Barriers to Incident Reporting among Nurses: A Qualitative Systematic Review

Incident reporting in health care prevents error recurrence, ultimately improving patient safety. A qualitative systematic review was conducted, aiming to identify barriers to incident reporting among nurses. Joanna Briggs Institute methodology for qualitative systematic reviews was followed, with data extracted using JBI QARI tools, and selected studies assessed for methodological quality using Critical Appraisal Skills Program (CASP). A meta-aggregation synthesis was carried out, and confidence in findings was assessed using GRADE ConQual. A total of 921 records were identified, but only five studies were included. The overall methodological quality of these studies was good and GRADE ConQual assessment score was “moderate.” Fear of negative consequences was the most cited barrier to nursing incident reporting. Barriers also included inadequate incident reporting systems and lack of interdisciplinary and interdepartmental cooperation. Lack of nurses’ necessary training made it more difficult to understand the importance of incident reporting and the definition of error. Lack of effective feedback and motivation and a pervasive blame culture were also identified.


Review Article
The World Health Organization defines an incident as "any deviation from the usual care that poses a risk of harm or causes injury to a patient, it includes errors, preventable adverse events, and hazards" (World Health Organization, 2005), These errors are one of the main persistent threats to patient safety (Bifftu et al., 2016;Chard, 2006). As shown in the Institute of Medicine (IOM) report "To Err is human," hundreds of people die every day due to preventable medical errors, which is much more than people who die because of airplane accidents (Institute of Medicine (US) Committee on Quality of Health Care in America, 2000). Economic losses of health care organizations are also a result of medical and nursing errors; more than 15 percent of health spending is wasted on costs of additional hospitalization, litigations, and other consequences of adverse events and poor quality (World Health Organization, 2018).
All professionals are bound to make mistakes from time to time, especially in a complex profession such as health care (Cleary et al., 2018), but a key step in improving the quality of care and enhancing the patients' safety is learning from these mistakes (Ellis & Abbott, 2019). One of the main ways of monitoring errors to prevent their recurrence is incident reporting (Hwang et al., 2012). An incident reporting system (IRS) is one of the bases of safe practice within an organization-having an effective IRS in the organization significantly increases the safety of the patients (Pham et al., 2010), it enables the organization to learn from past errors (Stavropoulou et al., 2015), and also helps identify potential risk and alerts staff to possible system failures in the organization (Pham et al., 2010).
In health care, "Silence Kills" (Ackerman, 2018;Wolf & Hughes, 2008); thus, both obligatory and voluntary reporting in hospitals are encouraged. While a well-reported, analyzed, and openly communicated incident helps everyone in the workplace understand the circumstances in which that incident occurred, learn from it, and thus prevent similar errors from occurring in the future (Galt & Paschal, 2009), nurses are reluctant to report medication errors, while they can identify them (Dirik et al., 2019).
Even though it has its drawbacks, incident reporting remains one of the most widespread strategies to improve patient safety (Macrae, 2016). While reporting is not just about speaking up, transparency about errors, or openness in communication, nor is it about filling up a form, it is about taking the first step in fixing a problem at hand and learning from it to avoid its recurrences, and voluntary reporting can really help identify health care systems' vulnerabilities (Vincent, 2011). Reporting of events is of little value unless the data reported were appropriately analyzed by skillful personnel (Wallace, 2010).
So it is pivotal that organizations stress on what comes after incident reporting, but first, patient safety incidents must be clearly defined and prioritized in order to deal with them according to their risk (Wolf & Hughes, 2008). Incidents can be classified into adverse events, which implies an unintended injury that results from medical management, mistakes/errors that are acts of commission or omission, near misses that are errors without a subsequent adverse event, and bad outcomes that are undesirable outcomes due to a disease process (World Health Organization, 2005). Reporting in the context of this article includes both selfreported errors and errors reported by other health care professionals, but not errors reported by automated software systems.
Organizations that have characteristics of transparency, openness in communication, and the learning from their mistakes tend to be much safer and more successful than those that do not (Reason, 2016). In a time where continuous quality improvement is essential for any organization to excel, every error that occurs-no matter how small or big it isshould be considered as an opportunity to improve. Incident reporting also helps organizations identify if there is a need for a process change or a system reform (Wolf & Hughes, 2008).
Potential future medical errors can best be prevented through reporting near misses (Wolf & Hughes, 2008). The reporting and analysis of near misses as well as errors can promote a "culture of learning" in health care organizations, creating an opportunity for all health care professionals to learn and prevent errors from occurring (Galt & Paschal, 2009). A "restorative" justice approach is better than a "retributive" justice approach, in that a restorative approach works toward understanding failure and fixing it rather than only focusing on the individuals involved. An organization with a "just culture" should not only look at the sharp end of the organization where frontline staff work and most accidents happen, but also dig deeper, hear from everyone, relate, and express remorse. There should be a less sense of blame and a more shared sense of trust, learning, and also accountability (Dekker, 2018), allowing people to be more engaged and be a part of the solution instead of fearing for their careers. These are all considered a part of the road toward making health care establishments highly reliable organizations (Sutcliffe et al., 2017).
There should be a process of justly engaging, listening to, and crying with the people involved in errors; "holding people accountable" does not always have to mean sanctions and punishment (Dekker & Hugh, 2014). Humans are fallible, and thus errors are inevitable, but working together to understand the complex high-risk system that health care workers work within will lead to a prevalence of a "culture of learning" and understanding why these errors occur and thus preventing them from occurring in the first place (Israelski, 2010). Hospitals should always work toward a design-driven improvement to help "design-out" potential mistakes (Fondahn et al., 2016) and help staff identify challenges in their work without assuming it is their fault. The focus should be on the system, to find out latent errors (Reason, 2000), and understand the resilience involved.
Health care regulatory bodies and oversight entities have recently started mandating that health care organizations should submit certain reports concerning adverse events; however, without institutionally based voluntary incident reporting, it is very difficult and challenging to catch errors (Wachter, 2012).
One in 10 hospitalized patients experiences an adverse event during their stay (Michel et al., 2017). Studies show that a great number of errors are made by nurses (Thomas, 2010), as 30 percent of nurses admit to committing at least one error every month (Hashemi, 2008), while nurses are less likely to report a medication error, especially near misses (Farag et al., 2017). The underreporting of adverse events and near mises remains high despite multiple reporting systems (Hamilton et al., 2018). Underreporting is considered a significant challenge facing error monitoring and error reduction efforts (Noble & Pronovost, 2010).
In order to start developing interventions that would encourage error reporting among nurses and increase their willingness to report, we first need to identify the barriers to nursing error reporting. A number of studies were conducted over the past 20 years in efforts to identify the factors contributing to error reporting by nurses and other health care professionals. Barriers identified include the lack of effective feedback to the reporter (Evans et al., 2006), as well as many other organizational and individual factors like the perceived negative administrative reactions and the culture of blame (Aboshaiqah, 2013). Systematic evidence explored attitudes or perceived barriers toward incident reporting of nurses mainly working in an acute care setting (Fung et al., 2012a), identifying that incident reporting vary across different study settings, with perceived barriers hindering the reporting process; thus, any potential intervention should be adapted to the different setting in order to increase reporting rates. Furthermore, Vrbnjak et al. (2016) systematically reviewed the literature until 2015 to identify barriers to nurses' reporting of medication errors and near misses, identifying that a "a non-blaming, non-punitive and non-fearful learning culture at unit and organizational level" is needed in order to enhance the reporting rates of medication errors and near misses. Efficiency of the reporting systems, management behavior, and nurses' education were the main areas where barriers were identified, while Rohde and Domm (2018) concluded that there is "minimal evidence clearly articulating nurses' clinical reasoning used to support medication errors." Considering the fact that underreporting of errors in health care is still common in many countries, including the UK, the United States, China (Gao et al., 2019), and South Korea (Lee, 2017), and barriers on incident reporting is a continuous evolving field of research, there is a need to further synthesize existing evidence on barriers on incident reporting not only on medication errors, as it has clear implications on patient safety, health care research, and health care policy. Although findings of quantitative, qualitative, and mixedmethod studies on barriers to nursing incident reporting appear to be somewhat similar, this review critically assessed the most relevant qualitative studies individually, synthesizing their findings to present a synthesized summary of the most important findings and conclusions, degerming the meaning of the phenomenon. The review contributes to nursing practice as it identifies the main barriers to nurses' reporting, researchers and policymakers can use these findings to develop strategies to encourage and increase reporting, leading to a more effective error reporting system in health care.

