Identification and Evaluation of Nursing Errors in Kowsar Hospital by SHERPA

1. BACKGROUND

Nowadays, human errors in the medical field are one of the most important and sensitive issues which lead to many unacceptable risks for societies and are considered a serious and inevitable threat to patients and hospital staff [1, 2]. Human errors are defined as the deviation of human performance from the specified rules and tasks, which exceed the acceptable level of the system and have an adverse effect on the system's performance. Obviously, these errors are due to human errors and the weakness of systems [3].

Regarding the nature of human errors in the medical field and the very devastating effects of these errors, patient's safety has become one of the global concerns. Patient safety involves reducing or eliminating harm to patients due to the wrong way of treatment and providing health services. Endangering patient safety is continually one of the inherent concerns in medical professions, although there are scattered warnings and recommendations in this regard, organized formal response to improve patients' safety has been limited [4, 5].

Previous studies indicate that human errors in medical occupations impose high financial and human losses on communities. Health workers' mistakes result in patient’s lack of trust, increased stress and complaints of patients [6, 7]. Institute of Medicine report in 2000 showed that between 44 and 98 thousand people die annually in US hospitals due to medical errors [8]. Another report of the institute of medicine in 2008 indicated that, on average, one million people around the world annually injure due to medical errors imposing 17 million dollars’ financial loss [6, 7]. According to WHO, one out of every 10 patients is affected by medical errors [7-10]. The good news is that 70% of medical errors are preventable [11].

Human errors occur for various reasons covering a combination of system and hidden errors [12]. Preventing errors will increase patient and staff's satisfaction, reduce surgical costs, and increase the credit of service providers. In most organizations, the costs of error prevention are significantly lower than the costs of events arising from it [13].

Nurses who carry out various activities in the hospital are one of the medical teams who might incur mistakes in their work. In a study conducted at the University of Pennsylvania, 30% of nurses reported having committed at least one error per day [14]. Various factors can cause medical errors [15], including fatigue [16], quality of work-life [17], age [18], level of education [19], degree of respect to safety principles [20], experience [21], leadership and organization [22], organizational structure [23], patients demographic characteristics [24] and changing work tasks [25].

Although developing countries are relatively facing more medical errors than developed countries [26], due to weak reporting systems and complicated legal requirements for creating a database, there is absolutely no precise estimation [27]. An Iranian study by Panjvini concluded that 16.7 percent of nurses committed medical [28]. In another Iranian study, 64.5% of nurses admitted that they had committed medical errors [29].

Different methods are used to identify human errors, including FMEA and RCA methods. FMEA (FAILUR MODE & EFFECTS ANALYSIS) is an analytical technique based on the law of “prevention before it happening” used to identify potential failure factors. Root cause analysis (RCA) is also a structured investigation and research process whose purpose is to know the real cause(s) of a problem and to find ways to eliminate this cause(s) [30]. The SHERPA technique is one of the error detection techniques which is designed to identify human errors in terms of type of error, likely outcomes, and control and prevention strategies. Identifying and analyzing medical errors provides an opportunity to share experiences for improving the safety of patients and employees [31]. This study was conducted to identify and evaluate human errors in the nursing job of Kowsar Medical Center, which is the only specialized and referral obstetrics and gynecology center in Qazvin Province. This hospital provides 15,000 inpatients and more than 36,000 outpatients care annually.

2. METHODS

This cross-sectional study was conducted to identify and evaluate human errors among nurses of Qazvin Kowsar Hospital in 2018. Several techniques are used to identify, evaluate and predict human potential errors in dynamic systems. Among these, the SHERPA technique is one of the most common methods of checking errors, which can be used to study human errors in terms of error type, probable outcomes, and control and prevention strategies and can be studied in a set of subtasks forming each job [32].

3. SHERPA STUDY

This technique was first created and developed by Emberi in 1986. Today, this method is widely used in hazardous industries, including nuclear, oil and gas, electricity transmission and distribution and petrochemical. Lyon et al. mentioned this technique as one of the seven techniques used to evaluate the performance of healthcare providers [33]. This method is performed in 8 stages, including analysis of task hierarchy, task classification, identification of human errors, outcome analysis, error retrieval, error analysis, crisis analysis and analysis of corrective strategies [34]. After analyzing the hierarchy of duties and implementing the eight stages of the SHERPA using the 882D MIL-STD standard, risks' classification was performed [35], and finally, practical solutions were proposed to reduce and control some of the identified errors.

