Journal of Critical Care 32 (2016) 170–174
Contents lists available at ScienceDirect
Journal of Critical Care journal homepage: www.jccjournal.org
The effect of a checklist on the quality of patient handover from the operating room to the intensive care unit: A randomized controlled trial Cornelie Salzwedel, MD a,⁎, Victoria Mai, MD a, Mark A. Punke, MD a, Stefan Kluge, MD, PhD b, Daniel A. Reuter, MD, PhD a a b
Department of Anesthesiology, Center of Anesthesiology and Intensive Care Medicine, University Hospital Hamburg-Eppendorf, Martinistrasse 52, 20246 Hamburg, Germany Department of Intensive Care Medicine, Center of Anesthesiology and Intensive Care Medicine, University Hospital Hamburg-Eppendorf, Martinistrasse 52, 20246 Hamburg, Germany
a r t i c l e
i n f o
Keywords: Postoperative period Patient handover Checklist Critical care Patient safety
a b s t r a c t Purpose: Handover of patient care is a potential safety risk for the patient due to loss of information which may result in adverse outcome. We hypothesized that a checklist for handover from the operating room (OR) to the intensive care unit (ICU) will lead to an increase of quality regarding information transfer compared with a nonstandardized handover procedure. Materials and methods: The study was conducted as a prospective, randomized trial in a university hospital. The quality of handovers with checklist was compared with handovers without checklist. Handovers were recorded by digital voice recorder and analyzed using an individual rating sheet for each patient. This enabled to discriminate between items that “must be handed over” (red items) and items that “should be handed over” (yellow items). Results: A total of 121 patient handovers from OR to ICU were included. Significantly more red items were handed over in the study group compared with the control group (study group: median 87.1%, 25-27 percentile 77.1%90.0%; control group: median 75.0%, 25-75 percentile 66.7%-88.6%; P b .01). Conclusions: This study gives first evidence that the use of a standardized checklist for patient handover from OR to ICU increases the quantity and quality of transmitted medical information. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Handover of patient care and responsibility is a potential safety risk for the patient due to loss of information with influence on patient outcome [1]. However, transfer of patient responsibility is unavoidable because care for the individual often outlasts one working shift or a patient is transferred between different departments in the hospital. During their treatment, patients undergoing major surgery are frequently exposed to multiple handovers. One of them is the handover from the operating room (OR) to the postanesthesia care unit or the intensive care unit (ICU). These patients often undergo major surgical procedures frequently associated with acute pathophysiologic deteriorations and furthermore exhibit extensive comorbidities. Their transfer is a highly complex work process. In consequence, many of those patients require complex treatment during the phase of handover, such as mechanical ventilation and/or hemodynamic support by continuous infusion of catecholamines, which requires constant attention from the care giving team. In the handover process, responsibility for these critically ill patients is completely transferred from one team to another one. Earlier
⁎ Corresponding author. Tel.: +49 40 7410 52415; fax: +49 40 7410 54963. E-mail addresses: [email protected] (C. Salzwedel), [email protected] (V. Mai), [email protected] (M.A. Punke), [email protected] (S. Kluge), [email protected] (D.A. Reuter). http://dx.doi.org/10.1016/j.jcrc.2015.12.016 0883-9441/© 2015 Elsevier Inc. All rights reserved.
studies have shown that postoperative handovers are often informal, brief [2–4], and frequently incomplete [5,6]. A medical error caused by insufficient transfer of information may lead to patient harm [7]. An analysis of adverse clinical incidents occurring in the recovery area showed that 14% of incidents happened because of communication failure [8]. To avoid loss of information during patient transfer, standardized protocols can help to increase the completeness of postanesthesia handover [6,7,9–11]. These types of protocols, such as a checklist, have been demanded for years by different medical associations and the World Health Organization [12,13]. Although many studies have dealt with health care handover, only few have focused on handover in the perioperative setting [14] and critically ill patients. Furthermore, it is not clear if handover checklists only increase quantity, for example, the number of single items handed over regardless of their relevance for the individual patient in this particular situation, or if they also increase their quality, for example, a gain of patient- and contextspecific information handed over. Therefore, the aim of the present study was to investigate the effect of the use of a checklist for postanesthesia handover in the ICU. Our primary end point was that a checklist will lead to an increase of quantity (less items will be omitted). Secondary end point was also quality (the individually important items will be handed over) of information transfer compared with non-standardized handovers.
