The patient safety concept is defined as preventing and reducing medical errors that could cause harm during the care journey.1 Patient safety is one of the crucial determinants and fundamental contributors to providing high-quality care.2,3 A report by the World Health Organization has raised global patient safety concerns, where 42.7 million adverse preventable events were annually reported worldwide.4 The harms resulting from preventable incidents threaten the level of safety and cause physical and psychological damage. Hence, interchangeably patient safety efforts at all levels should be focused on maintaining quality and safety in healthcare. Patient safety culture is one of the significant directions of enhancing safety. Efforts focused on identifying the impact of safety culture and its association with enhancing healthcare outcomes; accordingly, studies have shown a positive association between improving patient safety culture and reducing the number of adverse events reported in a healthcare organization5,6

The safety culture is linked to various healthcare outcomes such as patient satisfaction, medication error, and the willingness to report events at multiple levels. The culture is formed by individual and group values, attitudes, perceptions, competencies, and patterns of behavior that determine the commitment to an organization’s health and safety management.7

It has been evident that improving a blame-free environment for reporting and enhancing collaboration across the organization is directly affected by the strong safety culture on patient safety.8–10

Many factors can lead to a lack of communication, which has an impact on patient safety. Moreover, the blame culture and poor communication have been proven to be the most common factors contributing to poor patient safety culture in Arab countries.11 A study conducted in Saudi Arabia concluded that having a diverse workforce, cultural differences, and language barriers are major contributors to the lack of communication.12 Moreover, fear of blame was mentioned as one of the main obstacles to reporting incidents among nurses in Saudi Arabia.12

Sentinel events data reported by the Saudi Ministry of Health (MoH) from 2012 to 2015 were analyzed and showed that 91% were classified as preventable events.13 In response to this issue, Saudi Arabia initiated approaches to reduce and identify ways of improving the safety and quality of health care.14 Multiple initiatives have been implemented in Saudi Arabia aiming to measure the patient safety culture to target the area that needs improvement.14 The Saudi Patient Safety Center (SPSC) is one of the MoH initiatives to develop healthier care at the national level. One of the strategic goals is collecting and analyzing data to propose national legislation related to patient safety and improve the patient safety culture. The SPSC established a culture measurement project to quantify the culture of healthcare organizations within Saudi Arabia and identify priority areas to improve patient safety. The Agency for Healthcare Research and Quality (AHRQ) developed a diagnostic tool to assess hospital patient safety culture in July 2017, SPSC piloted the survey and then conducted the following 3 cycles: cycle 2 (2019), cycle 3 (2021), and cycle 4 (2022).15 This survey examines the patient safety culture of an organization from the perspective of hospital staff, including frontline clinical and nonclinical staff, pharmacy and laboratory personnel, hospital-employed physicians, hospital supervisors, managers, and administrators.

The project allows hospitals to establish baseline data with future reassessments facilitating measurement of changes in safety culture over time. Accordingly, a retrospective assessment is needed to evaluate the extent to which patient safety culture is a strategic priority and supports patient safety. Therefore, this study aimed to assess the patient safety culture trend in Saudi Arabia and indicate the extent to which safety changed over time. The ultimate objectives are to predict the factors contributing to the patient safety culture, identify improvement opportunities, and establish the positive response percentile for assessing future improvement efforts.

METHODS

This study is a retrospective analysis of the patient safety culture from a national database at the SPSC reported by the hospitals during a 4-year period from 2019 until 2022. In 2020, the data collection was suspended because of the urgency of the pandemic. The data were collected based on voluntarily submitted survey data from Saudi hospitals, and the consent form was obtained before the survey. It includes 358,313 healthcare provider and staff respondents who administered the survey using versions 1.0 and 2.0 between 2019 and 2022.

Participants

Governmental and private hospitals in Saudi Arabia, including military, academic, specialist, and MoH hospitals, were included. Participants were from different hospitals from all regions with different bed capacities, staff positions, levels of experience, and work areas.

The Patient Safety Culture Survey

The tool used to collect data was developed by the AHRQ16 to assess the patient safety culture.17 The Hospital Survey on Patient Safety Culture questionnaire has 2 versions (1.0 and 2.0) and dual-language versions (Arabic and English). Version 1.0 was piloted in 2018 and used in cycle 2 (2019) and cycle 3 (2021). Version 2.0 was translated and approved by the AHRQ and used in cycle 4 (2022). V1.0 consists of 42 items measuring 12 patient safety culture composites, whereas V2.0 consists of 32 items measuring 10 patient safety culture composites (Table 1). The items survey in both versions was a 5-point Likert scale of agreement (1 = “strongly disagree” to 5 = “strongly agree”) or a scale of frequency (1 = “never” to 5 = “always”); however, in V2.0, the (Don’t apply) was added.

