Granel et al. BMC Health Services Research (2020) 20:584 https://doi.org/10.1186/s12913-020-05441-w RESEARCH ARTICLE Open Access Nurses’ perceptions of patient safety culture: a mixed-methods study Nina Granel* , Josep Maria Manresa-Domínguez, Carolina Eva Watson, Rebeca Gómez-Ibáñez and Maria Dolors Bernabeu-Tamayo Abstract Background: There are relatively few qualitative studies concerning patient safety culture. Methods: We aimed to explore patient safety culture as perceived by the nursing staff in two public hospitals in Catalonia, Spain. A mixed-methods design was employed using a questionnaire, in-depth interviews, and non- participant observations. Results: Sixty-two percent of the nursing staff rated patient safety as “Acceptable” but was not higher because of work pressure and lack of resources as perceived by staff. “Teamwork within units” had the highest rate of positive responses, and “Staffing” had the lowest rate. Emergency units showed more negative results than the other two units. Conclusions: Safety incidents are not always reported due to fear of punishment, reflecting a lack of positive safety culture. It is necessary to design and implement strategies that promote a positive culture to avoid punitive responses and apply and evaluate these changes. Keywords: Nursing care management, Organizational culture, Patient safety, Safety management Background to promote, create, and maintain a positive patient safety In European countries with advanced healthcare sys- culture [5, 6]. An institution’s patient safety culture is tems, adverse events due to healthcare provided occur in the sum of individual and group values, attitudes, per- 8 to 12% of hospitalizations [1]. Adverse events have a ceptions, competencies, and behavior patterns that de- significant impact on patient morbidity and mortality termine the style of, competence in, and commitment to [2]. In Spain, 9–12% of patients treated in hospitals, in- safety management in the institution [7, 8]. Institutions cluding in emergency departments, suffer an adverse with a positive safety culture are characterized by com- event. Around 50–70% of these adverse events could be munication based on mutual trust, perceptions of the avoided by applying demonstrated safety practices [3]. importance of safety, and reliance on effective preventive Patient safety requires an organizational and multidis- actions. Staff and institutions recognize mistakes, learn ciplinary approach, in which nursing is fundamental. from them, and improve. Promoting a safety culture in Nurses can help prevent multiple avoidable adverse health systems is an international challenge [7]. events, such as medication errors, pressure ulcers, lack The most widely used instrument for assessing safety of information, falls, and nosocomial infections, among culture is the Hospital Survey on Patient Safety Culture others [4]. To improve patient safety, institutions need (HSOPSC) by the United States (US) Agency for Health- care Research and Quality (AHRQ) [8]. This question- * Correspondence: nina.granel@uab.cat naire assesses the perception of patient safety culture Nursing Department, Medicine Faculty, Universitat Autònoma de Barcelona, among hospital staff, including an institution’s values, Av. de Can Domènech, 737, 08193 Bellaterra, Barcelona, Spain © The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Granel et al. BMC Health Services Research (2020) 20:584 Page 2 of 9 beliefs, and standards; event reporting; communication; Catalonia. The hospitals had certain characteristics and leadership; and management [9]. This instrument has dimensions in common (same nurse/patient ratio and been validated in several European Union (EU) coun- overall patient numbers). The total beds per unit were tries: Finland, Iceland, Norway, Switzerland, and the between 28 and 30, having an average ratio of 1 nurse/ Netherlands [10]; Belgium [11]; Croatia [12]; France 10 beds in the Medicine and Surgery units. In the Emer- [13]; Italy [14]; the United Kingdom [15]; Portugal [16]; gency units each nurse was in charge of approximately 5 Spain [17]; and Sweden [18]. patients. The occupancy rate was 95–100% for all the Quantitative studies on patient safety culture, such as units. The units selected accounted for the highest per- the ones cited above, are abundant in the literature; centage of overall hospital nursing staff employed. however, qualitative or mixed-methods studies in this area are limited. In 2014, a thematic literature review Data collection was published in which only one qualitative study con- Permit from the hospitals’ research boards was obtained cerning safety culture was