Methods
A qualitative systematic review approach was followed, using The Joanna Briggs Institute (JBI) methodology for qualitative systematic reviews (Pearson et al., 2005;The Joanna Briggs Institute, 2014). The research question was framed using the Population/Participants, Phenomenon of Interest, Context/Setting, Study Design (PICoS) framework (Methley et al., 2014).

Search Strategy
A scoping pilot search on PubMed and CINAHL databases was conducted, so as to get a broad picture of the available literature and identify the most appropriate keywords (Tong et al., 2012). Following the initial search, the search queries were carefully tested and fine-tuned, using alternative terminology and synonyms that could be used to give more appropriate results by both authors. The multi-field search query was: (nurs* OR midwi*) AND (barrier* OR obstacle* OR hind*) AND (report* OR statement* OR claim*) AND (error* OR Incident* OR event* OR near miss* OR nearmiss* OR fault* OR mistake* OR mishap*) AND (qualitative OR interview* OR field stud* OR narrative* OR experience* OR opinion* OR view* OR perception* OR outlook* OR belief* or emotion* OR focus group* OR attitude*). A systematic search was conducted in five health care-related bibliographic electronic databases: PubMed, CINAHL, HMIC, PsycINFO, and EMBASE. All the identified papers were imported to "Mendeley" reference management system, and duplicates were removed. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) tool was utilized to undertake the studies' selection process. After the duplicates were removed, titles and abstracts of the remaining papers were read and examined against the eligibility criteria, and then a thorough screening of the full paper was carried out to identify studies that exactly match the set review criteria (Moher et al., 2009).

