Welch, Shari J. MD
Abnormal vital signs in the emergency department can be a harbinger of very bad things to come. A 1994 study at the State University of New York at Stony Brook correlated abnormal vital signs in the field with an increased likelihood of admission and death. (Acad Emerg Med 1995;2:863.) These findings were demonstrated again in 2008 at Groote Schuur Hospital in Cape Town, South Africa. (Emerg Med J 2008;25(10):674.) Sklar et al studied deaths of ED patients after discharge, and found that predictors of unexpected deaths included abnormal vital signs in the emergency department. (Ann Emerg Med 2007;49:735.)
But performance on the reassessment of abnormal vital signs varies widely in a national sampling. The Sullivan group found that staff did not reassess abnormal vital signs prior to discharge for 13.4 percent of more than 66,000 ED visits. (Ann Emerg Med 2006;48: Abstract 211.) Certain categories of chief complaints demonstrated even worse performance. Patients presenting with vaginal bleeding and abnormal vital signs, for instance, were discharged without reassessment of vital signs 33 percent of the time.
Clinicians' responses to abnormal vital signs in an emergency department are virtually unstudied, but a qualitative study done at the University of Western Sydney suggests that documentation and ineffective communication frequently kept the information about abnormal vital signs from reaching the attending physicians. (Aust Crit Care 2006;19:66.) Workload, distractions, and interruptions led to another segment of missed abnormal vital signs due to human factors. These authors recommend educational programs and improved communication networks.
Suzanne Stone-Griffith, RN, MSN, the former assistant vice president for quality at the Hospital Corporation of America (HCA), tackled this problem at her organization with a three-pronged strategy:
* Established clear parameters for documenting and repeating vital signs.
* Formalized nurse education and training.
* Set up auditing and a feedback loop to nurses.
Over time, this 157-hospital effort saw improvement at every location so that it is no longer considered a pressing clinical problem.
It is worth parsing the details of the three prongs. Some organizations require that all patients have vital signs retaken within 30 minutes of discharge. Maybe that is too many required repeated vital signs, and does not allow for focusing on the high-risk subsets of patients. Does the young patient with an ankle sprain really require repeat vital signs at discharge? For perspective, you should know what is going on outside the United States regarding vital sign recording. Ellen Weber, MD, a professor of clinical emergency medicine at the University of California, San Francisco, took a sabbatical to study ED operations at the University of Sheffield in Great Britain. She describes a different model from the one widely used here. Patients at the University of Sheffield are streamed into two categories at intake: major care and minor care. Patients being treated for minor injuries and illnesses are not required to have vital signs taken at all! And discharge instructions are given verbally in minor care. (Summit Exploring New Intake Strategies, Feb. 25, 2010, Salt Lake City.)
The Sullivan Group has online risk reduction courses that are effective and popular with physicians and nurses. These include education about which patient groups are particularly at risk for adverse events. Dan Sullivan, founder of the Sullivan Group and an emergency physician-attorney intensely interested in managing risk, said his group's analysis of adverse outcomes and malpractice claims pointed to a failure to recognize and act upon abnormal vital signs as a common cause of medical errors and patient injury. “Several [Sullivan Group] clients working on patient safety initiatives have prioritized the vital sign issue and created system solutions to eliminate this problem,” he said. “This has resulted in a dramatic reduction in vital signs-related bad outcomes.”
Finally, the Sullivan Group helped HCA in setting up a feedback loop to clinicians. The feedback loop is one of the most powerful tools in quality improvement. It has been used to improve door-to-physician times (Am J Med Qual 2011;26:138), handwashing (Am J Infect Control 2010;38:575), resuscitation performance (Curr Opin Crit Care 2010;16:196), and a whole host of ED-related behaviors. Almost any provider behavior can be changed by giving feedback. Nurses at HCA whose performance was not within parameters were given feedback and education. The collected data also informed policy decisions at HCA regarding vital signs.
Technology can be employed as a tool to improve compliance by “forcing functions” at various levels. If a nurse enters abnormal vital signs, for instance, she must indicate whether the physician was informed, or the computer will not let her proceed to the next charting page. Temperatures in adults and blood pressure readings in children are frequently omitted, incomplete vital signs would be captured electronically. The person documenting them would be forced to acknowledge and respond to them. A physician trying to flag a patient electronically for discharge could be presented with an icon to alert him that the last vital signs taken were abnormal, and he could not complete the discharge until the abnormal vital signs were reconciled.
Todd Taylor, MD, a physician executive with Microsoft and an expert on information technology, said technology plays an important role in managing critical information in the ED. “The issue of failing to check vital signs prior to discharge, or worse, failure to address abnormal ones that were taken, illustrates our lack of innovation in managing the ever-increasing data deluge in medicine,” he said. “Further, vital signs are but one indicator of the potential for bad things to occur downstream. What about the follow-up of labs and x-rays? In one study of discharge summaries, 87 percent failed to mention a pending lab test.” (J Gen Intern Med 2009;24:1002.) With the advent of robust health information systems, we can now produce more information than we have the human capacity to review.
So can computers help? In fact, they can. What do computers see that even highly trained professional do not? Everything, and they can correlate even unrelated variables in ways humans just never can. Emergency physicians and nurses are particularly good at pattern recognition, but computers do it more consistently, without bias, and automatically!
For many EDs struggling with this problem, particularly in busy, high-volume EDs, managing abnormal vital signs will need to be standardized and the process formalized. Where robust information technology systems exist, electronic cues could notify emergency physicians of abnormal vital signs, particularly if they are present at discharge. Where technology cannot be employed, a foolproof way to cue this critical finding must be employed. Qualified Emergency Specialists in Cincinnati used an effective low-tech system. When abnormal vital signs or any other impediment to safe discharge occurred, the staff would place a red paper flag on the chart for the physician to recheck the patient. For good measure, the nurses also handed the chart to the physician.
The point is, the problem of patients leaving the ED with incomplete or abnormal vital signs bedevils most emergency departments. As physicians, we should care about the problem for the safety of our patients and as a risk-management strategy. Though the heavy lifting to improve the process will have to come from nursing leadership, we should support these efforts wholeheartedly and support efforts to improve. When these efforts began at HCA in 2004, the organization had only 87 percent compliance with rechecking abnormal vital signs. With sustained effort over time following their three-pronged strategy, HCA can now boast 97 percent compliance with their abnormal vital signs policies. That is good news for the six million patients treated at HCA facilities each year.
Comments about this article? Write to EMN at firstname.lastname@example.org.
Dr. Welch is a fello...Image Tools
© 2011 Lippincott Williams & Wilkins, Inc.