Article Text
PDF
Abstract
Background Universal protocol implementation can be challenging in low-income or middle-income countries, particularly when providers work together across language barriers. The aim of this quality improvement initiative was to test the implementation of a colour-coded universal protocol in a Guatemalan hospital staffed by US and Guatemalan providers.
Methods From 2013 to 2016, a US and Guatemalan team implemented a universal protocol at a Guatemalan surgical centre for children undergoing general surgical or urologic procedures. The protocol was a two-step patient identification and documentation checklist, with the first step of all chart element verification in the preoperative area, after which a blue hat was placed on the patient as a visual cue that this was completed. The second step included checklist confirmation in the operating room prior to the procedure. We tested protocol implementation over three phases, identifying implementation barriers and modifying clinical workflow after each phase. We measured the error rate in documentation or other universal protocol steps at each phase and made modifications based on iterative analysis.
Results Over the course of programme implementation, we substantially decreased the rate of errors in documentation or other universal protocol elements. After the first phase, 30/51 patients (58.8%) had at least one error. By the third phase, only 2/43 patients (4.6%) had any errors. All errors were corrected prior to surgery with no adverse outcomes.
Conclusions Care teams of providers from different countries pose potential challenges with patient safety. Implementation of a colour-coded universal protocol in this setting can prevent and reduce errors that could potentially lead to patient harm.
- paediatrics
- surgery
- patient safety
- checklists
- audit and feedback
Statistics from Altmetric.com
Introduction
Problem description
Increasing interest has been focused on the use of surgical checklists and universal protocols to improve the safety of surgical care around the world.1–3 The Universal Protocol for Preventing Wrong Site, Wrong Procedure and Wrong Person Surgery was designed as a tool to confirm the site, procedure and person before any surgical procedure, and has been implemented in all of The Joint Commission (TJC)-accredited hospitals in the USA.4–6 Use of this checklist has been supported by the WHO Guidelines for Safe Surgery 2009, and is intended to improve perioperative mortality in a number of settings.7 8 The universal protocol includes preprocedure verification, surgical site marking and a surgical ‘time out’ prior to the procedure.4–6
Available knowledge
Despite the increased interest in patient safety and demonstrated benefits of the universal protocol, implementation of universal protocols has been limited in many low-income and middle-income countries (LMIC), although the reasons for lack of adoption are poorly defined.7 9 10 Common barriers to implementation of health programmes in low-resource settings include the complexity and costs of interventions, capacity of human resources to adopt new programmes, lack of leadership, weak health systems, and complex social, cultural and political contexts.11–14 Although tools to improve safety have the greatest potential to impact health in LMICs, implementation of surgical safety programmes in LMICs has been limited to date.15 16 Use of checklists in particular remains poor in LMICs, with compliance as low as 12%.17–19 Cultural differences may play a role in the reduced use of safety procedures. For example, in Thai culture, it is against societal norms to make marks on other people, which may constrain staff from marking the operative site before surgery.10 Identifying and addressing such differences would assist in protocol implementation.
To test the implementation of a universal protocol, we performed a quality improvement (QI) initiative at the Moore Pediatric Surgical Center (MPSC) (Guatemala City, Guatemala). The MPSC serves indigent children requiring surgical care from across Guatemala, with care provided by US and Guatemalan staff working together at all phases of surgical care.20 Previous work at the MPSC had shown concerns among hospital staff about surgical safety and preoperative checklists, specifically challenges with communication between providers across different languages, cultural biases and differences in hospital policies.21 In addition to these concerns, local staff at the MPSC raised concerns about visiting clinicians working in unfamiliar environments, which may increase the risk for wrong site, wrong procedure or wrong person procedures.
Rationale
To guide our analysis of the implementation of the universal protocol programme as well as to understand the context of care delivery, we used the Consolidated Framework for Implementation Research (CFIR).22 The CFIR is an established conceptual framework to systematically assess factors that may influence intervention implementation and effectiveness.23 We chose to use this framework for this project, as the CFIR merges constructs from several disciplines into a single framework that organises interacting domains which are directly relevant to the context of this study in Guatemala, including the universal protocol intervention itself, outer setting (economic, political and social context in Guatemala), inner setting (organisation structure, culture and internal climate of the MPSC), individuals involved and process (methods and approaches used in facilitating adoption of the universal protocol).