Inclusion and Exclusion Criteria
Types of Participants. Only studies where the participants were nurses were considered for inclusion in this review, and studies that explored the views of patients or health care professionals other than nurses were excluded. The choice to limit the review to the nurses' perceptions was made because nurses have a very significant role to play in the error reporting systems (Wolf & Hughes, 2008). There were no limitations on the roles/type of nurses, so studies that had participants from emergency care, acute care, ambulatory care, home care, hospice care, long-term care, or psychiatric health care were all considered.
Phenomena of Interest. There were no limitations on the types of the reporting incidents, so barriers to reporting near misses, medication errors, and sentinel events were all considered. Studies that addressed both barriers and facilitators to nursing incident reporting were also considered.
Context. The review included nurses working in any health care facility (e.g., primary, secondary, or tertiary) and in any country. All studies that were undertaken before the year 2000 were excluded, considering the fact that it was not until that year that patient safety and quality of care started to become a big part of health care research (Leape & Berwick, 2005;Stelfox et al., 2006). While multiple studies exist before 2000, the landmark IOM report recommended reforms that would expand reporting substantially (Leape, 2002), and many countries, such as the UK, Denmark, and Australia, developed a national IRS policy, even after 15 years of its launch, error reporting has not yet reached its full potential (Mitchell et al., 2016). Only studies published in English were included.
Type of Studies. This systematic review focused on qualitative studies that reported beliefs, opinions, attitudes, and experiences, through interactive data collection methods like interviews and focus group discussions. All quantitative and mixed-method studies were excluded. While the qualitative part of mixed-method studies, when clearly presented, have been included in qualitative systematic reviews, the value of mixed-method studies' findings is in the combination of data and explored separately without the types of mixed-method designs being taken into account (e.g., explanatory, exploratory, parallel, nested (embedded) designs [Shorten & Smith, 2017]) that might engage some bias. Qualitative studies make distinctive contributions to leadership studies (Bryman et al., 1996) and are much more appropriate and effective when studying perceptions and experiences (Parse et al., 1985;Pope & Mays, 2006), and hospitals have the best chance of tackling barriers that hinder reporting by using questionnaires for nurses to identify these barriers and work toward and encourage better reporting (Rutledge et al., 2018). Only studies published in peer-reviewed journals were considered. Systematic reviews, editorials, notes, commentary pieces, conference papers, and book chapters and theses were excluded.

Assessment of Methodological Quality
All included papers assessed for methodological validity. To determine the methodological quality of the paper, the Critical Appraisal Skills Program (CASP) tool for qualitative studies was used (Critical Appraisal Skills Programme, 2018). There was no weight to specific questions nor was there a scoring system for the checklist; however, each answer was explained and justified. There were no exclusions made based on the methodological quality of the studies, but they contributed to the interpretation of the overall findings (Dixon-Woods et al., 2004).

Data Extraction
The standardized qualitative data extraction tool-Qualitative Assessment and Review Instrument (QARI; The Joanna Briggs Institute, 2014)-was used to extract the data from the included papers. Data were extracted using the meta-aggregation approach, which is a multiphase process, as the JBI QARI tool includes two forms. The first one is where the general details of the study were extracted, including-but not limited to-citation details, phenomenon of interest, participants, setting, geographical and cultural information, study methods, and conclusions. The second part of data extraction-using JBI QARI (which is also considered the first step in data synthesis)-is where the findings that are relevant to the review questions were extracted. "Findings" are the statements and texts of interest to the review as published in primary studies. The included papers were read more than once until a good understanding of the studies, their context, and their outcome was gained. The findings which were reported in the papers as "categories" were extracted to the QARI form and accompanied by a verbatim illustration from the publication; this illustration was either a direct quotation from a study participant or, where not available, other supporting data from the author's words (The Joanna Briggs Institute, 2014). A level of credibility, based on the degree of evidence of the illustration, was assigned in each finding. The first author extracted the data, while the second author agreed to the extracted data.

Data Synthesis
The JBI meta-aggregation approach (The Joanna Briggs Institute, 2014) was used for data synthesis. In the process of meta-aggregation, there is no reinterpretation of the studies' findings or the generation of new outcomes; instead, the meta-aggregation approach aims at organizing evidence, by compiling categories articulated by the researchers of the primary studies, and creating synthesized findings that help avoid misunderstandings and misinterpretations (Florczak, 2019). The process of meta-aggregation is an explicit interpretive process that involves the extraction of studies' findings, with processes of categorization and synthesis (Hannes & Lockwood, 2011).
Every two or more related study findings were grouped together into "categories." They were grouped on the basis of similarity in meaning and concept, and only the credible (C) or unequivocal (U) findings were included in the categorizing process. These categories were then compared and synthesized to generate overarching synthesized findings (a set of statements that represent the best aggregation of categories and findings). Authors re-examined each finding to make sure the process was carried out exhaustively, and descriptions were developed for each category and synthesized finding of the review (Hannes & Lockwood, 2011).