identified [19]. Aside from a in March 2017, and data were collected from April to mixed-methods Iranian study [20] and an Ethiopian May 2017. The main researcher presented the project to cross-sectional study [21], no other qualitative studies the units’ supervisors, and they distributed the Spanish have been identified. Experts in patient safety culture validated version of the HSOPSC questionnaire [17] to have suggested that mixed-methods approaches, using the nursing staff who had been working in the hospital interviews or focus groups, can help researchers under- for a minimum of 1 year, along with a letter explaining stand safety culture in greater depth [19]. the project and covering informed consent. No exclusion The aim of this study was to explore patient safety cul- criteria were established. Once implemented, question- ture as perceived by the nursing staff in the Surgical, In- naires and consent forms were deposited securely and ternal Medicine, and Emergency departments of two anonymously in a box. To increase the response rate, public hospitals in Catalonia, Spain. The specific objec- weekly reminders were made by the supervisors and the tives were to: 1) Describe the strengths and weaknesses principal investigator to each work shift during the data of the perceived safety culture in the specific services collection period. Finally, questionnaires and consent provided, 2) Determine the perceived global patient forms were returned to the investigators. safety grade, and 3) Identify the number of incidents re- The HSOPSC contains 42 items classified into 12 di- ported during the last year. mensions (Tables 1 and 3). Nine items use a 5-point Likert scale to indicate the level of agreement (“strongly Methods disagree” to “strongly agree”), and the other three items This study examined the patient safety culture in two use a 5-point Likert to indicate frequency (“never” to “al- hospitals located in Catalonia through a convergent par- ways”). The instrument also measures sociodemographic allel design mixed-methods study following the guide- data, as well as the patient safety grade and number of lines recommended by Creswell [22]. This approach was incidents reported in the previous year. not for the purpose of comparing results, which is a The main outcome variable was positive response rate, common reason for conducting a parallel design; in- which was calculated as follows: the number of positive stead, the researchers included transformed data in the answers to each question in a dimension divided by the convergent analysis process. This integrative design total number of answers to all questions in the dimen- allowed us to maintain fidelity and separation among sion. Negatively-worded items were reverse-coded before one sequence of mixed-methods processes which re- calculation. A department was considered to show a sulted in quantitative findings and one qualitative se- good safety culture in each dimension or an area of quence which provided phenomenological context strength when a dimension’s percentage of positive re- around the data, both having the same priority and sponses was ≥75%; a low safety culture needing improve- remaining separate during the analysis. The findings ment was ≤50%. from these two methodological sequences were then To collect qualitative data, nine semi-structured in- unified and triangulated, which originated a set of data depth interviews were conducted, along with a total of to be interpreted, and an overall interpretation was per- 42 h of non-participant observation of nurses in the formed providing a greater understanding of the patient three units during their seven-hour shifts were carried safety culture at the participating hospitals units [22]. out by spending 2 days in each unit. The principal re- searcher wore the institutional uniform and was intro- Sample and setting duced to the staff on shift and answered research The study population consisted of nursing staff from the questions to avoid the Howthorne effect of observation. Internal Medicine, General Surgery, and Emergency The routines, performance, and actions of the nurses at units of two public hospitals from diverse regions of their workplace and the teamwork environment were Granel et al. BMC Health Services Research (2020) 20:584 Page 3 of 9 Table 1 Comparisons of patient safety culture dimensions between hospitals Safety culture dimensions Hospital 1 (n = 63) Hospital 2 (n = 46) p 1. Frequency of event reporting 43.92% (41.8) 42.75% (40.2) .885 2. Overall perceptions of safety 31.0% (28.0) 25.5% (24.4) .295 3. Supervisor/manager expectations and actions promoting safety 50.4% (33.1) 