Specific aims
The aim of this QI initiative was to implement a colour-coded preoperative checklist and universal protocol in a single paediatric surgical centre in Guatemala, and assess the impact of this protocol on the rate of medical documentation errors.
Methods
Our project followed the Standards for Quality Improvement Reporting Excellence (SQUIRE) 2.0 guidelines for study design and analysis. These guidelines provide a framework for the reporting of efforts to improve the quality, value and safety of healthcare, the use of theory in planning, implementing and evaluating improvement work, the context in which the work is done and the study of the intervention(s).24
Context
The MPSC is a 12 000 square foot facility with 3 operating rooms (OR), 6 preoperative and postanaesthesia care unit beds and 21 recovery beds. It employs 26 year-round staff, including physicians, nurses and OR technicians. We implemented this protocol in 148 children who underwent a range of paediatric general surgery and urology procedures during three separate 1-week periods between September 2013 and April 2016. Patient age ranged from 6 months to 17 years. All families spoke Spanish or a Guatemalan indigenous language as their primary language, and a trained interpreter was used to facilitate communication during all safety processes.
Before being scheduled for any procedures at MPSC, children recommended for surgery were evaluated in a preoperative clinic by a US surgeon working with a Guatemalan surgeon as well as a member of the US anaesthesia team. Documentation at the MPSC was in both English and Spanish. A surgical history and physical (H&P) and a preanaesthesia evaluation were completed in English, with the type and site of the planned procedure as well as the anaesthesia plan discussed with the family through use of a qualified interpreter. The patient and parent or guardian were then seen by a Guatemalan physician, who obtained the surgical and anaesthesia consent in Spanish based on the surgeon’s documentation in the H&P. This local physician was able to communicate efficiently with the family during the consent process and used a translator for other indigenous languages when needed. A US staff member took each child’s photograph, labelled it with his/her name and medical record number (MRN) and then attached it to the patient’s chart for identification purposes during the perioperative period.
Interventions
To implement this protocol, a US-based team worked closely with local partners at the MPSC to develop a timeline and training system to facilitate protocol adoption. Prior to implementation of this protocol, existing safety care protocols at the MPSC did not mandate any check of medical documentation prior to any surgical procedure. Our protocol was based on principles of the universal protocol as defined by TJC, including use of a two-step verification process (figure 1). In line with these principles, our protocol required that every patient undergoes surgical site marking and a ‘pre-procedure verification’ in the preoperative holding area (prior to entry into an OR) (step 1). After successful completion of the preprocedure verification, patients had their OR hats changed from red to blue as a visual cue to all staff that they were ready to be brought to the OR. In the OR, after induction of anaesthesia, but prior to the beginning of each procedure, each variable was reconfirmed with all OR staff using a second checklist as part of a ‘pre-surgical time-out’ (step 2).
Flow chart of two-step preoperative colour-coded universal protocol based on principles of The Joint Commission (TJC). Step 1: In the preoperative area, a red hat is placed on the child as a visual cue that preoperative verification is not complete. Preprocedure verification is performed with the patient, parent/guardian and members from the anaesthesia and surgical nursing teams with assistance of a medical interpreter. Following verification of identification, all documentation and site marking, a blue hat is placed on the patient as a visual cue that verification is completed. Step 2: In the operating room, a time-out is performed with the OR checklist confirmed by all team members prior to proceeding with the case. NPO, nil per os; OR, operating room.
On the day of each procedure, the patients were placed in a waiting area with a parent or guardian. All patients had an armband with their name and MRN and received a red hat as a visual cue to signify that they had not yet been cleared for the OR. While the patients were in the preoperative area, the surgeon checked all documents and marked the appropriate surgical site. At this point, all medical documents were reviewed together by a member of the US and Guatemalan nursing team and anaesthesia team alongside the patient and guardian with the assistance of an interpreter. This ‘pre-procedure verification’ consisted of verifying patient identifiers (name, MRN, picture), signed surgical and anaesthesia H&Ps, allergies, correct surgical procedure and marking, nil per os status and completion of surgical and anaesthesia consent. After successful completion of this verification step, a blue hat was placed on the patient as a visual cue to all members of the care team that the verification had been completed, signalling that the patient had no inconsistencies with the universal protocol and that he or she was ready to be taken to the OR. Any inconsistencies or errors with the verification process were addressed and corrected prior to the patient receiving a blue hat.