Assessment of Confidence in Review Findings
Confidence was established in the synthesized findings by using the GRADE ConQual tool (Munn et al., 2014), where the included papers were graded and ranked to determine the extent to which the findings were a reasonable representation of the phenomenon of interest. This was carried out through the assessment of dependability (i.e., reliability) by answering a set of questions linked to the quality and the appropriateness of the papers, and credibility (i.e., internal validity), which were assessed according to the authors' view of how well the review findings fit with data from the original studies (Munn et al., 2014).

Study Inclusion
A total of 921 records were identified (235 from PubMed, 183 from CINAHL, 28 from HMIC, 87 from PsycInfo, and 388 from EMBASE), and 422 duplicates were removed. Titles and abstracts of the remainder that consisted of 499 papers were reviewed by the first author. 12 studies included for full-text assessment for eligibility, resulting to five studies selected independently by both authors meeting the inclusion and exclusion criteria, as depicted in the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) flowchart ( Figure 1; Moher et al., 2009). Seven studies were excluded for the following reasons: two of the studies explored physicians' perceptions and not nurses' perceptions, while this systematic review only focused on nurses' perceptions; two of the studies used quantitative questionnaire approach, while this systematic review only focused on qualitative studies; one of the studies was a thesis and not a published, peer-reviewed article; one study included only the abstract as it was part of conference proceedings; and in one study, the abstract was available in English.

Assessment of Methodological Quality
The quality of each study was assessed using the CASP qualitative checklist. All five studies included clear descriptions of how they were conducted as a whole. The overall quality of the included studies were considered as relatively good; however, all the included studies failed to clearly address the researchers' relationships with the participants. Potential bias and the risk involving the influence of the researcher should have been discussed to allow for more reflexivity in the studies (Malterud, 2001). In relation to sampling, one of the five studies did not mention its recruitment strategy at all (Hashemi et al., 2012), and two of the five studies did not discuss data saturation (Hwang et al., 2012;Peyrovi et al., 2016). Table 1 depicts a summary of the overall methodological quality of the included studies.

Study Characteristics
All the five included studies were published within the past eight years (from 2012 to 2019). Coincidently, all five studies were conducted in developing Asian countries-two studies in Iran (Hashemi et al., 2012;Peyrovi et al., 2016), one in Korea (Hwang et al., 2012), one in Malaysia (Dyab et al., 2018), and one in China (Song & Guo, 2019). Four of the five studies were conducted in urban settings (Dyab et al., 2018;Hashemi et al., 2012;Peyrovi et al., 2016;Song & Guo, 2019), while one study did not mention specific locations (Hwang et al., 2012).
All participants in the five studies were nurses, in different age groups, and with varying years of work experience. A total of 234 nurses were included in these five studies, with majority of being female nurses; the exact female to male ratio is not available because one of the five studies did not mention the number of male and female nurses (Hashemi et al., 2012). Three studies included nurses working in different hospital wards and hospital clinics (Dyab et al., 2018;Hashemi et al., 2012;Hwang et al., 2012), and one study included nurses working only in the intensive care units of hospitals (Peyrovi et al., 2016), and one study included nursing interns only (Song & Guo, 2019).  In relation to sampling, two studies used a convenience and purposive sampling technique (Peyrovi et al., 2016;Song & Guo, 2019), one study used snowball sampling (Hwang et al., 2012), one study used both convenience and snowball sampling techniques (Dyab et al., 2018), and one study did not identify the sampling method used (Hashemi et al., 2012). All studies underwent ethical approval, with the nurses signing the written consent to participate in the study.
Two studies used focus group discussions to collect data from the participants (Hashemi et al., 2012;Song & Guo, 2019); two studies used in-depth semi-structured face-toface interviews with the nurses to collect the data (Dyab et al., 2018;Peyrovi et al., 2016); and one study used a methodological triangulation, where they used face-to-face interviews as well as emails to collect the data (Hwang et al., 2012). Four of the five studies conducted the interviews and focus group meetings in English (Hashemi et al., 2012;Hwang et al., 2012;Peyrovi et al., 2016;Song & Guo, 2019), while one study had a couple of interviews conducted in Malay (the native language of the participants) for nurses who could not express their thoughts in English, but translators worked closely with the researchers to get the nurses' exact views (Dyab et al., 2018; Table 2).

