Safety hazards in abdominal surgery related to communication between surgical and anesthesia unit personnel found in a Swedish nationwide survey

Background Many adverse events occur due to poor communication between surgical and anesthesia unit personnel. The aim of this study was to identify strategies to reduce risks unveiled by a national survey on patient safety. Methods During 2011–2015, specially trained survey teams visited the surgery departments at Swedish hospitals and documented routines concerning safety in abdominal surgery. The reports from the first seventeen visits were reviewed by an independent group in order to extract findings related to routines in communication between anesthesia and surgical unit personnel. Results In general, routines regarding preoperative risk assessment were safe and well- coordinated. On the other hand, routines regarding medication prior to surgery, reporting between the different units, and systems for reporting and providing feedback on adverse events were poor or missing. Strategies with highest priority include: 1. a uniform national health declaration form; 2. consistent use of admission notes; 3. systems for documenting all important medical information, that is accessible to everyone; 4. a multidisciplinary forum for the evaluation of high-risk patients; 5. weekly and daily scheduling of surgical programs; 6. application of the WHO check list; 7. open dialog during surgery; 8. reporting based on SBAR; 9. oral and written reports from the surgeon to the postoperative unit; and 10. combined mortality and morbidity conferences. Conclusion One repeatedly occurring hazard endangering patient safety was related to communication between surgical and anesthesia unit personnel. Strategies to reduce this hazard are suggested, but further research is required to test their effectiveness.

in the distribution of responsibilities and the hierarchy. Mistakes in the post-operative phase were also common, including missing information and excess of it. In interviews with the concerned staff, distracting stimuli in the environment, lack of standardized communication routines and primitive systems for exchange of information were identified as the main causes of mistakes [3]. In another study, the quality of reporting by anaesthesiologists was considered adequate in only 32 % of the cases. Despite the fact that in 96 % of the cases the patients handed over to post-operative care were stable, nurses in the postoperative unit were satisfied with the transfer reports in only 48 % of the cases [4]. Greenberg et al. found that most misunderstandings involved verbal communication between a single reporter and a recipient [5], particularly if a surgeon was involved.
Many mistakes are also made in theatre. By tradition, the staff working in theatre represents several disciplines, which makes communication more complicated [6]. Effective teamwork and communication are crucial to safety in the operating theatre, but communication is often complicated by differences in professional practices across disciplines and the ways in which professionals collaborate and may also be impaired by differences in language, attitude, and the level of education. Loyalty to the discipline may be greater than that to the team and to the common task at hand. Several studies have shown how the difficulties that arise when information is perceived differently by various members of the team cause tension, conflicts, and practical problems. Social barriers in the form of gender, ethnic background, and socio-economic differences may add to the divergent perception of information by different members of a team [7].
The present study therefore sought to analyse the reports from a national survey on patient safety to find out the ten most frequent hazards related to communications between anaesthetists and surgical unit personnel and to assess possible strategies to reduce the risks from such miscommunication.

Survey organization
To survey the routine procedures followed in Swedish surgical practice and to recommend any adaptations to the routine suited to each individual unit, a nationwide programme was started in 2011, titled Safe Abdominal Surgery. The programme offered every surgical unit in Sweden a visit by a survey team including 4-5 representatives from surgery and anaesthesiology. The programme focused exclusively on abdominal surgery; vascular, urological, and gynaecological surgeries were not included. The teams consisted of nurses and physicians who had undergone specialized training.
Altogether 56 units in Sweden were invited to participate in the programme. By the end of 2015, 43 of them will have been visited by a survey team. Written consent was obtained from all participating units prior to the revision. To start with, the participating units were requested to fill in a self-assessment instrument comprising 24 items on measures considered important to preventing adverse events in abdominal surgery (Fig. 1). The self-assessment instrument is presented in Appendix 1.
Data on admissions at each unit during 2012 were obtained from Statistics Sweden [8].
Following the self-assessment, each unit was visited by a survey team consisting of 4-5 members, representing both nurses and physicians from an independent unit, who had attended a two-day training programme. The training programme included an introduction in the aims and methods of the survey. The self-assessments of the units were also evaluated during the introduction.
The visits were based on interviews with personnel from different categories, including representatives from surgical and anaesthesiology units (residents and specialists in surgery and anaesthesiology). Nurses from the outpatient clinic and surgical wards and anaesthesia and surgery personnel from the operations room (OR) and post-operative care facility were also interviewed about the routines followed at their respective units. The head of the surgical department was responsible for assembling representatives from all personnel categories. Furthermore, the survey teams were instructed to address as many of the co-workers as possible in order to obtain information from unprepared staff members.
The observations and outcomes of the interviews were recorded in a standardized protocol based on the same items as those presented in Appendix 1. The report specified a certain number of routines that needed improvement at each unit.