TABLE 1 - Domain in Versions (1.0 and 2.0) Hospital Surveys on Patient Safety Composite Measures* HSOPS 1.0 HSOPS 2.0 D1 Teamwork within units Teamwork D2 Staffing Staffing and work pace D3 Organizational learning—continuous improvement Organizational learning— continuous improvement D4 Nonpunitive response to error Response to error D5 Supervisor/manager expectations and actions promoting patient safety Supervisor, manager, or clinical leader support for patient safety D6 Feedback and communication about error Communication about error D7 Communication openness Communication openness D8 Frequency of events reported Reporting patient safety events D9 Management support for patient safety Hospital management support for patient safety D10 Handoffs and transitions Handoffs and information exchange D11 Teamwork across units — D12 Overall perceptions of patient safety —

*Domains included in HSOPS 1.0 and 2.0 adopted from the Agency for Healthcare Research and Quality.

The SPSC established an electronic web-based survey platform and central repositories for survey data to lead and supervise the data collection and distribution.18 Each hospital’s authorized point of contact (POC) was required to register an account on the SPSC’s platform. All required information, such as the number of beds, and staff per department and facility, was entered by the POC. The SPSC team and hospitals were able to monitor the progress of data collection in real time and ensure anonymous participation and confidentiality. When each data collection cycle closed, the exclusion criteria recommended by the AHRQ were applied: nonverified registration, noncompleted registration, and hospitals with less than 10 completed surveys.

Regarding combining data from all cycles, although 2 versions (V1.0 and V2.0) were used, the relevancy of each item in each domain was revised and considered. Both survey versions have similar measuring items and are validated to measure patent safety culture. Some of the domain’s names were modified in V2.0; however, the concept of the subitems was slightly similar. All cycles were conducted in hospitals from different regions of Saudi. Accordingly, some domains were combined after revising the items, such as domain 1 in V2.0 (Teamwork) and 2 domains from V1.0 (Teamwork within units and Teamwork across units).

Statistical Analysis

IBM Statistical Package for the Social Sciences (SPSS), version 26.0 (Chicago, IL) for Windows, was used for the analysis, of descriptive statistics, including frequencies and percentages. The score of positive responses calculated responses percentage (agree/strongly agree), (always/most of the time), and negatively worded questions (disagree/strongly disagree) were considered in the calculation. Mean, median, and standard deviation (SD) were calculated for each item.

Analysis of variance statistical analysis was performed to demonstrate the differences between cycles and characteristics for the positive mean response by each domain. To generate a baseline and allow comparison of the hospital’s survey results with the aggregated findings from the database, the 10th, 25th, 50th, 75th, and 90th percentiles for both versions’ cycles were calculated to set the percentage of values. The general assumption was used where values within the 90 intervals from the lower limit 10th to the upper limit percentile are included in the range.19 The significance level was set at P < 0.05, and Tukey post hoc analysis was performed for the difference between groups.

RESULTS

In total, 358,313 surveys were used from 1008 hospitals representing the 3 cycles from all the regions of Saudi Arabia (Table 2). The characteristics of respondents in all cycles showed in (Table 2); nurses had the highest response rate, 46.2%, 32.3% were registered nurses, and 13.9% were either nurse practitioners, technicians, nurse educators, or midwives. Followed by physicians or consultants at 10.96%, respiratory therapists had the lowest response rate at 1.09%.