45.7% (34.2) .468 4. Organizational learning—continuous improvement 41.8% (33.3) 43.5% (35.0) .800 5. Teamwork within hospital units 70.2% (34.4) 63.6% (34.0) .319 6. Communication openness 55.6% (34.9) 37.0% (30.8) .005 7. Feedback and communication about error 27.0% (31.0) 32.6% (27.6) .322 8. Non-punitive response to error 39.7% (30.4) 35.5% (31.7) .489 9. Staffing 16.3% (16.8) 11.4% (13.6) .112 10. Hospital management support for patient safety 13.2% (23.6) 24.6% (33.2) .055 11. Teamwork across hospital units 45.6% (31.6) 48.4% (39.8) .679 12. Handoffs and transitions 54.4% (30.9) 59.8% (30.0) .363 observed; field notes were taken during the observations. safety culture. Atlas.ti version 7.5.12 for Windows was Participants for the interviews were selected purposively used. Quotations were coded by collection technique using critical case sampling approach [22], in which (OBS: non-participant observation and INT: in-depth nurses who were likely to provide the most information interview), individual participants’ numbers, and work were selected. The inclusion criterion was a minimum of units (MED: medicine, SURG: surgery, and EMERG: 3 years of work experience in the unit (“efficient” or “ex- emergency). pert” according to Benner’s theory) [23], with the object- ive of guaranteeing a deep knowledge of the services Results provided and the conditions in which they were pro- One-hundred and fifty-four questionnaires were distrib- vided. The main researcher during the non-participant uted to all nurses in the participating units in both hos- observation selected and invited nurses to set voluntarily pitals, and 109 were completed (response rate = 70.7%). the time to conduct the interviews outside the work area Ninety-three percent of the participants were women. in a meeting room facilitated by the hospital. Interviews Overall, 75 nurses (68.8%) had been working for more with three nurses from each participant unit were re- than 11 years. Internal Medicine units had a higher dif- corded until data saturation was reached [22] following ference in hours worked per week compared to Surgical an interview guide (Fig. 1). All the interviews were con- and Emergency units (Table 2). Interviews provided evi- ducted in Spanish, with each interview lasting 30–50 dence across all dimensions, but only the dimensions min. The transcripts of the interviews were reviewed by considered to be strengths or those requiring improve- the participants to ensure that their views were accur- ment were subjected to further analysis, which provided ately reflected. The audio-recorded interviews were tran- the most relevant results. scribed verbatim and translated into English by the The majority of the participants from which qualitative principal author. data were collected were women (8 out of 9), and had more than 10 years of clinical experience, and between 5 Data analyses and 17 years of experience in their units. SPSS 23.0 for Windows (Armonk, NY: IBM Corp.) was used to conduct descriptive statistical quantitative ana- Patient safety could be improved lyses. The significance level used in the bilateral con- Patient safety grade is shown in Table 2. Per the qualita- trasts was p ≤ .05. The categorical variables were tive data, most interviewees felt that patient safety could summarized with their absolute and relative frequencies, be improved and mentioned that lack of resources and and the continuous ones with their means and standard work pressure put the patients at risk. deviations. Pearson’s chi-squared test was used to test proportions, and variance analysis with Tukey’s post-hoc The nurse is giving medication to the patient; sud- correction or Dunnett’s T3 was used to test means, de- denly, she gets a phone call to explain the condition pending on the variance homogeneity. of a patient, who has to move from one unit to an- Participants’ responses to the interviews and observa- other. The nurse complains that she has heavy inter- tion field notes were coded in terms of the 12 categories ruptions. OBS 3 MED. established by the AHRQ that compose the patient Granel et al. BMC Health Services Research (2020) 20:584 Page 4 of 9 Fig. 1 Topic guide for the semi-structured interviews [You go home thinking] thank goodness, I’ve finished, The nurse from the morning shift calls another because I would have caused a disaster. Often you unit to explain the conditions of a patient to be leave with that feeling, or the feeling that you have admitted, she explains … that the medications forgotten things because you don’t have enough time! prescribed at night time haven’t been validated by INT 6 SURG. the night shift nurse on the computer