C. Salzwedel et al. / Journal of Critical Care 32 (2016) 170–174
2. Methods After approval of the ethics committee of the Medical Board of the City of Hamburg (PV4074) and the institutional workers’ council of the Hamburg-Eppendorf University Medical Centre and with written informed consent of the participating anesthesiologists and critical care physicians, 121 handovers of patients transferred from the OR to the ICU were included into this study. Because no specific personal or medical records directly related to patients were assessed, written informed consent from patients was waived by the ethics committee. The study was conducted as a prospective, randomized trial in a high-volume academic medical center in northern Europe from March 2013 until July 2013. The quantity and quality of handovers with use of a standardized checklist were compared with those of handovers without use of the checklist. The primary outcome was the percentage of items handed over from the caregiving anesthesiologist (resident) to the ICU physician and ICU nurse that were declared as “important to be handed over” by the supervising anesthesiologist (attending). The secondary outcomes were the percentage of items handed over from the caregiving anesthesiologist to the ICU physician which were declared as “should be handed over” by the supervising anesthesiologist and the duration of the handover. For that purpose, 2 documents were developed: First is a checklist, which was used in the study group for the handovers from the OR to the ICU. This checklist contained 13 categories of patient information, which should be addressed during patient handover (Appendix 1) by the anesthesiologist directly responsible for the patient (residents or board-certified staff anesthesiologists). Second is an assessment sheet for each patient enrolled in this study (both study and control group), which had to be filled out by the supervising anesthesiology attending responsible for the OR in which the surgery took place. In this institution, 1 attending is routinely responsible for supervision of a set of 4 ORs at a time (but not all of them with patients included in the study), so the attending has knowledge about patients and interventions, especially complications, but is not present in the OR during the entire procedure. Different to other studies (where all patient handovers had the same sheet for rating), this sheet was designed to create an individual rating scale for each patient to meet the different concerns of each patient and each intervention and therefore each individual handover. This assessment sheet contained 54 handover items and additional blank spaces that could be checked by the attending, depending on the patients’ characteristics and surgical intervention (Appendix 2). The supervising attendings were asked to work through the assessment sheet and decide for the individual patient which items had to be included in the handover. There were 2 rating categories: red items were defined as “must be handed over” to ensure patient safety, and yellow items were defined as “should be handed over” because it is additional useful information for the care giving team but omittance would not directly compromise patient safety. All surgical procedures in abdominal surgery, gynecology, urology, ear/nose/throat surgery, oral/maxillofacial surgery, orthopedic/trauma surgery, neurosurgery, spine surgery, neuroradiology, and vascular surgery on weekdays from 7:00 AM until 10:00 PM were checked for inclusion and exclusion criteria (Table 1). All handovers eligible to enter the study were consecutively included in the study. During the course of surgery, one of the investigators (VM or CS) went into the OR and handed out the assessment sheet to the responsible attending. The attending
Table 1 Inclusion and exclusion criteria Inclusion
Exclusion
Patient ≥18 y Postoperative transfer to ICU Written informed consent of caring anesthesiologist and critical care physician
Patient b 18 y Patient known on ICU Refusal of physicians to participate in study
171
then filled out the assessment sheet on the basis of patient data and her/ his knowledge of the course of surgery. They were not allowed to show the sheet to the caregiving anesthesiologist and were not informed about group allocation. At the end of surgery, the investigator randomized the caregiving anesthesiologist via closed envelop method to the study group or to the control group. The assessment sheet was collected from the responsible attending. On arrival in the ICU, either a checklist was handed out to use for the handover to the caregiving anesthesiologist (study group) or not (control group). All handovers were recorded by a digital voice recorder (VN-711PC, Olympus) positioned in the front pocket of the anesthesiologists’ scrubs. Handovers were later stored digitally and deleted from the voice recorder. The investigator waited outside the patient room until the handover was over to not disturb the usual workflow during postoperative handover. As it is routine in this institution, surgical handover was conducted by the surgeon independently from the postoperative anesthesiological handover. Usually, the surgical handover is conducted between surgeon and intensivist before the arrival of the patient. This is due to the workflow in the OR. Surgeons step off the OR table and go directly to the ICU, whereas surgical assistants, OR staff, and the anesthesia team transfer the patient from the OR table to a bed, transfer equipment, and respirator and therefore arrive in the ICU after the surgeon. This study investigated only the anesthesiological handover and not the handover conducted by the surgeon. If more than one handover of the same anesthesiologist was included in the study, the physician was only randomized into control or intervention group once and stayed in this group for all following handovers. This intervention was essential because if one anesthesiologist was randomized into the study group first, using the checklist for handover, and would be randomized into the control group for a second handover, this anesthesiologist would know the content of the checklist and would be biased during the following handovers. All recorded handovers were analyzed independently by 2 investigators (VM and CS). All items handed over by the anesthesiologist of the study and the control group were recorded on the assessment sheet by checkmark or handwritten comments. Ratings of both investigators were recorded in a digitalized sheet for later comparison. Investigators were not informed about group allocation of the participating anesthesiologist. After analysis of the handovers by the investigators, results of the assessment sheet filled out by the supervising attending were transferred into a digitalized study sheet and were compared with the results of the analysis of the investigators. The duration of the handover was recorded in seconds using a stop watch while listening to the recordings. 