TABLE 2 - Respondent Characteristics 2019 2021 2022 Total Percentage Hospital 250 366 392 1008 Total Respondents 77,732 134,924 145,657 358,313 Staff position  Attending/staff physician/consultant 7349 14,397 17,553 39,299 10.96  Registered nurse 29,193 46,708 39,839 115,740 32.30  Nurse practitioner/technician/nurse educator/midwife 7100 13,755 28,961 49,816 13.90  Technician (e.g., EKG, laboratory, radiology)/paramedics 5251 9,950 11,069 26,270 7.33  Resident physician/physician in training/student/trainee 4008 10,325 10,354 24,687 6.88  Pharmacist/pharmacy technician 3464 6527 5529 15,520 4.33  Duty charge/supervisor/manager/director/senior leader/executive/quality/risk management/infection control 4836 6286 9057 20,179 5.63  Physical, occupational, or speech therapist/psychologist/social worker 1396 2533 5129 9058 2.53  Physician assistant (nonphysician)/healthcare assistant 877 2134 1344 4355 1.21  Dietician 983 1940 1577 4500 1.25  Patient care asst/hospital aide/care partner 882 1847 1884 4613 1.28  Unit assistant/clerk/secretary 704 1613 4197 6514 1.81  Respiratory therapist 735 1491 1708 3934 1.09  Others* 10,954 15,418 7456 33,828 9.44 Work area  Other/s 24,995 24,089 13,857 62,941 17.56  Surgery 6682 8264 7252 22,198 6.19  Intensive care unit (any type) 6419 11,487 7606 25,512 7.12  Emergency department 5378 11,910 12,368 29,656 8.27  Medicine (nonsurgical) 5083 7484 7398 19,965 5.57  Many different hospitals units/no specific unit 4599 6337 9908 20,844 5.81  Pediatrics 4359 6595 9494 20,448 5.70  Obstetrics 4327 7444 7872 19,643 5.48  Pharmacy 3846 6432 4243 14,521 4.05  Laboratory 3627 7510 6568 17,705 4.94  Radiology 3076 5946 5594 14,616 4.07  Rehabilitation 2206 1765 3330 7301 2.03  Psychiatry/mental health 1692 2552 1503 5747 1.60  Anesthesiology 1443 2172 1816 5431 1.51  OPD NA 8140 8975 17,115 4.77  Different work areas† NA 27,530 37,873 65,403 18.25 Tenure in the work area  <1 y 13,478 18,721 23,027 55,226 15.41  1–5 y 55,529 63,615 66,438 185,582 51.79  6–10 y 34,140 30,818 34,194 99,152 27.67  ≥11 y 31,777 21,770 21,998 75,545 21.08 Tenure in the hospital  <1 y 5394 13,478 17,861 36,733 10.25  1–5 y 25,237 55,529 58,232 138,998 38.79  6–10 y 23,884 34,140 37,749 95,773 26.73  ≥11 y 23,217 31,777 31,815 86,809 24.23 Hours worked per week  <20 1448 2692 NA 4140 1.15  20–39 6594 11,455 NA 18,049 5.04  <30 NA NA 4485 4485 1.25  30–40 NA NA 40,588 40,588 11.33  >40 69,690 120,777 100,584 291,051 81.23

*Related to work areas not included within the survey stand.

†Different work areas as the V1.0 was updated.

EKG, electrocardiogram; NA, not available; OPD, outpatient department.

With regard to the work area, 8.27% of the respondents were from the emergency department, and 7.12 were from the intensive care units. Moreover, the highest response rate according to the experience was participants who had 1 to 5 years at 51.79%, and the majority worked 40 hours or more per week at 81.23%.

Table 3 shows the mean positive percent for all domains among the 3 cycles. However, there was no significant difference in the domain that focused on Management support for patient safety among the 3 cycles with respect to the different versions of the survey, indicating no improvement in the management domain across 3 years.

TABLE 3 - Composite Measure Results Average Percent Positive Response Domains 2019 2021 2022* P D1: Teamwork 69.71† 70.28‡ 79.86†‡ <0.000 D2: Staffing, staffing and work pace 31.73† 31.9‡ 46.86†‡ <0.000 D3: Organizational learning—continuous improvement 79.73† 80.54‡ 76.26†‡ <0.000 D4: Response to error 24.98† 26.19‡ 55.61†‡ <0.000 D5: Supervisor/manager expectations and actions promoting patient safety 63.69† 64.32‡ 70.84†‡ <0.000 D6: Communication about error 65.19† 65.27‡ 72.89†‡ <0.000 D7: Communication openness 53.04† 53.31‡ 64.61†‡ <0.000 D8: Reporting patient safety events 56.54 59.58 66.62 <0.000 D9: Management support for patient safety 63.61 64.32 62.92 0.35 D10: Handoffs and transitions 53.54† 55.56‡ 69.28†‡ <0.000 D11: Teamwork across units§ 59.38 60.16 NA 0.65 D12: Teamwork within units§ 80.25 80.74 NA 0.78 Overall perceptions of patient safety§ 57.35 59.13 NA 0.07 Average percent for all domains 57.42† 58.42‡ 66.58†‡ <0.000 Overall events reported  1–2 24.63 25.15 25.08 0.79  3–5 13.17 13.16 14.50 0.69  6–10 6.54 6.65 5.91 0.85  ≥11 5.57 6.06 6.22 0.09  None 50.05 48.98 48.29 0.95

P < 0.05 was considered to be significant.