After transport to the OR, to complete the universal protocol, all OR team members performed the second stage ‘pre-surgical time-out’ prior to beginning of procedure. During this ‘time-out’, all OR activities are stopped. The patient’s name and MRN were compared between the armband and chart, allergies were reviewed, procedure from the surgical consent and site marked were reviewed, use of preoperative antibiotics was confirmed (if indicated) and the surgical skin preparation was confirmed to be dry. All team members confirmed completion of the checklist and all discrepancies were resolved prior to proceeding with case.
Study of the interventions and measures
For this QI initiative, we studied the implementation of this colour-coded universal protocol using a pre/post implementation analysis in an iterative fashion over three phases (measured at baseline, during implementation and after implementation). During each phase of protocol implementation, we prospectively collected the rate of errors in the preprocedural documentation process. Errors were defined as any inconsistency or incongruence identified during the first or second-stage checklists. If there was an identified inconsistency, the details of such were described in a deidentified spreadsheet for each stage in the process by a member of the anaesthesia care team. All data were analysed at the end of each 1-week period.
During each phase of protocol implementation, we performed discussions with Guatemalan and US perioperative clinical staff as well as hospital leadership (n=22 total) using a series of focus group discussions. Each focus group included probing participants with semistructured open-ended questions to identify concerns among all local and visiting clinical staff about barriers and facilitations to use of this protocol. We used these focus groups to discuss the importance of minimising documentation errors and their potential for patient harm.
Over the course of three phases of protocol implementation, we modified the protocol based on focus group feedback. After the first phase (baseline), we did not change any preoperative protocol processes, such that we could identify whether the implementation of the colour-coded protocol itself, along with the focus group discussion of problems with protocol implementation, would impact the rate of documentation errors. Following the second phase of protocol implementation, we reviewed the rate of documentation errors and performed similar focus group discussions to define any barriers and facilitators to protocol implementation. Following this phase, we worked with MPSC staff and leadership to revise the preprocedural documentation process at the MPSC to fully implement the colour-coded protocol, including instituting several structural changes to the workflow to decrease the rate of errors. Following the third phase of this QI initiative, we again collected the rate of errors after these workflow changes.
Analysis
For implementation analysis, we included both a process and an outcome evaluation. We recorded and analysed comments from each focus group discussion, specifically identifying comments which reflected staff attitudes towards their surgical workflow and safety processes. We identified several exemplar comments, and used these comments to describe recurrent themes that emerge from data as well as the theory underlying the project.25 Two staff independently identified key findings for each focus group, and classified key findings within each domain of our theoretical model (outer setting, inner setting, individuals and process). Using these results, we summarised the barriers and facilitators to implementation of the universal protocol.
Ethical considerations
According to the policy activities that constitute research at our institution and the MPSC, this work met criteria for operational improvement activities, and was considered exempt from review by our institution’s Institutional Review Board.
Results
Clinical outcomes
During this QI initiative, we confirmed successful implementation of a colour-coded safety protocol at a paediatric surgical centre in Guatemala and found that implementation of this protocol over several phases in an iterative fashion resulted in a significant improvement in the perioperative safety process. In the first phase of this project (baseline), we identified a high rate of errors during use of baseline (existing) preprocedure verification processes, including 14/51 (27.5%) missing or incorrect patient identifiers, 1/51 (2.0%) surgical H&P with an incorrect surgical site and 19/51 (37.3%) missing or incorrect surgical consents. A total of 30/51 (58.8%) patients had at least one inconsistency necessitating clarification (figure 2). All missing or incorrect patient identifiers were resolved prior to patient transport to the OR. There were no inconsistencies or objections noted in any second-stage ‘pre-surgical time-out’ performed in the OR prior to the start of any surgical procedure and no adverse events.
Percentage of patients with errors identified at the ‘pre-procedure verification’. Missing or incorrect surgical consents fell from 37.3% to 2.3%, incorrect surgical site documentation on the surgical H&P fell from 2.0% to 0% and missing or incorrect patient identifiers fell from 27.5% to 2.3%. Overall rate of patients with at least one error fell from 58.8% to 4.7%. H&P, history and physical.