The tasks of nurses were divided into two parts: general tasks (related to all departments) and specialized tasks (related to specific departments of the hospital). The tasks were analyzed according to the Iranian nursing service standards reference book and the written tasks that nurses performed during the day, which were well known to the nurses participating in the study and according to what was done in the hospital, and the sub-tasks were determined. A total of 24 nurses, 3 in each department, were included in the study. The selection of the tasks was based on the opinion of the hospital officials, and included the most important tasks as well as the daily and routine tasks of the hospital nurses.We studied 32 tasks of nursing jobs, such as: General nursing tasks: patient admission, medication administration, CPR, nursing report registration, ward delivery, discharge with personal consent, and patient discharge. Specialized nursing tasks in the NICU department: connecting the ventilator to the patient, disconnecting the ventilator from the patient, caring for the patient under mechanical ventilation, suctioning, tracheostomy care, tracheal suction, and tracheal tube placement. Specialized nursing tasks in the ICU department: chest tube insertion, chest tube removal, patient monitoring, and NGT placement. Specialized nursing tasks in the neonatal department: admission of the newborn and placement of the umbilical catheter. Specialized tasks of nursing in the delivery department: anesthesia (spinal-epidural). Specialized nursing tasks in the operating room department: admitting the patient to the operating room, avoiding wrong connections of tubes, sounding (inserting a urinary catheter), and accepting the patient under observation. and changing the dressing. Specialized tasks of nursing in the emergency department: treatment of eclampsia in the emergency room and treatment of vaginal bleeding (first half of pregnancy) in the emergency room. Specialized tasks of nursing in the antenatal ward: sending the patient and taking an ECG from the patient. Specialized nursing tasks in the post-partum ward: placing a Foley catheter for hospitalized and atony patients. Following the formation of team, the required training was provided to the team members. After stating objectives and training the methodology using the hierarchical analysis method, tasks and subtasks of the nursing were determined, and the worksheets of SHERPA were completed through observation and interview. After that, worksheets were entered into Excel for analysis. Frequency and type of error (action: injection, retrieval: receiving information through the monitor, reviewing: guiding and managing a process, selection: choosing a route or other route according to medical command and communication: Conversation with Hospital departments) via worksheet SHERPA and risk level of errors using a matrix of classification of errors were specified.

We conducted interviews based on the Sherpa checklist and worksheet. The method’s reliability and validity have previously been approved [32]. This method has previously been applied in Iran’s healthcare setting and has been proven useful in recognizing errors [30, 36]

4. RESULTS

Thirty-two tasks were divided, and finally, 252 errors related to the activities of nurses were identified. Actions, communication, retrieval, revision and selection errors constituted 60%, 23%, 3%, 12% and 2% of the errors, respectivel (Table 1).

The cases were categorized based on risk level as follows: unacceptable risk with 16 errors, undesirable risk level 96 errors, acceptable risk level which needs revision with 107 errors, and acceptable risk level which does not need revision with 33 errors. As it is illustrated in Table 2, an acceptable risk, which needs revision, has the most percentage, and unacceptable errors have the least percentage (Table 2). Table 3 shows the frequency and percentage of errors in terms of the risk level.

Among retrieval errors, according to Table 4, the most non-revision errors from action errors were with 90 errors and the least non-revision errors of selection errors were 3 errors.

5. DISCUSSION

The most frequent errors of nurses were related to action errors with a frequency of 60%, which is consistent with Dastaran, Kermani, Krughdeh, Ghassemi [5, 36-40]. Revision errors came next, which happened due to the nature of the nursing job and its task.

Based on the level of risk, approximately 44% of the identified errors were unacceptable and undesirable; the unacceptable level of risk only allocated 6% of these errors, which are consistent with the study of Dastaran, Kermani and Krujdeh [5, 36, 37]. The highest “probability of errors” was for the possible errors and the least was for repeated errors, and the major part of “severity of error” was at the border level, and a lower percentage of them placed in the catastrophic and critical level, which is consistent with the study of Dastaran [5].

A significant part of the errors was identified at an undesirable risk level (38%) and acceptable with the need for revision (42%), which increases the importance and necessity of preventive and control measures. These measures include voluntarily reporting systems of error, improvement of diagnostic equipment and systems [39], guidelines, policies and related checklists [40]. Also, more attention to improving processes is recommended.

The risk levels of some errors were as follows: forgetting to assess the patient's respiratory conditions in the hospitalization task with a risk level of A3 (unacceptable); controlling the patient in terms of symptoms and immediate complications in the medication task with a risk level of A3 (unacceptable); control of vital signs in the task of connecting the ventilator to the client with risk level A3 (unacceptable); placing the baby in the right place in the chest for intubation with risk level D3 (acceptable with the need for revision); recording the actions taken in the patient file In monitoring the patient with risk level D3 (acceptable with the need for revision); lack of hand washing or disinfection in the task of placing the umbilical catheter with a risk level of B3 (unfavorable); and timely presence of all personnel in the task of delivering the ward with a risk level of C4 (acceptable without the need for revision).

Since a major part of errors do not have retrieval opportunities, and given that in the national accreditation system, lots of attention is paid to the issue of patient safety, it is considered as the national validation spirit. In addition to the measures and standards, another specific score is generally considered for hospitals. Therefore, the need for holding training courses at all levels of hospitals, especially at the level of senior managers, to justify and to gain their commitment to this issue is felt. Also, to prevent errors that had the most frequency, it is necessary to pay attention to standards, documents, hospital policies and in-service training.

In the general tasks of nursing, administering medication with a high-risk level and in the specialized tasks, connecting the ventilator to the patient, suctioning the trachea, monitoring the patient, and taking an ECG were among the high-risk tasks. However, the effective factors in the occurrence of errors were multi-factorial and related to people, equipment, and their organizations. In general, paying attention to the following items could be effective in reducing errors:

1-Equipment: Redesign or modification of the existing equipment, the establishment of standard reporting systems, and use of new medical technologies

2- Training: Employing experienced nurses and providing continuous training

3- Instructions: Providing new instructions, modifying old instructions, and preparing standard checklists

4- Organization: Making changes in the organization's policy, reviewing and improving management, and monitoring systems, stress management and standard teamwork

One of the limitations of this study is the lack of time to fully review all the duties of a nursing job and to study nursing errors in the hospital.

CONCLUSION

Most of the duties of nurses are related to provide healthcare and practical actions. Therefore, hospital authorities should pay more attention to functional errors and prioritize these errors in patient safety. It is recommended to include and implement specific measures in the field of patient safety culture in hospital standards to prevent errors. The SHERPA approach can be a good way to identify and evaluate medical errors.

REFERENCES