We’ve written multiple columns on the impact of interruptions and distractions on nursing and pharmacy and their relation to errors (see list at end of this column). In many cases physicians are the source of the interruptions. In fact, when a “Do Not Disturb” sign was piloted to assist medication administration by nurses (Pape 2005), all sorts of interruptions decreased with one exception: interruptions by physicians!
But physicians are also commonly on the receiving end as well – they are often being interrupted and those interruptions may lead to failure to return to tasks and other errors.
Carey Chisholm, MD and colleagues have written about the occurrence of interruptions on physicians for over a decade and have just published a new study on their occurrence in the emergency department setting (Chisholm 2011). The performed a time-motion study at 2 urban academic and 2 community emergency departments. Two-hour observation periods with 85 emergency physicians were included. They found that about half of each time period was spent on indirect patient care (eg. charting, reviewing test results, interacting with consultants, nurses, housestaff, etc.). Emergency physicians in the academic centers spent only 36 minutes of each two-hour period on average in direct patient care (those in community settings spent 41 minutes of each two-hour period on average), which included not only direct patient contact but also ordering tests and medications, doing procedures, or interpreting EKG’s. The remainder of the time was classified as personal and accounted for 6 minutes in the academic settings and 13 minutes in the community settings. All physicians cared for multiple patients simultaneously (median 5 for academic, 6 for community) but all experienced surges where they care for many more simultaneously. The median number of interpersonal interactions per two-hour period was 70 for academic physicians and 65 for community physicians.
So it’s pretty obvious that emergency physicians are both busy and multitasking and it should come as no surprise that they would experience interruptions frequently. Chisholm and colleagues found that emergency physicians in academic sites experienced a median of 12 interruptions per period and those at community sites a median of 6 interruptions per period. Of the interruptions, almost half resulted in breaks in task. These are interruptions that result in changing tasks. The authors note that both the times spent in direct and indirect patient care and the frequency of interruptions have changed little since they did similar surveys a decade ago (Chisholm 2000). That previous work had also shown that both the number of interruptions and breaks in task increased with the average number of patients being managed simultaneously.
Breaks in task are especially important because one may never return appropriately to the previous task. Even when using checklists (whether in healthcare or aviation or other industry) breaks in task may result in steps of a sequence being skipped or overlooked. That is one of the reasons that during critical activities pilots use the “sterile cockpit” concept and nurses or pharmacists use a similar concept wherein they flag themselves in some manner to prevent interruptions. In fact, a healthcare study done in Australia (Westbrook 2010a) showed that physicians failed to return to tasks 18.5% of the time after interruptions.
In the Westbrook study, 11% of tasks were interrupted (and 3.5% were interrupted more than once). The total time for tasks increased with interruptions. But, interestingly, when the authors corrected for a length of time of observation bias, they found that interrupted tasks were actually completed in shorter times! They speculated that physicians may be “catching up for lost time”. We would anticipate that such shortened duration tasks, rather than being examples of improved efficiency, might actually be especially prone to errors and omissions.
Another study of emergency physicians (Friedman 2005) showed emergency physicians at Toronto General Hospital were interrupted every 13.8 minutes on average (4.4 interruptions per hour) and that the rate of interruptions increased with increasing shift intensity. Half the interruptions were from nurses and a third from other physicians. While most interruptions did not require the physician to move to a new location, about 10% did require a move.
Some studies have shown phone calls and pages to be the most frequent type of interruption (Weigl 2011). Perhaps some of those could be avoided by making patient information available to the callers via other means. Of the face-to-face interruptions, it is not known how many involve patient care but Friedman et al. note that often the interruptions are of a personal or administrative nature rather than being related to patient care.
Emergency physicians were interrupted nearly three times more often than primary care physicians (Chisholm 2001) and experienced significantly more care interruptions, nonpatient interruptions, telephone interruptions, and breaks in tasks.