and, there- fore, she’s not aware if it’s been given to the pa- Emergency staff in both hospitals perceived a worse tient. They assume it’s been given without the patient safety grade than Internal Medicine and General night nurse’s confirmation and don’t report the in- Surgery staff (p = .010). Qualitative data were in accord- cident. OBS 1 EMERG. ance with these results. During the interviews, most nurses associated “inci- For example, I have an allergy reaction in the corri- dents” only with medication errors or issues and ver- dor without a monitor. Also, a 94-year-old lady in balized that, sometimes, they discussed them with the the corridor who, at any time, can become disorien- doctor. However, neither doctors nor nurses notified tated and fall from the stretcher. Then, what do you it in writing through the formal reporting process. want me to say? There are a lot of things going wrong every day related to patient safety. INT 5 I think that the majority get notified, but all? I don’t EMER. think so. If you realize, you tell the doctor; but, it doesn’t go any further. INT 4 MED. Incidents were not always reported Once, a nurse made a mistake; she administrated Incident reporting data are shown in Table 2. three 1 ml morphine ampoules when she had to ad- During the observations, other incidents were also de- minister 0.3 ml. She told the doctor, but they did not tected that were not reported by the staff: report it. INT 6 SURG. Granel et al. BMC Health Services Research (2020) 20:584 Page 5 of 9 Table 2 Patient safety grade, incident report and other work data between hospitals and units Hospital 1 Hospital 2 p Internal Medicine General Surgery Emergency p Total (n = 63) (n = 46) (n = 26) (n = 34) (n = 49) (n = 109) Patient safety grade .325 .010 Excellent/good 10 (16.1%) 12 (26.1%) 9 (34.6%) 4 (12.1%) 9 (18.4%) 22 (20.4%) Acceptable 39 (62.9%) 28 (60.9%) 14 (53.8%) 27 (81.8%) 26 (53.1%) 67 (62.0%) Poor 13 (21.0%) 6 (13.0%) 3 (11.5%) 2 (6.1%) 14 (28.6%) 19 (17.6%) Number of incidents reported .118† .049† None 50 (79.4%) 29 (63.0%) 14 (53.8%) 26 (76.5%) 39 (79.6%) 79 (72.5%) 1 to 2 incidents 12 (19.0%) 12 (26.1%) 6 (23.1%) 8 (23.5%) 10 (20.4%) 24 (22.0%) 3 to 5 incidents 1 (1.6%) 4 (8.7%) 5 (19.2%) 0 (0.0%) 0 (0.0%) 5 (4.6%) 6 to 10 incidents 0 (0.0%) 1 (2.2%) 1 (3.8%) 0 (0.0%) 0 (0.0%) 1 (0.9%) Working hours per week .039 .013 More than 40 h 11 (17.5%) 16 (34.8%) 12 (46.2%) 5 (14.7%) 10 (20.4%) 27 (24.8%) Career in the hospital .938 .185 1 to 5 years 12 (19.1%) 6 (13.0%) 3 (11.5%) 7 (20.6%) 8 (16.3%) 18 (16.5%) 6 to 10 years 9 (14.3%) 7 (15.2%) 6 (23.1%) 6 (17.6%) 4 (8.2%) 16 (14.7%) 11 to 15 years 19 (30.2%) 14 (30.4%) 8 (30.8%) 9 (26.5%) 16 (32.7%) 33 (30.3%) 16 to 20 years 8 (12.7%) 6 (13.0%) 1 (3.8%) 2 (5.9%) 11 (22.4%) 14 (12.8%) More than 21 years 15 (23.8%) 13 (28.3%) 8 (30.8%) 10 (29.4%) 10 (20.4%) 28 (25.7%) † p-value of the contrast between “none” versus “1 or more incidents” In Internal Medicine, a slightly higher percentage of By unit, dimension 5 showed an especially strong incident reporting was observed compared to the other percentage in Surgical units (83.1%); no other departments (Table 2). dimensions were considered strengths by unit (Table 3). Teamwork within units was the most valued dimension Dimension 5, “teamwork within units”, showed high per- Staffing rates and work pressure needed improvement centages of positive answers in the two hospitals (70.2 Dimension 9, “staffing,” showed the lowest percentages and 63.6%, respectively). Staff supported each other, (16.3 and 11.4%, respectively). Staff unanimously thought treated each other with respect, worked together, and that there were not enough personnel and that, some- usually helped their colleagues. times, patients’ care suffered because of the lack of personnel. This data were reinforced by the interviews The medical office is next to the nursing office, and and observations. that helps a lot. It contributes to good communica- tion. The communication with healthcare assistants Always, there is always pressure, if not the doctors, is good; we delegate work to them and … there is no the relatives … the workload … it is a lot. If you problem. INT 6 SURG. want to do things right, it is a lot. INT 3 MED. One nurse on shift asks other nurses and healthcare assistants if they have finished the patients’ hygiene “I am fed up,” “I am very tired,” “always the same.” rounds and if they needed help. OBS 3, 4 MED, and From these comments, you detect that staff are tired. 