vision statement released in November 2008 zeroed-in on supporting “clinicians and patients in making decisions at key points in care delivery.” Recommendations included: allowing national quality improvement priorities to shape the development of CDS tools; aligning quality measures development; CDS development; and, payment policy and evaluation programs. The vision statement projected the following timeline:

2008 to 2009: Pilot studies of standardized CDS are implemented, some of which are currently being realized through AHRQ grants and CMS pilot programs.

2009 to 2010: Best practices for patient-centric CDS are established.

2010 to 2011: Best practices for patient-centric CDS are incorporated into certification criteria.

2012: EHRs with CDS and other CDS tools are certified.

While timelines can often slip, these dates are consistent with the implementation dates of incentives in ARRA and work plans from the Certification Commission for Healthcare Information Technology (CCHIT). For its part, CCHIT has issued a new roadmap for expansion of its HIT certification programs, many of which will likely be critical for ARRA and other national reimbursement programs. The roadmap calls for the addition of certification criteria in Advanced Quality and Advanced CDS, with development scheduled to begin later this year.

The federal government is now committed to providing both incentives and disincentives for high and low quality care delivery.

Recommendations from the AHIC ad hoc CDS workgroup, advancing prior work from the 2006 Roadmap for National Action on Clinical Decision Support, focused on standardization of CDS data types and intervention elements for easier distribution and reuse; alignment of CDS targets with high-priority national quality targets; consideration of a national CDS repository to provide easier access to rules and interventions; collective CDS performance and experience measurement to make it easier to determine which interventions worked well and which did not; and, training on best practices in CDS implementation (all of these recommendations are available at the ONC Web site, www.hhs.gov/healthit). A number of forums for extending this work are in progress at this time.

The National Quality Forum (NQF) has been particularly active in exploring how to transform guidelines into effective practice through HIT. This has been a particularly difficult problem, considering the many stakeholders and processes involved in moving from a clinical guideline to specific, consistent measures and interventions that can be utilized and reported. NQF has recently formed an HIT department to further advance the potential of both quality reporting and CDS.

One critical element is NQF’s sponsorship of a second version of the Health Information Technology Expert Panel, which seeks to create a finite, manageable data set that allows for the consistent capture of information on quality across different vendor systems. NQF also has convened the National Priorities Partnership, bringing together 28 major national organizations to add focus and feasibility to these efforts, by identifying the highest-priority national healthcare performance and outcome targets.

Many other groups are contributing to the national efforts in both quality reporting and quality/safety enhancement through CDS. Some of these — by no means a complete list — include the American Quality Association alliance and the Hospital Quality Alliance; the National Committee on Quality Assurance; the Joint Commission; the American Medical Association’s Physician Consortium for Performance Improvement; the Scottsdale Institute; and, HIT organizations such as the American Medical Informatics Association, the Health Information Management Systems Society, and the e-Health Initiative.

Making CDS Work

The healthcare industry no longer regards CDS as a cure or panacea for every problem in healthcare. The question now is whether the healthcare industry can create a national blueprint and framework coupled with generalized case examples that allow healthcare professionals to move forward with CDS. Because many CDS ventures are difficult to extrapolate or generalize, healthcare professionals question if CDS results actually apply to their environment or situation. They also wonder if they can assume the multiple, but essential roles of CDS designer, champion, orchestrator, trainer and evaluator — all of which are needed to make CDS work.

In the next two years, the efforts of government, industry, and academic efforts are likely to focus less on specific alerts for clinicians and more on how CDS is working, in a variety of different forms, to achieve its promised results. A greater realization of: the value of different types of CDS (such as smart forms, order sets, targeted reference, and dashboards) in different situations; better recording of actual experience; forums to exchange practical lessons learned; and, the creation of sharable human-readable and machine-readable CDS formats, will go a long way toward helping CDS users learn from each other.

Such leadership and collaboration will be essential in order to understand where CDS can be most effective, most easily implemented and accepted, and most easily distributed to the diverse healthcare providers around the country. These collaborations are also starting to address applications of CDS to new problems, such as increased handoffs of care brought about by changes in staffing patterns and reduced residency hours.

CDS has a significant potential to facilitate improved quality, safer care and lower healthcare costs. In the past, the realization of that potential has all too often been perceived as a hit-or-miss operation, dependent on particular persons and environmental factors in a specific institution.

With the establishment of greater policy pressure, significant financial incentives, greater monitoring of quality performance, more communication among the many stakeholders in the quality enterprise, and continued development of new tools, resources and standards, both the necessity for CDS and its chance of success will continue to increase.