Findings
While a total of 56 findings were extracted, 13 of them were not included in the synthesis process as their credibility level was "unsupported." The remaining 43 findings were either unequivocal or credible. These 43 findings were aggregated into eight categories, and these categories were aggregated into four overarching synthesized findings (Table 3).
Synthesized Finding 1: There is a fear among nurses of the negative consequences of reporting, arising from their co-workers and supervisors. Three out of the five included studies (Dyab et al., 2018;Hashemi et al., 2012;Peyrovi et al., 2016) identified "fear of the repercussions of incident reporting" as a major theme because nurses usually mention it as the main barrier, whether that fear was fear of being stigmatized by their co-workers and supervisors, or fear of an inadequate response by the management in the form of a severe disciplinary action.
One of the main reasons nurses did not feel encouraged to report any patient safety incident was because they were afraid that they would get sued and face legal problems (Dyab et al., 2018;Peyrovi et al., 2016) and lose a lot of money (Hashemi et al., 2012). Another reason nurses underreported was the fear of being stigmatized for a long time by their co-workers (Peyrovi et al., 2016).
Nurses were also concerned about the reaction from management; negative reactions to error reporting from supervisors were common, as supervisors tend to blame and punish the nurses instead of offering support and working with them to reach the root causes of the error.
Synthesized Finding 2: There is a lack of the organizational support needed to foster an incident reporting milieu for nurses. The organization should be committed to supporting nurses by all means, including developing an effective reporting Fear of threats and legal action (U) If we report errors, they will be used against us. There will be no legal protection. They threaten us. Our problems will be examined in the Medical Council. Physicians will vote against us and in favor of their own interest. We are afraid of being ousted from our job. Continuous warning comes from the nursing office. (Hashemi et al., 2012, p. 3)

Fear
There is a fear among nurses of the negative consequences of reporting from their co-workers and supervisors Fear of economic losses (U) A nurse had to pay $8000 as blood money because of a mistake which was not only his fault. How much is our income to pay blood money. (Hashemi et al., 2012, p. 3) Fear of being stigmatized (U) Projection means you see staff around yourself or doctors that relate everything to that error you made. (Peyrovi et al., 2016, p. 3)

Fear of punishment (U)
[W]e believe that there is a series of punishments implemented following error reporting. (Peyrovi et al., 2016, p. 4) Fear of legal problems (U) They take a nurse to court! I don't know what happened then, but the nurse got in trouble, I fear that the same thing would happen to me. (Peyrovi et al., 2016, p. 4) Fear of organizational misconduct (U) If we report our errors, the authorities would have improper behaviour towards me! (Peyrovi et al., 2016, p. 4) Fear of disciplinary action (e.g., legal problems) (U) I fear from legal problems and disciplinary actions from the hospital. (Dyab et al., 2018, p. 8) Previous inadequate reactions from the organization (U) When we report our errors, we face harsh and unfair behaviors (warnings, threats, etc.).
Should an error be reported in such a milieu? (Hashemi et al., 2012, p. 4) Managerial misconduct Blame culture (C) Being at the sharp end, being named, blamed, and shamed, and being humiliated were expressed by the participants. (Hashemi et al., 2012, p. 4) Blame and punishment for the person involved in the incident (C) An organisational cultural practice of blaming and penalizing the department heads as well as those involved in medical errors was frequently reported. (Hwang et al., 2012, p. 284) Pointing fingers at nurses (U) Based on my experiences in intensive care, the nurse was scolded even if it was medical error considered as nursing errors! They first considered an error to be a nursing error, and then if they could not prove it, they considered it as medical error. (Peyrovi et al., 2016, p. 4) Lack of managerial support (U) Because whenever there was a problem, all wanted to be acquitted, there was selfishness and nobody wanted to support you! (Peyrovi et al., 2016, p. 4) Lack of attention to the cause of the error (U) Nobody ever asks the real reasons for nursing errors; nobody asks me if I was tired when I committed the error? Was there any excessive workload on you? What is your problem? Unfortunately, it's not important! (Peyrovi et al., 2016, p. 4) Failure to follow the origin of error (U) Because the origins of nursing errors remain unknown, there is no desire for reporting. (Peyrovi et al., 2016, p.4) Negative reaction from in-charge nurses (U) The sister will monitor me more. (Dyab et al., 2018, p. 9) Absence of effective feedback (U) No one goes through all the errors and give me a feedback. (Dyab et al., 2018, p. 9) Work pressure (U) We are really busy; a great number of patients and a limited number of staff. We do not have time for reporting the errors and being involved in the process of error reporting. Doing something for the patient in this little time is my concern. (Hashemi et al., 2012, p. 4) Inadequate organizational reporting system There is a lack of the organizational support needed to foster an incident reporting milieu for nurses Lack of time (U) We are really busy; a great number of patients and a limited number of staff. We do not have time for reporting the errors and being involved in the process of error reporting. Doing something for the patient in this little time is my concern. (Hashemi et al., 2012, p. 4) Time-consuming reporting process (C) These factors include the personnel's lack of time and the reporting process' being timeconsuming. (Hashemi et al., 2012, p. 4) No assurance of anonymity (C) These incident reporting systems required reporters to identify themselves, participants were often dissuaded from reporting incidents. (Hwang et al., 2012, p. 284) (continued) Findings (barriers to nursing error reporting) Illustration Categories