2, 6 SURG. INT 4 MED. However, difficulties existed concerning teamwork be- There is work pressure in some shifts with less staff. tween nurses and doctors. The workload is the same, and supervisors can see that you can’t manage it, and you ask for extra staff I think communication is good between doctors and and they don’t arrive, perhaps because nobody’s nurses, but it also depends with whom you communi- available or because it’s not possible at that mo- cate better, you know, who to address to. INT 6 SURG. ment. INT 2 SURG. Granel et al. BMC Health Services Research (2020) 20:584 Page 6 of 9 Table 3 Comparisons of patient safety culture dimensions between units Safety culture dimensions Internal Medicine General Surgery Emergency p (n = 26) (n = 34) (n = 49) 1. Frequency of event reporting 53.9% (41.2) 42.2% (41.2) 38.8% (40.4) .312 2. Overall perceptions of safety 26.9% (22.3) 36.0% (26.2) 24.5% (28.2) .139 3. Supervisor/manager expectations and actions promoting safety 50.0% (32.4) 48.5% (33.1) 47.5% (35.1) .953 4. Organizational learning—continuous improvement 51.3% (30.2) 52.0% (35.0) 31.3% (32.2) .007† 5. Teamwork within hospital units 74.0% (27.8) 83.1% (25.2) 53.1% (37.4) < .001‡ 6. Communication openness 52.6% (34.2) 52.9% (33.9) 41.5% (34.3) .235 7. Feedback and communication about error 35.9% (28.2) 32.4% (33.3) 23.8% (27.2) .190 8. Non-punitive response to error 32.1% (30.5) 44.1% (34.5) 36.7% (28.2) .308 9. Staffing 12.5% (14.6) 16.9% (18.2) 13.3% (14.5) .479 10. Hospital management support for patient safety 26.9% (32.7) 15.7% (23.5) 15.0% (28.9) .191 11. Teamwork across hospital units 55.8% (31.9) 44.1% (33.2) 43.9% (35.2) .304 12. Handoffs and transitions 64.4% (28.4) 58.1% (29.3) 51.5% (32.0) .210 † Tukey post-hoc contrast: differences between Emergency and Internal Medicine (p = .035) and between Emergency and General Surgery (p = .015) ‡ Tukey post-hoc contrast: differences between Emergency and Internal Medicine (p = .021) and between Emergency and General Surgery (p < .001) The following attitudes were detected through obser- The lack of teamwork with doctors was confirmed repeat- vation: excessive concern related to lack of time for edly, supporting the significant difference in dimension 5. tasks, irritability, tension, and tiredness. I miss the communication we had before the system The safety culture of the emergency units slightly was digitized because we used to learn a lot, and it differed from other hospital units allowed teamwork with the doctors. INT 5 EMERG. Emergency units showed lower positive response rates on 10 out of the 12 dimensions, with significant dif- When everything was digitized, communication ferences on dimensions 4 “organization learning—con- levels decreased. For example, with medication tinuous improvement” (p = .007) and 5 “teamwork changes, there are doctors who communicate them, within units” (p < .001). Nurses in the Emergency but there are others who don’t; you have to go be- units detected a lack of resources to apply and evalu- hind them to see the changes and the plan they have ate preventive measures to avoid mistakes, and appro- for the patient. INT 1 EMERG. priate measures were not applied to prevent mistakes from reoccurring. They did not receive enough train- The Catalan hospitals studied had globally similar safety ing in these areas and emphasized a lack of resources culture to allow improvements. There was no significant difference between hospitals except on dimension 6 “communication openness”; the We have 20 h per year for training, which is very lit- staff at Hospital one spoke more freely and questioned tle. We have done a two-hour training on patient their supervisors more (Table 1). safety. It focused on errors, incidents … I left with However, it was not possible to verify this through the the feeling that it was good, but the problem is lack qualitative data, owing to the diversity of responses and of resources. INT 1 EMERG. some discomfort among the interviewees in speaking about errors. They did not always answer questions In Emergency units, staff had their own routine: nurses clearly. had patients assigned and worked independently. Al- though they worked as a team when necessary, they Some staff keep it quiet, and others report that sometimes did not manifest mutual support and respect. things happen. INT 2 SURG. We discuss everything with our supervisor without be- The staff have a lack of care sometimes. There are ing afraid because there is enough trust. INT 3 MED. those who are more involved and others who are not. This is a service with a lot of staff and movement; it The interviewees stated that this lack of communica- is easier to leave your position. It is not like a ward. tion openness was related to perceived culpability and INT 1 EMERG. the fear of punishment. Granel et al. BMC