3. Statistical analysis Sample size calculation for the primary study end point was based on an aimed difference in the information items handed over of 20% between the control and the intervention group. With a power of 80% and a statistical significance of P b .05, a total of 116 patients had to be included, 58 per group. Data were analyzed using Microsoft Excel 2010 (Microsoft, Redmond, WA) and SigmaPlot 12.0 (Systat Software, Hamburg, Germany). The percentage of demanded items handed over was calculated for every handover. All data were tested for normality distribution (Shapiro-Wilk). For comparison of differences of the specific items between the groups, the χ2 test or Fisher exact Test was applied depending on sample size. For comparison of overall items (red and yellow), subgroup analysis, and duration of the handover, the t test or Mann-Whitney rank sum test for not normally distributed data was used. Statistical significance was designated as P b .05. 4. Results The handovers of 134 patients were included in this study. Two handovers were missed because of overlapping handovers. One hundred thirty-two handovers were recorded. From those, 11 handovers
172
C. Salzwedel et al. / Journal of Critical Care 32 (2016) 170–174
had to be excluded from analysis because of missing assessment sheets or technical problems with the recording of the handover. Finally, 121 handovers were included in the analysis (Fig. 1), with 60 handovers in the intervention group and 61 in the control group. After first assessment, both investigators rated 92.9% of items identical. All differing items were reviewed together with a third investigator, discussed, and rated once more. For all items (100%), agreement could be reached. Fifty-one different anesthesiologists conducted the handovers, ranging from 1 to 12 handovers per anesthesiologist (2.4 ± 2.1). None of the contacted anesthesiologists declined to participate in the study. The attendings claimed a total of 2053 red items (“must be handed over”) to be handed over by the caregiving anesthesiologist, with an average of 17 items (with a range of 4 to 31) per patient. In addition, 356 yellow items (“should be handed over”) were indicated by the attendings, for an average of 3 items (with a range of 0 to 11) per patient. In the study group, a median of 87.1% (percentile 25-75: 77.1%-90%) of the red items was handed over, whereas in the control group, it was a median of 75.0% (percentile 25-75: 66.7%-88.6%) (Fig. 2), representing a highly statistical significant difference between both groups (P = .005). A median of 60.0% (percentile 25-76: 36.7%-100%) of yellow items was handed over in the study group, which did not differ statistically significantly from the control group with a median of 50.0% (percentile 25-75: 33.3%-69.0%) of items per handover (P = .203) (Fig. 3). Detailed analysis of each item showed that only the items “age” (P = b.001), “fluids” (P = .006), and “mental status preoperatively” (P = .011) were statistically significantly handed over more often in the intervention group than in the control group. We observed no statistically significant difference with all other items. The following items from the list of handover items were never or rarely selected by any attending and are therefore not included in the analysis: “Dentition,” “Dental trauma,” “Malignant hyperthermia,” “Reaction on blood transfusion,” “ST-segment changes/suspicion of myocardial infarction,” “Pulmonary embolism,” “Cardiopulmonary resuscitation,” “Pulmonary obstruction,” “Stoma,” “Hematuria,” “Suspicion of urinary tract infection,” “Suspicion of cerebral ischemia,” “Seizure,” and “Intracranial pressure”. The duration of the handover did not differ significantly (study group: median 208 seconds, 25-75 percentile 142-276 seconds; control group:
Fig. 2. Red items handed over. Comparison of percentage of red items handed over in the study group compared with the control group. Minimum value, first quartile, second quartile (=median), third quartile, and maximum value are shown. *Statistically significant.
median 174 seconds, 25-75 percentile 115-255 seconds; P = .201) between the 2 groups. Subgroup analysis did not reveal a statistically significant difference of handover quality between male and female anesthesiologists (P = .412) or clinical experience of the anesthesiologist of 1 to 2 years compared with anesthesiologists with clinical experience of more than 2 years (P = .976).
5. Discussion The results of this study give first evidence that the use of a checklist for medical handovers of patients from the OR to the ICU not only increases the quantity of information transported but also increases the quality, for example, the amount of patient- and context-specific information transmitted.
Fig. 1. Inclusion of handovers.
C. Salzwedel et al. / Journal of Critical Care 32 (2016) 170–174
Fig. 3. Yellow items handed over. Comparison of percentage of yellow items handed over in the study group compared with the control group. Minimum value, first quartile, second quartile (=median), third quartile, and maximum value are shown. NS indicates not statistically significant.
Several studies have shown that the use of a checklist for postoperative handovers of patients increases the quantity of items handed over [6,7,9–11], most of them focusing on handover from OR to postanesthesia care unit or handover from OR to the ICU of pediatric cardiac surgery. Our data support these findings, showing that the amount of important information handed over increased by 13% with the use of a checklist for patient handover (75% in the control group compared with 87.1% in the study group). Research if this quantitative increase of information transfer also affects the quality of patient handover has been rare. With this study, we were able to show that handovers of patients with complex surgery or numerous comorbidities, and therefore scheduled postoperative admission to ICU, improve with the use of a standardized handover protocol. Because of an individual rating sheet to be considerate of the differences of patients’ diagnoses and interventions, we were able to point out for each individual patient whether items with important patient- and/or context-specific information were omitted, or whether items were omitted because of minor or no importance in this specific patient/case. As an example, the item “known allergies” transfers patient- and context-specific information of highest relevance in a patient with known allergies. In this case, this item