*The HSOPS 2.0 was used.

† and ‡ indicate the difference based on post hoc test.

§The HSOPS 1.0 was used.

Details of HSOPS 1.0 and HSOPS 2.0 composite measures, respectively, are as follows: D1: Teamwork within units, teamwork; D2: Staffing, staffing and work pace; D3: Organizational learning—continuous improvement; D4: Nonpunitive response to errors, response to error; D5: Supervisor, manager, or clinical leader support for patient safety; D6: Feedback and communication about errors, communication about error; D7: Communication openness; D8: Frequency of events reported, reporting patient safety events; D9: Management Support for Patient Safety, Hospital Management Support for Patient Safety; D10: Handoffs and transitions, handoffs and information exchange.

NA, not available.

Although there was a slight increase in the average positive response, no significant difference between C2 and C3 in teamwork across units, teamwork within units, and the overall perceptions of patient safety domains were found.

Notably, no significant difference was observed in the positive response for reporting events among the 3 cycles (Fig. 1). Concerning the average patient safety rating, the overall patient safety grade was rated as excellent or very good by 60% to 70% of respondents in all cycles, whereas cycle 4 was significantly lower in the positive response than the other cycles, while the fair and poor were higher in cycle 4 (Fig. 2).

FIGURE 1:

Average percentage of reporting patient safety events for all cycles. C2, 2019; C3, 2021; C4, 2022.

FIGURE 2:

Average patient safety rating for all cycles. C2, 2019; C3, 2021; C4, 2022.

Table 4 demonstrates the average positive response by the bed capacity for hospitals with 50 to 100, 101 to 200, 201 to 300, 301 to 500, and more than 500 beds. The significant difference was mainly between C4 and the other cycles across most of the domains. However, the Organizational learning—continuous improvement domain was significantly lower in C4 across all hospital types, which means no improvement was observed in the last cycle. In hospitals with 301 to 500 and more than 501 beds, C3 was significantly higher than C4. In hospitals with 301 to 500 beds, there was a significant improvement between the 3 cycles in the Communication openness domain with 42%, 49.2%, and 60.9%, respectively. Moreover, in the same bed capacity hospital, a difference was observed between C2 and C4 with a 12% improvement in the Reporting patient safety events. Meanwhile, in the hospital with more than 500 beds, there was a significant improvement in the Reporting of patient safety events between C2 and C3 at 5.8% and between C2 and C4 at 10.1%. Notably, the Management support for the patient safety domain showed no significant difference across the 3 cycles among all types of hospitals. With regard to the overall positive percentage, hospitals with 101 to 200 beds had no difference across all cycles. Meanwhile, the hospital with 301 to 500 beds showed a significant increase in the average positive response for all domains across all cycles (Table 4). Percentiles in Table 5 were derived from the positive response for both versions’ data sets; accordingly, there was a slight increase in the median 50th percentile.

TABLE 4 - Domains by Hospital Characteristics (Bed Capacity) Bed Capacity 50–100 101–200 201–300 301–500 >501 Variable n (Mean) n (Mean) n (Mean) n (Mean) n (Mean) Year of Cycle 2019 (n = 55) 2021 (n = 81) 2022 (n = 87) 2019 (n = 31) 2021 (n = 35) 2022 (n = 42) 2019 (n = 24) 2021 (n = 34) 2022 (n = 36) 2019 (n = 125) 2021 (n = 198) 2022 (n = 207) 2019 (n = 15) 2021 (n = 18) 2022 (n = 20) D1 73.3 73.2 82.0*† 68.9 68.3 79.5*† 66.3 65.7 77.4*† 60.6 65.6 74*† 64 65 74.8*† D2 32.8 33.5 48.2*† 32.2 31.5 46.7*† 31.9 30.6 45.6*† 30.5 30.7 42.7*† 29 30.7 43.2*† D3 82.1 81.6 77.7*† 79.2 79.7