Health Services Research (2020) 20:584 Page 7 of 9 Well, I guess it’ll be for fear of reprisals or for … the differ between Spain and other EU countries: “manager person above you … you know, he can blame you in support for patient safety” and “staffing” were rated the some way or, if it’s not that, what other reason can lowest of all the dimensions in Sweden [32]; however, there be? If nothing happens, no one finds out. INT 3 “none—punitive response to error” and “manager sup- MED. port for patient safety” were rated the lowest, just below “staffing,” in Portugal [16]. One French study indicated Discussion “none—punitive response to error” as the lowest [13], The mixed-methods research provided an overview of while an Italian study reported “teamwork among hos- the state of patient safety culture among staff in two pital units” just as low as “staffing.” [14] Catalan hospitals and elucidated a deeper understanding The characteristics of organizations with a positive of the perceptions of nurses in these hospitals and, by safety culture [33]—such as trust in the effectiveness of extension, in Spanish hospitals generally. Most partici- preventative measures, communication based on trust, pants had more than 11 years of nursing experience, and team perceptions about the importance of safety— which according to previous studies, indicates that they are areas in which to improve in the institutions studied. have good knowledge of the institutions and their safety In no case had the evaluation of staff or institutions been culture [24, 25]. attempted. The main purpose of previous work has been Elevated concerns about insufficient time to complete to understand survey results in order to propose future tasks, irritability, tension, and tiredness are signs of improvement strategies. stress [26], which, in turn, can be associated with work Our findings can assist nurse managers to identify the pressure while on shift and, for some nurses, working weaknesses and strengths in relation to hospital safety overtime. Differences in staff ratios and work pressure issues based on nursing perceptions and behaviors. influence on institution’s safety culture [21]. Thus, Nurse managers could develop and implement specific nurses did not rate safety culture higher than “accept- strategies to improve patient safety, including providing able” as a result of work pressure and perceived lack of education related to patient safety culture to the staff resources. members (including managers). Managers could also It has been previously described in the literature that promote a positive culture through communication by certain risky behaviors can become norms under work encouraging staff to notify incidents and avoiding puni- pressure [27]. Furthermore, the lack of open communi- tive responses. cation in notifying the hospital through formal reporting of avoidable adverse events because of fear of punish- Limitations ment and perceived culpability [28], along with con- Strategies to ensure rigor such as building a research firmed misreporting detected through the observations team, elaborating guides for data collection, defining and and with acknowledgment in interviews that there are obtaining adequate participation and reaching data sat- unreported events, qualifies the low number of reported uration were achieved. A limitation of this study is its incidents. focus on nurses, who comprised around 40% of health- Four studies have been conducted in Spain using the care staff in the studied hospitals. Future mixed- HSOPSC [6, 17, 29, 30]. All indicated that “teamwork methods studies must be conducted to obtain evidence within units” was the dimension with the most positive from other professionals and form a complete view of results and “staffing” was the lowest, matching the the hospitals’ patient safety culture in Catalan Hospitals. current results. This suggests that the strengths and weaknesses of patient safety culture are similar through- Conclusions out Spanish hospitals, and they have not changed over Teamwork within units is considered to be a valuable time despite the fact that Spain has several different re- quality in these institutions globally; nevertheless, staff gional healthcare systems. numbers and work conditions could be improved. Spe- A 2011 study showed low percentages of positive re- cifically, incidents are not always reported by nursing sponses in 30 Emergency departments in Spain [29]. A staff for fear of punishment, pointing to a lack of positive comparable result was found in the present study and in a safety culture. Therefore, we assert the need to work on previous paper [17], where distinct