therefore represents information that must be handed over, and omission would increase patient’s risks. However, in a patient with no allergic predispositions, transfer of this item does not add specific information that would lower the patient-individual risk. We further analyzed if any specific items were disproportionally omitted in each group. When comparing those omissions between both groups, we found differences only for 3 items (“age,” “fluids,” and “mental status preoperatively”), again showing a high heterogeneity. We suppose that this result is an effect of the high heterogeneity of patients included, with regard to their diagnoses, surgical procedures, and comorbidities. Because of the study protocol, we were not allowed to document any directly patient-related data and cannot prove this hypothesis at that point. However, an effect of different communication skills and personal manners of the medical team members with regard to handovers likely influenced those findings. This again strengthens the meaningfulness to follow a standardized checklist for handovers both in routine and in emergencies as well as in medical education. The use of the checklist did not lead to a change in the duration of handover. This finding correlates well with the results of other similar trials [11]. However, there are several limitations to this study: Overall, we detected poor compliance with the checklist. The intervention group
173
still omitted a total of 18.5% of red items that were claimed to be handed over by the attendings. This might also be the result of the great differences in perception of different physicians of which information is important and which is not. In the case of the content of patient handover, it is not possible to define right and wrong. Every handover and also the patient assessment sheet filled out by the attending remained a subjective evaluation. Furthermore, we were not able to verify whether attendings had complete knowledge about the patient and the surgical procedure. All involved staff were informed that we were conducting a study about postanesthesia handovers in the ICU and that handovers were recorded. This may have led to an increased effort of personnel to perform well during handover. We believe that a greater improvement might have been observed if the control group participants had not been aware of the recording of the handover. Of course, this was not possible because of study design. Finally, and most importantly, omissions of verbal information do not implicitly lead to more clinical incidents or reduced patient safety. Based on the study design, we were not able to address this question. Further studies need to assess how far measurable clinical outcome can be influenced by the intelligent use of checklists for patient handovers from the OR to ICU. A further interesting aspect that should be investigated in another approach would be how the receiving ICU team (physician and nurse) perceived the handovers conducted with our without standardized protocol. Another future goal for patient safety could be to implement intelligent checklists into digital patient files that automatically collect important clinical information, for example, from premedical history or surgical records throughout the hospital stay, and filter the important and relevant information for a successful and complete handover. 6. Conclusion Patient handover is a complex intervention, and the use of a standardized checklist for handover from OR to ICU increases the quantity and quality of transmitted medical information. It should be implemented in all perioperative settings to improve patient safety. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jcrc.2015.12.016. Funding This investigator-initiated and investigator-driven study was funded by departmental resources only. Conflict of interests None. Acknowledgments We acknowledge all anesthesiologists and critical care physicians of the University Hospital Hamburg-Eppendorf that participated in this study. References [1] Saager L, Hesler BD, You J, Turan A, Mascha EJ, Sessler DI, et al. Intraoperative transitions of anesthesia care and postoperative adverse outcomes. Anesthesiology 2014;121:695–706. [2] Horn J, Bell MDD, Moss E. Handover of responsibility for the anaesthetized patient—opinion and practice. Anaesthesia 2004;59:658–63. [3] Nagpal K, Arora S, Abboudi M, Vats A, Wong HW, Manchanda C, et al. Postoperative handover: problems, pitfalls, and prevention of error. Ann Surg 2010;252:171–6. [4] Smith AF, Pope C, Goodwin D, Mort M. Interprofessional handover and patient safety in anesthesia: observational study of handovers in the recovery room. Br J Anaesth 2008;101:332–7. [5] Milby A, Böhmer A, Gerbershagen MU, Joppich R, Wappler F. Quality of postoperative patient handover in the post-anesthesia care unit: a prospective analysis. Acta Anaesthesiol Scand 2014;58:192–7.
174
C. Salzwedel et al. / Journal of Critical Care 32 (2016) 170–174
[6] Salzwedel C, Bartz H-J, Kühnelt I, Appel D, Haupt O, Maisch S, et al. The effect of a checklist on the quality of post-anesthesia patient handover: a randomized controlled trial. Int J Qual Health Care 2013;25:176–81. [7] Boat AC, Spaeth JP. Handoff checklists improve the reliability of patient handoffs in the operating room and postanesthesia care unit. Paediatr Anaesth 2013;23:647–54. [8] Kluger MT, Bullock MF. Recovery room incidents: a review of 419 reports from the Anaestehtic Incident Monitoring Study (AIMS). Anaesthesia 2002;57:1060–6. [9] Joy BF, Elliott E, Hardy C, Sullivan C, Backer CL, Kane J. Standardized multidisciplinary protocol improves handover of cardiac surgery patients to the intensive care unit. Pediatr Crit Care Med 2011;12:304–8. [10] Karakaya A, Moerman AT, Peperstraete H, Francois K, Wouters PF, de Hert SG. Implementation of a structured information transfer checklist improves postoperative data transfer after congenital cardiac surgery. Eur J Anaesthesiol 2013;30:764–9.
[11] Moller TP, Madsen MD, Fuhrmann L, Ostergaard D. Postoperative handover: characteristics and considerations on improvement. A systematic review. Eur J Anaesthesiol 2013;30:229–42. [12] Leotsakos A, Zheng H, Croteau R, Loeb JM, Sherman H, Hoffman C, et al. Standardization in patient safety: the WHO High 5s project. Int J Qual Health Care 2014;26: 109–16. [13] Rosenthal C, Balzer F, Boemke W, Spies C. Patient safety in anesthesiology and intensive care medicine. Measures for improvement. Med Klin Intensivmed Notfmed 2013;108:657–65. [14] Petrovic MA, Aboumatar H, Baumgartner WA, Ulatowski JA, Moyer J, Chang TY, et al. Pilot implementation of a perioperative protocol to guide operating room-to-intensive care unit patient handoffs. J Cardiothorac Vasc Anesth 2012; 26:11–6.