Another study (Jeanmonod 2010) showed that emergency physicians are interrupted more often in certain activities than others. For example, they were interrupted during charting or reviewing data about 50% of the time. Bedside interruptions were less common (26%) but had a negative impact on patient satisfaction. The majority of interruptions were initiated by another physician or nurse. Unlike the above studies, these authors found physicians rarely changed tasks after an interruption.
The most recent Chisholm study was based on data collected in 2006 so there remain unanswered questions about the impact of new technologies on interruptions. It might be anticipated that more widespread availability of CPOE and electronic medical records might reduce many of the interruptions related to obtaining patient-related information, such as telephone calls. However, those same technologies may introduce other types of interruptions (eg. overzealous use of rules-based alerts and reminders, need to fumble through multiple screens, etc.). And the impact of text messaging is unknown. While focused use of text messaging could conceivably promote asynchronous communication that might be expected to reduce interruptions, experience in the real world outside of healthcare suggests that text messaging creates far more interruptions and distractions.
The concept of the “no interruption zone” that is commonly used by pharmacists during medication preparation or by nurses during medication administration may also be used for physician activities in certain settings. Bright yellow laminated signs were used at a New Mexico hospital to ensure that no one entered patient rooms during central line insertion (Hill 2010).
Our August 25, 2009 Patient Safety Tip of the Week “More on the Impact of Interruptions” cited an excellent article on the impact of interruptions on medication administration errors (Westbrook 2010b). That article also had a good discussion of strategies to minimize interruptions and future directions for research on reducing the occurrence of interruptions.” highlighted some excellent studies on interruptions and distractions in healthcare. In that we provided references to some good statistics about the frequency and nature of interruptions and distractions and listed some recommendations about how to avoid them. We also discussed some of the cognitive research on how memory works and how interruptions affect what stays in your “working memory”, a key concept in error occurrence. Our May 4, 2010 Patient Safety Tip of the Week “
Prior Patient Safety Tips of the Week dealing with interruptions:
Pape TM, Guerra DM, Muzquiz M, et al. Innovative Approaches to Reducing Nurses’ Distractions During Medication Administration. Journal of Continuing Education in Nursing 2005; 36(3): 108-116 May/June 2005
Chisholm CD, Weaver CS, Whenmouth L, Giles B. A Task Analysis of Emergency Physician Activities in Academic and Community Settings. Ann Emerg Med 2011; published ahead of print January 31, 2011
Chisholm CD, Collison EK, Nelson DR, Cordell WH. Emergency Department Workplace Interruptions Are Emergency Physicians “Interrupt-driven” and “Multitasking”? Academic Emergency Medicine 2000; 7(11): 1239–1243
Chisholm CD, Dornfeld A, Nelson DR, Cordell WH. Work interrupted: A comparison of workplace interruptions in emergency departments and primary care offices. Ann Emerg Med 2001; 38(2): 146-151
Westbrook JI, Colera E, Dunsmuir WTM, et al. The impact of interruptions on clinical task completion. Qual Saf Health Care doi:10.1136/qshc.2009.039255
Friedman SM, Elinson R, Arenovich T. Emergency Physician Work, Communication and Interruptions: A Human Factors Approach. Israeli Journal of Emergency Medicine 2005; 5(3): 35-42
Weigl M, Müller A, Zupanc A, et al. Hospital doctors' workflow interruptions and activities: an observation study. BMJ Qual Saf qshc.2010.043281Published Online First: 6 January 2011
Jeanmonod R, Boyd M, Loewenthal M, Triner W. The nature of emergency department interruptions and their impact on patient satisfaction. Emerg Med J 2010; 27: 376-379
Hill P. What is the initiative? CLABSI Prevention: Assure no one entered the patient room during a bedside central line insertion. New Mexico Hospital Association. Patient Safety Newsletter. January 2011
Westbrook JI., Woods A, Rob MI., Dunsmuir WTM, Day RO. Association of interruptions with an increased risk and severity of medication administration errors. Archives of Internal Medicine. 2010, 170(8): 683-690.