After the first phase of protocol implementation demonstrated a high rate of errors, we shared these results with all US and Guatemalan staff. During these discussions, almost all team members were surprised at the high error rate, and local staff expressed appreciation for the opportunity to improve the safety processes. Local leadership and staff recognised the implications of current safety practices to result in potential wrong patient, wrong procedure or wrong site surgery. At that stage, we introduced the concept of using a colour-coded universal protocol to potentially decrease the rate of documentation errors. However, local staff as well as hospital leadership did not feel the need to implement any change in the workflow processes, as most staff vocalised doubt that these errors would be repeated due to a state of increased awareness. A consensus decision was made not to change practices, and all staff were informed that we would repeat this data collection to assess the impact of the exposure of the protocol itself on the rate of documentation errors.
After the second phase of protocol implementation, without any change in existing safety workflow, we again found a high rate of errors identified during the ‘pre-procedure verification’, including 15/54 (27.8%) missing or incorrect patient identifiers, 0/54 (0%) surgical H&P with an incorrect surgical site and 16/54 (29.6%) missing or incorrect surgical consents (figure 2). A total of 29/54 (53.7%) patients had at least one inconsistency necessitating clarification. There were no inconsistencies or objections noted in any second-stage ‘pre-surgical time-out’ performed in the OR prior to the start of any surgical procedure and no reported adverse events. These data were again shared with US and Guatemalan healthcare providers in focus group discussions. During these discussions, both Guatemalan and US providers were surprised at the high error rate. In discussion at this stage, staff were in agreement that changes in workflow were indicated. Local leadership and staff opted to include the colour-coded two-stage universal protocol into all future perioperative workflows, as well as use of one additional step of a written preprocedure verification checklist for use in the preoperative clinic to support proactive identification and rectification of any potential issues at that time.
After the third phase of protocol implementation (after adoption of a colour-coded universal protocol), we found a markedly lower rate of documentation errors, including 1/43 (2.3%) missing or incorrect patient identifier, 0/0 (0%) surgical H&P had an incorrect surgical site and 1/43 (2.3%) had missing or incorrect surgical consents. A total of 2/43 (4.7%) patients had at least one inconsistency necessitating clarification (figure 2). There were no inconsistencies or objections noted in any second-stage ‘pre-surgical time-out’ performed in the OR prior to the start of any surgical procedure and no reported adverse events.
One year after protocol implementation, MPSC leadership confirmed that this protocol had been integrated into hospital policy and was practised consistently in clinical workflow. Internal audit over a 1-week period showed that 0/58 (0%) patient had any documentation errors.
Implementation analysis
Implementation analysis based on focus group discussions demonstrated several themes which defined the barriers and facilitators to adoption of a colour-coded universal protocol (figure 3). We classified these themes in line with the CFIR framework. Several factors were operational in the outer setting of healthcare, including a long-standing social resistance to healthcare change, a hierarchical structure to OR dynamics and a political context which may limit healthcare change. Representative exemplar comments from the nursing staff included:
Our directors want to direct all of the change…they do not ask us our opinions.
Identification of the barriers and facilitators to implementation of a colour-coded universal protocol in Guatemala. Implementation analysis based on focus group discussions and summarised using the theoretical framework of Consolidated Framework for Implementation Research (CFIR). OR, operating room.
We found several barriers operational within the inner setting, including a system which is not used to healthcare change as well as one which needs to be flexible to the systems of teams visiting from many different institutions, each with their own culture and safety policies. Representative comments from both leadership as well as nursing staff included variations on the following:
It is hard to change our safety policies with each different visiting team.
Over time, the individuals involved (physicians, nurses and administrative staff) were quite receptive to healthcare change, although process changes were delayed from the second to the third stage of the implementation process until it became recognised by all stakeholders that individual initiative was not enough to induce process improvement. It was noted that physician acceptance appeared critical for nursing staff acceptance, consistent with the hierarchical nature of the Guatemala health system.26 Finally, the implementation process itself required iterative improvement over multiple phases and recognition of the value of colour coding to simplify confirmation of the completion of the protocol, which allowed for optimal local input into all stages of the process. Representative comments from nursing, physician and leadership staff confirmed the ease of the perioperative process, including comments such as:
Now we can see how easily the color-coding helps with our checklist. It is easy to do and makes sense…
Discussion
Performing surgical care in an environment with staff from multiple countries, particularly with language and/or cultural barriers, poses substantial challenges with patient safety.27 28 In almost all settings, patient safety can be improved with the use of the universal protocol as a preprocedure checklist. In order to successfully implement a universal protocol, the entire team must be able to understand, accept and use the protocol effectively for maximal benefit.29 In this study, we found that implementation of a multistep, colour-coded universal protocol allowed us to identify inconsistencies or errors with patient identification, surgical procedures and surgical site that could potentially lead to patient harm. During our implementation process, we confirmed that iterative implementation of this process with local staff feedback at each phase facilitated protocol adoption. Finally, implementation of such a protocol has the ability to significantly impact other upstream processes, in this case modification of the documentation process in the preoperative clinic such that any documentation inconsistencies and errors could be resolved prior to the day of surgery.
