Patient Safety: Is Technology Enough?
Since the 1999 publication of the Institute of Medicine’s “To Err is Human: Building A Safer Health System,” patient safety has become an industry buzzword and a critical area of focus for healthcare providers. This study shocked the industry with the reported large number of near misses and adverse events that threatened the health of patients nationwide. According to the report, errors result in the deaths of 44,000 to 98,000 hospitalized patients and greater than 1 million injuries per year in the United States.
Since the 1999 publication of the Institute of Medicine’s “To Err is Human: Building A Safer Health System,” patient safety has become an industry buzzword and a critical area of focus for healthcare providers. This study shocked the industry with the reported large number of near misses and adverse events that threatened the health of patients nationwide. According to the report, errors result in the deaths of 44,000 to 98,000 hospitalized patients and greater than 1 million injuries per year in the United States.
Equally concerning were the results of a 2003 Rand Health study, which concluded that patients only received 55 percent of recommended care for preventative care, acute care and care for chronic conditions. The reality appears to be even worse when we examine the degree of underreporting, which is often as much as a factor of 20, according to a 1995 study entitled, “The incident reporting system does not detect adverse drug events: A problem for quality improvement.” This underreporting could be due to shame, fear of punishment or lawsuits, the time required to complete the reporting or the lack of recognition of the seriousness of potential injury.
In addition to the obvious negative impact on patient safety, medical errors also burden healthcare institutions with significant financial costs. On average, one adverse drug event adds $2,000 or more to the total hospitalization cost, according to a 2004 study published in the Journal of the American Medical Informatics Association entitled, “A Trial of Automated Safety Alerts For Inpatient Digoxin Use With Computerized Physician Order Entry.” Per year, that becomes $2 billion in incremental hospital expenses according to a 2003 study published in the Journal of Nursing Care Quality. The complete picture is even worse once one includes additional costs of medication errors including malpractice insurance premiums and reductions in employee productivity.
Detection and Analysis
As an industry, healthcare committed to, and has succeeded in, identifying the sources of these medical errors. According to “Patient Safety: Achieving a New Standard for Care” (2004), an analysis of various reports revealed that errors were commonly associated with the following:
- Poor understanding and application of drug therapy (30 percent)
- Incomplete or inaccurate knowledge and application as it relates to patient factors that affect drug therapy (29.2 percent)
- Incorrect mathematical calculations or data input (calculations, decimal points, units) (17.5 percent)
- Miscommunication of drug dosing (incorrect drug name, dosage form or abbreviation) (13.4 percent)
But understanding where and when these errors happen is only half of the solution. One must also understand why these errors happen. Simply put, these errors occur because we depend on the frailty of human memory and performance–aspects that can change daily, or even from hour to hour as providers become tired, stressed, preoccupied or interrupted in mid-thought. The risk associated with this lapse in consistent, quality care is amplified by the prevalence of extremely powerful drugs in the critical care environment, thereby setting the stage for a simple mistake to lead to serious consequences.
Furthermore, these providers have been hindered by a medical record system that is limited to static data captured on paper, handwritten and verbal instructions and non-centralized storage locations. Thus, clinicians rarely have timely access to complete, accurate and decipherable patient data at the point of care.
Optimize the Environment
Now that we know what, when and why these errors are occurring, how do we prevent them from happening in the future? First, we must examine the care delivery environment and make the changes necessary to prevent future medical errors from occurring. Healthcare organizations often begin with the implementation of computer physician order entry (CPOE), an electronic prescribing system that removes the potential for legibility errors, automates dosage calculations, analyzes possible drug interactions, and verifies drug nomenclature.
CPOE also contributes to broader clinical decision support systems by supplying updated, drug-specific information and other recent research that could impact the clinician’s treatment plan, as well as evidence-based order sets for the most common hospital and emergency department diagnoses, procedures and illnesses–capabilities that are not possible with paper-based systems.
CPOE is typically a component of a larger electronic medical record solution, allowing physicians to integrate their electronic prescribing with patient lab results, transcriptions and other aspects of patient medical records. This integrated system for managing, accessing and sharing patient data addresses many of the causes of patient errors.
Beyond the IT Promise
The movement toward integrating technology solutions into the healthcare delivery environment has been a paradigm shift for the industry as well as for each individual organization. During this metamorphosis, healthcare executives must be cognizant of important variables that can impact short and long term success.
Many organizations underestimate the complexity of deployments for healthcare technologies such as CPOE or EMR. These adaptable systems can be customized to fit the unique workflows of the environment, which can vary widely between two departments such as Emergency and ICU, let alone between different facilities. Moreover, the degree of sophistication and automation of the IT solutions already in place can also cause significant variance in workflows. These technical issues add to the challenges inherent to a technology implementation.
However, the largest hurdles are not always technical in nature. Clinicians have proven to be highly sensitive to changes in their workflow and will often have difficulty coping with new technologies, especially if they appear to slow down or otherwise complicate the delivery of patient care. Hospitals and other healthcare organizations are well aware that a widespread lack of adoption can cripple the rollout of valuable healthcare technologies.
For many organizations, it is easier and more effective to alter existing workflows to build in quality rather than to create new teams and new workflows. These enlightened organizations view this transition as a transformational process wherein the implementation team identifies the workflow changes needed and manages those changes by including clinicians in every step of the process. Important aspects include the completion of a comprehensive analysis of the clinician workflow prior to software selection; inclusion of clinicians in the planning and implementation process; early solicitation of clinical champions; and, open communications about usability issues.
The importance of the implementation process, particularly the integration of clinical champions, cannot be underestimated. In many instances, the same well-established CPOE product provided by a specific vendor might be successfully deployed at five hospitals, yet result in failure at five other hospitals due to poorly planned implementation or lack of champions.
The financial aspects of any implementation also are critical to success. Those who have lived through a technology deployment understand that the process can be slow, costly and frustrating if expectations regarding timelines and ROI have not been clearly communicated and commitments clearly outlined between the vendor and the organization.
Lastly, the lack of industry standards in the data, workflow and care delivery continues to be an impediment to the adoption and implementation of healthcare technologies that enable organizations to reduce medical errors and, thereby, improve patient safety. The demand for healthcare data standards, in conjunction with systemwide interoperability, only continues to increase as the nation endeavors to create a national health information infrastructure.
The healthcare industry has made great strides in identifying sources of medical errors. Indeed, many organizations have already reduced the human source of errors by implementing electronic prescribing systems, integrated electronic medical records and expanded clinical decision support systems. Less obvious, but equally important to the success of these implementations is the attention given to clinician adoption, usability issues, workflow customization and realistic ROI expectations. Through this holistic strategy of focusing on each variable that influences patient safety, the healthcare industry will continue its transformation into a safe, high-quality healthcare delivery system.
JF Lancelot is president of technology and operations for San Diego-based CliniComp International. Contact him at [email protected].