Review of the revision reports
A preliminary retrospective review of the survey reports pointed to a repeating pattern of problems in communication between the surgical staff and the personnel from anaesthesia. To examine these observations more systematically, a group was commissioned by the Patient Insurance LÖF [9] to evaluate the reports and to enumerate all the problems related to patient safety in the routines involving communications between the surgical staff and the personnel from anaesthesia. This group was also multi-professional, comprising nurses and physicians from surgical and anaesthesia units.
The present study is based on the observations and conclusions from the survey teams. The reports compiled between 2011 and 2013 from 17 surgical units were reviewed, and the ten most frequent hazards related to communications between personnel of the surgical and the anaesthesia units were selected for retrospective evaluation. Each of these hazards was subdivided into 1-6 issues, and safety systems and routines at each of the units were assessed based on these issues.
Adverse events were defined as unintended experiences related to a surgical procedure. Risks were defined as potential sources of adverse events that could have been avoided by using better routines. Safety hazards were defined as potential sources of all adverse events, both avoidable and unavoidable. Mistakes were defined as adverse events due to inappropriate method of care and, therefore, potentially preventable. As the survey did not involve any new intervention or other interference with the health care, no application to ethical committee was sent.

Data from the review reports
The reports were assessed using a predefined template with the following questions. The questions were in Swedish; an English translation is given here. The review analysed to identify the ten most frequent issues Survey report according to predefined protocol Internal evaluation based on predefined self-assessment instrument Other risk assessment systems parallel to ASA Adequate documentation of all assessments 4. Is the patient's status optimal prior to surgery?
Routines to ensure that nothing is neglected 5. Is appropriate medication confirmed prior to surgery and anaesthesia? Documentation available to anaesthesia as well as surgical-unit personnel 6. Routines for clear peri-operative communication WHO checklist and/or any other system for crew resource management used consistently and correctly 7. Is peri-and post-operative pain management clearly specified? Clear routines Documentation available to anaesthesia as well as surgical-unit personnel 8. Is exchange of information ensured when patients are transferred from post-operative unit and intensive care unit? Clear and consistent routines defined 9. Is correct exchange of information ensured between the medical units that a patient passes through during in-patient treatment and the units responsible for the patient after discharge? Documentation available to anaesthesia as well as surgical-unit personnel Routines for documenting all measures taken Principles of situation background assessment recommendation (SBAR) used consistently in communication between different units 10.Are there routines for providing feedback on adverse events or systematic safety problems that could be improved? Mortality and morbidity conferences comprising anaesthesia and surgical unit personnel In the template, each of the issues listed under the subheadings was rated on the following scale: 1. Unsatisfactory 2. Partly fulfilled 3. Satisfactory 4. Not assessable

Statistical analysis
To account for inter-observer reliability, three randomly selected reports were reviewed by two independent reviewers and, based on these assessments, intra-class correlation coefficients were estimated. Internal consistency between different items for each respective unit was estimated using Cronbach's alpha. Missing values were replaced with the means of each unit. The number of adverse events reported to the Swedish Patient Insurance as a percentage of total number of admissions was tested for correlation against the mean score to test whether the outcome of the revisions could predict inadequate safety routines.

Results
The results of the 17 survey reports are shown in Fig. 2. The units comprised six university hospitals, four regional hospitals, and seven local hospitals. The median number of admissions during 2012, including admission that did not proceed to surgery, was 22 954 and ranged from 5400 to 105 924. The median number of adverse events reported to the Swedish Patient Insurance during 2007-2014 was 77 and ranged from 23 to 223. Data on adverse events from 2015 are not yet available. The median number of approved reports during the same period was 25 and ranged from 6 to 89.
In general, the routine procedures related to preoperative risk assessment were safe and well-coordinated. There were, however, some problems in the routines followed during pre-operative medication and those that involved reporting between different units and in the systems for reporting and providing feedback on adverse events. Several survey reports did not provide reliable information on the discharge from post-operative care units and intensive care units: communications between these two units and the receiving units are therefore difficult to evaluate.
In the review reports, a number of problems were mentioned repeatedly (Table 1). Although it is impossible to quantify these observations or to evaluate the effectiveness of the suggested measures to improve safety, some patterns could be identified. The most frequently recorded recommendations are listed in Table 1. The WHO checklist was not used consistently at the units, but it was not possible to determine which of the elements in the checklist were neglected.
Cronbach's alpha for all items at the 17 units was 0.501, and the intra-class correlation coefficients for the reports assessed by two independent observers are given below.