Contents lists available at ScienceDirect
Journal of Critical Care journal homepage: www.jccjournal.org
The effect of a checklist on the quality of patient handover from the operating room to the intensive care unit: A randomized controlled trial Cornelie Salzwedel, MD a,⁎, Victoria Mai, MD a, Mark A. Punke, MD a, Stefan Kluge, MD, PhD b, Daniel A. Reuter, MD, PhD a a b
Department of Anesthesiology, Center of Anesthesiology and Intensive Care Medicine, University Hospital Hamburg-Eppendorf, Martinistrasse 52, 20246 Hamburg, Germany Department of Intensive Care Medicine, Center of Anesthesiology and Intensive Care Medicine, University Hospital Hamburg-Eppendorf, Martinistrasse 52, 20246 Hamburg, Germany
a r t i c l e
i n f o
Keywords: Postoperative period Patient handover Checklist Critical care Patient safety
a b s t r a c t Purpose: Handover of patient care is a potential safety risk for the patient due to loss of information which may result in adverse outcome. We hypothesized that a checklist for handover from the operating room (OR) to the intensive care unit (ICU) will lead to an increase of quality regarding information transfer compared with a nonstandardized handover procedure. Materials and methods: The study was conducted as a prospective, randomized trial in a university hospital. The quality of handovers with checklist was compared with handovers without checklist. Handovers were recorded by digital voice recorder and analyzed using an individual rating sheet for each patient. This enabled to discriminate between items that “must be handed over” (red items) and items that “should be handed over” (yellow items). Results: A total of 121 patient handovers from OR to ICU were included. Significantly more red items were handed over in the study group compared with the control group (study group: median 87.1%, 25-27 percentile 77.1%90.0%; control group: median 75.0%, 25-75 percentile 66.7%-88.6%; P b .01). Conclusions: This study gives first evidence that the use of a standardized checklist for patient handover from OR to ICU increases the quantity and quality of transmitted medical information. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Handover of patient care and responsibility is a potential safety risk for the patient due to loss of information with influence on patient outcome [1]. However, transfer of patient responsibility is unavoidable because care for the individual often outlasts one working shift or a patient is transferred between different departments in the hospital. During their treatment, patients undergoing major surgery are frequently exposed to multiple handovers. One of them is the handover from the operating room (OR) to the postanesthesia care unit or the intensive care unit (ICU). These patients often undergo major surgical procedures frequently associated with acute pathophysiologic deteriorations and furthermore exhibit extensive comorbidities. Their transfer is a highly complex work process. In consequence, many of those patients require complex treatment during the phase of handover, such as mechanical ventilation and/or hemodynamic support by continuous infusion of catecholamines, which requires constant attention from the care giving team. In the handover process, responsibility for these critically ill patients is completely transferred from one team to another one. Earlier
⁎ Corresponding author. Tel.: +49 40 7410 52415; fax: +49 40 7410 54963. E-mail addresses: [email protected] (C. Salzwedel), [email protected] (V. Mai), [email protected] (M.A. Punke), [email protected] (S. Kluge), [email protected] (D.A. Reuter). http://dx.doi.org/10.1016/j.jcrc.2015.12.016 0883-9441/© 2015 Elsevier Inc. All rights reserved.
studies have shown that postoperative handovers are often informal, brief [2–4], and frequently incomplete [5,6]. A medical error caused by insufficient transfer of information may lead to patient harm [7]. An analysis of adverse clinical incidents occurring in the recovery area showed that 14% of incidents happened because of communication failure [8]. To avoid loss of information during patient transfer, standardized protocols can help to increase the completeness of postanesthesia handover [6,7,9–11]. These types of protocols, such as a checklist, have been demanded for years by different medical associations and the World Health Organization [12,13]. Although many studies have dealt with health care handover, only few have focused on handover in the perioperative setting [14] and critically ill patients. Furthermore, it is not clear if handover checklists only increase quantity, for example, the number of single items handed over regardless of their relevance for the individual patient in this particular situation, or if they also increase their quality, for example, a gain of patient- and contextspecific information handed over. Therefore, the aim of the present study was to investigate the effect of the use of a checklist for postanesthesia handover in the ICU. Our primary end point was that a checklist will lead to an increase of quantity (less items will be omitted). Secondary end point was also quality (the individually important items will be handed over) of information transfer compared with non-standardized handovers.