Synthesized Findings
No integrated, dual reporting system (C) Insufficient or partial integration between newly computerised reporting systems and conventional reporting systems seems to prevent efficient incident reporting in busy clinical situations. (Hwang et al., 2012, p. 284) Difficulty to report multi-department involved incidents (C) Reporting systems had difficulty in integrating the reports that involved multiple departments and providing timely information to the person responsible for managing incident reporting. (Hwang et al., 2012, p. 284).
Low reporting rate (U) The number of reporting near misses is particularly low. (Hwang et al., 2012, p. 282) Heavy workload (C) These barriers include heavy workload. (Dyab et al., 2018, p. 8) Lack of time (U) We do not have time for reporting. It is a long story and takes much time. (Dyab et al., 2018, p. 8) Tiredness (U) Sometimes, we are tired. Once we are tired we decide not to report. (Dyab et al., 2018, p. 8) Lack of confidentiality (U) I prefer to fill anonymous form [. . .] Because I feel shy and would not work further. Also, I would feel sorry for the patient. So, I prefer to fill the form without names. (Dyab et al., 2018, p. 9) The inconvenience of the reporting system (U) We have to report it (the event) twice, which makes me fidgety. (Song & Guo, 2019, p. 202) No benefit from reporting (U) Subsequent solutions are my concern. The reporting system should be associated with patient information e.g., HIV or TP, which would allow me to determine what to do next. (Song & Guo, 2019, p. 202) Lack of cooperation from clinical departments (C) Other perceived barriers included interdepartmental conflict and lack of interdepartmental cooperation, incident reporting being used only in the nursing department. (Hwang et al., 2012, p. 284)

Lack of cooperation
Middle-level managers' lack of knowledge and skills of patient safety and incident reporting (C) Department managers can control reporting processes by accepting or rejecting a specific incident report in a formal reporting line. Their awareness and attitude towards incident reporting directly influences the reporting practices of their staff. (Hwang et al., 2012, p. 282) Lack of physician's reporting and participation (C) The lack of physicians' participation in incident reporting was described as a problem. (Hwang et al., 2012, p. 282) Social influence (U) At first, we did not know how to deal with these problems; we often followed the teacher's guidance. (Song & Guo, 2019, p. 202) The weakness of knowledge and nursing skills (U) Someone may not be aware of his mistake and does not report it. Nurses have little knowledge in this field. The educational system has shortcomings. (Hashemi et al., 2012, p. 3)

Lack of Knowledge
Some nurses are not well trained to take part in the incident reporting system Lack of knowledge (U) No one informed me of the criteria (for what should be reported). . . .The error tends to be reported if the medicine entered the patient's body since the patient must then be observed closely. (Song & Guo, 2019, p. 202) Lack of staff's knowledge and skill related to incident reporting (C) [M]ost hospital staff lack the knowledge and skills required for effective incident reporting (e.g., knowledge of sentinel events and near misses). (Hwang et al., 2012, p. 282) The ambiguity of the notion of error (U) [M]ost hospital staff lack the knowledge and skills required for effective incident reporting (e.g., knowledge of sentinel events and near misses). (Hwang et al., 2012, p. 282) Something might be right in my opinion, but you or the head nurse considers it as an error! We should know what an error is! For example, if a patient is drugged with delay due to the crowdedness of the wards, is it an error? (Hashemi et al., 2012, p. 4) Ambiguity of an "error" Late reporting (U) The first barrier is the omission of reporting incidents. (Hwang et al., 2012, p. 282) Low severity of error (U) When an error with no serious outcome is practiced and everything is under control, no report is needed. (Hashemi et al., 2012, p. 4)  Unwillingness to accept the responsibilities of the error (U) The personnel are different; it depends on an individual's sense of responsibility and commitment to work. (Hashemi et al., 2012, p. 3)

Unmotivated and Uncommitted
Nurses Nurses lack the motivation needed to ensure their active participation in the incident reporting system Saving professional reputation (U) If I tell the doctor that such and such nurse administers mannitol, what can we do for the patient now? He/she pessimistically looks at the nurses as if the nurse is really careless and intentionally makes wrong choices. (Peyrovi et al., 2016, p. 32) Embarrassment (U) Facing the embarrassment from my family and friends is tough. They will blame us. (Dyab et al., 2018, p. 8) Personal Feelings The feeling of uncertainty, dishonor and shame (U) If I reported all nursing safety events, hospital recruiters will refuse to hire me, It was humiliating. I still feel embarrassed when I remember the mistake. (Song & Guo, 2019, p. 202) Lack of system accessibility (UN)  system. Most studies showed that the IRS in the organization did not encourage nursing error reporting (Dyab et al., 2018;Hwang et al., 2012;Peyrovi et al., 2016;Song & Guo, 2019); they discussed the inconvenience of the error reporting processes as one of the barriers, as well as the lack of confidentiality and anonymity in reporting (Dyab et al., 2018). Also in relation to the inadequacy of the reporting system, nurses from more than one study mentioned lack of time and heavy workload as reasons for errors going unreported (Dyab et al., 2018;Hashemi et al., 2012).
Synthesized Finding 3: Some nurses are not well trained to take part in the incident reporting system. Most of the included studies evidenced a lack in the skills, knowledge, and training required to take part in effective incident reporting (Hashemi et al., 2012;Hwang et al., 2012;Song & Guo, 2019). Lack of cooperation was identified as one of this finding's categories; however, only two out of the five included studies supported that category (Hwang et al., 2012;Song & Guo, 2019). Furthermore, some nurses believed that the definition of error was vague, or the outcome of an error defined the need of reporting (Hashemi et al., 2012;Hwang et al., 2012).
Synthesized Finding 4: Nurses lack the motivation and commitment needed to ensure their active participation in the incident reporting system. Studies showed that some nurses were unwilling to accept the responsibility of an error as they did not have a reason to be committed to their work (Hashemi, 2008;Peyrovi et al., 2016). On the individual level, nurses chose not to report their errors for different personal reasons. The feeling of shame, dishonor, and embarrassment were some common barriers (Dyab et al., 2018;Song & Guo, 2019). Nurses felt that committing and reporting an error might result in stigmatization (Song & Guo, 2019).