Discussion
The nationwide survey showed that in Sweden, the routines for communication between the personnel of anaesthesia units and those of surgical units may involve safety risks. Although efforts have been made in recent years to prevent mistakes in surgical procedures by defining secure routines and by managing the different teams better, many departments continue to lack proper routines for preventing systematic and recurring hazards due to poor communication between the personnel of anaesthesia units and those of surgical units. The recommendations in Table 1 may reduce the risk of adverse events, although further studies are required to prove that the recommendations are effective.
Routine procedures for risk assessment prior to surgery were considered adequate in the majority of units. Precautions to prevent post-operative pain were satisfactory;  Fig. 2 Assessment ratings from 17 survey reports. The figure shows the distributions of all subscales for each issue related to communications between surgical personnel and anaesthesia personnel. Because data were missing for some subscales, the total does not add up to 100 % for each scale. When the outcome was not assessable from the reports, the unit was excluded from the analysis however, the routines for communication between different departments needed improvement in most of the hospitals. The survey reports did not provide enough data to draw definite conclusions regarding the routines for ensuring correct diagnoses and intended procedures, proper medication and care prior to surgery, peri-operative communication and feedback on adverse events or systematic safety problems; when the reports did provide enough data, the routines for peri-operative medication and interdisciplinary conferences turned out to be either poor or lacking altogether.
Most of the recommendations presented in Table 1 have been evaluated earlier and are now part of safety improvement programmes in many hospitals. Some new and untested routines were also encountered during the review. However, because the main purpose of the study was to identify safety hazards that could be avoided by following well-established measures, only those recommendations that may gain general acceptance are included. Although every survey team included representatives from the fields of surgery and anaesthesia, the mix of anaesthetic nurses, theatre nurses, surgeons, and anaesthesiologists varied from team to team. Since the experiences of these professionals from different fields may have been different, the survey report may also have been influenced by personal views of the team members. Also, even if the standardized protocol was used by all the teams, it was difficult to abstract all data in a uniform way. However, despite the uncertainties related to some issues in the survey reports, the inter-observer reliability, when reviewed retrospectively, was high.
The methods used in the present study were semiquantitative. The data were derived from a wide range of Swedish surgical units and selection bias cannot be ruled out. The programme was designed to survey Swedish surgical health care in general, which makes it impossible to assess the effectiveness of each of the strategies presented above. Some of the suggested strategies have been tested in larger studies [10,11], but evidence-based support for these is still very limited. Furthermore, despite firm evidence, there may be reluctance to implement new routines [12]. There was no way of validating the responses in the self-assessment instruments with full security, but at the visits information interviews were undertaken with as many of the staff members as possible, including those were not prepared for the visits in advance.
At the time of the analysis, less than a third of all Swedish surgical units had had participated in the study. More units have participated since. Nevertheless, it is possible that the units that participated during the first three years were those keener than the rest to bring about improvements.
Checklists in medical practice may be an effective way to ensure that no step in a surgical process is missed so as to avoid adverse events that may occur when personnel from two different unitssurgery and anaesthesia, for exampleare required to work together [13]. The SURPASS checklist accompanies the patient at each stage of the surgical process [14] and is completed by members of the team in the pre-operative ward, the operating theatre, the post-operative unit, and the surgical ward. Another way of improving the surgical process and minimizing the risk of adverse events owing to miscommunication between surgical personnel and personnel of the anaesthesia unit is ERAS, the enhancedrecovery-after-surgery programme. The programme was designed to attenuate the stress response during surgery, mainly by focusing on the need for parenteral analgesia and intravenous fluids. It is based on a checklist and continuous feedback to all personnel on their performance. Adopting the ERAS programme has lowered morbidity by up to 50 % and shortened the surgery-related hospital stay after major abdominal surgery by about 30 % [15][16][17].
The WHO checklist was created to improve safety in the operating theatre. The purpose of the checklist is to ensure that all conditions are optimal for patient safety and that all the personnel present are identifiable and accountable. The checklist includes 19 items divided into three critical phases of the surgical procedure: before anaesthesia is induced, before the skin incision is made, and before the patient leaves the operating theatre. The results of adopting the checklist have been evaluated in large cohort studies [18,19] and its use was found to reduce the risk of serious adverse events considerably. Another tool is the SBAR frame (short for situation background assessment recommendation), which is designed to elicit concise and focused information using standardized questions, and can be applied in any situation in which information about a patient is required to be communicated [20].
The present study was performed based on the assumption that checklists improve surgical safety. Although checklists have been reported to reduce hazards [13,19,21], they have also been questioned [22]. It is difficult to test safety routines because of the Hawthorne effect, i.e. the very fact that their outcome is going to be recorded may induce the intervention group to perform better. If implemented mechanically, checklists may create a false sense of security and decrease alertness, which can help in dealing with unforeseen hazards. The implementation of checklists should thus be followed by a careful evaluation of their impact on patient safety. The Safe Abdominal Surgery programme, for example, will be evaluated using data on adverse outcomes from gallstone, hernia, and colorectal cancer registries.