C. Salzwedel et al. / Journal of Critical Care 32 (2016) 170–174
2. Methods After approval of the ethics committee of the Medical Board of the City of Hamburg (PV4074) and the institutional workers’ council of the Hamburg-Eppendorf University Medical Centre and with written informed consent of the participating anesthesiologists and critical care physicians, 121 handovers of patients transferred from the OR to the ICU were included into this study. Because no specific personal or medical records directly related to patients were assessed, written informed consent from patients was waived by the ethics committee. The study was conducted as a prospective, randomized trial in a high-volume academic medical center in northern Europe from March 2013 until July 2013. The quantity and quality of handovers with use of a standardized checklist were compared with those of handovers without use of the checklist. The primary outcome was the percentage of items handed over from the caregiving anesthesiologist (resident) to the ICU physician and ICU nurse that were declared as “important to be handed over” by the supervising anesthesiologist (attending). The secondary outcomes were the percentage of items handed over from the caregiving anesthesiologist to the ICU physician which were declared as “should be handed over” by the supervising anesthesiologist and the duration of the handover. For that purpose, 2 documents were developed: First is a checklist, which was used in the study group for the handovers from the OR to the ICU. This checklist contained 13 categories of patient information, which should be addressed during patient handover (Appendix 1) by the anesthesiologist directly responsible for the patient (residents or board-certified staff anesthesiologists). Second is an assessment sheet for each patient enrolled in this study (both study and control group), which had to be filled out by the supervising anesthesiology attending responsible for the OR in which the surgery took place. In this institution, 1 attending is routinely responsible for supervision of a set of 4 ORs at a time (but not all of them with patients included in the study), so the attending has knowledge about patients and interventions, especially complications, but is not present in the OR during the entire procedure. Different to other studies (where all patient handovers had the same sheet for rating), this sheet was designed to create an individual rating scale for each patient to meet the different concerns of each patient and each intervention and therefore each individual handover. This assessment sheet contained 54 handover items and additional blank spaces that could be checked by the attending, depending on the patients’ characteristics and surgical intervention (Appendix 2). The supervising attendings were asked to work through the assessment sheet and decide for the individual patient which items had to be included in the handover. There were 2 rating categories: red items were defined as “must be handed over” to ensure patient safety, and yellow items were defined as “should be handed over” because it is additional useful information for the care giving team but omittance would not directly compromise patient safety. All surgical procedures in abdominal surgery, gynecology, urology, ear/nose/throat surgery, oral/maxillofacial surgery, orthopedic/trauma surgery, neurosurgery, spine surgery, neuroradiology, and vascular surgery on weekdays from 7:00 AM until 10:00 PM were checked for inclusion and exclusion criteria (Table 1). All handovers eligible to enter the study were consecutively included in the study. During the course of surgery, one of the investigators (VM or CS) went into the OR and handed out the assessment sheet to the responsible attending. The attending
Table 1 Inclusion and exclusion criteria Inclusion
Exclusion
Patient ≥18 y Postoperative transfer to ICU Written informed consent of caring anesthesiologist and critical care physician
Patient b 18 y Patient known on ICU Refusal of physicians to participate in study
171
then filled out the assessment sheet on the basis of patient data and her/ his knowledge of the course of surgery. They were not allowed to show the sheet to the caregiving anesthesiologist and were not informed about group allocation. At the end of surgery, the investigator randomized the caregiving anesthesiologist via closed envelop method to the study group or to the control group. The assessment sheet was collected from the responsible attending. On arrival in the ICU, either a checklist was handed out to use for the handover to the caregiving anesthesiologist (study group) or not (control group). All handovers were recorded by a digital voice recorder (VN-711PC, Olympus) positioned in the front pocket of the anesthesiologists’ scrubs. Handovers were later stored digitally and deleted from the voice recorder. The investigator waited outside the patient room until the handover was over to not disturb the usual workflow during postoperative handover. As it is routine in this institution, surgical handover was conducted by the surgeon independently from the postoperative anesthesiological handover. Usually, the surgical handover is conducted between surgeon and intensivist before the arrival of the patient. This is due to the workflow in the OR. Surgeons step off the OR table and go directly to the ICU, whereas surgical assistants, OR staff, and the anesthesia team transfer the patient from the OR table to a bed, transfer equipment, and respirator and therefore arrive in the ICU after the surgeon. This study investigated only the anesthesiological handover and not the handover conducted by the surgeon. If more than one handover of the same anesthesiologist was included in the study, the physician was only randomized into control or intervention group once and stayed in this group for all following handovers. This intervention was essential because if one anesthesiologist was randomized into the study group first, using the checklist for handover, and would be randomized into the control group for a second handover, this anesthesiologist would know the content of the checklist and would be biased during the following handovers. All recorded handovers were analyzed independently by 2 investigators (VM and CS). All items handed over by the anesthesiologist of the study and the control group were recorded on the assessment sheet by checkmark or handwritten comments. Ratings of both investigators were recorded in a digitalized sheet for later comparison. Investigators were not informed about group allocation of the participating anesthesiologist. After analysis of the handovers by the investigators, results of the assessment sheet filled out by the supervising attending were transferred into a digitalized study sheet and were compared with the results of the analysis of the investigators. The duration of the handover was recorded in seconds using a stop watch while listening to the recordings. 