Assessment of Confidence in Review Findings
Confidence in the results were also evidenced using the GRADE ConQual tool (Munn et al., 2014). The JBI ranking scheme (The Joanna Briggs Institute, 2014) was established, and review findings were scored and assessed according to it. Table 4 presents the ConQual assessment results as well as a summary of the review findings.

Discussion
Existing systematic evidence on barriers to reporting of medication errors and near misses highlighted the need for further research on the topic (Vrbnjak et al., 2016), while an earlier systematic review on barriers to incident reporting among nurses was limited to nurses working in acute care settings only (Fung et al., 2012b). This review takes a more diverse approach in terms of both clinical settings and the types of incidents, aiming to capture generic and recurring safety issues (Holmström et al., 2012).
There seems to be an abundance in evidence, supporting the findings of this review. Fear of negative consequences was the most supported barrier identified in this review, as it was the most frequently described and the strongest perceived barrier in 13 different studies that were carried out to address the issue of barriers to incident reporting (Chiang, 2005;Chiang et al., 2010;Copping, 2005;Fathi et al., 2017;Fernandez et al., 2010;Kingston, 2011;Lin & Ma, 2009;Mansouri et al., 2019;Mayo & Duncan, 2004;Petrova, 2010;Qin et al., 2015;Yumei et al., 2016;Yung et al., 2016).
The review identified that managerial misconduct was also a barrier to incident reporting among nurses, as previous poor reactions to errors from the organization led to a decrease in nursing reporting. Organizations that adapt a "culture of blame" are more likely to have low reporting rates (Almutary & Lewis, 2012;Hesari et al., 2015;Kingston, 2011;Mansouri et al., 2019). Kingston (2011) andMansouri et al. (2019) identified "the existence of a culture of blame" as a barrier to error reporting among nurses. Hesari et al. (2015) identified that authorities tend to focus on the person who made the error and not how to learn from it, which discourages nurses from reporting.
Lack of appropriate support from the organization to the nurses who report their errors was another barrier identified in this review, which is similar to the main findings of six other studies (Azadi et al., 2011;Elder et al., 2008;Evans et al., 2006;Handler et al., 2007;Mostafaei et al., 2014;Sanghera et al., 2007). Lack of effective feedback was identified as a main barrier in multiple studies (Evans et al., 2006;Mostafaei et al., 2014), and nurses felt it was futile to report (Brubacher et al., 2011) as there were no follow-up investigations (Azadi et al., 2011).
In terms of the IRS of the organization, in many studies (Brubacher et al., 2011;Elder et al., 2008;Fathi et al., 2017;Fernandez et al., 2010;Kingston, 2011;Mostafaei et al., 2014;Moumtzoglou, 2010), nurses mentioned the difficulties in the incident reporting process as a major barrier to reporting, which confirms the review's findings. A difficult reporting process (Kingston, 2011) that does not respect the nurses' heavy workload (Fernandez et al., 2010;Moumtzoglou, 2010) and time constraint (Brubacher et al., 2011;Elder et al., 2008) is what makes nurses not report. Other inconveniences in the incident reporting system (such as the lack of confidentiality) were also considered among the barriers to nursing error reporting (Grant & Larsen, 2007;Jeffe et al., 2004).
This review found that error underreporting among nurses can simply be a response to the lack of the appropriate nurses'
training required to implement an effective reporting system. While many studies argue that the reason nurses do not report their errors is the absence of adequate education about the reporting process, some nurses do not know what exactly constitutes an error (Fathi et al., 2017). Mostafaei et al. (2014) identified "lack of clear definition of what to report" as a barrier to nurses' reporting. The following seven studies attributed error underreporting among nurses to barriers like the ambiguity of the notion of error and lack of nurses' knowledge and skills related to incident reporting (Brubacher et al., 2011;Fathi et al., 2017;Hesari et al., 2015;Kreckler et al., 2009;Mansouri et al., 2019;Mostafaei et al., 2014;Wakefield et al., 1996). Nursing administration, organizational leaders, and other relevant stakeholders should try to develop an atmosphere where nurses report their errors honestly and without fear, while providing them with the necessary training and educational programs on patient safety and incident reporting, and giving them feedback after investigating the systems' errors rather than blaming the individuals. A thorough review was undertaken to help develop an effective form of feedback to the reporting of safety issues in healthcare (Benn et al., 2009), identifying different modes of incident reporting feedback. The Safety Action and Information Feedback from Incident Report (SAIFIR) framework identified that effective feedback relies on timely corrective actions, and broad dissemination of information among other things such as appropriateness of feedback delivery and frontlines empowerment (Benn et al., 2009).
While the background section mentioned the rationale for undertaking this review, which included clear evidencebased implications for the quality and patient safety improvement science and research, the fundamental purpose of reporting errors is to learn how to improve (Firth-Cozens, 2002); identifying and facing the barriers to reporting shall help create an organization where error reporting is culturally acceptable, ultimately leading to improving the safety of both the patients and the staff (Pham et al., 2010). A "just culture" (Marx, 2005) that minimizes punishments and has the right balance of safety and accountability could help achieve this (Dekker, 2016b). A significant amount of resources should be allocated by the leadership to intensively analyze these errors and apply the lessons derived from error reporting to the health care system.