Interpretation
The use of surgical checklists varies widely around the world. In Guatemala, the WHO-supported Surgical Safety Checklist is not a mandate or requirement for any hospital, nor is it required for accreditation from the Ministry of Health, although some local hospitals have adopted the protocol.13 Specifically, Delgado Hurtado et al noted in Guatemala that gaps in the knowledge and awareness of the WHO Surgical Safety Checklist can negatively affect the correct use of the protocol.13 In contrast, at the MPSC, we confirmed the ability to implement a colour-coded universal protocol, demonstrated that it can reduce the risk for potential errors and demonstrated sustained integration of this protocol into hospital policy and practice. The challenge at the MPSC, similar to other settings which involve providers from different countries or institutions, is to maintain the professional identity of all stakeholders, and respect institutional and organisational leadership. Our experience has shown that a colour-coded universal protocol can be conducted consistently within this care setting, despite the large number of groups involved in clinical care at MPSC.
Potentially the most helpful part of our implementation analysis was the value of successfully introducing a new healthcare programme in an iterative fashion over multiple phases, such that local staff can be engaged and lead the implementation process itself rather than simply be recipients of new healthcare programmes. We found that by measuring the rate of medical documentation errors over multiple phases as the issues of patient safety and perioperative processes were gradually introduced into the local care setting, local staff were able to recognise the need to change workflow to improve patient safety. Although this process took more time and effort than simply dictating a new healthcare programme, this process was critical to acceptance of the need for this programme by local staff and leadership.
Limitations
There are several limitations in our study. First, we focused our outcomes on documentation errors, which may differ from actual adverse outcomes related to preoperative checklists. However, the universal protocol preprocedure checklist is designed to minimise the risks for potential for medical errors, and therefore any analysis of protocol use is unable to demonstrate whether any documentation errors would have actually led to patient harm. However, as perioperative checklists are a reliable tool to help prevent such errors in other similar settings, it is reasonable to assume that the protocol itself can reduce errors in this context. Second, since this study was performed at a single centre, the generalisability of implementation of a colour-coded safety protocol across other settings in Guatemala or LMICs is unclear. Each hospital in a healthcare system should identify how to use such a tool in order to minimise medical errors within the context of their local environment. Further analysis should identify the generalisability of tools such as this to other LMIC or low-resource settings to improve patient safety.
Conclusion
In all parts of the world, the universal protocol can serve as a useful tool to enhance patient safety to minimise the potential for wrong patient, wrong site or wrong procedure surgery. In order for it to be successful in environments limited by language barriers and different perioperative practices, the entire perioperative team has to be committed and engaged in the implantation of tools that are accepted by all staff. Implementation of a colour-coded Universal Protocol in these settings may help with adoption of programmes to reduce patient harm and improve system-based weaknesses in patient safety around the world.
Acknowledgments
The authors would like to acknowledge the following people and organizations for their assistance and support of this initiative: Bethany Bedford, Lisa Biltz, Caitlin Curtis, Lisa Einhorn, Emily Funk, Stephanie Jones, Milena Kaleab, Monica Mohe, Rebecca Motykiewicz, Jennifer Neifeld, Megan Nute, Shelly Pecorella, Crystal Pennington, Dayna Seguin, Jill Smith, Casey Vera, Jamie Wardlow, Sara Zucco, The Shalom Foundation, and Mending Kids International
References
Footnotes
BMT and ST contributed equally.
Competing interests None declared.
Ethics approval IRB.
Provenance and peer review Not commissioned; externally peer reviewed.