3. Statistical analysis Sample size calculation for the primary study end point was based on an aimed difference in the information items handed over of 20% between the control and the intervention group. With a power of 80% and a statistical significance of P b .05, a total of 116 patients had to be included, 58 per group. Data were analyzed using Microsoft Excel 2010 (Microsoft, Redmond, WA) and SigmaPlot 12.0 (Systat Software, Hamburg, Germany). The percentage of demanded items handed over was calculated for every handover. All data were tested for normality distribution (Shapiro-Wilk). For comparison of differences of the specific items between the groups, the χ2 test or Fisher exact Test was applied depending on sample size. For comparison of overall items (red and yellow), subgroup analysis, and duration of the handover, the t test or Mann-Whitney rank sum test for not normally distributed data was used. Statistical significance was designated as P b .05. 4. Results The handovers of 134 patients were included in this study. Two handovers were missed because of overlapping handovers. One hundred thirty-two handovers were recorded. From those, 11 handovers
172
C. Salzwedel et al. / Journal of Critical Care 32 (2016) 170–174
had to be excluded from analysis because of missing assessment sheets or technical problems with the recording of the handover. Finally, 121 handovers were included in the analysis (Fig. 1), with 60 handovers in the intervention group and 61 in the control group. After first assessment, both investigators rated 92.9% of items identical. All differing items were reviewed together with a third investigator, discussed, and rated once more. For all items (100%), agreement could be reached. Fifty-one different anesthesiologists conducted the handovers, ranging from 1 to 12 handovers per anesthesiologist (2.4 ± 2.1). None of the contacted anesthesiologists declined to participate in the study. The attendings claimed a total of 2053 red items (“must be handed over”) to be handed over by the caregiving anesthesiologist, with an average of 17 items (with a range of 4 to 31) per patient. In addition, 356 yellow items (“should be handed over”) were indicated by the attendings, for an average of 3 items (with a range of 0 to 11) per patient. In the study group, a median of 87.1% (percentile 25-75: 77.1%-90%) of the red items was handed over, whereas in the control group, it was a median of 75.0% (percentile 25-75: 66.7%-88.6%) (Fig. 2), representing a highly statistical significant difference between both groups (P = .005). A median of 60.0% (percentile 25-76: 36.7%-100%) of yellow items was handed over in the study group, which did not differ statistically significantly from the control group with a median of 50.0% (percentile 25-75: 33.3%-69.0%) of items per handover (P = .203) (Fig. 3). Detailed analysis of each item showed that only the items “age” (P = b.001), “fluids” (P = .006), and “mental status preoperatively” (P = .011) were statistically significantly handed over more often in the intervention group than in the control group. We observed no statistically significant difference with all other items. The following items from the list of handover items were never or rarely selected by any attending and are therefore not included in the analysis: “Dentition,” “Dental trauma,” “Malignant hyperthermia,” “Reaction on blood transfusion,” “ST-segment changes/suspicion of myocardial infarction,” “Pulmonary embolism,” “Cardiopulmonary resuscitation,” “Pulmonary obstruction,” “Stoma,” “Hematuria,” “Suspicion of urinary tract infection,” “Suspicion of cerebral ischemia,” “Seizure,” and “Intracranial pressure”. The duration of the handover did not differ significantly (study group: median 208 seconds, 25-75 percentile 142-276 seconds; control group:
Fig. 2. Red items handed over. Comparison of percentage of red items handed over in the study group compared with the control group. Minimum value, first quartile, second quartile (=median), third quartile, and maximum value are shown. *Statistically significant.
median 174 seconds, 25-75 percentile 115-255 seconds; P = .201) between the 2 groups. Subgroup analysis did not reveal a statistically significant difference of handover quality between male and female anesthesiologists (P = .412) or clinical experience of the anesthesiologist of 1 to 2 years compared with anesthesiologists with clinical experience of more than 2 years (P = .976).
5. Discussion The results of this study give first evidence that the use of a checklist for medical handovers of patients from the OR to the ICU not only increases the quantity of information transported but also increases the quality, for example, the amount of patient- and context-specific information transmitted.
Fig. 1. Inclusion of handovers.
C. Salzwedel et al. / Journal of Critical Care 32 (2016) 170–174
Fig. 3. Yellow items handed over. Comparison of percentage of yellow items handed over in the study group compared with the control group. Minimum value, first quartile, second quartile (=median), third quartile, and maximum value are shown. NS indicates not statistically significant.
Several studies have shown that the use of a checklist for postoperative handovers of patients increases the quantity of items handed over [6,7,9–11], most of them focusing on handover from OR to postanesthesia care unit or handover from OR to the ICU of pediatric cardiac surgery. Our data support these findings, showing that the amount of important information handed over increased by 13% with the use of a checklist for patient handover (75% in the control group compared with 87.1% in the study group). Research if this quantitative increase of information transfer also affects the quality of patient handover has been rare. With this study, we were able to show that handovers of patients with complex surgery or numerous comorbidities, and therefore scheduled postoperative admission to ICU, improve with the use of a standardized handover protocol. Because of an individual rating sheet to be considerate of the differences of patients’ diagnoses and interventions, we were able to point out for each individual patient whether items with important patient- and/or context-specific information were omitted, or whether items were omitted because of minor or no importance in this specific patient/case. As an example, the item “known allergies” transfers patient- and context-specific information of highest relevance in a patient with known allergies. In this case, this item