An effective IRS should include all type of incidencenear misses, at-risk behaviors, hazardous conditions, and other types of adverse events. Learning should be communicated throughout the organization and the health care industry (Galt & Paschal, 2009), and every relevant stakeholder and health professional (physicians, nurses, and allied healthcare professionals) should be involved in the process of reporting. While an IRS is usually successful when it is voluntary, nonpunitive, and protected, this type of working environment allows innovation and learning to flourish (Dekker, 2016a). The process of reporting is of vast importance in health care because when frontline staff report any type of error that took place in their department, this helps the organization understand what is going wrong, and, therefore, work on preventing the recurrence of this error in any other department, and ultimately helping improve the patients' outcomes (Dekker, 2018).
Many health care accrediting organizations such as The Joint Commission (Joint Commission International, 2017), regulatory bodies, and other health care oversight entitiessuch as the Patient Safety Organizations (Elkin et al., 2016)-have recently started mandating that health care organizations submit certain reports concerning adverse events that occur in the hospital, along with the analysis performed to reach the root cause of that event and the action taken to prevent its recurrence. However, without institutionally based voluntary incident reporting, it is very difficult and challenging to catch and learn from all errors (Wachter, 2012). Many hospitals have started working on quality improvement projects and other strategies that aim to increase the "patient safety culture" and reporting in the hospitals (Gleeson et al., 2019;Parmelli et al., 2012) A complete meta-aggregation schematic/overview flowchart was developed, which contains all the review processes to identify the issue, find the appropriate research studies, categorize, and synthesize findings. The categories' descriptions, synthesized findings, and recommendations for practice were also included in the flowchart as per the JBI guidelines (The Joanna Briggs Institute, 2014; Figure 2).
While every effort was made to plan and undertake a comprehensive search strategy, there are some potential limitations for this review. The finally selected search terms might have led us to missing publications. However, every effort was made to include them, using PICoS, mnemonic, while designing the search query and pilot searches performed. In addition, only studies in English language were included, which may have led to some bias too.
The small number of studies is certainly a limitation in itself to conclude and generalize results, but at the same time, it demonstrates the need for further research in this topic. The review included five studies representing only four countries, all of which happened to be developing countries, but the fact that all the review's findings were strongly supported by a lot of quantitative and mixed-method worldwide studies compensates for the limited number of the included studies and their countries of origin. A previous systematic review in 2016 included a majority of studies conducted in Western countries (Vrbnjak et al., 2016), also confirming the findings of this review, while a global study on error reporting systems and barriers in 16 different countries from Africa, Australia, Europe, and North America showed a major commonality between perceived barriers to incident reporting (Holmström et al., 2012). Furthermore, a review in primary care identified that there is a lack of safety culture (Lainer et al., 2015), while studies supported the findings of this review partially, both in primary care (Michel et al., 2017) and in long-term care settings (Vaismoradi et al., 2020).
The methodological quality of the included studies was relatively good, apart from the relationship between the researcher and the participants; none of the five studies addressed potential bias in the research, and one study lacked a sampling strategy (Hashemi et al., 2012), which might influence the quality of the findings of this review. No studies were excluded according to their methodological quality assessment, which is considered a standard practice (Dixon-Woods et al., 2007).
The importance of incident reporting in health care is undisputed, being a very important pillar in improving health care systems and ensuring patient safety. This review identified several barriers to incident reporting among nurses. Fear of negative consequences-such as blame and legal issues-was the most prominent barrier and the main reason contributing to the underreporting of nursing errors. Barriers also included managerial misconduct, inadequate incident reporting processes, and the lack of the necessary training and motivation required to implement an effective IRS. This review findings agreed with previous systematic evidence, showing that there is still a need to develop effective interventions and policies to overcome these barriers and encourage incident reporting among nurses. Changes should be implemented in the system with a cultural shift from individual blaming and pointing fingers to awareness, trust, sharing, and learning.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.