departments were positive safety culture in the participating hospitals compared. Of note, percentages were lower in Emergency through specific strategies of improvement. The quanti- departments in both studies than in other departments. tative results from the HSOPSC mostly match those of In most other European studies, the dimension “team- other studies carried out in Spain and the qualitative work within units” has been considered a strength, ex- data analyzed; however, there are differences when com- cept in some eastern European countries [13, 16, 18, 31]. paring results with those from hospitals in other Euro- Regarding dimensions needing improvement, the results pean countries. Granel et al. BMC Health Services Research (2020) 20:584 Page 8 of 9 Abbreviations 9. Sorra JS, Dyer N. Multilevel psychometric properties of the AHRQ hospital AHRQ: Agency for Healthcare Research and Quality; EMERG: Emergency; survey on patient safety culture. BMC Health Serv Res. 2010;10(1):199. EU: European Union; HSOPSC: Hospital Survey on Patient Safety Culture; https://doi.org/10.1186/1472-6963-10-199. IBM: International Business Machines; INT: In-depth interview; MED: Medicine; 10. European Society for Quality in Health Care. Use of Patient Safety Culture NY: New York; OBS: Non-participant observation; SPSS: Statistical Package For Instruments and Recommendations. Aarhus: European Union Network for Social Sciences; SURG: Surgery; US: United States Patient Safety;2010. 11. Vlayen A, Hellings J, Claes N, Abdou EA, Schrooten W. Measuring safety Acknowledgements culture in Belgian psychiatric hospitals: validation of the Dutch and French Not applicable. translations of the hospital survey on patient safety culture. J Psychiatr Pract. 2015;21:124–39. https://doi.org/10.1097/01.pra.0000462605.17725.48. 12. Brborović H, Šklebar I, Brborović O, Brumen V, Mustajbegović J. Authors’ contributions Development of a Croatian version of the US Hospital survey on patient NG and MDB made substantial contributions to the conception and design safety culture questionnaire: dimensionality and psychometric properties. of the project and NG performed the data acquisition. JMM and NG Postgrad Med J. 2014;90:125–32. https://doi.org/10.1136/postgradmedj- analyzed and interpreted data. RG and CW were involved in drafting the 2013-131814. manuscript and revising it critically for important intellectual content. All 13. Occelli P, Quenon J-L, Kret M, et al. Validation of the French version of the authors read and approved the final manuscript. hospital survey on patient safety culture questionnaire. Int J Qual Health Care. 2013;25(4):459–68. https://doi.org/10.1093/intqhc/mzt047. Funding 14. Bagnasco A, Tibaldi L, Chirone P, et al. Patient safety culture: an Italian No funding or sources of support in the form of grant has been received to experience. J Clin Nurs. 2011;20:1188–95. https://doi.org/10.1111/j.1365-2702. conduct the research. 2010.03377.x. 15. Waterson P, Griffiths P, Stride C, Murphy J, Hignett S. Psychometric Availability of data and materials properties of the hospital survey on patient safety culture: findings from the The datasets used and/or analysed during the current study are available UK. Qual Saf Health Care. 2010;19:e2. https://doi.org/10.1136/qshc.2008. from the corresponding author on reasonable request. 031625. 16. Eiras M, Escoval A, Grillo IM, Silva-Fortes C. The hospital survey on patient safety culture in Portuguese hospitals: instrument validity and reliability. Int Ethics approval and consent to participate J Health Care Qual Assur. 2014;27(2):111–22. https://doi.org/10.1108/IJHCQA- All ethical principles of biomedical scientific research promulgated in the 07-2012-0072. Declaration of Helsinki were respected. Approval for the study to be conducted on behalf of all the study sites was obtained from the Comitè 17. Ministry of Health. Analysis of the patient safety culture in the hospital d’Ètica d’Investigació Clínica del CST (Ref. number: 03–17–103-013) and from setting of the Spanish National Health System. Reports, Studies and the Comitè de Recerca del FPHAG (Ref. number: 20176001). Written consent Research. http://www.msc.es/organizacion/sns/planCalidadSNS/docs/ was obtained from all study participants. Analisis_cultura_SP_ambito_hospitalario.pdf. Published 2009. Accessed 20 Nov 2016. 18. Nordin A, Wilde-Larsson B, Nordström G, Theander K. Swedish hospital Consent for publication survey on patient safety culture—psychometric properties and health care Not applicable. staff’s perception. Open J Nurs. 2013:41–50. https://doi.org/10.4236/ojn.2013. 