therefore represents information that must be handed over, and omission would increase patient’s risks. However, in a patient with no allergic predispositions, transfer of this item does not add specific information that would lower the patient-individual risk. We further analyzed if any specific items were disproportionally omitted in each group. When comparing those omissions between both groups, we found differences only for 3 items (“age,” “fluids,” and “mental status preoperatively”), again showing a high heterogeneity. We suppose that this result is an effect of the high heterogeneity of patients included, with regard to their diagnoses, surgical procedures, and comorbidities. Because of the study protocol, we were not allowed to document any directly patient-related data and cannot prove this hypothesis at that point. However, an effect of different communication skills and personal manners of the medical team members with regard to handovers likely influenced those findings. This again strengthens the meaningfulness to follow a standardized checklist for handovers both in routine and in emergencies as well as in medical education. The use of the checklist did not lead to a change in the duration of handover. This finding correlates well with the results of other similar trials [11]. However, there are several limitations to this study: Overall, we detected poor compliance with the checklist. The intervention group
173
still omitted a total of 18.5% of red items that were claimed to be handed over by the attendings. This might also be the result of the great differences in perception of different physicians of which information is important and which is not. In the case of the content of patient handover, it is not possible to define right and wrong. Every handover and also the patient assessment sheet filled out by the attending remained a subjective evaluation. Furthermore, we were not able to verify whether attendings had complete knowledge about the patient and the surgical procedure. All involved staff were informed that we were conducting a study about postanesthesia handovers in the ICU and that handovers were recorded. This may have led to an increased effort of personnel to perform well during handover. We believe that a greater improvement might have been observed if the control group participants had not been aware of the recording of the handover. Of course, this was not possible because of study design. Finally, and most importantly, omissions of verbal information do not implicitly lead to more clinical incidents or reduced patient safety. Based on the study design, we were not able to address this question. Further studies need to assess how far measurable clinical outcome can be influenced by the intelligent use of checklists for patient handovers from the OR to ICU. A further interesting aspect that should be investigated in another approach would be how the receiving ICU team (physician and nurse) perceived the handovers conducted with our without standardized protocol. Another future goal for patient safety could be to implement intelligent checklists into digital patient files that automatically collect important clinical information, for example, from premedical history or surgical records throughout the hospital stay, and filter the important and relevant information for a successful and complete handover. 6. Conclusion Patient handover is a complex intervention, and the use of a standardized checklist for handover from OR to ICU increases the quantity and quality of transmitted medical information. It should be implemented in all perioperative settings to improve patient safety. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jcrc.2015.12.016. Funding This investigator-initiated and investigator-driven study was funded by departmental resources only. Conflict of interests None. Acknowledgments We acknowledge all anesthesiologists and critical care physicians of the University Hospital Hamburg-Eppendorf that participated in this study. References [1] Saager L, Hesler BD, You J, Turan A, Mascha EJ, Sessler DI, et al. Intraoperative transitions of anesthesia care and postoperative adverse outcomes. Anesthesiology 2014;121:695–706. [2] Horn J, Bell MDD, Moss E. Handover of responsibility for the anaesthetized patient—opinion and practice. Anaesthesia 2004;59:658–63. [3] Nagpal K, Arora S, Abboudi M, Vats A, Wong HW, Manchanda C, et al. Postoperative handover: problems, pitfalls, and prevention of error. Ann Surg 2010;252:171–6. [4] Smith AF, Pope C, Goodwin D, Mort M. Interprofessional handover and patient safety in anesthesia: observational study of handovers in the recovery room. Br J Anaesth 2008;101:332–7. [5] Milby A, Böhmer A, Gerbershagen MU, Joppich R, Wappler F. Quality of postoperative patient handover in the post-anesthesia care unit: a prospective analysis. Acta Anaesthesiol Scand 2014;58:192–7.
174
C. Salzwedel et al. / Journal of Critical Care 32 (2016) 170–174
[6] Salzwedel C, Bartz H-J, Kühnelt I, Appel D, Haupt O, Maisch S, et al. The effect of a checklist on the quality of post-anesthesia patient handover: a randomized controlled trial. Int J Qual Health Care 2013;25:176–81. [7] Boat AC, Spaeth JP. Handoff checklists improve the reliability of patient handoffs in the operating room and postanesthesia care unit. Paediatr Anaesth 2013;23:647–54. [8] Kluger MT, Bullock MF. Recovery room incidents: a review of 419 reports from the Anaestehtic Incident Monitoring Study (AIMS). Anaesthesia 2002;57:1060–6. [9] Joy BF, Elliott E, Hardy C, Sullivan C, Backer CL, Kane J. Standardized multidisciplinary protocol improves handover of cardiac surgery patients to the intensive care unit. Pediatr Crit Care Med 2011;12:304–8. [10] Karakaya A, Moerman AT, Peperstraete H, Francois K, Wouters PF, de Hert SG. Implementation of a structured information transfer checklist improves postoperative data transfer after congenital cardiac surgery. Eur J Anaesthesiol 2013;30:764–9.
[11] Moller TP, Madsen MD, Fuhrmann L, Ostergaard D. Postoperative handover: characteristics and considerations on improvement. A systematic review. Eur J Anaesthesiol 2013;30:229–42. [12] Leotsakos A, Zheng H, Croteau R, Loeb JM, Sherman H, Hoffman C, et al. Standardization in patient safety: the WHO High 5s project. Int J Qual Health Care 2014;26: 109–16. [13] Rosenthal C, Balzer F, Boemke W, Spies C. Patient safety in anesthesiology and intensive care medicine. Measures for improvement. Med Klin Intensivmed Notfmed 2013;108:657–65. [14] Petrovic MA, Aboumatar H, Baumgartner WA, Ulatowski JA, Moyer J, Chang TY, et al. Pilot implementation of a perioperative protocol to guide operating room-to-intensive care unit patient handoffs. J Cardiothorac Vasc Anesth 2012; 26:11–6.
