Patient handoffs can always be improved
What happens when a hospital patient's physician goes off duty and another physician assumes responsibility for the patient? Or when care is transferred to the next nursing shift? How about when a patient is transferred from intensive care to another unit?
Too often, an efficient and safe patient handoff does not occur, probably because health care professionals, unlike air traffic controllers and others who perform vital handoffs, do not receive adequate training in how to communicate during these transfers of responsibility and across different information systems, says Richard M. Frankel, PhD, professor of medicine at the Indiana University School of Medicine in Bloomington, and a research scientist at the Health Services Research and Development Center on Implementing Evidence-Based Practice, Richard L. Roudebush Veterans Affairs Medical Center and the Regenstrief Institutes Inc.
The solution, he says, is to teach physicians and nurses the proper handoff process, using a model based on principles of adult learning, effective feedback, and clinical experience.
A 2005 study by Frankel and his colleagues determined that poor communication in medical practice is one of the most common causes of medical error.1 A precise patient handoff from one physician or nurse to the next is critical to patient safety and care, says Frankel, a medical sociologist who studies physician communication.
"The safest method of transferring responsibility for a patient is a face-to-face handoff in which the physician or nurse going off duty talks directly with the physician or nurse coming on duty," Frankel says. "Computerized medical records can facilitate face-to-face handoffs. Body language and other crucial factors are lost when the handoff is done over the phone, and a written handoff may be difficult to read doctors have notoriously poor penmanship errors especially in numbers or decimal places are easy to make; and written notes are open to misinterpretation or misplacement."
Seek uniformity in handoffs
Uniformity in handoffs should be a primary concern for risk managers, Frankel says. The organization should have a standardized method for patient handoffs that is used by all caregivers, and in all clinical areas, he says. Unfortunately, only about 8% of medical schools teach effective handoff techniques to medical students, Frankel says.
"There is good evidence outside of health care in high-reliability organizations that when handoffs occur, they happen systematically, in the same way every time," Frankel says. "In health care, we see people doing it differently all the time. It's not because they don't care, but because they think their way works. It might work, but the inconsistency across the organization creates a patient safety risk."
Frankel offers this example he has seen in one hospital: In one unit, the day shift nurse makes an audio tape when he or she is going off shift, describing the patients and what they will need, and then the nurse covers the patients while the incoming nurse listens to that tape. The outgoing nurse is available to answer any questions before leaving. But for the midnight shift, the outgoing nurse makes the tape recording and then leaves immediately. There is no opportunity to ask questions of the outgoing nurse.
"So, there we have a variation by shift in the way in which the handoff is being done," he says. "That kind of unwanted variation is a red flag for adverse events and near misses."
An obstacle in health care is the lack of communication across roles, Frankel says. Nurses tend to devise their own procedures and physicians do the same, and too often there is no discussion of how to coordinate and consolidate practices, he says. Getting them together to discuss handoff procedures and agree on a standardized system would be a big step forward, he says.
Frankel also recommends studying incident reports to look for instances in which the handoff was insufficient.
"About 80% of all adverse incidents can be traced back to a communication breakdown in a handoff somewhere along the line of patient care," he says. "The incoming person isn't very active, doesn't ask a lot of questions, the outgoing person is in a hurry to get on to whatever they're doing next, is vague about the patient's needs, and then you have a prescription for adverse outcomes."
Use 'talk-back' format
Frankel endorses a concept commonly used in aviation called the "teach back" or "talk back." The idea is that the party receiving information repeats it to confirm that he or she received it correctly. If an air traffic controller gives an order to an airliner pilot to climb to 5,000 feet and turn to a certain heading, the pilot doesn't just say "OK" and proceed to do that. The pilot repeats the information back, confirming that he or she heard the information and giving air traffic control a chance to correct it if the original communication was in error.
"It's a positive feedback loop that creates real two-way communication between the parties, rather than one person speaking and assuming that the information is being received and understood correctly," Frankel says. "It has reduced the number of accidents in aviation, and I think it could do the same in health care if we made it a standard part of handoffs."
Handoffs also are risky during transitions of care, such as when a patient's care is transferred from a hospital to a physician practice or vice versa, says Amy Boutwell, MD, MPP, content director at the Institute for Healthcare Improvement (IHI), the nonprofit organization based in Cambridge, MA, that works with hospitals around the world to improve patient care.
"One thing we've learned is that we must not consider the handoff a one-way communication process from one care provider to another," Boutwell says. "The handoff cannot just be about one provider handing over the patient and passing on the necessary information. This has to be a two-way communication, which is the goal when you're talking about a handoff between two people in the same organization also. But when you're passing on patient care from one organization to another, in physically separate places, it becomes more difficult to ensure that you're getting that meaningful two-way exchange of information."
Track readmission rates
Frankel and Boutwell both suggest tracking readmission rates and adverse events as a way to monitor the effectiveness of patient handoffs. Boutwell suggests those data can be most useful when applied to a group, such as a hospitalist group within the hospital, rather than individuals. The data should be provided to the group to show how their handoff processes affect patient care, and how the group's rates differ from others within the organization.
Boutwell says one of the first ways to improve handoffs is to make them a priority. Too often, she says, clinicians know the right way to ensure a proper handoff, but they get discouraged by the barriers they encounter. They may try to call the home care manager or the family caregiver for a meaningful exchange of information, but they can't reach the other party so they just sign the discharge papers and move on.
"We have to make proper handoffs a priority the same way we make other issues a priority in health care. It has to become something that is not just a good thing to do when you have the time and there are no obstacles, but rather a priority that must be carried out properly no matter the difficulties you encounter," she says. "That can require a culture change, and risk managers can play a role in changing the way everyone in the organization looks at this issue."
Reference
1. Solet DJ, Norvell JM, Rutan GH, Frankel RM. Lost in translation: Challenges and opportunities in physician-to-physician communication during patient handoffs. Acad Med 2005; 80:1,094-1,099.
Sources
For more information about improving patient handoffs, contact:
Amy Boutwell, MD, MPP, Content Director, Institute for Healthcare Improvement, Cambridge, MA. Telephone: (617) 710-5785.
Richard Frankel, PhD, Professor of Medicine, Indiana University School of Medicine, Telephone: (317) 490-0150. E-mail: rfrankel@iupui.edu.