38A006. Competing interests 19. Pumar-Méndez MJ, Attree M, Wakefield A. Methodological aspects in the The authors declare that they have no competing interests. assessment of safety culture in the hospital setting: a review of the literature. Nurse Educ Today. 2014;34:162–70. https://doi.org/10.1016/j.nedt. Received: 30 April 2019 Accepted: 18 June 2020 2013.08.008. 20. Abdi Z, Delgoshaei B, Ravaghi H, Abbasi M, Heyrani A. The culture of patient safety in an Iranian intensive care unit. J Nurs Manag. 2015;23:333–45. References https://doi.org/10.1111/jonm.12135. 1. Rafter N, Hickey A, Condell S, et al. Adverse events in healthcare: learning 21. Wami SD, Demssie AF, Wassie MM, Ahmed AN. Patient safety culture and from mistakes. QJM. 2015;108(4):273–7. https://doi.org/10.1093/qjmed/ associated factors: a quantitative and qualitative study of healthcare hcu145. workers’ view in Jimma zone hospitals, Southwest Ethiopia. BMC Health 2. World Health Organization. Exploring patient participation in reducing Serv Res. 2016:1–10. https://doi.org/10.1186/s12913-016-1757-z. health-care-related safety risks. Geneva: WHO; 2013. doi:978–92–890-0294-3. 22. Creswell JW, Plano Clark VL. Designing and Conducting Mixed Methods 3. Ministry of Health. Patient Safety Strategy of the National Health System Research of John W. Creswell and Vicki L. Plano Clark. 3rd ed. Thousand 2015–2020. Ministry of Health, Social Services and Equality; 2015. p. 1–128. Oaks: Sage; 2018. https://doi.org/10.1016/S0025-7753(08)76454-6. 23. Benner P. From novice to expert. New Jersey: Prentice Hall; 2001. 4. Pérez Beltrán A. Ruiz Frías Gil S, López Ruiz M, et al. the measurement of 24. Ammouri AA, Tailakh AK, Muliira JK, Geethakrishnan R, Al Kindi SN. Patient patient safety culture: a literature review. Revista de Enfermería. 2013;36(9): safety culture among nurses. Int Nurs Rev. 2015;62(1):102–10. https://doi. 50–5. org/10.1111/inr.12159. 5. Davins J, Oliva G, Álava F, Navarro L, Vallès R. Vision and evolution of patient 25. Chang HY, Friesner D, Lee IC, et al. Openness to experience, work safety in Catalonia. Med Clin. 2014;143:1–2. https://doi.org/10.1016/j.medcli. experience and patient safety. J Nurs Manag. 2016;24(8):1098–108. https:// 2014.07.004. doi.org/10.1111/jonm.12414. 6. Mir-Abellán R, Falcó-Pegueroles A, Puente-Martorell ML. Attitudes towards 26. American Psychiatric Association. Diagnostic and Statistical Manual of Mental patient safety culture in a hospital setting and related variables. Gac Sanit. Disorders. 5th ed. Washington, DC: Author;2013. 2016;31(2):145–9. https://doi.org/10.1016/j.gaceta.2016.07.019. 27. Rocco C, Garrido A. Patient safety and safety culture. Revista Médica Clínica 7. Ministry of Health. Analysis of the patient safety culture of the primary care las Condes. 2017;28(5):785–95. professionals of the National Health System. Reports, Studies and Research. 28. Anglès R, Linás M, Alerany C, García MV. Incident reporting system and https://www.seguridaddelpaciente.es/resources/documentos/2014/ management of incidents: implementation and improvement actions ANALISIS%20DE%20LA%20CULTURA%20DE%20SEGURIDAD%20%20DEL%2 derived for patient safety. Med Clin. 2013;140(7):320–4. 0PACIENTE%20EN%20ATENCI%C3%93N%20PRIMARIA.PDF. Published 2014. 29. Gama ZA, Oliveira AC, Hernandez PJ. Patient safety culture and related Accessed 20 Nov 2016. factors in a network of Spanish public hospitals. Cadernos de Saúde Pública. 8. Sorra J, Famolaro T, Yount N, Smith S, Wilson S, Liu H. Hospital survey on 2013;29(2):283–93. https://doi.org/10.1590/S0102-311X2013000200015. patient safety culture: 2014 user comparative database report. Rockville: 30. Roqueta Egea F, Tomás Vecina S, Chanovas Borrás M. Patient safety culture Agency for Healthcare Research and Quality; 2014. in 30 Spanish hospital emergency departments: results of the Agency for Granel et al. BMC Health Services Research (2020) 20:584 Page 9 of 9 Healthcare Research and Quality's Hospital Survey on Patient Safety Culture Emergencias. Revista de La Sociedad Española de Medicina de Urgencias y Emergencias. 2011;23:356–63. 31. Wagner C, Smits M, Sorra J, Huang C. Assessing patient safety culture in hospitals across countries. Int J Qual Health Care. 2013;25:213–21. https:// doi.org/10.1093/intqhc/mzt024. 32. Danielsson M, Nilsen P, Rutberg H, Arestedt KA. National study of patient safety culture in hospitals in Sweden. J Patient Saf. 2017. https://doi.org/10. 1097/PTS.0000000000000369. 33. Sorra J, Gray L, Streagle S, et al. AHRQ Hospital Survey on Patient Safety Culture: User’s Guide AHRQ Publication No. 15–0049-EF (Replaces 04–0041). Rockville: Agency for Healthcare Research and Quality; 2016. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.